Custody Decisions in Social and Cultural Contexts: In-Depth and Focus Group Interviews with Nineteen Judges in Taiwan CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS: IN-DEPTH AND FOCUS GROUP INTERVIEWS WITH NINETEEN JUDGES IN TAIWAN HUNG-EN Liu I. INTRODUCTION AND RESEARCH PURPOSES ---- 227 II. METHODOLOGY ----------------------------- 229 III. OVERVIEW OF THE LEGAL SYSTEM AND FAMILY PROCEEDINGS IN TAIWAN -------------------- 231 IV. FAMILY CASES: FINDINGS FROM THE INTERVIEWS 233 A. JUDGES' GENERAL ATTITUDES ABOUT AND OBSERVATION OF FAMILY CASES -------------------------------- 233 1. General Reasons Why Some Judges Disliked or Felt Uncomfortable With Family Cases ------------------- 233 2. General Reasons Why Some Judges Felt Dedicated to Family Cases --------------------------- 237 B. REASONS WHY THE NEW LAW DOES NOT WORK AS EXPECTED --------------------------------------- 237 1. Some Judges' Reluctance to Exercise Discretion and Power of Investigation --------------------- 237 2. Judges' Lack of Thorough Legal Knowledge or Training 239 3. Some Judges' Lack of Comprehensive, Non-Legal Training ------------------------------- 241 4. Some Litigants 'Lack of Legal Knowledge and Judges' Over-Focus on Divorce Itself ---------------- 242 C. CULTURAL VIEWS AND OTHER CONSIDERATIONS REGARDING DIVORCE ----------------------------- 245 Asst. Professor of Law, National Taipei University. JSD, JSM, Stanford Law School; LLM, LLB, BSc, National Taiwan University. The author would like to thank Professors Lawrence Friedman and Rogelio Perez-Perdomo for their invaluable comments. Part of this research was sponsored by Chiang Ching-Kuo Foundation for International Scholarly Exchange. A preliminary version of this article forms part of the author's JSD dissertation at Stanford Law School. COLUMBIA JOURNAL OF ASIAN LAW 1. Traditional Ideas, New Ideas, and Their Influence on the Practices ------------------------------ 245 2. Differences Between Urban and Rural Areas ------ 250 V. CUSTODY DECISIONS: JUDGES' ATTITUDES AND BELIEFS, THOUGHTS ON THE NEW STANDARD, AND CUSTODY DETERMINATION FACTORS ---------- 251 A. WHAT JUDGES THINK AND FEEL ABOUT DETERMINING CUSTODY --------------------------------------- 251 B. CUSTODY AS PARENTS' RIGHTS FROM SOME JUDGES' PERSPECTIVE ------------------------------------ 255 C. JUDGES' GENERAL FEELINGS ABOUT THE "BEST INTERESTS OF THE CHILD" STANDARD ------------------------- 256 D. FACTORS USED IN DETERMINING CUSTODY AND THEIR IMPORTANCE ------------------------------------- 257 1. Age of the Child and the "Tender Years Doctrine "- 257 2. Sex of the Child -------------------------- 258 3. Sex of the Parent-Some Judges' Preference for Mothers ------------------------------- 259 4. Wishes of the Child- ----------------------- 260 5. Needs of the Child's Personality Development ---- 263 6. Moral Character and Performance of the Parents-- 264 7. "Occupation and Economic Resources of the Parents" and "Child Support" ---------------------- 265 8. Interview Report of Social Workers ------------ 267 9. Continuity, Stability, or Primary Caretaker -------- 276 10. Support from Relatives --------------------- 76 11. Domestic Violence ------------------------ 78 E. THE "ALL-OR-NONE CUSTODY" DECISION PATTERN ---- 279 1. Judges' Attitudes About Visitation ------------- 280 2. Judges' Attitudes About Joint Custody ---------- 281 3. The "All-or-None Custody" Tradition and the Sense of "Face " -------------------------------- 282 F. OTHER FINDINGS --------------------------------- 284 1. The Current Process ofAssigning Judges to a Specialized Family Division ----------------- 284 2. Whether Mothers Received Custody More Often Than Fathers -------------------------------- 286 VI. FURTHER DISCUSSION -------------------------------------- [17:2 288 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 227 A. WHEN THE NEW LAWS MEET THE JUDGES WITH OLD SKILLS AND OLD SCHEMAS ------------------------------- 288 B. ADVERSE EFFECTS OF CURRENT JUDICIAL PRACTICES --- 295 1. Problems of Using Economic Criterion and the "All-or-None Custody" Decision Pattern ---------- 295 2. Problems ofJudges' Over-Focus on Divorce Itself While Overlooking the Importance of Custody Issues ---- 299 VII. CONCLUSION -------------------------------- 302 I. INTRODUCTION AND RESEARCH PURPOSES In 1996, Taiwan adopted the "best interests of the child" standard to substitute for the presumption of paternal custody in deciding child custody cases.' Five years later, the researcher collected a sample of Taiwan's court decisions on divorce cases involving custody disputes. By using the method of content analysis, the sample of cases was quantitatively analyzed in order to analyze how the courts applied the "best interests of the child standard."2 The findings attest that Taiwan's court decisions on child custody cases actually reflect many cultural ideas, such as stereotyped gender roles, family autonomy, a sense of "face," all-or-none custody, and the tradition of parents' long-term financial support for their children. The varying socio-economic climate of Taiwan across districts and the lack of public welfare programs also clearly affect judges' custody decisions. Moreover, this study finds that since 1996, custody has been overwhelmingly awarded to mothers, whereas before 1996 fathers were favored by the courts. 3 Article 1055-1 of the Civil Code is the core of the 1996 Amendments with regard to the best interests of the child. According to this article, the court shall, but is not limited to, consider the following factors to decide a custody arrangement for the child's best interests: (1) age of the child; (2) sex of the child; (3) number of children of the parents; (4) health of the child; (5) wishes of the child; (6) needs of the child's personality development; (7) age of the child's parents; (8) occupation and economic resources of the parents; (9) moral character and performance of the parents; (10) health of the parents; (11) living conditions and life styles of the parents; (12) wishes and attitudes of the child's parents as to his custody; (13) parent-child relationship and affection; (14) relationship and affection between the child and other people who live with him. In addition, the article provides that the court could consider the interview report of social workers. See CIVIL CODE art. 1055-1 (Taiwan ROC). 2 Hung-En Liu, Mother or Father: Uho Received Custody? The Best Interests of the Child Standard and Judges' Custody Decisions in Taiwan, 15 INT'L J.L. POL'Y & FAM. 185 (2001). 3 A more thorough analysis was later presented in the author's doctoral dissertation: Hung-En Liu, Custody Decisions in Social and Cultural Contexts 18-72 (2002) (unpublished J.S.D. dissertation, Stanford University School of Law). COLUMBIA JOURNAL OF ASIAN LA W This article is a continuation of the previous study through interview research with judges. Following the quantitative examination of court decisions, the researcher found that it was necessary to conduct interview research with judges for several reasons. First, Taiwan's court decisions do not always elaborate the judge's reasoning in detail. Many commentators have harshly criticized the ambiguity and incompleteness of reasoning often seen in Taiwanese judges' decisions.' In fact, some judges often use "Li Gao" (model decision drafts) and just fit the names and facts of the cases into the drafts without changing the reasoning part while writing their decisions.' Given that the reasoning of the decisions analyzed in the previous study was not complete in many cases,' apparently we need to collect further data in order to reveal how judges actually applied the best interests of the child standard. Second, even if the reasoning in the court decisions seemed to be complete and detailed, it is highly doubtful whether the justifications written in the decisions actually revealed the judges' whole decision-making processes and all the factors they had considered. The judges are limited by statutes and legal language, and they may write socially and legally acceptable justifications in order to cloak some biases of their own used in deciding the cases. Interview research can help draw out some of the other factors considered. Third, interview research provides a unique opportunity, which cannot be provided by quantitative analysis, to let judges tell their experiences in their own terms and put the cases they decided into the appropriate context. Through their own descriptions and ideas-not necessarily legal language and ideas-the interviews can further reveal some hidden thoughts and considerations that were in their (sometimes subconscious) minds when making decisions. Meanwhile, context is important because how judges decide a case and why they do it in that way may be a product of some specific circumstances, which usually are not written into the decisions. Fourth, interview research can provide a great deal of information about judges' personal backgrounds and how that 4 Chiu et al., infra note 5, at 118. 5 Lien-Kung Chiu et al., A Study on the Disclosure of Discretionary Evidence and Reasoning of Court Decisions, 166 CHINESE L.J. 118 (1997); Hung-En Liu, Legal Studies through Psychological Approaches: Perspectives of Family Law Issues 147-63 (1996) (unpublished LL.M. thesis, National Taiwan University School of Law) (on file with National Taiwan University School of Law). 6 In the process of analyzing these cases, the researcher found that many court decisions did not make their reasoning full and clear. Many judges merely "mentioned" some factors but never explained how these factors should be valued and why they disregarded other factors. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 229 influences their decisions. Fifth, interview research with judges can help to verify and further explain the findings of the quantitative analysis of court decisions. II. METHODOLOGY In June and July of 2001, the researcher conducted in-depth interviews with ten judges in Taiwan. All of these interviews were conducted in-person and on a one-on-one basis, except one interview that was conducted with two judges at once. Most of these interviews lasted one and a half hours, although one lasted only fifty minutes and another lasted two hours. Participants were selected through non-random snowball sampling;7 that is, the researcher contacted and interviewed some judges, and then asked them to refer him to their colleagues or friends as potential interviewees. The same procedure was repeated until the target number often interviewees had been reached. In addition to the in-depth interviews, the researcher also conducted a focus group interview with nine other judges. One of the judges volunteered to organize this group, and the interview lasted an hour. This focus group interview was conducted after the researcher had finished two in-depth, one-on-one interviews. Though the interview sample was not intended to be statistically representative of all judges in Taiwan, the researcher tried to include interviewees with different characteristics such as sex, marital status, where they sat to hear cases, and experience in hearing family cases in order to get a general range of judges who decide family cases. Meanwhile, in addition to interviewing judges who were very willing to be interviewed, the researcher also tried to interview judges who were reluctant to be interviewed at first because of the potential significance of this reluctance.' Of the nineteen interviewees, nine were men and ten were women, and ten were married and nine were single. They were from ten different courts-nine district courts and an appellate court. Four of the 7 This research used the snowball sampling technique for several reasons. First, there had been almost no interview research with judges in Taiwan; because judges were totally unfamiliar with this kind of research and also because of the conservative atmosphere in the Taiwanese judiciary, they could be extremely suspicious of a stranger who tried to interview them. Second, random sampling was not feasible because the researcher could not obtain a list of all Taiwanese judges' names and the courts where they heard cases in the first place. Finally, random sampling and interviews following it would have been cost and time prohibitive for this research. 8 For example, it might indicate that they lacked confidence in their decisions and/or that they were more conservative. COLUMBIA JOURNAL OF ASIAN LA W judges heard cases in urban areas, six heard cases in rural areas, and nine heard cases in both urban and rural areas or in suburban areas. All judges had decided custody cases, but their experience ranged from six months to ten years; the average length of experience was two years.9 Six judges were in a specialized family division handling only family cases, but they had also heard property cases before being assigned to the family division; the other thirteen judges were in ordinary civil courts that decided both family cases and property cases. In all of the interviews, participants were asked a series of semi-structured open-ended questions. Though the researcher had prepared a list of questions to ensure that all areas of interest were covered,'" the questions were not asked in a strict order. The researcher asked questions according to the flow of each interview and was not confined to the prepared questions. If the participant wanted to discuss other facts or personal experience, the researcher would give him or her the opportunity to do so in order to reveal something which the researcher otherwise might not have known. The researcher might also follow up on interesting responses when appropriate and ask for examples when the answers were too abstract or vague. During the interview process, the researcher encouraged the judges to reveal their own experiences and stories, points of view, and explanations of how and why they made custody decisions in a, if any, specific way. The researcher also encouraged them to illustrate the situations they themselves identified as problematic in the judicial process and to describe how they dealt with these situations and why they chose these ways. If time permitted, they were also asked to describe and comment on the behavior and attitudes of their colleagues and other people, such as litigants and lawyers, who were involved in the judicial process. The researcher did many things to try to assure that the judges would feel comfortable to talk about what they really thought. Although one purpose of the interviews was to verify the findings of the previous analysis of court decisions, the researcher avoided mentioning any of these findings directly during the interviews. The researcher also avoided making any comment on any of the interviewees' responses. Before the start of all interviews, the researcher sent the judges a signed 9 The interviewees' length of experience was not long because of at least two reasons. First, according to the judicial administration's assignment method, almost all judges hearing family cases would be reassigned to other fields within two or three years. Second, it seemed that younger judges usually were more likely to accept the researcher's requests for interviews. '0 A copy of the interview protocol can be found in the appendix of Liu, supra note 3. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 231 letter which stated that their remarks would be used only for research purposes and that they could choose to keep their remarks confidential and anonymous. All interviews were conducted at locations the judges chose themselves. With permission from the participants, all interviews were recorded on audiotapes. The tapes were then transcribed, analyzed to yield key analytical themes, and translated into English from Chinese. All quotes included in this article are verbatim, although some quotes have been edited for clarity. Because some judges were unwilling to reveal their participation, though others had no objection to do so, the researcher decided not to reveal the names of any of the judges, and thus, all interviews are treated anonymously. III. OVERVIEW OF THE LEGAL SYSTEM AND FAMILY PROCEEDINGS IN TAIWAN This overview introduces Taiwan's legal system and family proceedings to provide background information for discussing how Taiwan's judges apply the "best interests of the child" standard in divorce cases involving custody disputes. As Taiwan is a civil law country, comprehensive legal codes usually are the primary source of law, and judges seldom create legal norms themselves. Most of Taiwan's major laws, such as the Civil Code and the Criminal Code, were basically "transplanted" from Japan, Germany, and Switzerland in the early twentieth century after the Ching Dynasty was overthrown. " Nevertheless, some recent pieces of legislation and amendments, such as the "best interests of the child" standard and the Domestic Violence Prevention Law, clearly make reference to laws in the United States.'2 Taiwan's first law degree is at the undergraduate level. Partly because of the civil law system's emphasis on codified laws, legal education and academia focus mainly on logic, abstract concepts, and law codes. Taiwan's judges are selected by a national examination that anyone finishing the degree can take, so most judges are only between twenty-five and thirty years old when they start to hear cases; in fact, 11 Herbert H. P. Ma, The Sources and Structure of Modern Chinese Law and the Chinese Judicial System, in TRADE AND INVESTMENT IN TAIWAN: THE LEGAL AND ECONOMIC ENVIRONMENT IN THE REPUBLIC OF CHINA 27 (Herbert H. P. Ma ed., 2d ed. 1985); Tai-Sen Wang, Studies on Taiwanese Legal History, NAT'L TAIWAN U. L.J., Dec. 1993, at 1, 2-14. 12 FEHNG-SHIAN GAU, A MONOGRAPH ON THE DOMESTIC VIOLENCE PREVENTION LAW 85-87 (1998); Liu, supra note 5, at 181. COLUMBIA JOURNAL OF ASIAN LAW being a judge may be their first job after college. In addition to law, almost none of Taiwan's judges have ever received any training in other disciplines. 3 In Taiwan, there are two types of divorce: consensual divorce and judicial divorce. The Civil Code recognizes consensual divorce which only has to be registered at administrative agencies by the parties themselves."' If a couple cannot reach a mutual agreement to divorce, they may file for judicial divorce based on statutory grounds, but only the spouse who is not responsible for the existence of the grounds may file the petition. The judicial divorce is primarily fault-based, with the exception of a few no-fault grounds; the statutory grounds for divorce include bigamy, adultery, ill-treatment, an attempt at murdering the plaintiff, desertion, disappearance, incurable loathsome diseases, incurable insanity, and imprisonment. The major no-fault exception, added in 1985, is a "significant matter that makes maintenance of marriage difficult," but only the party not responsible for this matter may file for divorce."I The new law concerning child custody issues adopts the "best interests of the child" standard and lists the factors that a court should consider while determining custody. The new law also, for the first time in Taiwan's legal history, introduces the terms and arrangements of visitation and joint custody.'6 In most jurisdictions in Taiwan, family cases are decided by an ordinary Civil Court that also handles property cases. By the end of 2001, only Taipei District Court and Taichung District Court had a Family Division specializing in family cases. While in principle all civil cases follow an adversary procedure, Part IX of the Code of Civil Procedure clearly distinguishes between the procedure for handling family cases and the procedure for handling property cases. In short, because family cases are usually concerned with public interests such as child welfare, the new law enacted in 1999 allows more discretion to judges and authorizes them to investigate on their own motions in order to make proper decisions. For example, article 575-1 stipulates that "the court may take into consideration the facts not alleged by the parties, and the court shall make necessary investigation into the facts and 13 See generally Howard S.H. Shyr, Legal Education with Duo Faqade of Profession and Humanities, NAT'L TAIWAN U. L.J., Oct. 1999, at 179 (arguing the necessity of reforming legal education in Taiwan). 14 CIVIL CODE art. 1050 (Taiwan ROC). 15 CIVIL CODE art. 1052 (Taiwan ROC). 16 CIVIL CODE art. 1055, 1055-1 (Taiwan ROC); see also supra note 1. [17:2 2004] CUSTODYDECISIONS IN SOCIAL AND CULTURAL CONTEXTS 233 evidence."' 7 By contrast, in property cases, "a party shall bear the burden of proof with respect to the fact he alleges in his favor."'8 For another example, article 572-1 clearly provides that "the court, on its own motion, may determine custody and visitation according to the best interests of the child and is not confined to the parties' allegations."'9 In contrast, in property cases, "the court may not give a judgment on any matter not mentioned by the parties."20 IV. FAMILY CASES: FINDINGS FROM THE INTERVIEWS A. Judges' General Attitudes About and Observation of Family Cases Although the 1999 amendments to the Code of Civil Procedure allow more discretion to judges and authorize them to make investigations on their own motions in order to pursue the best interests of the child, it appears that some judges do not like to exercise their discretion and power of investigation. In the interviews, before asking specifically about custody and the "best interests of the child" standard, the researcher asked the judges about their general feelings about and observation of family cases. The results give us some clues about why judges would or would not like to exercise that discretion and power. In addition, their observations and experience may also provide us with some social and cultural contexts that could influence their attitudes and affect their decisions. 1. General Reasons Why Some Judges Disliked or Felt Uncomfortable With Family Cases Several judges explicitly expressed how and why they disliked family cases. They described that litigants always raised many issues that were not legally important from the court's perspective, and that it was a waste of the court's time and energy. As judge A, a male judge sitting in an ordinary civil court, put it with a forced smile: 17 Code of Civil Procedure art. 575-1 (Taiwan ROC). 18 Code of Civil Procedure art. 277 (Taiwan ROC). 19 However, "the parties should have opportunities to express their opinions before the court makes the decisions." Code of Civil Procedure art. 572-1 (Taiwan ROC); CIVIL CODE art. 1055 (Taiwan ROC). 20 Code of Civil Procedure art. 388 (Taiwan ROC). COLUMBIA JOURNAL OF ASIAN LAW I have been a judge for almost three years, but I really dislike hearing family cases more than hearing other cases. I think it is very difficult to determine what happens and who is right or wrong in a family. Besides, really, hearing family cases is always hearing noisy quarrels, from the beginning till the end. After seeing so much stuff like this, it affects our own mood." Some judges especially pointed out that they thought property cases were better for them because these cases were "clearer" and therefore they usually did not need to exercise discretion or investigate themselves. Judge D4, a male judge sitting in an ordinary civil court, explained that: I think ordinary civil cases are clearer because the most important disputes and resolutions are based on the law. Although family cases are also based on rights of claim provided in the law, what the litigants dispute usually is not the rights of claim but the facts, facts that are extremely hard to prove. Family affairs are so trivial, and it is so hard to meet the burden of proof, and then the parties just tend to become emotional when in the courtroom. It is so difficult to keep order in the court. So I dislike family cases unless they are the kind of cases that are very clear. For example, [according to the law] one can file for divorce if his or her spouse has been sentenced to more than three years' imprisonment or has been sentenced to imprisonment for an infamous crime-this condition is very clear, and the case is so easy to decide, not like another condition "intolerable ill-treatment from the other spouse. ' "22 As noted in the previous study, Taiwanese judiciary and legal academia have a long rooted tradition of legal positivism, and judges believe that they should stick strictly to statutes and prefer not to create or supplement law themselves. 23 This tradition clearly appeared in some 21 Interview with Judge A in Taiwan (June 29, 2001). 22 Interview with Judge D4 in Taiwan (July 12, 2001). 23 HUI-HSIN CHEN, STUDIES OF ISSUES IN FAMILY LAW 285-88 (1993); Liu, supra note 5, at 165-67, 181-82. This phenomenon might be due to the following reasons: the underdeveloped judiciary within the semi-authoritarian government until 1980s and the insufficient resources distributed to the [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 235 judges' reasons for not liking family cases. Another male judge sitting in an ordinary civil court illustrated that: Divorce cases are troublesome. Many litigants come to tell you lots of their facts, but when you ask them "On which ground provided in article 1052 of the Civil Code would you like to file for divorce," they cannot give you an answer.... You will have to ask him, "Are you filing for divorce based on this (or that) section of the article 1052?" And you may even need to turn to the specific page containing that article in a book of law codes for him and ask him to check. . . . I had a case. . . . The litigants took all different kinds of trivial stuff out for quarreling, and I felt so annoyed by them when hearing the case! I just felt very annoyed, because I couldn't find the legal grounds and couldn't fit the facts into the conditions set by the law.24 Judge J mentioned an interesting story related to this pattern of thought: There was a litigant in my courtroom, and in his petition document he said, "Judge X's fate is just too good, and he will never understand our lowly commoners' lives. So he always decides cases according only to law codes, and he is not capable of solving my problems at all."25 In fact, during the interview process, it was obvious many judges felt uncomfortable without having a book of law codes in front of them when they discussed family law and cases with the researcher. They would stop and request a book of law codes even though the researcher judiciary and the excessively heavy caseload of judges. Meanwhile, because the whole modem legal system in Taiwan was basically transplanted from the West, while the law codes and legal technique were very unfamiliar, legal scholars and judges had no choice but to focus on definitions of words in the law codes and stick to them. The tradition of legal positivism has been therefore strengthened. See Jane Kaufman Winn, Relational Practices and the Marginalization of Law: Informal Financial Practices of Small Businesses in Taiwan, 28 LAW & SOC'Y. REv. 193, 201-05 (1994); Yung-Chin Su, Taiwan's Social Change and the Development of Legal Scholarship, in SYMPOSIUM OF CONTEMPORARY PRESTIGIOUS LEGAL SCHOLARS 551, 560-64 (Chinese Law Journal ed., 1996). 24 Interview with Judge F in Taiwan (July 15, 2001).25 Interview with Judge J in Taiwan (July 20, 2001). Judge X here was another judge hearing this litigant's previous trial; he was not an interviewee in this research. COLUMBIA JOURNAL OF ASIAN LA W was just asking a simple question such as, "According to your experience, what are the differences between family cases and property cases?" "What do you think about the 'best interests of the child' standard?" or "Do you send every one of your custody cases to a social worker for evaluation?" These questions were about their personal experiences or opinions, but some judges still asked to check law codes before answering. Some judges felt uncomfortable with family cases, not because they were annoyed by litigants' emotional, trivial quarrels, or because they could not make decisions based only on the statutes, but because they did not believe they were competent in this field. In fact, because Taiwan's judges are usually selected by a national examination following completion of a college legal degree, most of these new judges are single and do not know much about marriage, divorce, and childrearing. As a young female judge sitting in a jurisdiction of both rural and urban areas put it: Sometimes when handling divorce cases, I felt pretty embarrassed. My feeling was that: as a young person without any marriage and childrearing experience, how could I be capable of telling them what they should do with their marriage or children? Sometimes I felt confused and did not know what my standpoint should be, and I doubted whether I had the experience and capability to tell them these kinds of things and decide their cases. All I can say was that I had learned these law codes, and legal education gave me some legal ideas and trained me to think and judge in this way. So sometimes I had a guilty conscience [because of my incompetence] .26 All of these judges who disliked or felt uncomfortable with family cases sat in ordinary civil courts and handled both property cases and family cases. Since the sample for interviews was not representative, we cannot be certain whether this was just a coincidence. 26 Interview with Judge D6 in Taiwan (July 12, 2001). [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 237 2. General Reasons Why Some Judges Felt Dedicated to Family Cases Of all the interviewees, judges who sat in a specialized family division tended to have different attitudes toward family cases. Some of these judges were very dedicated to family cases. They thought their job was worthwhile and profoundly meaningful to both the society and themselves. Judge B 1 explained that: Handling family cases makes you feel more related to lives, you know. You feel you are really touching people and participating in a person's life, growth, or development, unlike ordinary civil or criminal cases-I especially hate ordinary civil cases! For money, some people count every single cent, . . . you would feel so bored by these cases .... But family and juvenile cases are different. You may feel ... you are standing at the turning point of his life, and you can help him grow or progress, but you can also give him up and let him sink. You get involved in his life, and I feel that is a huge responsibility. You can choose to be a careless judge, but you can also choose to be a serious judge. Sometimes I think in a Buddhist way: if you plant the evil seed [because you don't do your job], one day you will get the evil fruit yourself.27 B. Reasons Why the New Law Does Not Work as Expected Despite the attempt of the legislature to promote the child's best interest in child custody cases, many reasons still hinder the legislature's good intentions. 1. Some Judges' Reluctance to Exercise Discretion and Power of Investigation Some judges-all of which sat in ordinary civil courts-directly identified why they hesitated to exercise the discretion and power 27 Interview with Judge BI in Taiwan (July 11, 2001). COLUMBIA JOURNAL OF ASIAN LA W authorized by the law. Apparently, their heavy caseload is one important reason. As Judge A put it in the interview: Researcher: Suppose that in a case neither of the litigants mentions the custody issue, but from their documents or statements you find that there is a child. Would you then investigate and determine the custody issue on your own motion? Would you ask them about it? Judge A: I don't think I would in every case. I think... To be honest, sometimes this is just making trouble for myself! . . . [E]ven though the law says that we may decide according to the best interests of the child on our own motions, usually we would not do so. . . .One consideration is that the caseload is heavy enough! Usually, we don't want to add more burdens on ourselves." Judge G made a very candid comment, saying, "Which judge would want to be such a meddler?... If you do not petition for custody, I can simply decide the divorce. Why should I bother to ask or care about the custody?"29 Judge H concurred with the idea that the heavy workload is a reason, but she also thought that judges should not intervene in family affairs so much. Meanwhile, the difficulty of deciding custody might also make judges hesitant. She said, What is my opinion about the old saying "Qing Guan Nan Duan Jia Wu Shi" [even honest and smart government officials will find it very difficult to judge and decide on family matters]? Of course it is true. Certainly I think it is very difficult to judge a couple's right and wrong doings .... From our judges' standpoint, we may think. . . one less case is always better than one more case. If they petition for only divorce, and they don't mention custody, why should we interfere so deeply? Besides, for us, to decide custody is a much 28 Interview with Judge A, supra note 21. 29 Interview with Judge G in Taiwan (July 16, 2001). [17:2 2004] CUSTODY DECISIONS 1N SOCIAL AND CULTURAL CONTEXTS 239 more stressful thing .... They can discuss the custody issue themselves after the divorce.0 All of the judges mentioned above recognized that the new law authorizes them to determine custody according to the best interests of the child even if neither of the parents petitions for custody. However, strikingly, as will be seen in the next section, some judges even did not know the existence of new statutes. 2. Judges' Lack of Thorough Legal Knowledge or Training In one interview, a very experienced judge who had heard family cases for more than ten years insisted that judges were not allowed to determine custody if no parent petitioned for it. When the researcher reminded her that the new law revised the adversary procedure in family proceedings and authorizes judges to determine custody on their own motions, this judge claimed that she knew the new law and was certain there was no such provision in it, so the researcher must be wrong. She claimed that custody was a matter that should rely on the parents' own agreement, a matter that the court should not interfere in. Finally, the researcher showed her the specific provisions of the new law in a book of law codes. The judge read the provisions repeatedly and appeared very surprised to know of their existence.3 Apparently, even though the new law had been enacted for more than five years when the interview was conducted, this judge still used the logic of the old law to handle custody cases. It appeared that the habit of applying the old law and the idea of family autonomy were ingrained in her mind, so she could not recognize or believe the new ideas and new procedures in the new law. If someone assumes that every judge who is supposed to or is even required to know the new law is actually familiar with it, he or she would be very disappointed upon learning the truth. 32 A similar scenario happened when some judges were asked their experience before 1996. Of the nineteen interviewees, three had heard custody cases both before and after 1996. When asked about the 30 Interview with Judge H in Taiwan (July 20, 2001). 31 Interview with Judge G, supra note 29. 32 In fact, in several different interviews, many judges appeared to be unfamiliar with the new law. See infra text accompanying note 80 and note 113. COLUMBIA JOURNAL OF ASIAN LAW differences in judicial practices after 1996, all three judges gave similar answers. According to what they described, before 1996, it was an absolute adversary system-"no petition, no decision." Judges could not make any decision beyond what the litigants petitioned for. Because of the paternal presumption, custody usually was not an issue in a divorce case. If the mother did not petition for it, the judge did not even need to mention a custody arrangement in the decision because the father would receive the custody automatically. "Before 1996, we almost always handled only the part of whether to grant divorce. Family cases then were pretty simple," said Judge B2.11 However, the 1993 Child Welfare Amendment Act had partially revised the presumption of paternal custody and had authorized judges to award custody on their own motions according to the child's interests. Past research found that very few judges had applied this Act to determine custody; before 1996, almost all judges still applied only the paternal presumption in the Civil Code to decide divorce cases.' In addition to judges' conservativeness, one possible explanation was that judges were not familiar with or even did not know the Child Welfare Act because it was not included in the curriculum of legal education or training.5 In the interviews, the researcher asked these three judges specifically about whether they were certain that before 1996 it was an absolute adversary system and whether there was any law authorizing judges to determine custody on their own motions. All of them promptly gave a similar answer: "No, at that time we judges could not determine custody on our own motions. There was absolutely no such law." 6 They apparently knew nothing about the existence of the 1993 Child Welfare Amendment Act, or they had simply forgotten about it. 33 Interview with Judge B2 in Taiwan (July 11, 2001). 34 Liu, supra note 5, at 57-182.35 Id.; Yueh-Mi Lai, The Application of the Norm of Parental Rights in the Child Welfare Act 117 (1998) (unpublished M.A. thesis, Graduate Institute of Child Welfare, Chinese Culture University, Taiwan). In fact, Judge J asserted, "'In the court, Child Welfare Act is a law that many judges hate to apply. They despise this law because it was drafted only by some outsiders-advocates of child welfare. Besides, the quality of this legislation is pretty poor." Interview with Judge J, supra note 25. 36 Interview with Judge B2, supra note 3359; Interview with Judge G, supra note 29; Interview with Judge C in Taiwan (July 11, 2001). [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 241 3. Some Judges' Lack of Comprehensive, Non-Legal Training Many judges pointed out that with family cases, compared to property cases, the legal issues are relatively simple, but determining the facts is much more difficult. Even though the main problem in most family cases is how to determine the facts, because almost all of Taiwan's judges have no training in other disciplines such as psychology, social work, and sociology, and given that Taiwan's legal education mainly focuses on logic, abstract concepts, and law codes, it is not surprising that many judges are only capable of relying on law codes to decide family cases. Judge F made an interesting remark: Before I started hearing cases, in school and even in the Training Institute for Judicial Officers,37 I had never been taught about how to actually investigate facts and question litigants or witnesses. When I first sat on the bench, I did not even know what the first question I should ask was; neither did I know what the second question should be. I needed to grope for everything myself. In the beginning of my being a judge, all litigants were guinea pigs for me to let me practice. 8 Another dedicated judge, one of very few judges knowledgeable in the field of family cases, further explained the situation: The most significant difference between family cases and property cases is its continuous nature. Property cases are done after you pronounce judgment, but family cases are not. Family cases involve [litigants' and children's] psychology, mental status, and many societal considerations, and there are many issues that cannot be solved with legal relations .... Compared with the ones in ordinary civil cases, the legal relations in family cases are much simpler .... I think the intriguing part is the reality and phenomena-the problems the litigants 37 After passing the national examination, all prospective judges (and prosecutors) need to enter the Training Institute for Judicial Officers for a one and a half year training period. They will be officially appointed as judges after finishing the training. Interview with Judge F, supra note 24. COLUMBIA JOURNAL OF ASIAN LAW actually face. In ordinary civil cases, we let the litigants argue with each other and then apply the law; whether our decisions can actually solve their problem is not the point for us. But I think when you handle family cases, the point is to help the litigants solve problems, and you cannot simply apply the statutes to solve their problems. •.. Till now, very few judges have these ideas as I do... . In my session, I listen carefully to what the litigants say and let them tell more details. Because you don't know their actual lives and problems, of course you should give them more opportunities to talk. But many judges think it's wasting time. They want to hear legal relations and legal issues, not these .... Many people just want to make a decision; they think "Why should I care about your trivial matters?". . . Generally speaking, they don't have much knowledge of social reality, psychology, and social work, so how can you blame them? These considerations simply never enter their minds because they have no idea about their existence. They don't even know that they may need to seek an expert's assistance, because they have no idea! ... Knowledge of statutes is all what they got; of course they will rely only on statutes and think [handling cases] faster is better. 9 4. Some Litigants' Lack of Legal Knowledge and Judges' Over-Focus on Divorce Itself Several judges remarked that many litigants in family cases lacked legal knowledge relevant to their cases. Judge A, who sat in a jurisdiction including both suburban and rural areas, reported that "[in our jurisdiction,] from our observation, most litigants in family cases come from the lower middle class or lower class; usually, they can't afford an attorney, and they don't know how to submit claims or proceed. 4 ° The reason more litigants come from the lower class in family cases rather than in property cases may be a person usually needs to be able to afford property before he or she can have a property case, but marriage or divorce disputes may happen to people from any social class. 39 Interview with Judge J, supra note 25. 40 Interview with Judge A, supra note 21. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 243 According to these judges' descriptions, many litigants did not even know they could (and maybe should) settle custody issues in court along with the divorce. Some of them thought they might need to file another lawsuit for the custody after the divorce, and other litigants thought, "It's not necessary because the child has already been living with me for a long time, so I don't need a custody decision from the court."' In fact, as noted earlier, some judges also did not want to "remind" the litigants of the custody issue because of heavy workloads, hesitance to intervene in family affairs, and worries about the difficulties in deciding custody. For judges hearing both property cases and family cases, there might be another reason: they were accustomed to following an adversary procedure in handling civil cases; even though the new law adds some non-adversary procedures to family proceedings and authorizes more discretion and power to judges, they still had a "habit" of following the adversary procedure. Judge F, sitting in an ordinary civil court in a rural area, recalled a true story: Once one of my colleagues did not decide custody in a divorce case, and he just granted the divorce. He never asked the couple whether they had a child. But the court later found out from documents that there was a child. The Chief Judge told him to correct the decision because custody was a matter that a judge should exercise his discretion on in order to decide in the child's best interests. But my colleague thought, "Civil cases should follow an adversary procedure; when the plaintiff did not petition for custody, why should I handle it?" 2 In this story the judge apparently treated family cases in the same way he treated property cases, even though the "law in books" says the opposite. In fact, many interviewees expressed the same attitude, basically meaning, "If you don't petition, I won't decide" about issues in family cases. On one hand, judges were more familiar with the adversary system. On the other hand, since the legal relations in family cases were less complicated and the litigants might lack the knowledge to submit claims, by following this procedure they could handle these cases more effortlessly. It is no wonder that in the interviews a few 41 Interview with Judge C, supra note 36. 42 Interview with Judge F, supra note 24. COLUMBIA JOURNAL OF ASIAN LAW judges described family cases as "simple," "effortless," or even "cases that don't need much brain." 3 While some litigants did not know to raise the custody issue in a divorce suit, according to the interviewees' observation, far more litigants did not know that they could petition for child support, alimony, or compensation. Several judges stated the same thing: when the litigants-especially the litigants in rural areas--did not retain an attorney, almost none of them knew to raise these issues in a divorce case. Again, many judges assumed the attitude: "If you don't petition, I won't decide." We will discuss the child support issue later. A few judges admitted frankly, "In a divorce case, I only focus on divorce itself and don't pay much attention to the other parts, such as custody."44 When asked why, Judge F explained, The law establishes clear legal grounds and conditions for divorce. Since in our decisions we need to point out a specific ground for divorce, of course we need to carefully examine the facts and find the ground that meets them .... After you decide the divorce part and can write a sound decision about it, you can write the custody part desultorily because the law does not establish a clear standard [for custody decision]. The [best interests of the child] standard is very vague, and we have social workers' reports that we can copy into our decisions. To be honest, whether a judge would carefully decide the custody is up to his conscience. To sum up, the interview results show that many judges, especially judges who sit in ordinary civil courts, were over-focused on divorce itself but usually ignored the very issues directly deriving from divorce. They ignored issues that would be very important to post-divorced families for several reasons: the tradition of legal positivism, the fact that judges were not equipped to determine the complicated facts in family cases, judges' heavy workloads, judges' hesitance to intervene in family affairs, judges' habit of following an adversary procedure, and litigants' lack of legal knowledge. 43 For example, see Interview with Judge G, supra note 29; Interview with Judge E in Taiwan (July 13, 2001). 44 Interview with Judge F, supra note 24; Interview with Judge K in Taiwan (July 20, 2001). 45 Interview with Judge F, supra note 24. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 245 C. Cultural Views and Other Considerations Regarding Divorce 1. Traditional Ideas, New Ideas, and Their Influence on the Practices In traditional Chinese society, divorce was shameful for the individuals, the whole family, and even their relatives, because it meant that all of them failed to fulfill their culturally imposed obligations-the individuals failed to keep their marriage intact in order to raise children and support their elders, while all their family and relatives failed to persuade them to remain in the marriage. 6 Even people who were not their family or relatives were expected to help the couple remain in the marriage, an expectation that can be clearly captured in the old saying "Quan He Bu Quan Li" (always encourage others to reconcile, not to divorce). Divorce was a taboo in traditional Chinese society, and the shame and stigma were even more serious for divorced women. 7 Interestingly, without being asked, several judges mentioned some phenomena related to these traditional ideas. Judge C described, Many litigants come to the court without a lawyer. They come here themselves, but in fact ... You know, many lawyers are unwilling to enter a courtroom for family cases, because they think handling these [divorce] cases is wicked and may bring them bad luck. So they prefer to only write petition documents for the litigants. The documents are written by lawyers, but the lawyers wouldn't come to the court. 8 Not only did these ideas affect some lawyers, but they were in some judges' mind too. Judge A made a vivid remark: Older generations may take divorce as a huge thing. Some elders may say, "You decide to let them divorce? 46 Chiu-Tao Chun, A Study of the Practices of Divorce, 9 ANNUAL REVIEW OF JUDICIAL RESEARCH 261, 262 (1989); PING-CHEN HSiUNG, HISTORY OF CHINESE CHILDREN 260 (2000); Pao-Ching Lu, A Study of Divorced Women's Social Adaptation in Taipei City 2 (1979) (unpublished M.A. thesis, Graduate Institute of Sociology, National Taiwan University). 47 Chun, supra note 6, at 2. 48 Interview with Judge C, supra note 36. COLUMBIA JOURNAL OFASIANLAW Oh, Yao-Shou [you shorten your own life]!""' They think it will damage your integrity . . . In district courts, usually we judges are relatively young. After our decisions are appealed to an appellate court, maybe [older] judges there may have a firmer attitude of"Quan He Bu Quan Li" (always encourage others to reconcile, not to divorce), and then our viewpoints would be very different. . . . Though article 1052 of the Civil Code provides several legal grounds for divorce, it allows a lot of discretion to judges. What is "intolerable ill-treatment from the other spouse?" What is a "significant matter that makes maintenance of marriage difficult?" There is a lot of room for judges' personal values and beliefs. Personal attitudes toward marriage will affect judges' decisions. Some will grant divorce, but some others will not. ° Many relatively young judges explicitly expressed that they did not believe in fault-based divorce and traditional ideas. Even though the current divorce law is mainly fault-based, with the exception of a few no-fault grounds, these younger judges clearly preferred the no-fault divorce system. In fact, they asserted that they might adopt the broadest explanation of the no-fault exception and try to apply it as much as possible. Apparently, they embraced the modem concepts of marriage and divorce, concepts that Taiwan's younger generation follows today.' Judge H explained that, in reality, the fault-based grounds sometimes do not make much sense: According to current law, to grant divorce, there must be specific legal grounds, but we in fact adopt a rather lenient interpretation [of the law]. ... While the litigants have already ripped each other's faces off in the courtroom and things have become ugly, if you still stick to the strict interpretation [of the law], you can only maintain the mere appearance of a marriage relation. I'd 49 In the Taiwanese dialect South Min, Yao-Shou is a vivid expression meaning "Your life will be shortened because of what you did!" 50 Interview with Judge A, supra note 21. 51 Chun, supra note 46, at 262-63. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 247 rather let them finish their relation, and maybe they will be able to start a new life.52 Some judges clearly resorted to their personal beliefs about marriage in hearing divorce cases. For example, Judge A said, "We young judges all think that if the couple can't get along we should let them divorce .... Only when both parties are willing, could there be a marriage. If one party already can't stand it anymore, I think it's inhumane to force them to remain together."53 A few judges directly criticized the current law because they believed the fault-based system might cause more problems. As Judge K put it, I don't encourage divorce.... But... Let me express my opinion of the current law: the conditions for divorce are kind of too strict. ... In some ways I manage to ease the conditions. I think when two people cannot get along, even though you deny the divorce, still, they will not live together. In many cases, the couple has been separated for a long time when one of them files for divorce. If you don't grant the divorce, you will allow the unstable legal status to continue. On documents they still are a married couple, but in fact they don't live together and have no actual marriage relation. This will produce more problems .... If the court grants the divorce, at least it will have the opportunity to decide custody and visitation arrangements for them. . . .If you deny the divorce, their legal status will be disconnected with their real everyday lives. It's nonsense to deny the divorce. They in fact already need to settle custody, visitation, and child support issues, but they can't if you don't let them divorce, because these rights and legal relations are based on the marriage's dissolution .... Love is a mutual thing. When only one party is willing to, it's not possible to maintain the relationship." 52 Interview with Judge H, supra note 30. 53 Interview with Judge A, supra note 21. 54 Interview with Judge K, supra note 44. COLUMBIA JOURNAL OF ASIAN LAW Nevertheless, a few judges noticed that no-fault divorce might be a disadvantage to the one who is more vulnerable and has fewer resources in a marriage relation. Usually, in Taiwan, wives are more vulnerable because they have less or even no income.5 For example, Judge DI also preferred no-fault divorce, but he frankly said that whether the wife is the plaintiff would significantly affect his decision: I found that if the wife petitioned for divorce, I almost always granted the divorce. But if the husband was the plaintiff, it was not necessarily so. In fact, I granted husband's petition much less often. Ha! ... I am a male judge myself, but I am stricter with male plaintiffs who petition for divorce. 6 Judge K, who had criticized the current fault-based system, also brought up this consideration: Sometimes dissolution of the marriage could be obviously unfair to one party. . . . For example, the husband has a mistress in Mainland China, and he has transferred all his property to that mistress. Then he sues the wife for divorce based on the fact that they have not lived together for a long time. You can see the unfairness is so obvious [if we grant his divorce petition], and it damages the integrity of justice and fairness. Considering that many people in Taiwan have limited legal knowledge and don't know how to protect themselves, and considering that marriage still could ensure some traditional women's rights, under these circumstances, I may not grant the husband's divorce petition. To be frank, I let the wife at least have [the ability to agree or disagree to] divorce as a bargaining chip with the husband. She still can agree to divorce if the husband compensates what she will lose because of the divorce and transfers the property back. 7 55 Liu, supra note 3, at 170-7 1. 56 Interview with Judge Dl in Taiwan (July 12, 2001). 57 Interview with Judge K, supra note 44. Some judges who were not interviewees for this research also used this method to retain the wife's bargaining chip. See Taiwanese Law Association, Transcript of Seminar and Panel Discussion on Divorce Law, 17 LAW FORUM 313 (1996). [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 249 In the previous study the researcher has discussed the old saying "Qing Guan Nan Duan Jia Wu Shi" (even honest and smart government officials will find it very difficult to judge and decide on family matters), which implies the idea of family autonomy. 8 When asked about their opinions of this old saying and family autonomy, the interviewees had differing viewpoints. Some judges were in favor of it. For instance, many interviewees thought that a judge should not intervene when the parties could reach an agreement on custody themselves. They preferred to let the parties negotiate on their own. Even though the litigants had petitioned for custody, still some judges might encourage the litigants to withdraw the petition and negotiate themselves outside of the courtroom.59 By contrast, some, but very few, judges did not always embrace the idea of family autonomy; they thought the court should intervene to protect the child's interests in custody issues, especially since the child had no advocate in the proceedings.6 Ideas related to gender preference also emerged from the interviews. Traditional Chinese society has always had the ideas of "Chuan Zong Jie Dai" (a man should continue the family line by producing a male heir) and "Zhong Nan Qing Nu" (valuing men or sons higher than women or daughters).6' Several judges similarly described how these ideas appeared in some divorce cases: sometimes the couple (especially the father) might fight for the custody of boys more bitterly. When there were both a son and a daughter in the family, sometimes the father wanted custody of only the son, and the mother wanted only the daughter.62 One judge asserted that these phenomena could be more often seen in rural areas.63 In fact, a few judges themselves also tended to award a son's custody to the father and a daughter's custody to the mother. We will discuss it further in a later section. 58 Liu, supra note 2, at 187 (2001). 59 For example, see Interview with Judge G, supra note 29; Interview with Judge D5 in Taiwan (July 12, 2001); Interview with Judge BI, supra note 27; Interview with Judge B2, supra note 33. 60 For example, see Interview with Judge J, supra note 25. 61 HSIUNG, supra note 46, at 321. 62 Interview with Judge A, supra note 21; Interview with Judge C, supra note 41; Interview with Judge G, supra note 29; Interview with Judge H, supra note 30; Interview with Judge K, supra note 44; Interview with Judge J, supra note 25. 63 Interview with Judge A, supra note 21. COLUMBIA JOURNAL OF ASIAN LA W 2. Differences Between Urban and Rural Areas A few judges interviewed had heard cases in both rural and urban areas. A few others grew up in urban areas themselves but now sat in rural courts. When asked whether they noticed any differences between rural and urban areas in terms of, for instance, the features of litigants and family cases, most of them gave an affirmative answer. For example, Judge A, who sat in a jurisdiction that included both suburban and rural areas, described what happened in a case: Once I had a case in which the mother filed for divorce and petitioned for custody rights. She had three or maybe four children, but she only asked for the daughter's custody. When I asked her why, she replied-I remember her reply very clearly, "A son always belongs to the father, so I shouldn't take his son away!" Maybe. . . the traditional idea of Chuan Zong Jie Dai, the idea that a man should continue the family line by producing a male heir, is still ingrained in our rural areas. The idea is still pretty influential here.'4 As to parents' attitudes and ideas about children and custody, many judges noticed differences between rural and urban areas. Judge D I, who grew up in Taipei and now sat in a jurisdiction that included both urban and rural areas, gave an example of the difference: I really find that people in the rural area.., there is a gap between the knowledge levels and ideas of theirs and urban people's. Many parents here think children are their property... You can tell from the words and terms they almost always use. They very often speak of children as they speak of "things." They very often say that: the child is "yours" or is "mine," so I want to give him away to you or you can have nothing after I get him.65 64 Interview with Judge A, supra note 21. The phenomenon here observed by Judge A might be related to why we found that rural judges were less likely to award custody to mothers than were urban judges. It is plausible to predict that, in rural areas, when mothers fought less often for custody of sons, judges would less often award custody of sons to them. 65 Interview with Judge D1, supra note 56. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 251 Almost all of the judges who had heard cases in both rural and urban areas remarked about the same phenomenon: the litigants in urban areas usually had higher education levels and more legal knowledge, and it was more common for them to retain an attorney. A judge sitting in an urban area roughly estimated that of all the family cases that entered her courtroom, at least half of the litigants retained an attorney, while another judge sitting in a rural area said in only one or two out of every ten cases had he seen an attorney.6 A highly experienced judge, who had sat in many different urban and rural areas, also asserted that the quality of rural attorneys was relatively worse; generally speaking, they did not make as many legal claims and arguments as did urban attorneys, and the legal documents they wrote usually had more mistakes. This judge also asserted that urban litigants were less likely to reconcile with the opposite parties because their attorneys usually did not want them to reconcile outside of the court. Furthermore, given that the rural litigants were more conservative, had less legal knowledge, and did not retain an attorney often, and also given that the rural attorneys made fewer, and maybe worse, legal claims and arguments, not surprisingly, several judges asserted that the custody cases in rural areas were less complicated and the disputes between the parties were less bitter. 68 A rural judge commented that child support, alimony, and property division might not become heated issues since the litigants usually did not have much property in rural areas.69 V. CUSTODY DECISIONS: JUDGES' ATTITUDES AND BELIEFS, THOUGHTS ON THE NEW STANDARD, AND CUSTODY DETERMINATION FACTORS A. What Judges Think and Feel About Determining Custody As noted earlier, some judges-especially those sitting in ordinary civil courts who insist on the adversary procedure-thought 66 Interview with Judge E, supra note 43; Interview with Judge F, supra note 24. 67 Interview with Judge G, supra note 29. 68 Interview with Judge G, supra note 29; Interview with Judge D9 in Taiwan (July 12, 2001); Interview with Judge F, supra note 24. 69 Interview with Judge A, supra note 2 1. COLUMBIA JOURNAL OF ASIAN LAW custody was relatively simple or less complicated for them to decide. In fact, in a divorce case, they paid much more attention to whether to grant divorce rather than custody and child support. Judge F described how he and some of his colleagues might write custody decisions: If the judge thinks custody is just a minor issue, . . . he may think the custody part is "very" easy to write. He can copy whatever the social worker says, separate two children and award custody of one to the father and the other to the mother, or just check whose fault caused the divorce and award custody to the other spouse. Then he can simply copy all the factors listed in article 1055-1 and say "I believe this decision meets the best interests of the child standard." We can find so many decisions.. • including some of my own decisions because I have a heavy caseload [and had no time to do better], decisions that state "After considering the report of a social worker, and after considering the child's wishes, the child's needs, the parents' moral character, their economic resources, et cetera, et cetera, I believe the custody should be awarded to this person." But, in fact, there is no "reasoning"; they just directly jump to this "conclusion." . . . They just "mention" all these factors but don't say anything about "how" they actually considered this or that factor in this case according to this or that fact.... Why do some judges do so? Because if I write the decision in this way, it appears that I have considered all these factors; if the decision goes to the appellate court afterward, the appellate court can't say I haven't considered this or that factor. Then, how judges actually consider the factors or exercise discretion exists only in their minds, not in the decisions. . . . In the custody decision part, usually you don't need to explain your reasoning because it's just too difficult to explain. But the divorce part is different; you need to explain why the very facts meet a specific legal ground or condition. The difference is that the law doesn't say you need to [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 253 meet a specific ground or condition before you can receive custody.7 In contrast, a few other judges expressed a different attitude toward custody decisions. They thought custody was extremely difficult to decide because the facts in custody cases are often more complicated than the facts in ordinary civil cases. In custody cases they need to consider not only the past and present facts, but also to predict what may happen in the future. Judge H explained, When the couple themselves can't reach an agreement on the child custody arrangement, of course the judge needs to bear the burden to decide it. But, when we bear the burden, we feel so much pressure from it, because.. . how can you tell what the "best interests of the child" is? I am really very scared! The factors listed in the law are just something that appears to be objective ... According to my experience, in this respect, what a judge can do is very limited, because we don't have many chances to contact the child-maybe just once, and it's only ten minutes or at most thirty minutes.... How can you really find out his interaction with the parents in a courtroom? It's not possible.... Sometimes social workers' reports are incomplete. How can we make decisions based on such kind of reports? . . . Really, sometimes, I really ask myself: do the decisions I make actually meet the best interests of the child? To be frank, sometimes I worry a lot. . . . I feel so much pressure on my mind. I really suspect that, with insufficient information and data, the decisions I make may be wrong and I can ruin the child's future .... In particular, in Taiwan, once custody is decided, it's very rare to see a modification of the custody decision afterward.7 Some other judges expressed their frustration. They wanted to make better custody decisions, but they could not because of limited resources and heavy workloads. Judge D2 stated that, 70 Interview with Judge F, supra note 24. 71 Interview with Judge H, supra note 30. COLUMBIA JOURNAL OF ASIAN LAW Custody cases have a special feature: they are not merely legal issues; many terms stipulated in the laws concerning custody are uncertain, and judges are expected to exercise discretion to decide. . . . For example, the "best interests of the child" . . . is very difficult for us to decide; even though we have marriage experience, it's still very difficult .... unless you have some staff or experts who help you investigate,... [but] the current support system [for investigation] is entirely insufficient.... So even if you are dedicated to custody cases, even if you want to be a meddler who really cares, you still feel it's very difficult to make good decisions. We have the pressure of heavy caseloads and other pressure.... It's not just whether you are dedicated and willing to spend your time and energy. We need some system reform. . . . I have a strong feeling of helplessness!72 Some previous studies have discussed how a judge's gender, personal beliefs, values, and life experiences might affect his or her custody decisions.73 Several judges frankly admitted that it was true. For example, Judge A said that: After having become a father myself, I really feel that men and women are quite different. Women really are more patient to take care of children. We men always think that: since women can do it better, we should just give the job to them. When we hold or hug a baby, we are just trying to have fun. Ha! ... This is my personal experience.74 Judges of different genders might have different observations and viewpoints. An interesting scenario happened when Judge B 1 (a male) and Judge B2 (a female) were interviewed together. They were colleagues sitting in the same rural court. Their conversation was that: 72 Interview with Judge D2 in Taiwan (July 12, 2001). 73 Hui-Hsin Chen, Changes in the Family Status of Chinese Women from a Legal Perspective, in PROCEEDINGS OF CONFERENCE ON THE CHINESE FAMILY AND ITS ETHICS 227, 228 (Center of Chinese Studies ed., 1999); Liu, supra note 3, at 54-72. 74 Interview with Judge A, supra note 21. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 255 Judge B 1: In our jurisdiction, there is a problem: we don't have any women's growth group. . . . Some mothers play the role of victims.... They don't think of advancing themselves in life. Sometimes we think that maybe they will be more suitable to be the custodian of the child after receiving proper education, but we can't find this kind of group to help them.... I think, after divorce, these women should try to develop themselves. We should let them grow... Judge B2: I think some fathers need [to grow] too! Judge BI (laugh): My god! My god! Judge B2: Some fathers need to grow more than mothers do! Judge B 1: But I saw so many mothers who only knew to cry and complain ... They didn't think of developing themselves in life. Judge B2: But fathers were just indifferent to the marriage! They were worse! ... They were in more need of growth!75 B. Custody as Parents' Rights from Some Judges' Perspective In ancient China, society viewed the child as an economic asset or even property, and the head of the household (usually the father) held complete control of the child." In the context of this tradition, not surprisingly, child custody in Taiwan was always identified as a "right" belonging to the parents. Before 1996-the year in which the new custody law was enacted, almost all law books or court decisions referred to custody as "Jian Hu Quan," which means "custody right."77 However, the new law directly changes the old term of custody and refers to custody as "parental rights and obligations to the child." 8 This change is of course necessary because the new law adopts the "best interests of the child standard" that shifts the focus of custody disputes 75 Interview with Judge BI, supra note 27; Interview with Judge B2, supra note 59.76 TUNG-TSU CHIU, CHINESE LAW AND CHINESE SOCIETY 16 (1994). 77 For example, see YEN-HUI TAI & TUNG-HSIUNG TI, CHINESE FAMILY LAW 257 (2d ed. 1988); 69 Taishang 3676 (Taiwan Supreme Court, 1980); 71 Taishang 1568 (Taiwan Supreme Court, 1982); 69 Taishang 675 (Taiwan Supreme Court, 1980). 78 CIVIL CODE art. 1055 (1) (Taiwan ROC). COLUMBIA JOURNAL OF ASIAN LAW from the issue of who has the right to custody to what kind of custody arrangements will serve the child's best interests. The child's rights and each parent's obligations are highlighted in the new law. In all the interviews, the researcher intentionally avoided using the old term "Jian Hu Quan" (custody right). Instead, the researcher used "Jian Hu" (custody) or the complete phrase "parental rights and obligations to the child." Strikingly, almost all judges interviewed automatically changed back to the old term and used "custody right" in their answers or descriptions. Similarly, asking about visitation, the researcher intentionally avoided using "visitation right"; instead, the researcher used only "visitation" or the complete legal phrase in the new law "meeting and interacting with the child."79 However, many judges still used "visitation right." The term they (unconsciously) chose to use might reflect that they still tended to consider custody as a "right" belonging to parents, a tendency that the following findings of the interviews seemed to attest to. First, as noted earlier, in divorce cases, some judges still explicitly followed the adversary procedure; when the litigants did not petition for custody, they would not remind them of this issue and would not make any decision regarding it. Obviously, they considered custody as similar to a property right that the owners could freely decide whether to claim or not. Second, even though the new law highlights parental obligations following divorce, still, only a few judges specifically mentioned any of the parental obligations. When asked about the non-custodial parent's obligation of child support following divorce, several judges appeared to be unfamiliar with it. A judge sitting in a rural area even needed to look up in a book of law codes to make sure of what it was. Some judges admitted, "I have never awarded child support in a divorce case."8 Seemingly, for some judges, the "obligation" aspect of custody did not enter their minds often, and they tended to consider the "right" aspect of custody only. C. Judges' General Feelings About the "Best Interests of the Child" Standard Before discussing some specific factors to be considered in determining the best interests of the child, the researcher asked the judges about their general feelings about this new standard. Several said 79 CIVIL CODE art. 1055 (5) (Taiwan ROC).80 Interview with Judge B 1, supra note 27; Interview with Judge F, supra note 24. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 257 it was very abstract and vague. A judge described it, saying that "determining the child's best interests was like making a work of art, and we could not establish universal criteria for the best ways to do it." 8' Another judge said, "Everybody has his own definition of the child's best interests, but there must be one general definition that our society in reality accepts most."' 2 As noted earlier in this section, a few other judges expressed the fear, pressure, and frustration they felt while determining the unknown, uncertain best interests of the child.83 Though the law lists many factors to be considered in determining the best interests of the child, several judges pointed out that they would consider only a few specific factors and tended to ignore others. As Judge E and Judge H put it: "No matter how many factors are listed in the law and whether the standard is abstract, what I will consider is just these specific few factors."' "There are so many factors listed in the law. Of course I won't consider all of them in every case. In fact I can't remember all of them. Ha!"85 Interestingly, the specific factors that each of these judges considered were quite similar. All of them mentioned "wishes of the child" and "interview report of social workers." Most of them mentioned "economic resources of the parents," "support from relatives," and "age of the child." This fmding verifies the fmdings of the quantitative analysis of court decisions.86 D. Factors Used in Determining Custody and Their Importance 1. Age of the Child and the "Tender Years Doctrine" The quantitative analysis of court decisions has found that more than one-fourth of all the decisions analyzed had considered "age of the child"; strikingly, whenever the court had considered this factor, the custody was awarded to the mother, a phenomenon very similar to the "tender years doctrine" in the legal history of the United States.87 The 81 Interview with Judge K, supra note 44. 82 Interview with Judge C, supra note 41. 83 See supra note 71, 72 and accompanying text. 8 Interview with Judge E, supra note 43. 85 Interview with Judge H, supra note 30. 86 Liu, supra note 3, at 28-32. 87 Liu, supra note 2, at 195. COLUMBIA JOURNAL OFASIANLAW interview results show that many judges supported the idea of the "tender years doctrine." Most of the judges interviewed believed that the mother is the parent ideally and inherently suited to care for children of a "tender age." Judge H, a female judge who was a mother herself, explained, I tend to award custody of younger children to mothers, because ... This idea may come from my personal life experience. You know, taking care of younger children is very tiring! You need to pay attention to many, many trivial things, and the children need to receive attentive care. Usually, mothers can do better, and fathers are not capable enough. Many of these judges mentioned an old Supreme Court decision made in 1928, which stated that custody should be awarded to the father except when the child is still in his or her infancy.88 Apparently, the "tender years doctrine" had existed in Taiwan's judicial practices before 1996, and many judges were familiar with it. However, a few judges did not take the "tender years doctrine" for granted. They thought the court still should consider who the actual caretaker of the child has been. As Judge C and Judge J put it: "Indeed, usually mothers are more capable of caring for children, but this is a social reality caused by gender inequality. Our society always assumes that women should be caregivers of children. The responsibility is always imposed on women."89 "The concern should be who the caregiver has been. If the mother has been the caregiver, I may award the custody of the infant to her, but mothers do not have priority simply because of their gender."go In addition to paying attention to tender aged children, some judges paid special attention to adolescents when considering the factor of "sex of the child." 2. Sex of the Child The previous analysis of court decisions showed that though many judges had not explicitly considered "sex of the child," when they 88 16 Shang 1105 (ROC Supreme Court, 1928). 89 Interview with Judge C, supra note 41. 90 Interview with Judge J, supra note 25. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 259 did do so, they preferred that fathers received boys' custody and mothers received the girls'. 9' The interviews verify that a few judges did tend to determine custody in this "boys for fathers" and "girls for mothers" pattern, especially when the boys or girls were in their adolescence. When asked why they tended to award custody of boys to fathers and custody of girls to mothers, their reasons were very similar-they emphasized the importance of gender role identification. Judge C explained, After a boy enters high school, he will be physically energetic, and he may need a basketball field or a sports ground to exercise. Unless the mother is athletic, she can't accompany the boy in exercising in a sports field. If the boy lacks his father's company during his high school period, he will be very pitiful and become very feminized. If the child is a girl, she should live with the mother because she needs her mother to tell her about menses, sex, pregnancy, and contraception. It will be a misfortune for her if she lives with the father." In fact, how judges considered "sex of the child" might interplay with some cultural ideas and social customs. As previously noted, because the ideas of "Chuan Zong Jie Dai" (a man should continue the family line by producing a male heir) and "Zhong Nan Qing Nu" (valuing men or sons higher than women or daughters), some fathers wanted to receive only the custody of their sons while the mothers might not dispute it.93 Not surprisingly, the court might tend to award a boy's custody to the father and a girl's custody to the mother when both the litigants seemed to agree on this arrangement. 3. Sex of the Parent-Some Judges' Preference for Mothers Even if there was no child of a tender age involved, some judges still explicitly showed their preference for awarding custody to the mothers. Since "sex of the parent" is not a factor listed in the law, from 91 Liu, supra note 3, at 32. 92 Interview with Judge C, supra note 41. 93 See supra note 61-63 and accompanying text. COLUMBIA JOURNAL OF ASIAN LA W the perspective of gender equality, theoretically, judges should not prefer one parent over the other solely on the basis of the sex. However, in reality some judges obviously still did. For example, both Judge E (a female) and Judge Dl (a male) made similar remarks: "When the conditions and status of both parents are similar, and they seem to be equally suitable for custody, of course I will award custody to the mother.... Mothers are more attentive."' "Unless the mother is very unsuitable for custody, I may tend to award custody to the mother."95 By contrast, no judge expressed any preference for fathers in the interviews. However, a few judges claimed that they did not have any preference for either mothers or fathers. 4. Wishes of the Child Perhaps partly because the law stipulates that judges shall consider the child's wishes when he or she is over seven years old, almost all judges stated in the interviews that "wishes of the child" was a very important consideration for them in determining custody. This result is consistent with the findings of the quantitative analysis of court decisions, findings that "wishes of the child" was one of the factors most often considered in custody decisions.96 Most of the judges stated that, in order to make the child feel safe and comfortable in expressing opinions openly and honestly, they would send the parents out of the courtroom when they asked the child about his or her wishes. Considering that the child might feel conflicting loyalties between parents, a problem that could traumatize the child and affect the reliability of his or her statements, avoiding asking the child in front of the parents might be necessary. 91 However, a few judges described that they sometimes asked the child about his or her wishes when the parents were in the same room. For instance, Judge C mentioned a case she had heard, "When I asked the child about her wishes, she felt embarrassed to answer in front of her parents. They were in the same open court because of my heavy caseload and limited time." 98 Similarly, Judge H stated that she might not send the parents out of the courtroom if the children did not request it in front of the parents; 94 Interview with Judge E, supra note 43. 95 Interview with Judge D1, supra note 56. 96 Liu, supra note 3, at 28-3 1. 97 Barbara L. House, Considering the Child's Preference In Determining Custody: Is It Really In The Child's Best Interest? 19 J. JUV. L. 176, 178-86 (1998). 98 Interview with Judge C, supra note 36. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 261 she also stated that she would ask the parents' opinions about whether they should be sent out. She said, "I may ask both parties' opinions. If the child is older,... I would ask the child whether he or she wants the parents out in order to talk to me privately. If the child does not think the parents' presence will bother him or her, it may not be necessary to send the parents out." Even though the parents might be out of the courtroom when the judges interviewed the child, afterwards, parents would still have the opportunity to know the opinions that the child expressed. Judge A stated that he might ask the parents to come in after interviewing the child, and then, in front of the child, he would tell them what the child had said and ask their opinions about it.'" Meanwhile, Judge K pointed out that, according to the law, the parents and their attorneys can request to read the record of the interview any time. 101 Several judges revealed that, when interviewing the child, they would adopt a more relaxed style. For example, they might first chat with the child about his/her clothes or the cartoons he/she liked; only after the child felt comfortable to talk with them, would they ask further about his or her wishes as to the parent he or she wanted to live with.02 One judge mentioned that her court had a special "children's room," which was decorated as a warm and relaxed environment where judges could interview children, but she also mentioned that the room was used only occasionally. 13 Another judge stated that occasionally she might interview the child in her office. All of the other judges seemed to conduct the interviews solely in courtrooms. Though the law only requires judges to ask the child's opinions when he or she is over seven years old,"° many judges stated that they still would ask children under the age of seven years if he or she could clearly express his or her wishes. Some of these judges noticed that urban children usually mature at an earlier age than do rural children, so they might consider the children's wishes even though they were under seven years old. 105 99 Interview with Judge H, supra note 30. 100 Interview with Judge A, supra note 21. 101 Interview with Judge K, supra note 44. 102 Interview with Judge BI, supra note 27; Interview with Judge B2, supra note 59; Interview with Judge F, supra note 24; Interview with Judge C, supra note 41 103 Interview with Judge H, supra note 30. 104 Non-Contentious Matters Law art. 71-6 (Taiwan ROC). 105 See supra notes 64-69 and accompanying text. COLUMBIA JOURNAL OF ASIANLAW Although social workers almost always included the child's wishes in their reports, all the judges said they would ask the child again themselves. If what the child told a social worker differed from what he or she told the judge, one judge clearly stated that she would make the decision according to her own interview result and discard the social worker's description."° A few judges noted that what the child expressed might be affected by some of the parents' actions. For example, the parent with whom the child lived might "coach" the child in saying specific things to the judges; some parents might "bribe" the child with gifts or treat him/her extremely well in order for the child to favor them. One judge even mentioned that some parents might "threaten" the child and demand that he or she say favorable things about them. All of the judges who had noted these situations stated that they would be very careful while asking the child's wishes and that they would not stop at the child's first few responses. However, other judges seemed to be unaware of these possible situations that complicated what the child's real wishes were; some of them would just expect a direct and honest answer from the child. Not only would almost all the interviewees consider the child's wishes in determining custody, but they also asserted that they almost always respected the child's wishes and awarded custody in accord with those wishes. This is consistent with the findings of the quantitative analysis of court decisions. 7 Meanwhile, three judges specifically pointed out that the older the child was, the more important his or her wishes were, while other judges did not mention that they might place different weight on the wishes of a child based on the child's age. Only two interviewees had ever determined custody out of accord with the child's wishes. In one case, according to Judge H's description, she decided against the child's wishes because the conditions and status of the parents differed sharply, and the child had chosen the obviously inferior parent. She thought this child was too young to know what was best for him.'°8 In another case heard by Judge F, the child was a sixteen-year-old girl; theoretically, she was old enough to persuade the judge to respect her wishes. However, the judge also disregarded her choice based on the similar rationale: the child might not know what was best for her. 106 Interview with Judge E, supra note 43. 07 Liu, supra note 2, at 196 (2001). l08 Interview with Judge H, supra note 30. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 263 The second case above is interesting because it includes some discussions on the choice of lifestyle and career path, discussions related to the judge's personal values and beliefs. In this case, the girl wanted to live with her father, who did not oppose her relationship with her boyfriend and her current job-"Bin Lang Xi Shi," a job that required the girl to wear only sexy (and semi-transparent) underwear on the street to promote selling the local snack, Bin Lang, which is a snack purchased primarily by lower class consumers. Meanwhile, the father did not have a stable job, and girl had to support him sometimes. By contrast, the mother opposed the girl's relationship with her boyfriend because she thought the girl was too young. The mother also opposed the job, hoped that the girl would return to school, and was willing to pay her tuition. The judge concluded that the mother cared about the child more than the father did and that eventually, the girl would realize her mother's love after she became more mature.0° Several judges noted that, in their experience, most children chose to live with their mothers. This is consistent with the findings of the quantitative analysis of court decisions."10 These judges asserted that the phenomenon might be caused by the fact that most children were taken care of by their mothers, who almost always had better relationships with their children. 5. Needs of the Child's Personality Development In the previous study the researcher has found that though "needs of the child's personality development" is a factor listed in the law, the court seldom considered it in determining custody."' When asked in the interviews how they considered this factor, most of the judges said they had absolutely no idea what it was and seldom considered it in their decisions. Several judges remarked that even social workers also seldom elaborated on this factor. However, a few judges still tried very hard to describe what the "needs of the child's personality development" could be. One judge thought that it referred to stability of care, relationships, and social milieu. Several other judges asserted that it was related to the gender role identification and the decision pattern of "boys for fathers, girls for mothers," a pattern that they supported. Another judge believed that it 19 Interview with Judge F, supra note 24. 110 Liu, supra note 3, at 28-31. 111 Id. COLUMBIA JOURNAL OF ASIAN LA W related to the parents' personality, philosophy of life, and standard of value. In addition, two judges mentioned the parents' criminal record, their addiction to drinking or drugs, and the ways in which they disciplined and educated the child. 6. Moral Character and Performance of the Parents Judgments on the parents' "moral character and performance" obviously relate to some subjective beliefs and standards of value. The analysis of court decisions found that not many judges had explicitly considered this factor in their decisions," 2 and the interview results seem to be consistent with this finding. One judge thought the parents' moral character and performance were not necessarily relevant to the child's best interests. Some other judges appeared to be unfamiliar with this factor. In fact, one judge did not even believe that it is listed in the law until the researcher showed her the law code."3 Nevertheless, several judges mentioned some situations they thought might relate to this factor: family violence behavior, addiction to drugs, and criminal record. Interestingly, sex-related conduct, such as working in the sex industry (not necessarily as a prostitute), adultery, and homosexuality were referred to as bad moral character and performance by several judges. In addition, a judge thought that the parents' moral character and performance were related to their standard of value and philosophy of life. When asked about adultery, Judge C said "Adultery violates the moral principles of our society .... We almost never award custody to the one who commits adultery." "I However, other judges who mentioned adultery did not completely agree with Judge C. For instance, Judge J asserted, "Adultery is not relevant to taking care of children!""' 5 Meanwhile, coincidentally, at least three judges expressed the same opinion: only when the child was seriously disturbed or traumatized by the adultery would they think it was relevant to the child's best interests. 2 See supra Table 8. 113 Interview with Judge G, supra note 29. 114 Interview with Judge C, supra note 41. 115 Interview with Judge J, supra note 25. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 265 7. "Occupation and Economic Resources of the Parents" and "Child Support" When asked about how they actually determine the best interests of the child, most interviewees immediately brought up the factor "occupation and economic resources of the parents," even though the researcher never mentioned it. This is consistent with the findings of the quantitative analysis of court decisions: "occupation and economic resources of the parents" was the factor most often considered in addition to "social workers' reports.""' 6 However, many judges added a supplementary remark soon after they mentioned this factor, a remark which basically meant that, as Judge K put it, "This is a factor to be considered, but it is not absolutely important.""' 7 Before the researcher asked why, most of these judges further explained it as follows: The new law provides that each parent's obligations to support the child will not be changed after divorce, so the economically less competent parent could receive custody and ask the economically more competent parent to pay child support. As to the consideration of the parents' economic resources, two judges asserted that things had changed since 1996 because of the new law. As Judge G put it: Before 1996, it was almost impossible for an economically less competent parent to receive custody rights because there was no law regarding non-custodial parents' obligations of child support. In judicial practice, the custodial parent almost always had parental rights and obligations of child support at the same time. But the new 1996 law provides that each parent's obligations to support the child will not be changed after divorce. Things seem to have changed because of the new law... . Nowadays, [in determining custody] the importance of economic resources is not so absolute anymore." 8 From another perspective, and without relating to child support, many judges explained why they did not think this factor was absolutely important. They asserted that although they considered this factor often, 116 Liu, supra note 3, at 28-30. 117 Interview with Judge K, supra note 44. 118 Interview with Judge G, supra note 29. See also Interview with Judge C, supra note 41. COLUMBIA JOURNAL OF ASIAN LA W it did not mean that they would tend to award custody to the parent with more economic resources because the point should be to have "enough" economic resources, not "more" economic resources. As Judge H explained it: Of course the parents' occupation and economic resources are important! If you don't have enough economic resources, how can you raise a child? It's very expensive to raise a child nowadays! But I don't require you to have a very good job or a very high income; I only require you to have enough to cover the everyday living expenses of you and the child.... You don't need have to have a higher income than the other parent."9 Here the judges' own explanation verifies what the researcher surmised in the previous quantitative study.'20 Intriguingly, though many judges did mention that the new law entitles the custodial parent to receive child support from the ex-spouse, most of the interviewees said that in reality they seldom or never awarded child support in divorce cases. Here we see another sharp contrast between the "law in books" and the "law in action." Moreover, one judge insisted that a judge was not allowed to award child support in a divorce case because, according to the law, the litigants needed to file an independent suit for it after the divorce case, even though the new law clearly says the opposite.'2 ' Again, we find that not all judges who were supposed to or are required to know the new law were actually familiar with it. When asked why they seldom or never awarded child support, the judges gave different answers. One judge laughed and said, "Unless the litigant petitions for it, I don't want to make trouble for myself. Ha! ... If I don't need to consider child support, I can finish this case sooner. If she hasn't petitioned, I want to pronounce the decision before she can petition."'22 Most other judges said that they seldom or never awarded child support because the litigants seldom or never petitioned for it, either because of their lack of legal knowledge and some social/cultural customs. When the researcher asked whether they would "remind" the 119 Interview with Judge H, supra note 30. 120 Liu, supra note 2, at 204 (2001). 121 Code of Civil Procedure art. 572-1 (Taiwan ROC); Non-Contentious Matters Law art. 71-6 (Taiwan ROC). 22 Interview with Judge E, supra note 43. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 267 economically less competent parent of child support, almost all of them answered that they would not. Many of them said, "If the litigant really needs it, she or he can petition for it after the divorce anyway."'23 In addition, a few judges commented that a child support order was merely a piece of paper; they did not believe that a piece of paper would actually be effective, which made judges consider child support less.'24 8. Interview Report of Social Workers Almost all of the judges raised some issues concerning "social workers' reports" immediately after they were asked how they actually determined the best interests of the child. Evidently, all of them thought that social workers' reports played a significant role in their decision-making processes. However, the judges appeared to have contradictory and mixed feelings about the social workers and their reports. To begin, almost all judges stated that they very much relied on social workers' reports because they themselves lacked the necessary knowledge, time, opportunities, and resources for determining the best interests of the child. They claimed that they had to rely on social workers' reports because it was one of only a few things they could cling to in the current system. As Judge D2 explained: [T]he "best interests of the child".., is very difficult for us to decide ... unless you have some staff or experts who help you investigate. It's absolutely impossible to know the truth merely according to what you see and what the litigants say in the courtroom. . . .But the current support system [for investigation] is entirely insufficient. The court doesn't have any staff for this investigation, so all we can rely on is the outside social workers .... Judges seldom refute the social workers' opinions. . . .So in fact there is no such thing as "to decide according to the best interests of the child"; in fact, it's just "to decide according to a social worker's report." The truth is just that simple.'25 123 For example, see Interview with Judge E, supra note 43; Interview with Judge K, supra note 44; Interview with Judge A, supra note 21. 124 See Liu, supra note 2, at 213 (2001). 125 Interview with Judge D2, supra note 72. COLUMBIA JOURNAL OF ASIAN LA W As noted earlier, many judges thought they never received sufficient (if any) training for determining the best interests of the child. Judge K asserted that, compared with social workers, judges were basically incapable of determining custody, stating that, "Judges are just a rubber stamp." '126 Many judges stated that they would send every custody case to social workers. One of them explained "I send every case to the social workers, even though in the end I may not use or mention the reports in my written judgment.... I send every case to them because I can't go to the litigants' home to investigate for myself. They can be my eyes and ears." '127 However, some other judges stated that, according to the Child Welfare Act, they were only required to send cases that involved children under twelve years old,'28 so sometimes they did not send a case involving a child over twelve to a social worker. Two judges explained that children over the age of twelve usually could clearly express their wishes and describe their relationships with their parents, so social workers' reports might not be necessary. Ironically, though so many judges claimed they relied on social workers' reports to determine custody, almost all of them strongly questioned the creditability and reliability of the reports. They pointed out many problems of the social workers and the reports in the current system. To begin, the quality of the reports were often unacceptable, and the contents of the reports were often incomplete. Judge D2 heatedly described, [T]hey interview only once or twice, so the reports they make are superficial. . . . Their evaluation is usually pretty vague, and you don't know what their actual opinions are .... You can't find the recommendations based on their professional background. So in the end you still have to count on yourself to make a decision... • Every time I see the reports, I have a strong feeling of helplessness! The social workers still tell me to exercise discretion according to the best interests of the child, but I don't know what I am expected to do! .. . And I don't know whether they write something partially; I can't tell 126 Interview with Judge K, supra note 44.127 Interview with Judge F, supra note 24. 128 Child Welfare Act art. 41 (Taiwan ROC). [17:2 2004] CUSTODYDECISIONS IN SOCIAL AND CULTURAL CONTEXTS 269 if they do. If they add something into the reports, judges will be easily influenced. Basically, what a judge will write in his decision is what the social worker writes in the report. Judges seldom refute the social worker's opinion. . . .Because judges do not have time and opportunities to go out of the courtroom to interview as the social workers do, and because the judges can't say as much as the social workers can, it's almost impossible to refute the reports. 29 According to the judges' description, some social workers wrote the reports hastily. Judge K described, "Once I asked the litigant how the social worker interviewed him. He told me that the social worker came to interview him for only about twenty minutes and then just left. You see, the social worker could write out some stuff just based on that."'30 One judge even mentioned that she knew some social workers might not interview the litigants face-to-face; sometimes they might just telephone the litigants and ask some questions. These factors all lent to judges' extreme dissatisfaction with the reports they received. Currently, all of the courts send custody cases to the departments of social services of the counties or cities where the litigants live. Most departments of social services have their own social workers who are government officials who conduct the investigations and interviews, while others entrust the job to private child welfare organizations.' Even now, there has been no consensus among the different counties or organizations on the procedures and contents of the investigations and interviews. Different entities nearly always produce reports in different formats with dissimilar contents.'32 Because many couples had already separated and lived in different areas before they filed for divorce, most judges had received reports from departments of social services from several different municipalities. All of these judges remarked that in addition to the formats and contents of the reports, the quality of the reports also varied to a great extent from municipality to municipality, a problem that was caused by the insufficiency of qualified social workers in some areas, 129 Interview with Judge D2, supra note 72. 130 Interview with Judge K, supra note 44. 131 Bun-Fu Liu, The Current Situations of the Custody and Adoption Investigations Provided by County Governments, in PROCEEDINGS OF THE SEMINAR ON THE PRACTICES OF ADOPTION CASES IN TAjwAN 7, 11 (Children's Bureau Ministry of Interior ed., 2001). ' 32 Id. at 12. COLUMBIA JOURNAL OF ASIAN LA W especially rural areas. In addition, among the urban areas, Taipei City surpassed the other urban areas. Judge H, who sat in Taipei County, which bordered Taipei City, illustrated, Whenever I receive the reports, I really feel that the discrepancies between the resources of Taipei City and Taipei County are very striking. Taipei County has insufficient personnel and budget, so they entrust the job to a private organization .... Usually, the organization just makes a report of merely two pages. An interview report of merely two pages! It's very simple and crude. You tell me how I can judge based on a report so simple and rude? It's very difficult! The information is insufficient. In contrast, when we send a case to Taipei City, they will write a very formal report in a pile of papers. They write it in detail.... You can see from the report they visited and interviewed more than once, and they may describe what happened in each of their visits and interviews, what the child looked like, how they tried to click with the child and explore his wishes, and how the child interacted with the mother and the father. It's quite detailed. At least they write down many more things that we can take into consideration. So sometimes I sigh and think: how come Taipei County is behind Taipei City so much? . . . Taipei City's reports are usually concluded with a clear recommendation, but Taipei County's reports always have a very vague conclusion and ask the judge to decide the case himself according to the best interests of the child.'33 Ironically, although Taipei County's judges harshly criticized the quality of the reports in their own county, some judges sitting in more rural areas praised Taipei County's reports because, compared with their own counties' reports, they were much better. Almost all of the judges remarked on the imbalance of social service resources between different areas in Taiwan; according to their description, it seemed that the more urbanized the area was, the more funds and 133 Interview with Judge H, supra note 30. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 271 qualified social workers it had.'3 However, it should be noted that although almost all of the judges outside of Taipei City praised Taipei City's reports, the judges in Taipei City themselves still thought the reports were "often not useful, .... not really objective, .... often unprofessional.""'3 They thought that even Taipei City still seriously lacked necessary funds and competent social workers. Even if we put aside the problems of the quality and reliability of reports, another serious problem still exists. Many judges asserted that when the two litigants lived in different areas, it was virtually impossible to compare the reports made by different social workers in different areas to decide who the more suitable parent was. There was no basis for comparison because the two reports were made according to different procedures, by different workers, and with different contents and points. Each worker interviewed only one parent, knew almost nothing about the other parent, and had no opportunity to compare the child's interactions with both the father and the mother. As previously noted, the custody cases were sent to local governments; since each local government has its own jurisdiction, this problem of bureaucracy was inevitable. Not surprisingly, many judges claimed that this was a serious problem because, even if the quality of the reports was good, they still could not figure out which parent was more suitable to be the child's custodian. Many judges pointed out the same serious problems that seemed to exist in all municipalities. They found that many social workers just recorded whatever the litigants said to them in the reports; they might not verify whether it was true, and they might not add any of their comments on it. Given that the litigants usually tried to exaggerate the other party's disadvantages and his or her own qualities, the reports might contain a great deal of conflicting hearsay and incorrect information. As Judge B2 remarked, "Why would I need you to record what the litigants say? The litigants probably will say it again in the courtroom. The point is to observe things that I can't see [in the courtroom], things like his behavior, his living environment, and his attitude towards and interaction with the child.' 136 134 However, in the interviews, one (and only one) rural judge seemed to be satisfied with the social workers' reports in his county. See Interview with Judge F, supra note 24. 135 Interview with Judge E, supra note 43; Interview with Judge K, supra note 44 ; Interview with Judge J, supra note 25. 136 Interview with Judge B2, supra note 59. COLUMBIA JOURNAL OF ASIANLA W 1 In fact, several judges mentioned that because the social workers had very limited time and funds, and also because they might lack professional knowledge and experience, it was understandable why some of them made reports in this inappropriate way. Furthermore, since the court had no control over the social workers, who were either employees of local governments or on the staff of private organizations, the social workers did not need to worry much about what the court thought about their reports. The social workers might not have much incentive to burden themselves because the court and the litigants did not pay for their investigations and interviews. Government social workers would receive the same salaries no matter how many reports they made and what the quality of the reports was; for the staffs of private organizations, the local governments paid them only basic fees. Meanwhile, the social workers did not need to come to the courtroom to explain their reports to the judge, the litigants, or the attorneys. Even if the judges requested their presence in the court, as Judge K described, "They would try everything they could to avoid showing up in the courtroom, especially showing up in front of the litigants."'37 Social workers were not able to make complete and reliable reports for several reasons. First, they were not well trained. Recent research in Taiwan shows that social workers themselves were confused by what the "best interests of the child" was and what factors they should consider. They had no objective standard to rely on, so they might base their reports on their own values and beliefs. 38 Second, most social workers, especially the ones who were government employees, also had to handle other cases such as women's protection, elder's support, and administrative chores in addition to the custody cases. Their heavy workloads simply prevented them from having enough time and energy to focus only on custody cases. Third, social workers 137 There were several reasons the social workers were reluctant to present in front of the litigants. First, they were afraid that they might offend one of the litigants if they claimed the other was a more suitable parent than he/she was. The litigant might even threaten them. Second, they did not have time to go to the court for every report they made because of heavy workloads. Third, they might not be confident about their own reports so they would try to avoid being questioned by the judge, the litigants, and the attorneys. The first two reasons emerged from the discussion in a national seminar attended by nearly a hundred social workers and judges hearing family cases. The researcher attended this seminar and heard the discussion in person. See PROCEEDINGS OF THE SEMINAR ON THE PRACTICES OF ADOPTION CASES IN TAIWAN (Children's Bureau Ministry of Interior ed., July 11-12, 2001). 138 Hon-Yei Yu, CPS Workers'Decision-Making in Child Protection, 28 NCCU J. SOC. 81 (1998); Yu-Lun Huang, What Are the Best Interests of the Child? Child-Protection Social Works' Decision-Making in Sending Abused Children Home After Settlement and Its Factors Behind Decision-Making 78-83 (2000) (unpublished M.S.W. thesis, National Taiwan University). [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 273 might not be the only people to blame because judges themselves usually asked the workers to finish the reports within a week. Many social workers complained about this time constraint and claimed that it was impossible to finish a complete and correct report in such a short time. 39 Nevertheless, several judges remarked that the reports were unreliable and sometimes useless not only because of the problems mentioned above, but also because of the fact that the parents could manipulate the process. Judge G narrated a true story: In this case, when both the parents lived in Taipei, a social worker went to investigate and interview them and concluded that the mother was more suitable for being the child's custodian. She described that the father had some inappropriate behavior. Shortly after that, the father moved to Taichung and took the child with him, so a Taichung social worker went to interview and made another report, but this worker concluded that the father treated the child very well. I thought it was so weird.... I asked the Taichung worker to investigate and interview again, but this worker called my clerk and said, "Can I not go? The father threatened me very often." At that moment I suddenly realized that the report was problematic! ...So social workers may worry about their safety, and it may influence the correctness of their reports.... This was the first time I had ever thought of this problem."' In addition to possible threats from the parents, some judges noticed that the parents might play-act in front of the social worker. As Judge B2 put it: "The social worker tells you he will come today, so you can make everything orderly at home just today and say all the beautiful but untrue words to him.""' Moreover, as noted earlier, some parents might coach, bribe, or even threaten the child to influence what he or she will say to the social worker. Contradictorily, even though so many judges strongly doubted the reliability of social workers' reports, they still often quoted the 139 Liu, supra note 131, at 21. 40 Interview with Judge G, supra note 29. 141 Interview with Judge B2, supra note 59. COLUMBIA JOURNAL OF ASIAN LAW reports in their written judgments. Some judges claimed they almost always decided in accordance with the social workers' conclusions-if the reports had clear conclusions. However, it seems that many judges quoted the reports merely to justify and legitimatize their decisions, and they might arbitrarily select what they need from the reports to "fit" their already-made decisions. For example, Judge F frankly described, When writing a decision, I have a habit of putting the summary of the conclusion of the social worker's report in the first paragraph-but of course, if I quote the report, it means that his opinions are consistent with mine; otherwise, I am making trouble for myself because I will have to write another paragraph to refute his opinions.... Though the social worker may write many things in the report, as far as I know, most judges read only the conclusion part.... Why? Because the judges only need to quote the conclusion about which parent is more suitable as the custodian, and then they can say, "You see, it's the social worker who says so.' "142 Another judge also stated that, to her knowledge, some judges might "never mention the social worker's report in their written judgments if it is inconsistent with their opinions, or, of all the opinions stated in the report, they [might] ignore the ones that are different from theirs and quote only the ones consistent with theirs."'' 3 Similarly, in talking about the phenomenon that many social workers did not reach a clear conclusion in their reports, Judge A made an interesting comment with a mysterious smile: "In fact it's better if they don't make a clear conclusion. Otherwise, if we want to make a decision different from their conclusion, we will have to elaborate reasons for not taking their conclusion." 1' In fact, how some judges actually took into consideration or quoted the social workers' reports was clearly illustrated in a true story told by a social worker who publicly shared her experiences in a national seminar: "In several cases, before I even finished the reports, judges called and told me that they had already 142 Interview with Judge F, supra note 24.143 Interview with Judge C, supra note 41. 14 Interview with Judge A, supra note 21. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 275 made their decisions; they said the only thing they still needed was my report.' 45 Among all the thoughts and descriptions emerging from the interviews, there was one last point worth mentioning here concerning social workers' reports. In the previous study the researcher had pointed out that the law seems to restrict judges to considering only social workers' reports. By comparison, in many other countries, other professionals including psychologists and psychiatrists can be called upon to conduct custody evaluations."* In the interviews, the researcher asked the judges whether they had ever considered calling for other professionals' assistance, but all the judges except one promptly replied that they never had and probably never would. When asked why, all of them gave similar reasons. First, the court did not have any funds for hiring the other professionals, and according to the current law the litigants did not need to pay for it. Namely, since departments of social services of local governments would pay for social workers' reports, obviously social workers were more "convenient" for the court. Second, the mental health profession is still underdeveloped in Taiwan. Many judges stated that they did not know to which institutions or professionals they could send the cases; even if they did, they doubted that these institutions or professionals were capable of making custody evaluations. Intriguingly, several judges appeared to be completely ignorant about why child psychology or child psychiatry could be related to determining the best interests of the child. One judge said, "I don't know how a psychologist or psychiatrist can help in determining custody. I don't know in which respect they can provide assistance."'47 Another judge could only think of one occasion in which she might send the case to a psychologist or psychiatrist: "Maybe when the litigant is psychotic, I would send him to a hospital for mental evaluation."'48 The interview results verify the findings of the analysis of court decisions in the researcher's previous study. On the one hand, because the "best interests of the child" is too vague and it is difficult to decide, judges were not confident or competent in making custody decisions, so they had to rely more and more on social workers. On the other hand, as discussed in the previous study, Taipei City has many more resources 145 The researcher attended this seminar and heard the discussion in person. See PROCEEDINGS OF THE SEMINAR ON THE PRACTICES OF ADOPTION CASES IN TAIWAN (Children's Bureau Ministry of Interior ed., July 11-12, 2001). 146 Liu, supra note 2, at 212 (2001). 147 Interview with Judge A, supra note 21. 148 Interview with Judge G, supra note 29. COLUMBIA JOURNAL OFASIANLA W for social services than any other area in Taiwan. '49 The interviewees' perceptions of the availability and reliability of social worker services accurately reflect the imbalance of resources among Taiwan's different areas. From the analysis of court decisions, we have seen that Taipei's judges considered social workers' reports more often than did the judges from Pingtung, a more rural area. Considering that both the availability and reliability of social worker services were more limited in rural areas, it is not surprising that rural judges did not take into account the social workers' reports as often as did the urban judges. 9. Continuity, Stability, or Primary Caretaker As previously noted, though "continuity, stability, or primary caretaker" is not listed in the law, the quantitative analysis of court decisions still shows that the court often took it into account. In the interviews, almost all judges asserted that this was an important consideration for them. They also noticed that social workers' reports discussed this factor very often. Interestingly, a few judges were accustomed to considering it, so they took it for granted that it must be listed in the law; when the researcher told them it is not listed, they were very surprised. Most of the judges who thought this factor to be important remarked that it is in the child's best interests not to break the continuity and stability of the social milieu and caregiving for the child. Some of them also pointed out the importance of the emotional attachment between the child and his or her primary caretaker. A few judges noticed that usually the primary caretaker of the child was the mother, so considering this factor they might be more likely to award custody to the mother. The interview results verify what we have found in the quantitative analysis of court decisions.'50 10. Support from Relatives "Support from relatives" is not listed in the law either. However, several judges mentioned it themselves before the researcher did. The other judges who had not mentioned it on their own, and who were later asked their opinion about it, also asserted that it was an important 149 See Liu, supra note 2, at 197 (2001). '50 Id. at 198-99. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 277 consideration for them. Some interviewees noticed that it was a factor very often discussed by social workers in their reports. In the previous study, when talking about "support from relatives," the researcher mainly referred to the support in caring for the child. Surprisingly, in the interviews, two judges referred to "support from relatives" as financial support for the custodial parent. In fact, they mentioned this factor when explaining what they would do if the more suitable parent did not have enough economic resources. Another judge said he would refer the parent to a social worker for public assistance. It is interesting because none of those judges thought of obtaining child support from the non-custodial parent, which can be ordered and enforced by the court first; instead, their first notion was to let the custodial parent seek financial support from her or his own relatives. On the one hand, their attitude seems to reflect the traditional idea of a private welfare system, which bases an individual's well-being on a family unit and a network of relatives;'-' on the other hand, it may indicate that these judges followed the "all-or-none custody" tradition in Taiwan's society, a tradition to which we shall return later. When asked why they thought this factor to be important, many judges pointed out that today usually both men and women are employed, so they need others to care for their children when they are at work. Interestingly, all judges who specifically mentioned this sat in jurisdictions that included urban areas. Some of them related this factor to another factor, "economic resources." Judge H commented that "Of course you need to go to work after divorce; otherwise, how can you support the family? Then 'who will help to take care of the child' becomes a big problem.' ' 52 Several judges explicitly expressed their opinions that, in terms of looking after the child, relatives are better than nannies and Philippine maids. '"I A few interviewees specifically mentioned grandmothers but never mentioned grandfathers. Interestingly, almost no judge ever mentioned childcare centers or daycare programs for children of tender ages; two judges mentioned that support from relatives would be especially important before the child reached 151 See Liu, supra note 2, at 187, 200 (2001). 152 Interview with Judge H, supra note 30; Interview with Judge E, supra note 43. 153 Nowadays, in Taiwan, some middle and upper class families hire maids from the Philippines and Indonesia to take care of their children and household chores, because fewer and fewer Taiwanese are willing to be maids and these foreign maids are less expensive. This phenomenon is more common in urban areas. See EXECUTIVE YUAN, TAIWAN ROC, REPORT ON FERTILITY AND EMPLOYMENT OF MARRIED WOMEN IN TAIWAN AREA 7 (2000). COLUMBIA JOURNAL OF ASIAN LAW schooling age. As noted in the previous study, judges' responses may reflect the traditional way of relying only on relatives to take care of children and the social reality that childcare centers and daycare programs are insufficient in Taiwan. '1 The interview results verify that many judges still consider this factor even though it is not listed in the law; seemingly, urban judges were more aware of the fact that more and more women are employed today. 11. Domestic Violence As noted in the previous study, quantitative analysis of court decisions shows that while determining the best interests of the child, some judges seemed to think that spouse abuse incidents were unimportant or at least less important than child abuse incidents. 55 However, the newly enacted Domestic Violence Prevention Law provides a rebuttable presumption that it is not in the child's best interests to award custody to the abusive parent-no matter whether he or she is the perpetrator of spouse abuse or child abuse. 6 In the interviews, the researcher intended to explore how the judges consider domestic violence incidents and how they perceived and applied this law while determining custody. All the interviewees were aware of the rebuttable presumption in the new law. Compared with the laws of the "best interests of the child" standard and the specialized family proceedings, the judges were more familiar with the Domestic Violence Prevention Law. Several mentioned that this familiarity was because domestic violence had received a lot of attention from society, and the Judicial Yuan had held a series of training programs for judges to help them learn this specific new law. By contrast, no such training program had ever been held for the laws of the "best interests of the child" standard and the specialized family proceedings. Though the law clearly provides that the perpetrator is presumed unsuitable for custody, no matter whether he or she is the perpetrator of spouse abuse or child abuse, many judges clearly weighed child abuse and spouse abuse differently in their considerations. Most of them I54 A survey shows that, while they are working, 62.5% of Taiwanese mothers count on their parents or relatives to care for their minor children, but only 15.7% of them send their children to daycare centers. See TAIWAN PROVINCE GOV'T, SURVEY OF TAIWANESE WOMEN'S STATUS AND LIVES 92-93 (1993). See also Liu, supra note 3, at 12-13, 38-40. '55 See Liu, supra note 2, at 200-02. 156 1998 Domestic Violence Prevention Law art. 35 (Taiwan ROC). [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 279 mentioned that the perpetrator might abuse only the spouse but not the child, so he or she might still be a suitable parent if the other things he or she could provide the child were favorable. However, none of them except one judge noticed that the child might not only suffer from the violence against him but also suffer from witnessing spousal abuse. It appears that many judges were not fully aware of why the law also presumes that the perpetrator of spouse abuse is unsuitable for custody.'57 E. The "All-or-None Custody" Decision Pattern One of the most striking findings of the quantitative analysis of court decisions is the judges' decision pattern of "all-or-none custody." That is, judges tend to grant custody to only economically competent parents and would not award child support to them at the same time. The court supposes that the parent who receives custody should be the one-and the only one-who has the obligation to support the child. Usually, the custodial parent has all of the parental rights and obligations, while the non-custodial parent has none of them. None of the judges awarded joint custody and they seldom gave visitation orders. Partly because of some cultural ideas and social customs, Taiwan's judges have had this decision pattern since the era of the old law, and the new law has not really changed it. ' As we have seen, the interview results show that, on one hand, most interviewees did tend to grant custody to only economically competent parents; on the other hand, it was true they seldom or never award child support in custody cases. In order to verify and further explore the finding of the "all-or-none custody" decision pattern, the following sections will discuss judges' attitudes about joint custody, visitation, and the relevant cultural ideas and social customs. Note that during the interviews, in order not to affect or lead the judges' responses, the researcher avoided mentioning the."all-or-none custody" and any relevant cultural ideas or social customs. The following statements from the judges were made before the researcher gave any hint about the finding of "all-or-none custody" decision pattern. 157 See Liu, supra note 3, at 40-45; Naomi R. Cahn, Civil Images of Battered Women: The Impact of Domestic Violence on Child Custody Decisions, 44 VAND. L. REv. 1041, 1055-58 (1991). 158 See Liu, supra note 2, at 202-06, 212-14 (2001). COLUMBIA JOURNAL OF ASIAN LAW 1. Judges' Attitudes About Visitation Many judges stated that, unless the litigants petitioned for it, they would not give a visitation order in a custody decision. These judges believed this was a matter which should be discussed and arranged by the parents themselves, not only because visitation arrangements need to consider the parents' intentions and the coordination and cooperation between them, but also because a court order might be infeasible or too inflexible for them. However, of all the interviewees, compared with the number of the judges who had awarded child support, there were more judges who had determined visitation arrangements; a few judges had determined visitation arrangements on their own motions. Several of the judges who had determined visitation arrangements on their own motions mentioned that the new law clearly allows the court to do so. By contrast, they believed that the new law does not provide whether the court could award child support without a litigant's petition; therefore, they would not award child support on their own motions. Once again, we find Taiwanese judges' tendency to "never do anything that the law does not 'extremely clearly' say judges can do." ' 9 Because the law clearly provides that the court could determine visitation arrangements, it seems to change the judicial practice a little bit: before the new law was enacted, no empirical research had ever found any visitation order in court decisions; now we can find some-though not many. Note that still many judges asserted that they would not give a visitation order unless the litigants petitioned for it. When asked why and under what circumstances they would give a visitation order, some judges replied that they believed the non-custodial parent should have the "right" to visit the child. A few judges noted that the child should have the right to meet and interact with both parents too, because children need the love of both parents' love in order to aid in the child's development. One judge said she would consider giving a visitation order if the social worker's report recommended it. Interestingly, several judges coincidently mentioned that they would give a visitation order only ifthe litigants appeared to be 159 As noted earlier, in fact, on some occasions, even though the law clearly allows the court to make decisions on their own motions, many judges still hesitate to do so. See supra text accompanying notes 28-45. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 281 furious with each other and it seemed impossible that they could agree on a visitation arrangement themselves. A few judges noted that many litigants, especially those not represented by an attorney, did not know they could petition for visitation. As Judge A put it: "The litigants' ideas are simply that: custody right belongs to one party, and the other party will have almost nothing. Many people still have these ideas."'" Without any hint given by the researcher, the "all-or-none custody" tradition in Taiwan's society emerged spontaneously from the judge's dialogue. 2. Judges' Attitudes About Joint Custody All the interviewees except one had never awarded joint custody; in fact, they did not think they would ever award joint custody in the future. The only judge who had awarded joint custody asserted that it was extremely rare for her to do so. When asked their thoughts about joint custody, almost all immediately pointed out that joint custody is totally infeasible in Taiwan, and they provided similar reasons as follows. First, they asserted that in their courtrooms they had never found any couple who could really maintain a friendly and cooperative relationship after divorce, a relationship that is necessary for joint custody if we expect it to actually work. Note that Taiwanese law recognizes consensual divorce which only has to be registered at administrative agencies, so usually the couples who go to the court for divorce are those who are in a bitter dispute. Intriguingly, many judges spontaneously remarked that, compared with things in Western societies, in Taiwan's society and culture usually it is unimaginable to see a divorced couple maintain a friendly and cooperative relationship; divorced couples usually "rip the face off' and completely stop contacting each other or, worse, become antagonists. Judges' observations and descriptions here are consistent with the findings of past empirical research and what the researcher surmised in the previous study. 161 Second, many judges mentioned that joint custody would be infeasible and very inconvenient because, according to the law, the custodian is the child's statutory agent when the child carries out a juristic act, such as a contract. Meanwhile, administratively, the 160 Interview with Judge A, supra note 21. 161 See Liu, supra note 2, at 203, 214 (2001). COLUMBIA JOURNAL OF ASIAN LAW custodian is required to represent the child whenever the child applies for a passport or visa, admission to a public school, etc. If both parents are the child's custodians but they do not live together and may not always agree with each other, these things will not go smoothly for the child and the parents may need to go to the court often for decisions on their disagreement. 16 2 Third, some judges asserted that a sound visitation arrangement could also serve the purpose of maintaining the child's relationship with the non-custodial parent, so joint custody is not necessary. A few judges remarked they might award sole custody to one parent but let the other parent have frequent or long-term visitation with the child, arrangements such as letting the child live with the non-custodial parent every weekend or every summer. In short, the interview results verify the findings of the quantitative analysis of court decisions: Taiwan's judges almost never awarded joint custody. In fact, we find that many judges intentionally avoided awarding joint custody because, based on their observation of social reality, they strongly believed that joint custody could not actually work in Taiwan. Social customs and cultural ideas about how a divorced couple would behave toward each other obviously played a role in many judges' decision-making processes. 3. The "All-or-None Custody" Tradition and the Sense of "Face" Surprisingly, the "all-or-none custody" tradition spontaneously and frequently emerged from the interviews when the judges remarked on child support and visitation issues. As previously noted about visitation, a judge clearly explained that many litigants did not know they could petition for visitation because they took for granted that the non-custodial parent "will have nothing." Judge K made a vivid observation on how many litigants thought about this: When they first entered the courtroom for custody disputes, many litigants were so tense because they believed that if they couldn't receive custody they would 162 A judge made an interesting comment: "If I award them joint custody and later they can't agree with each other on these things, they will come back to the court again to ask me to decide for them. It's so troublesome. It's like I am increasing my own trouble. So I want to simply award sole custody, and then that's it." See Interview with Judge E, supra note 43. [17:2 2004] CUSTODY DECISIONS 1N SOCIAL AND CULTURAL CONTEXTS 283 lose the child forever . . . they would not see the child again. I had to convince them not to think in that way... • I tried to tell them the non-custodial parent could have visitation. 63 Similarly, discussing child support, many judges made remarks on the "all-or-none custody" tradition in the society. In fact, several judges made almost identical observations: Many people think that after divorce I don't want to have anything to do with the other party. If we are going to break up the relationship, we should break it up completely and totally. There shouldn't be any connection left. If I receive the child's custody, even if I can't afford to raise him myself, I don't need the court to award child support from the other party. I just don't want to have anything to do with that person. I would count on my own relatives' support to raise him. ,64 As noted in the researcher's previous study, in Chinese or Taiwanese society, divorce means "to rip the face off' (Si Po Lian) and "to lose face" (Diu Lian), and it is a dishonorable and unpleasant incident for both spouses and their families. 165 In the interviews, many judges mentioned that the idea of "face" played an important role in the process of divorce and custody disputes. Some of them observed that, because of "face," many single parents never wanted to receive child support or other assistance from their ex-spouses because taking money or assistance from other people, especially someone with whom you have a broken relationship, means "losing face" and is extremely humiliating." Results of past empirical research clearly show the practice of these cultural ideas in Taiwan's society. Usually, after divorce ex-spouses do not contact each other; the non-custodial parent will just leave the custodial parent alone to be completely responsible for the child, and the custodial parent will not want to ask assistance from the 163 Interview with Judge K, supra note 44. 164 Interview with Judge F, supra note 24. See also Interview with Judge H, supra note 30; Interview with Judge E, supra note 43; Interview with Judge A, supra note 21. 165 See Liu, supra note 2, at 203 (2001). 166 See also Taiwanese Law Association, supra note 57, at 323. COLUMBIA JOURNAL OF ASIAN LA W ex-spouse. According to a national survey, as many as 96.2 % of divorced parents never or seldom contacted their ex-spouses; only 16.2 % of non-custodial parents visited their children about once a month or more frequently, and 81.6 % of them never or seldom visited their children.1 61 Studies also show that many single parents count on their parents or relatives, not their ex-spouses, for financial support,'68 just as some judges expected the single parents to do.'69 The interview results show that the judges were aware of the social custom and cultural tradition of "all-or-none custody" and might make decisions in accordance with it. However, some judges made decisions in accordance with it not because they intended to do so or because they especially approved of it. In fact, some judges assumed the "no petition, no decision" attitude and did not like to determine custody, child support, and visitation on their own motions. Since most litigants still followed the "all-or-none custody" tradition, it is little wonder these judges' decisions would be in accordance with it. F. Other Findings 1. The Current Process of Assigning Judges to a Specialized Family Division As noted earlier, until the end of 2001, only Taipei District Court and Taichung District Court had a Family Division specializing in family cases.'7 ° Judges in this division would hear and decide only divorce, custody, adoption, and other family cases. Many interviewees, including several judges sitting in ordinary civil courts, remarked that such a specialized family division is necessary because of the different features and proceedings between family cases and property cases.' One might expect that these judges in a specialized family division had special training or background knowledge in this field. However, the interview results show a different picture. Of all the six interviewees sitting in a specialized family division, only two had had 167 CHING-FU CHANG, THE CURRENT STATUS OF SINGLE PARENT FAMILIES AND ITS POLICY IMPLICATIONS 120-24 (1995). 168 See, e.g., WAN-I LIN & WEN-LI CHIN, A STUDY ON SINGLE-PARENT FAMILIES IN TAIPEI AND RESPONSIVE POLICIES 111-32 (1992); TAIWAN PROVINCE GOVERNMENT, SURVEY OF THE NEEDS OF TAIWANESE SINGLE-PARENT FAMILIES 12 (1996). 169 See supra text accompanying note 151. 170 In addition, Nantou District Court had an experimental division specializing in both juvenile and family cases. 171 See supra text accompanying notes 21-39. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 285 specialized training or expertise in the field of family law before they were assigned to the division. The other four judges, as Judge C put it, were unfamiliar with family law and family proceedings when they were first assigned: Assigning me to the family division was merely an administrative accident .... I was like a layman in the field of family law when I first started. Though I had studied the statutes of family law to prepare for the national qualifying exam [for judges], I had never touched this field again. So in the beginning I knew almost nothing. ... But I had to learn and be familiar with it in a short time .... After being a judge in the family division, I still almost never had any opportunity to receive any training in this field.'72 More seriously, after these judges gradually gain expertise from first-hand experience of hearing many family cases, they will be transferred to other divisions or courts. Many interviewees bitterly complained of the current judicial administrative practice that rearranges a judge's position every two years or so. Because of this judicial administrative practice, whenever a judge inexperienced in this field finally feels familiar with family law and proceedings, he or she will be replaced by another totally inexperienced judge, and everything will start from the beginning again. Hearing this judicial administrative practice, the researcher was surprised because obviously it is inefficient and unprofessional. According to the judges' descriptions, this practice was based on a resolution of the "work assignment meeting" attended by all judges; because many judges want to have an opportunity to hear family cases, the meeting passed a resolution which provided that, during the reassignment process, judges who have never heard family cases will be considered first for assignment to the family division. When asked why many judges wanted to have an opportunity to hear family cases, the interviewees gave two explanations. On one hand, most judges want to have the opportunity to hear every kind of case to expand their experience and expertise in their careers as judges. Some judges want to do so because they plan to retire into private legal 172 Interview with Judge C, supra note 41. COLUMBIA JOURNAL OF ASIAN LAW practice in the future; more experience and expertise in more different fields of law certainly will help their private legal practice. On the other hand, as noted earlier, many judges-especially the judges who insisted on an adversary procedure and relied only on law codes-believe that family cases are relatively "simple" and "effortless"; so if they are assigned to the family division, they could relax a little fiom the heavy workloads of the other cases. Judge G told a true story that vividly described what some judges thought: "When I first came to this court, a judge was transferred to the family division because she was pregnant. She went to the family division because the workload there was light, and she could relax a little during the pregnancy.' ' 73 2. Whether Mothers Received Custody More Often Than Fathers A striking finding of the quantitative analysis of court decisions is that, after 1996, judges much more often awarded custody to mothers than fathers. Furthermore, judges in urban Taipei were much more likely to grant custody to mothers than were judges in rural Pingtung.'74 In the interviews, before the researcher ever asked or gave any hint, three judges spontaneously asserted that nowadays they more often granted custody to mothers than fathers. Several other judges who had not mentioned it on their own and after being asked whether mothers or fathers received custody more often in their decisions also noted that mothers received custody more often than fathers. All of these judges above sat in jurisdictions that included urban areas. By contrast, two judges asserted that mothers did not necessarily receive custody more often in their decisions, and both of these two judges sat in rural areas. Though the sample for interviews was not representative, these interview results seem strikingly consistent with the findings of the quantitative analysis of court decisions. When asked why they had more often awarded custody to mothers than fathers, the judges provided several explanations. Several mentioned that mothers usually were the primary caretakers and had better relationships with the children. A few judges noticed that, when asked their wishes, most children chose to live with their mothers. These are all consistent with the findings of the quantitative analysis of court decisions. In addition, one judge claimed that she found most 173 Interview with Judge G, supra note 29. 174 Liu, supra note 2, at 192-94 (2001). [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 287 social workers' reports favored the mothers, but another judge in another court claimed that it was not necessarily true. Also, two other factors may also result in more mothers receiving custody. First, today there are more female and young judges with ideas of gender equality on the bench than before.' 75 Second, more and more women file for divorce and petition for custody because they have independent incomes and also because, given that the paternal presumption was abrogated, they now have a greater chance of receiving custody. 76 Many judges mentioned the phenomenon that in divorce cases most plaintiffs were the wives. These judges' observations are consistent with official statistics, which show that today more than two-thirds of divorce cases are filed by the wife. '71 Some judges observed that, in the divorce cases filed by the mothers, the mothers were more likely to receive custody because many of these cases were ex parte cases based on the fathers' desertion. Meanwhile, they noticed that many wives filed for divorce because they were being abused by the husbands; on this occasion, partly because of the rebuttable presumption favoring the victims of spouse abuse,'78 mothers usually received custody. Moreover, as noted earlier, some judges explicitly showed their preference for mothers. For instance, both Judge E (a female) and Judge D1 (a male) made similar remarks: "When the conditions and status of both parents are similar and they seem equally suitable for custody, of course I will award custody to the mother. . . . Mothers are more attentive.' ' 1 9 "Unless the mother is very unsuitable for custody, I may tend to award custody to the mother.""'8 By contrast, in the interviews, '75 Since the early 1990s female students have become the majority at many law universities in Taiwan; since 1997 more women than men have entered the training process to become trial judges. See Liu, supra note 3, at 56-57. 176 See supra note 177 and accompanying text. 177 JUDICIAL YUAN, TAIwAN ROC, JUDICIAL STATISTICS 8-62 (2001). In fact, the proportion of the divorce cases filed by the wife has increased strikingly in the past decade. In 1991, only 56.6% of divorce cases were filed by the wife; in 1996, the proportion increased to 66.5%; in 2000, the proportion reached 70.4%. Many commentators believe that this phenomenon relates to the fact that nowadays more and more women have independent incomes so they are not afraid to leave unfortunate marriages. Some others believe that it may relate to the fact that many wives do not have enough resources or the capability to bargain with their husbands outside of courtrooms so they resort to the court. This study believes that the new law, which gives women a greater chance than before to receive custody, may also play a role. See JUDICIAL YUAN, TAIWAN ROC, 1998 ANNUAL REPORT OF JUDICIAL AFFAIRS 161 (1999); Chao-ju Chen, Rights, Legal Reform, and Indigenous Feminism: The Example of Divorce in Taiwan, 62 NAT'L CHENGCI U. L. REv. 25, 44 (1999). 178 See supra note 156 and accompanying text. 179 Interview with Judge E, supra note 43. 150 Interview with Judge DI, supra note 56. COLUMBIA JOURNAL OF ASIAN LA W no judge expressed any preference for fathers, though a few judges claimed they did not have any preference for either mothers or fathers. Interestingly, two rural judges spontaneously mentioned the phenomenon of "foreign brides" or "mainland brides" in Taiwan. During the past decade, more and more Taiwanese men, especially rural or lower middle class men, have attempted to find their brides in Mainland China, Vietnam, and Indonesia. These men with lower incomes usually found it difficult to marry Taiwanese women. Currently there are about 210,000 mainland brides and foreign brides in Taiwan. 8' Owing to cultural differences, many of these international marriages end in divorce. These two rural judges observed that, in these divorce cases, usually it was the husband filing for divorce because the wife decided to leave Taiwan or had already deserted him, so usually it was the father who received custody. This may partly explain why some rural judges were more likely to award custody to the fathers than were some urban judges. VI. FURTHER DISCUSSION A. When the New Laws Meet the Judges with Old Skills and Old Schemas As has been seen, some things have changed since 1996, but most have not or have changed merely a little. Concerning the most striking changes, both the quantitative analysis of court decisions and in-depth interviews find that today many judges more often award custody to mothers than fathers, a phenomenon that is in sharp contrast to the situations before 1996. As previously noted, this phenomenon can be ascribed to several factors. One is that many judges associate the "best interests of the child" with the traditional perception and stereotype of women's role as the inherently better caretaker. Another is that most judges observe the social reality that in most cases mothers are the primary caretakers and have better relationships with their children, a social reality that also leads to most children wishing to live with their mothers. With respect to whether fathers or mothers more often receive custody, in appearance it is the new law that has changed the judicial practice in Taiwan. However, we should note that some traditional ideas and beliefs and the social reality affected by them still play an important I8 Thousands of Men Can't Find Brides ifNot "Developing Out, "CHINA TIMES ON-LINE (Taiwan), at http://tw.news.vahoo.com/2001/12/13/society/ctnews/2907121 .html (last visited Dec. 13, 2001). [17:2 2004] CUSTODYDECISIONS IN SOCIAL AND CULTURAL CONTEXTS 289 role in the process. The new law itself does not cause all of the change; neither does social change or female or young judges.'82 In fact, because the "best interests of the child" is vague and uncertain, it provides a stage for the traditional ideas and social customs such as women's role as the inherently better caretaker. This stage gives these ideas and customs an opportunity to show their influences through the behavior and beliefs of the legal actors-the judges, the litigants, the attorneys, or even the social workers. After the new law was enacted, it is not surprising to find things have changed, because we did expect some differences. What may be more interesting is why some things changed so little or even never changed, though the new law clearly intended to change them. We have seen that the "all-or-none custody" decision pattern still persists; judges almost never awarded joint custody, seldom granted child support, and only occasionally determined visitation arrangements. In divorce cases, many judges still focused only on whether to grant divorce and paid little attention to the other issues such as custody and child support. Meanwhile, though the new law explicitly expects judges to exercise their discretion to pursue the best interests of the child and make proper decisions without being confined to the parties' allegations, most judges still followed the adversary procedure and hesitated to intervene in family affairs, which in their minds include the child's custody. Admittedly, family privacy should be respected, but considering that many litigants lack necessary legal knowledge and that the child has no advocate in the proceedings, judges should at least "remind" the parents of some important issues and investigate the facts relevant to the child's interests. However, this research shows that many judges never did these things; in fact, some of them did not know they had the power, while some others intentionally avoided exercising it because they did not want to "make trouble for themselves." This portion of the study argues that the new law does not make all the changes expected because the law is being carried out by legal actors who have old skills, old thoughts, and old habits. Since this research deals primarily with court decisions and judges' decision-making processes, the following paragraphs will focus on the role of judges. To begin with, as discussed earlier, Taiwanese judiciary and legal academia have a long rooted tradition of legal positivism, a tradition of making decisions and discussions based only on statutes. 182 See Liu, supra note 3, at 54-62. COLUMBIA JOURNAL OF ASIANLAW The legal education mainly focuses on logic, abstract concepts, and law codes. The interview results clearly show that this tradition and training style result in many judges clinging to law codes; not only do these judges dislike exercising discretion when there is no clear criterion, but they may also be incapable of determining the complicated facts related to the child's interests and the litigants' actual lives. More seriously, they do not have enough time and motivation because of heavy caseloads, and they do not have sufficient resources such as reliable social workers to assist them. Not surprisingly, many judges prefer staying in the old and familiar adversary system and rely only on law codes to "keep everything simple." In Taiwan, more and more commentators notice that many new laws transplant some new ideas or establish some new systems without any prior investigation and preparation of how to actually implement them. Many new laws were deemed nearly impossible to implement after they were enacted because, except for the "law in books," everything is still the same-there are no new resources, no new personnel, no new training, but just the same old system and same people with old ideas and old skills. 183 Apparently, the new child custody laws discussed here are confronted with the same problem. The new laws expect judges to determine the facts relevant to the child's interests, but judges are not well equipped to do so. Note that, as many judges pointed out in the interviews, to determine custody, judges need not only to consider past andpresent facts, but also to predict what may happen to the child in the future; meanwhile, facts relevant to the child's interests and the litigants' family lives are extremely difficult to prove. From a judge's point of view, the facts in custody cases are much more complicated and difficult to determine than the facts in ordinary civil cases. We have seen that some judges really wanted to exercise the power and discretion to make better decisions, but they simply could not because they lacked the necessary background knowledge and training; when they turned to social workers for professional assistance, they found the workers often were also untrained and unreliable. Little wonder that these judges described their feelings as "frustrated," "helpless," or even "scared." It is noteworthy that both the analysis of court decisions and the interview results seem to indicate that judges in specialized family 183 See Shao-Liang Liu, Transplantation of Law and Social Change, TAIPEI B. J., Apr. 1994, at 7-10; Hui-Lin Shee, The Meanings and Limitations of the Law of Legal Protection Against Sexual Exploitation of Children, TAIPEI B. J., Mar. 1998, at 38, 46. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 291 divisions tended to be more familiar with the contents and purposes of the new laws, and they tended to appreciate the special features and proceedings of family cases. By contrast, many judges sitting in ordinary civil courts tended to treat custody cases in the same way they treated property cases; as noted earlier, some of them were almost ignorant of the contents and purposes of the new laws. The discrepancies between the judges in specialized family divisions and the judges in ordinary civil courts could be ascribed to several reasons. To begin with, many of these new laws had not been enacted when the judges were still in law school or in training; given that there are so many new statutes or amendments every year, it is virtually impossible to ask the judges to be familiar with every single piece of legislation. Considering that family cases constitute only about 3 % of all cases in ordinary civil courts, 1" little wonder that judges in these courts pay far less attention to laws relating to family cases. Also not surprisingly, these judges may be so accustomed to the adversary procedure they follow in handling 97% of their cases, that they automatically (and unconsciously) treat family cases in the same way they treat other civil cases. In contrast, the judges in family divisions need to apply only the laws relating to family cases, so naturally they would be much more familiar with the contents and purposes of the new custody laws. They would notice the different features and proceedings of family cases because that is why they are in a "specialized" division. However, as we have seen, even some judges in specialized family divisions continued to adopt a few traditional attitudes in the old law to handle custody cases. They might still consider custody as a "right" belonging to parents; they seldom awarded child support and might still assume the "no petition, no decision" attitude to handle child support and other issues. Although the heavy caseloads and limited resources mentioned above could partly explain this phenomenon, this study contends there may be a more fundamental theory that could explain why both the judges in specialized divisions and the judges in ordinary civil courts may still tend to adopt some old attitudes when they apply the new laws. The "schema theory" in cognitive psychology may provide an explanation. This theory intends to reveal the cognitive processes of how individuals take in, understand, remember, and apply new 184 In 1998, Taiwan's trial courts received 1,368,616 civil cases in total, but only 45,314 of them were family cases. Family cases constituted only 3.3 1% of all civil cases. See JUDICIAL YUAN, TAIWAN ROC, 1998 ANNUAL REPORT OF JUDICIAL AFFAIRS 163, 168 (1999). COLUMBIA JOURNAL OF ASIANLAW information from the outside world. According to this theory, "schemas" (or "schemata") are mental structures or frameworks based on an individual's previous experiences and knowledge; new information is perceived, comprehended, interpreted, and remembered through one's schemas. This theory contends that learning and memory are not a "storage-retrieval" or reproductive process, but a reconstructive process because our existing schemas would affect our perception, comprehension, interpretation, and memory of the new information. If the new information is inconsistent with or unfamiliar to the schemas, very often it may be omitted or changed to fit into the existing schemas. 8 The concept of "schemas" was first introduced by Fredric Bartlett in 1932. In his classic study of reconstructive memory, he used stories from a Native American folktale that was unfamiliar to his English participants. Bartlett read the exotic stories aloud to these participants and later asked them to recall the stories. Intriguingly, he found the information in their recall was changed in many ways from the original stories. The participants tended to (unconsciously) omit the material unfamiliar to them and the material inconsistent with their existing understandings. He observed, "any element of imported culture which finds very little background in the culture to which it comes must fail to be assimilated." ' 6 Meanwhile, when the input information was not totally but relatively unfamiliar to the participants, they tended to (unconsciously) change it into the relatively familiar. Bartlett gave a very interesting example in his book: In the stories "peanut" was mentioned, but peanut was an uncommon type of nut in England when the study was conducted; many participants, then, recalled it as "acorn," which was the most common form of nut in England, even though "acorn" had never been mentioned in the stories. 87 In addition to the omission and transformation, Bartlett also found that the participants might use their existing experiences and knowledge to "fill in" gaps in the plots of the stories; that is, they might use what they had already known to add information to rationalize the plots. For example, Bartlett's original stories included the phrase "That Indian has been hit," but some participants recalled it as "an Indian is 185 ROBERT L. SOLSO, COGNITIVE PSYCHOLOGY 180, 212, 247-74 (6th ed. 2000). 186 FREDRIC C. BARTLETT, REMEMBERING 125 (1932), quoted in William F. Brewer, Bartlett's Concept of the Schema and Its Impact on Theories of Knowledge Representation in Contemporary Cognitive Psychology, available at http://www-instruct.nmu.edu/osychology/hwhitake/content/ bartlettschema.htm (last visited Mar. 3, 2002). "8 See Brewer, supra note 186. [17:2 2004] CUSTODYDECISIONS IN SOCIAL AND CULTURAL CONTEXTS 293 killed" or "he had been wounded by an arrow" even though whether he died or was wounded by an arrow had never been mentioned in the original stories.'88 Applying this theory in our research, we may find that old schemas affected how many judges perceived, comprehended, interpreted, and applied the new laws. As previously noted, the concept that "custody is a right belonging to parents" and the tradition of "all-or-none custody" have been rooted in Taiwanese culture and society for hundreds of years; in fact, they had been part of the Civil Code and judicial practices since the Republic of China was established nine decades ago too. Similarly, the attitude "no petition, no decision" for any civil suits and the tradition of "not interfering with family affairs" were deeply ingrained in many judges' minds because they had been taught and trained to adopt such attitudes and apply such traditions. The interview results indicate that almost all of the new provisions judges tended to omit or seldom apply are the provisions inconsistent with these old concepts or traditions. For instance, many judges seldom or almost never applied the provisions of visitation and child support from the non-custodial parent, provisions that contradict the "all-or-none custody" tradition which assumes that the non-custodial parent would not have any rights and obligations to the child after divorce. Many judges seldom or never exercised the discretion allowed by the new law to decide custody issues on their own motions; on the contrary, they stuck to the adversary procedure of "no petition, no decision." In fact, we have seen that one judge insisted that she knew the new law and was certain it did not authorize judges to determine custody on their own motions because custody was a family matter with which the law would not possibly interfere. When the researcher finally showed her the provisions in a book of law codes, she was extremely surprised.'89 Obviously, she unconsciously used her previous understandings to change the unfamiliar/inconsistent into the familiar/consistent, a process that is very similar to Bartlett's research participants who changed "peanut" into "acorn" in their recall. How judges used the term "custody right" is another example of how they unconsciously applied their schemas to "omit the unfamiliar/inconsistent" or to "change the unfamiliar/inconsistent into the familiar/consistent." As noted earlier, even though the new law 188 Id. 189 See supra text accompanying note 3 1. COLUMBIA JOURNAL OFASIANLAW specifically changes the old term of custody and refers to custody as "parental rights and obligations to the child"; still, in the interviews, almost all judges automatically changed it into the familiar and used the old term "custody right" while they omitted the obligation aspect of custody. Seemingly, the law had changed, but these judges' schemas had not. According to the research of many cognitive psychologists, an individual's schemas can be changed when he or she realizes that the new information is in conflict with existing schemas and there is a need to accommodate the schemas to fit the new data. Such cognitive conflict or "disequilibrium," as Jean Piaget named it, should be manifested so the new information cannot be simply assimilated into the old schemas. '1 Some psychologists suggest that in order to change individuals' schemas, we sometimes need to provide an interactive environment to "provoke" conflicts so individuals can clearly realize the differences and conflicts between their prior knowledge and the new information. Otherwise, if we simply provide new information or new knowledge to the individuals and expect them to automatically change their old conceptions and old beliefs, quite often we would be very disappointed.9 ' In Taiwan, transplanting new ideas or new institutions from other countries into the legal system is common. 19 Unfortunately, everyone simply assumes that after the new laws are promulgated, the legal actors-especially judges-will know them and apply them because the old laws have been replaced. This research finds that the reality is far different from expected. In the first place, some judges never become familiar with the new laws even if they have been in existence for many years. In the second place, though many judges know or think they know the new laws, they comprehend them incorrectly because they omit some provisions or transform the meanings of these provisions to fit the new laws into their old schemas. The result is that the new ideas or new institutions are not really implemented; instead, through schemas and behavior of some judges and other legal actors, old ideas and old beliefs sometimes use the new laws as a stage or an agent to show their influence. 190 Jeremy Roschelle, Learning in Interactive Environments: Prior Knowledge and New Experience, in PUBLIC INSTITUTIONS FOR PERSONAL LEARNING 37, 38-42 (John H. Falk & Lynn D. Dierking eds., 1995). '9' Id. at 44-51. 192 Liu, supra note 183, at 7-9. [17:2 2004] CUSTODY DECISIONS 1N SOCIAL AND CULTURAL CONTEXTS 295 This finding suggests that, in addition to solving the problem of heavy caseloads and insufficient resources, we also need to educate judges about the special features and purposes of the new laws. It may be just a fantasy to simply assume that all judges would suddenly and automatically know and comprehend the new laws, which are quite distinct from the long-rooted traditions and their old legal knowledge. Judges should be encouraged to discover and articulate their old concepts and beliefs and further discuss the differences between the new laws and the old laws. Through discovery and realization of cognitive conflicts, they may change their old schemas and correctly comprehend the contents and purposes of the new laws. From the perspective of "law in action," we should remember that when the new laws meet judges with old skills, old resources, and old schemas, the new laws can make only limited changes. Judges may still resort to their old beliefs, old knowledge, and old habits to explain and apply the new laws because they are accustomed to the old laws or because that is all they have. B. Adverse Effects of Current Judicial Practices 1. Problems of Using Economic Criterion and the "All-or-None Custody" Decision Pattern As we have seen, judges usually emphasize the economic resources of the parents and adopt the "all-or-none custody" decision pattern. These judicial practices are problematic for many reasons. To begin, using the economic criterion may harm the child's best interests, for it tends to disregard the importance of the psychological and emotional needs of the child. By using economic resources as a necessary factor, custody may be awarded to an unfit or even dangerous father only because the mother does not have enough income, even though she may have better parenting capabilities.'93 Second, the "all-or-none custody" decision may harm the child's psychological development because it may cause the child to lose all contact with the non-custodial parent." Moreover, it may make 193 See Liu, supra note 2, at 204-05. 194 Researchers have found that, after their parents' divorce, children may be able to adjust better if they have continued contact with the non-custodial parents and them. See James H. Bray, Psychological Factors Affecting Custodial and Visitation Arrangements, 9 BEHAV. SCI. & LAW 419, at 432 (1991); JUDITH S. WALLERSTEIN ET AL., THE UNEXPECTED LEGACY OF DIVORCE 294-312 COLUMBIA JOURNAL OF ASIAN LAW custody disputes more bitter and fierce because the litigants believe that losing the case means losing the child forever in all respects. "I Ironically, while so many judges complain about their heavy caseloads, 196 many of them may not notice that their "all-or-none custody" decision pattern may in fact prolong and intensify the custody disputes in their courtrooms, a result that may make their caseloads heavier. Third, using parents' economic resources and occupation as a necessary factor in determining custody is unfair to mothers, not only because of the structured gender wage gap in society, 97 but also because of the fact that many women have sacrificed their career opportunities to raise their children and take care of their families. The economic criterion improperly ignores married women's contributions to their families other than wage earning. According to official surveys, in 1998, while the labor force participation rate of married men (spouse present) and cohabiting men was 80.9 %, the rate of married women (spouse present) and cohabiting women was only 46.5 %.18 More than 76.8 % of unemployed women were not employed because they needed to "take care of family" or "look after household affairs" (46.9 % and 29.95 % respectively). "I To be more specific, 38.7 % of married women (spouse present) did not go to work because they needed to take care of their children. :°° The unsound legislation and judicial practice also have very adverse effects outside courtrooms. In Taiwan, a large percentage of divorce and relevant custody disputes have never entered any court, because consensual divorce only has to be registered at administrative agencies; in this case, parents can decide their own custody arrangements by an agreement. 2°' However, just as Mnookin and Kornhauser suggest, "the legal rules. . give each parent certain claims based on what each would get if the case went to trial," and "the (2000); ELEANOR E. MACOBY & ROBERT H. MNOOKIN, DIVIDING THE CHILD: SOCIAL AND LEGAL DILEMMAS OF CUSTODY 287-88 (1992). 195 See supra note 163 and accompanying text. 196 For example, see supra notes 28, 30, 70, 72 and accompanying text. 197 In 1998, the average monthly income of female employees was NT 27,401 dollars, while male employees' average monthly income was NT 37,596 dollars. See EXECUTIVE YUAN, TAIWAN ROC, REPORT ON THE MANPOWER UTILIZATION SURVEY OF TAIWAN AREA 128-29 (1998). '9' Id. at 2-3. 199 MINISTRY OF INTERIOR, TAIWAN ROC, SURVEY OF WOMEN'S LIFE STATUS IN TAIWAN AREA 86 (1998). 200 MINISTRY OF INTERIOR, TAIWAN ROC, SURVEY OF WOMEN'S LIFE STATUS IN TAIWAN AREA 188 (1993). 201 CIVIL CODE art. 1050 (Taiwan ROC). [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 297 outcome that the law will impose if no agreement is reached gives each parent certain bargaining chips. ' 202 Many studies attest that the phenomenon of "bargaining in the shadow of the law" is also apparent in Taiwan. For example, several studies prior to 1996 indicate that the old law obviously put wives at a great disadvantage in seeking divorce and child custody outside courts, as well as inside courts. 2°3 The decision pattern of "all-or-none custody," one that uses finances as a necessary factor and does not award child support, disproportionately reduces the mother's bargaining power. Because the mother usually has fewer economic resources than does the father, it is highly predictable that the mother would avoid entering a court and make concessions in property division and alimony to receive custody.2" The mothers who care about and love their children the most would be most willing to accept an inferior bargain; i.e., they will be "punished" more. While divorce typically leads to a decline of economic status,0 5 this decision pattern further reduces the mother's opportunity to receive fair property division and alimony, not to mention that it has blocked the mother and the child's possibility of receiving child support from the father in the first place. The welfare of both the child and the mother in a divorced single-mother family may be severely damaged because of the concomitant financial difficulties. Moreover, because they are afraid of losing their children, many mothers who do not have enough economic resources may give up the idea of seeking divorce in the first place, even if they are victims of spouse abuse or their marriages have become intolerable. The mothers to whom their children matter most might be punished by losing an 202 Robert Mnookin & Lewis Kornhauser, Bargaining in the Shadow of the Law: The Case of Divorce, 88 YALE L.J. 950, 968 (1979). 203 See, e.g., HuI-HS1N CHEN, STUDIES OF ISSUES IN FAMILY LAW 271-72 (1993); Tsung-Lo Huang, A Comparative Study of Consensual Divorce, NAT'L TAIWAN U. L.J., Dec. 1980, at 183; Yi-Wei Huang, The Procedure ofDivorce and Child Custody Decisions, CHAO YANG L. REV., Aug. 1991, at 12, 20. 204 Hui-Hsin Chen, Pursuing the Gender Equality from the Perspective ofLaw, 49 NAT'L CHENGCHI U. L. REV. 85, 94-95 (1993). A similar phenomenon can also be seen in the U.S. See MARY A. MASON, FROM FATHER'S PROPERTY TO CHILDREN'S RIGHTS 129 (1994); Garska v. McCoy, 278 S.E.2d 357 (1981). 205 In particular, the economic status of single mothers with children drops drastically, partly because of the gender wage gap. According to an official survey made by the Taiwan Province Government in 1995, the average household income of mother-headed single-parent families was only equal to 32.9 % of the average income of all families in Taiwan Province. In fact, divorced women's economic status is further exacerbated not only by the law, which sets a very strict rule for seeking alimony, but also by the judicial practice in which judges seldom award child support. Note that most divorced women in Taiwan never obtain any alimony or child support. See TAIWAN PROVINCE GOVERNMENT, SURVEY OF THE NEEDS OF TAIWANESE SINGLE-PARENT FAMILIES 7, 32 (1996); Liu, supra note 3, at 180-81. COLUMBIA JOURNAL OFASIANLAW opportunity to improve their well-being or to merely protect themselves. Many of Taiwan's researchers and lawyers have attested to this phenomenon.2 Some may argue that the explanations in this study are somewhat self-contradictory, because while this study has proved that mothers usually have fewer economic resources and that judges tend to use finances as a determinative factor in deciding custody, it does not explain why judges award custody to mothers much more often than fathers. However, this specious critique misreads this study's findings and explanations. On the one hand, the findings do not indicate, and this study has never argued, that judges tended to use finances as a determinative factor; but the findings do show that judges tended to use finances as a necessary factor. 2°7 Namely, parents who had more economic resources than their ex-spouses did not necessarily always receive custody, but, it is clear that judges tended not to award custody to the parent who did not have enough financial resources. The interview results clearly indicate that, in many judges' consideration, having enough economic resources was not sufficient, but it was necessary. On the other hand, the result of a "double burden" on women can rebut the seemingly plausible critique. The "double burden" means that women need to work outside their homes to ensure their financial ability and to play the role of inherent caretakers of their children at the same time.0" From many judges' point of view, this double burden may be the best arrangement, because it accords with both their belief in women's inherent qualities of being a better caretaker and their emphasis on the importance of economic resources in deciding custody. Not only so, but the whole legal system and governmental policy tend to 20 Huang, supra note 203, at 20; Yu-Bao Lin, It's Easy to Get Married But Difficult to Get Divorced, in PROCEEDINGS OF THE 1998 CONFERENCE OF WOMEN'S POLICY ISSUES (visited Mar. 12, 2000), available at http://taiwan.vam.org.tw/nwc/nwc3/papers/forum445.htm. In fact, a similar phenomenon has also been found in the U.S.; see MASON, supra note 204, at 17. See also Wenmay Rei, In the Name of the "Best Interests of the Child"-A Study of Child Custody in Divorce Decisions, NAT'L TAIWAN U. L.J., Apr. 1999, at 245, 268. 208 Because gender stereotyping is ingrained, women's labor force participation and concomitant economic independence do not necessarily enable them to be liberated from the duty of caring for children; instead, many of today's Taiwanese women have to perform their duties both inside and outside their homes. See Lang-Wen Huang, The Study of Ideology on the Division of Housework for the Married Men and Women, SOOCHOW J. SOC. WORK, Mar. 1998, at 81, 98; Liu, supra note 2, at 209-10; Liu, supra note 3, at 8-10. [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 299 reinforce this double burden on women, due to traditional ideas and resorting to the private welfare system based on a family unit. 29 However, this double burden on mothers obviously limits women's choices and possibilities in their lives and careers. Just as Margaret Mead comments, the emphasis on the concept of the mother-child relationship is "a mere and subtle form of anti-feminism in which men-under the guise of exalting the importance of maternity-are tying women more tightly to their children than has been thought necessary since the invention of bottle feeding and baby carriages."2 ' This double burden also marginalizes women in the labor market and thus limits their abilities to improve economic status, because they usually need to be involved in informal employment to take care of their children.1 Many empirical studies consistently attest that the consideration of caring for children significantly affects women's choice ofjob; women tend to work informally when they have young children.212 Women's double burden is even worsened by the lack of public childcare centers and the high cost of private daycare programs. Usually, women need to rely on their relatives (in most cases it is the child's grandparents) to care for their children while they are at work. As we have seen, judges' decisions also considered and reflected this social reality.213 2. Problems of Judges' Over-Focus on Divorce Itself While Overlooking the Importance of Custody Issues Both the analysis of court decisions and the interview results clearly indicate that many judges over-focus on whether to grant divorce but overlook the custody issues, including visitation and child 209 See Liu, supra note 2, at 209. In fact, before the late 1990s, Taiwan's legal system and governmental policy intentionally reinforced the perception of women's inherent role of being a mother and nurturing children. Legislation and policy continually connected women with child welfare; this policy has even been directly stipulated in Constitution. See generally Li-Ju Lee, Law and Social Norms in a Changing Society: A Case Study of Taiwanese Family Law, 8 S. CAL. REV. L. & WOMEN'S STUD. 413, 422-39 (1999); Liu, supra note 3, at 186-87. 210 Margaret Mead, Some Theoretical Considerations of the Problems of Mother-Child Separation, 24 Am. J. ORTHOPSYCHIATRY 24 (1954). 211 Yu-Hsia Lu, Economic Development and Married Women's Employment in Taiwan: A Study of Female Marginalization, J. POPULATION STUD., July 1994, at 107, 121.2 12 Id.; Wen-Yin Chien & Chemg-Tay Hsueh, The Employment of Married Women in Taiwan: Its Patterns and Causes, J. POPULATION STUD., Apr. 1996, at 113, 127; Yu-Hsia Lu, Married Women's Informal Employment in Taiwan, 2 PROC. NAT'L SCI. COUNCIL, ROC 202, 209-10 (1992). 211See supra text accompanying notes 151-154; Liu, supra note 3, at 38-40. COLUMBIA JOURNAL OF ASIAN LA W support, involved in a divorce case.1 4 Many judges, especially some judges in ordinary civil courts, believe that the litigants can "discuss the custody issues themselves after the divorce" so the court does not need to intervene; in fact, some judges believe that they should not intervene because they still insist on an adversary procedure and the idea of family autonomy.25 These judicial practices are extremely problematic for the following reasons. First, as pointed out by some judges, many litigants in family cases clearly lack legal knowledge. On one hand, many of the litigants do not even know they can settle custody issues in court along with the divorce; on the other hand, some of them mistakenly think that, since the child has already been living with them, and their spouses do not mention custody issues in court, they do not need to petition for custody and will receive it as a matter of course.2 6 In these cases, it is highly doubtful that the litigants will actually "discuss the custody issues themselves after divorce" because they may not know they still have to. As a result, many things may remain uncertain as both the court and the parents may not really settle the custody issues. Second, precisely because the custody issues may remain uncertain after a court grants the divorce, disputes may arise sometime after the divorce."' Not only may the disputes surprise the parents who assumed that they have received custody, but the disputes may enter courts again and become part of judges' new caseloads. In fact, compared with settling these issues along with the divorce, settling these disputes after the divorce as new lawsuits may add more burdens on the court system. It may add more burdens because the judges hearing the new lawsuits may not be the same judges hearing the divorce cases, and they may need to investigate the similar facts related to the litigants' family lives again. Even if the new lawsuits will be heard by the same judges hearing their divorce cases, a coincidence unlikely to happen under the current system,2 ' the new lawsuits will inevitably bring new proceedings occupying more ofjudges' time while the same litigants must go to courts again. So ironically, while some judges refuse to determine custody on their own motions and do not want to "remind" the litigants of the custody issues because they try not 14 See supra text accompanying notes 28-45, 122-123; Liu, supra note 2, at 203, 206. 215 id. 2:6 See supra text accompanying notes 40-41. 217 In fact, in the interviews one judge did mention some actual cases of this kind. See supra note 41. 218 See supra text accompanying notes 172-173. [17:2 20041 CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS 301 to "make trouble for themselves, ' 2'9 what they do may, in fact, create more trouble for the whole court system and increase the caseloads of their colleagues and themselves. Third, even if the parents know they could settle custody issues on their own, when judges overlook the custody issues in divorce cases, they are overlooking the child's best interests. Note that the new law authorizes judges to investigate facts and determine custody on their own motions for the child's well-being. 20 That is, judges are expected to be "gatekeepers" who protect the child's interests while there is no advocate for the child in divorce proceedings. As discussed earlier, according to the "best interests of the child" standard and the new law, custody is not a private right that can be arbitrarily handled or determined by the parents. Admittedly, family privacy should be respected, but judges should at least try to remind the parents of custody issues and review the parents' agreement on custody to ensure that it does not harm the child's well-being. When judges overlook the custody issues and thoughtlessly allow the parents to make or not make arrangements on their own, clearly they treat custody as a private right belonging to the parents, an attitude that is obviously inconsistent with the purpose of the new law. 22' Fourth, these judicial practices would be disadvantageous to the litigant who is more vulnerable and has fewer resources in the marriage relation. We have seen that in Taiwan it is usually the wife who is more vulnerable because she has less or even no income. The traditional ideas and customs even encourage wives to depend financially on their husbands. Not surprisingly, in most cases, divorce is more devastating for the wives than for the husbands, so compared with the husbands, the wives often have fewer bargaining chips. As noted by several judges in the interviews, when a woman's husband wants to leave her, the diction to agree to a divorce may be her only leverage in bargaining for child custody, child support, property division, and alimony.122 If judges thoughtlessly grant only the divorce and assume that the litigants can settle these issues themselves after the divorce, the court has in fact deprived some wives of their only bargaining chip--whether to agree to divorce. When the husbands attempt to "get rid of' the wives as soon as 219 See supra text accompanying notes 28-30. 220 See MINISTRY OF JUSTICE, TAIWAN ROC, THE DRAFTING PROCESS OF THE 1996 AMENDMENTS TO THE FAMILY BOOK OF THE CIVIL CODE 252 (1996). 22 See supra text accompanying notes 77-80. 222 See supra text accompanying notes 55-57. COLUMBIA JOURNAL OF ASIAN LA W possible, this judicial practice may give the husbands a shortcut to getting out, while at the same time leaving the wives helpless and powerless. To sum up, the current judicial practices and attitudes over-focus on whether to grant divorce but improperly overlook the custody and property issues derived from the divorce. These judicial practices and attitudes not only may harm the child's well-being, but they may harm women's interests both during and after the divorce process. Some judges' attitudes clearly reflect that they just want to process more cases more quickly, but they do not think in a problem-solving way and disregard the litigants' actual circumstances. Ironically, exactly because they do not think and make decisions in a problem-solving way, their course of action may in fact create heavier caseloads for the court system. Problems other than the dissolution of marriage itself may be unresolved, caseloads may be increased, and the litigants and the children may be harmed. While in appearance the judges make decisions within the allowance of the "law in books," their attitudes and decision patterns in fact make the "law in action" operate in the opposite direction of what the new law expects. VII. CONCLUSION Cultural and social circumstances may significantly influence judges' explanations and applications of the "best interests of the child" standard. The findings of this research confirm that Taiwan's court decisions of child custody cases and the judicial practices actually reflect many cultural ideas, such as stereotyped gender roles, family autonomy, and a sense of "face." Meanwhile, the social and economic background of Taiwan, such as the social custom of "all-or-none custody" and the different social conditions and resources in different districts, also affect judges' custody decisions. After the "best interests of the child" standard substituted the presumption of paternal custody in 1996, judges predictably awarded custody to mothers much more often. In fact, the change has been so dramatic that custody is now overwhelmingly awarded to mothers, a development in stark contrast with the preference for paternal custody before 1996. This study argues that judges usually award custody to mothers not only because of the gender-neutral standard, but also because judges combine traditional ideas and social customs in determining the best interests of the child. On one hand, judges tend to [17:2 2004] CUSTODYDECISIONS IN SOCIAL AND CULTURAL CONTEXTS 303 adopt the stereotyped gender roles and assume that women are inherently better caretakers of children. On the other, many judges cling to the tradition of considering economic competence as a necessary factor in deciding custody, a tradition that has been adopted since the era of the old law. In the meantime, this study finds that judges rarely award child support to the custodial parents, almost never grant joint custody, and seldom give visitation orders. Even though there has been a new law providing that each parent's obligations to support the child will not be changed after divorce, and even though there has been another new law introducing the possibility of joint custody and visitation arrangements, still this "all-or-none custody" decision pattern has not been changed. The predominant preference for mothers in custody decisions may not really be a victory for women. Judges' preference for the gender-stereotyped role of women, along with the emphasis on economic competence, imposes a double burden on women. Divorced single mothers today are expected to assume the role of an inherent caretaker of children and to work outside their homes to support the family single-handedly since judges rarely order child support and alimony from non-custodial fathers. From many judges' points of view, this double burden on single mothers may be the best arrangement because it accords with both their belief in women's inherent qualities of being a better caretaker and their emphasis on the importance of economic resources-both comprise a main part of their ideas of what the best interests of the child should be. However, these judicial practices (i.e., the "all-or-none custody" decision pattern and the preference for a double burden on single mothers) have very adverse effects on the child's well-being and gender equality. These practices cause the child to lose all contact with the non-custodial parent and thus may harm the child's psychological development. Secondly, appearing to emphasize the best interests of the child, these practices tie women more tightly to their stereotyped social roles and thus limit their career possibilities. Lastly, while divorce typically leads to a decline of economic status, given that there is a gender wage gap and that Taiwan still has few public welfare programs to support single-parent families, these practices further worsen postdivorce single-mother families' economic status. Since the divorce rate is increasing and judges today overwhelmingly award custody to mothers, the current judicial practices have been in fact creating more economically insecure single-mother families because the court still COLUMBIA JOURNAL OF ASIAN LAW tends to follow the tradition of not awarding child support to single custodial parents.223 Furthermore, this study finds that many judges-especially those sitting in ordinary civil courts-still follow the adversary procedure to handle family cases as they handle property cases. These judges usually focus only on whether to grant divorce but overlook the importance of custody and property issues derived from the divorce. They are ignorant of the facts that they may put women at a severe disadvantage in bargaining outside the courtrooms. They also neglect their role as the "gatekeeper" who protects the child's interests when there is no advocate for the child in divorce proceedings, a role that is provided by the new law. In general, this study finds that most judges still tend to use their old beliefs, old knowledge, and old skills toexplain and apply the new law. One might expect that, in a civil law country like Taiwan, judges' states of mind would not play an important role because they are expected to apply the legal codes "mechanistically"--that is, they are expected to be a "technician" who finds the appropriate legal code and then adapts the facts of the pending case to it.224 However, the results of this study show that sometimes this expectation may be just a myth. In order to ensure divorced single-mother families' well-being, this study suggests that Taiwan should develop both a public welfare system and an institution of child support from non-custodial parents to assist them. To pursue both the child's best interests and gender equality, judges should stop using economic competence as a necessary factor in determining custody. If the more suitable parent is economically less competent, we should use both public welfare programs and private child support from the non-custodial parent to assist her or him. Moreover, judges should determine visitation arrangements more often and pay attention to the child's psychological and emotional needs in addition to material needs. While taking note of the importance of the public welfare system, Taiwan's policymakers and researchers should not disregard the issue of the private child support institution and the problems of current judicial practices. 223 Liu, supra note 3, at 179-86. Note that the divorce rate in Taiwan has been increasing significantly during the past few decades. From 1984 to 2000, the divorce rate in Taiwan grew from 1.0 to 2.4 divorces per 1,000 population. Although this number is not as high as the ones in the U.S. and some European countries, it was the highest in Asia. See EXECUTIVE YUAN, TAIWAN ROC, SOCIAL INDICATORS OF TAIWAN AREA 7 (2000). 224 For the "mechanistic approach" in civil law countries, see ARTHUR T. VON MEHREN & JAMES R. GORDLEY, THE CIVIL LAW SYSTEM 1138-42 (2d ed. 1977). [17:2 2004] CUSTODY DECISIONS IN SOCIAL AND CULTURAL CONTEXTS. 305 The results of this research also indicate that when new laws meet legal actors with old resources, old skills, and old schemas, the new law existing in books can make only limited changes. In order to reform the judicial practices, this research proposes that we should establish a comprehensive family court system and train some judges to be specialized in this field. Also, we should pour more resources into the system and train specialized workers to conduct custody evaluations. Furthermore, we should try to change legal actors' old schemas by evoking discussions on the problems caused by the outdated traditions and judicial opinions. Only after attending to the "law in action" can we possibly achieve the goals of the new law-to pursue both the best interests of the child and gender equality.