Not long after immersing myself in this project, I began to visualize Sisyphus going to divorce court. His fate is an apt metaphor for the protracted and sometimes futile uphill struggle of China’s mostly female divorce plaintiffs, whose petitions will almost certainly fail at first – even in cases involving domestic violence, regardless of the severity of the allegations or the strength of the evidence.Footnote ¹ Many plaintiffs give up on litigation, either resigning themselves to staying married to their abusers or pursuing divorce through civil government channels outside the court system. Of those who do return to court, most will eventually succeed, albeit sometimes only after multiple attempts and long delays.

My key tasks in this book are to trace the origins and chronicle the consequences of this highly institutionalized practice of denying first-time petitions (Reference HeHe 2009), which I call the “divorce twofer” because a court typically grants a divorce only after trying the same case twice. Obtaining a divorce after two (or more) attempts is no bargain for litigants, but, as we shall see later, denying divorce petitions has helped judges in a variety of ways. Courts and judges have enjoyed divorce litigation’s “two for the price of one” quality, for which female plaintiffs have paid dearly. The divorce twofer’s benefits to courts and judges have come at the expense of gender justice.

As I studied tens of thousands of courts’ written divorce decisions, I was struck both by the high prevalence of domestic violence allegations and by judges’ tendency to ignore them. I was surprised by the ubiquity of judges’ brazen and inscrutable disregard for plaintiffs’ well-documented claims of domestic violence. I was mystified by how commonly judges denied divorce petitions on the grounds that mutual affection had not broken down and reconciliation remained possible despite admissible evidence of horrific spousal abuse. A remarkable feature of Chinese court rulings to deny divorce petitions is the overwhelming extent to which they are based on judges’ arbitrary assessments of the strength of the marital foundation and speculative prognostications about litigants’ reconciliation prospects. For this reason, the divorce twofer extends unabated to cases involving domestic violence.

I was equally amazed to find that judges’ tendency to deny first-attempt divorce petitions had increased dramatically beginning in the mid-2000s. In some ways, the contemporary struggle is a throwback to the Mao era when divorce was notoriously difficult (Reference TsuiTsui 2001:105–06). To my dismay, I discovered that, among all divorce-seekers, women have been hugely disadvantaged not only in their prospects of obtaining a divorce on the first try but also in gaining child custody. Women have borne the brunt of this judicial clampdown on divorce. Their formidable difficulties thus harken even further back to China’s imperial days (Reference BakerBaker 1979:45; Reference Honig and HershatterHonig and Hershatter 1988:206).

These parallels to earlier periods, however, are strictly confined to courts. The Sisyphean character of divorce litigation stands in stark contrast to a relatively quick and simple administrative pathway to uncontested divorce in the Civil Affairs Administration, which accounts for the vast majority of China’s divorces. Indeed, the liberalization in 2003 of this extrajudicial pathway helped triple China’s crude divorce rate within 15 years. A prerequisite of divorcing outside of court, however, is mutual consent – and agreement on all terms. Most divorce cases brought to court, therefore, are contested. A considerable share of them have been filed by women making allegations of domestic violence.

Why have courts become averse, and increasingly so, to granting first-attempt divorce petitions? Why have judges remained so unmoved by domestic violence allegations? Why have women’s divorce litigation outcomes been so much worse than men’s? These are the questions I set out to answer in this book.

In my quest for answers, the first place I looked was China’s family laws. Widely dubbed “breakdownism,” the ultimate legal standard for divorce is the “breakdown of mutual affection.” Strictly according to the law, judges can grant divorce petitions only if their mediation efforts fail to reconcile the couple and they determine that “mutual affection has indeed broken down.” Breakdownism is analogous to no-fault divorce elsewhere in the world insofar as judges can apply it to grant a unilateral divorce petition on the grounds of irreconcilable differences. If mediation fails, a judge need only take a plaintiff’s claim of the breakdown of mutual affection at face value to grant a divorce. Chinese judges almost never apply the law this way, however. More often than not, they take a defendant’s unwillingness to divorce as proof that mutual affection has not broken down.

The law provides additional divorce standards. A judge is supposed to grant a plaintiff’s divorce petition if the defendant fails to show up and mediation cannot be carried out, or if the litigants satisfy a physical separation test. Most importantly, statutory wrongdoing – including domestic violence – automatically establishes the breakdown of mutual affection. Any one of a series of fault-based legal tests known collectively as “faultism” automatically satisfies the breakdownism standard and therefore provides sufficient grounds for an adjudicated divorce. Again, however, judges rarely apply the law this way.

On paper, Chinese family law adheres to global legal norms concerning women’s rights in general and protections against domestic violence in particular. Since the 1980s, the law has fully empowered judges to grant a divorce on the fault-based grounds of domestic violence. Although the term “domestic violence” debuted in Chinese law in 2001, earlier legal provisions extended protections – particularly to women and children – against “maltreatment” and “abuse,” and provided the right to divorce on this basis. Ambiguities in the law, however, have also provided a way out for judges disinclined to grant a divorce. Most judges tend to privilege breakdownism over faultism. Rather than affirming the breakdown of mutual affection on the basis of statutory wrongdoing, judges tend to do the opposite: they sideline plaintiffs’ fault-based claims and rule to preserve abusive marriages by determining that the litigants’ marital discord can be fixed and that mutual affection has therefore not completely broken down.

Denying divorce petitions solves a lot of problems for judges, who face pressures from many sources, perhaps the least of which is the law. Chinese judges are rewarded and punished according to how well they support the court system’s dual imperative to maximize judicial efficiency and minimize social unrest. Like Reference LipskyLipsky’s (2010) “street-level bureaucrats,” Chinese judges take advantage of their considerable discretion to bend and reinterpret formal rules. As street-level bureaucrats, they have developed unofficial “routines and simplifications” not only to complete their relentless work tasks but also to maximize their scores on measures their superiors use to evaluate their work performance (Reference LipskyLipsky 2010). In so doing they have produced informal de facto rules that deviate from official de jure rules.

The divorce twofer emerged as one of judges’ creative coping strategies. By helping overworked judges to close cases quickly and thus to clear their oppressive dockets, the divorce twofer is a docket-shrinking machine. Most of the divorce petitions that judges swiftly deny on the first attempt do not come back to court as second-attempt petitions. Moreover, the cases that do come back are less fraught and contentious – and are thus easier to dispose of and less likely to lead to “extreme incidents” of violence and unrest. Finally, marital preservation supports China’s political ideology of family harmony as a means of maintaining social stability.

Street-level bureaucrats also save time and effort by making snap judgments guided by prevailing stereotypes and biases (Reference LipskyLipsky 2010). Chinese judges sort litigants into cultural categories of credibility and deservingness in part according to patriarchal cultural beliefs. Because they deem women’s claims to be less credible than those of men, judges attach less weight to women’s allegations of domestic violence than to their alleged abusers’ denials. They use batterers’ apparent contrition as evidence of reconciliation potential and thus as grounds for denying victims’ divorce petitions. They support the rural patriarchal order by granting child custody – particularly of sons – to fathers. Judges also fear for their own personal safety lest they upset a defendant with a history of violence. To some degree, judicial decision-making occurs in the shadow of threats of violent retribution.

In recent years, the annual number of contested divorce petitions adjudicated by Chinese courts has exceeded half a million (Ministry of Civil Affairs of China, various years), at least one-quarter of which involve claims of violence and other forms of abuse (Reference Chen and DuanChen and Duan 2012; Reference LiLi 2015b). Such cases, usually filed by women, usually result in a court ruling to preserve the marriage (Ministry of Civil Affairs of China, various years; Reference XuXu 2007). My empirical analyses of the written court decisions of almost 150,000 divorce adjudications spanning eight years in two Chinese provinces, Henan and Zhejiang, show that courts’ long-standing practice of denying divorce requests on the first attempt (Reference HeHe 2009) has intensified since the mid-2000s, and that China’s judicial clampdown on divorce has disproportionately impacted women. They also show that when they do grant divorces, courts favor fathers over mothers with respect to child custody, in part because women who flee domestic violence often leave their children behind. Men who beat their wives are thus rewarded with child custody.

The tragic 18-year saga of He Jie, a woman from Dingxi County in Gansu Province, offers a preview of almost every theme of this book about the struggle of Chinese divorce litigation.

Curtain-Raiser

He Jie’s husband, Zhang Dong, began to beat her soon after they registered their marriage in 1986. His violent temper did not wane following the birth of their son. When their son was six months old, Zhang Dong’s beating left He Jie collapsed on the floor with a ruptured eardrum and urinary and bowel incontinence. When she got up after Zhang Dong demanded that she return to work, he beat her again. He Jie’s screams alerted the neighbors, who reported the situation to her parents, who in turn rushed her to the hospital. In 1987, as a consequence of this episode, He Jie filed her first divorce petition with the Dingxi County People’s Court.Footnote ² While awaiting her trial, she left her son behind and stayed with a relative in the provincial capital of Lanzhou, where she looked for work. Zhang Dong traveled to Lanzhou to express his remorse. He pledged never to repeat his offenses, and if he did, to agree to divorce and provide economic compensation. He also begged He Jie’s parents to persuade her to give him another chance, which they did. In consideration of Zhang Dong’s contrition, the court denied He Jie’s divorce petition on the grounds that mutual affection had not completely broken down.

The very next day after the court’s adjudicated denial, Zhang Dong brutally attacked He Jie. Later, in 1988, he dumped a basin of foot-washing water over her head and, wielding a cleaver, chased her out of their home. Not knowing where else to go, she returned to her parents’ home. That same year, He Jie filed her second divorce petition. She also sought the assistance of the local branch of the All-China Women’s Federation and the local People’s Congress, both of which attempted to persuade her that countless couples experience the same thing, that physical fights are no big deal. Afterward, He Jie declared to Zhang Dong that she would move to Lanzhou and look for work while awaiting the court’s ruling. After the court denied her second petition, Zhang Dong traveled to Lanzhou to retrieve her.

In 1993, the court denied He Jie’s third divorce petition after yet another convincing display of remorse by Zhang Dong. In 1996, Zhang Dong chased He Jie again with a cleaver. This time, as she was trying to escape through the front door, he caught her by grabbing her hair. When he held the knife against her neck and moved it back and forth on her skin, she nearly lost three fingers when she tried to push the blade away. Her fingers remained attached by a small amount of sinew. Although she was bleeding profusely, he prevented her from going to the hospital. Only by pretending to use the bathroom was she able to escape to the hospital, where her fingers were reattached.

The fourth time she filed for divorce, He Jie was more determined than ever to succeed. She reasoned that if she used medical records as evidence of Zhang Dong’s abuse, the court would be unable to use “mutual affection has not broken down” to deny her petition. Zhang Dong wrote a “pledge letter” admitting his mistakes, promising never to repeat them, and begging for one more chance. Under enormous pressure – from Zhang Dong’s work unit, which wrote a formal statement and affixed its official red seal to vouch for his commitment to become a better person; from He Jie’s older brother, who was moved by Zhang Dong’s gestures; and from her precarious employment situation at her own work unit, which had started laying off employees – He Jie relented and withdrew her divorce petition.

A few years later she did indeed get laid off. After Zhang Dong was also let go by his work unit shortly afterward, he regularly got drunk and beat her. In 2002, Zhang Dong was arrested for hiring a prostitute. After He Jie bailed him out of jail, he beat her. After a few more years of abuse, He Jie resumed plans to file her fifth divorce petition. One day, in May 2005, when she returned home to discover Zhang Dong drinking with a friend, she ran to her mother’s home, where she spent the night in order to avoid another beating. Several hours after He Jie returned home the following morning, Zhang Dong notified her mother that she had killed herself by drinking rat poison. He rushed her to the emergency room where she was pronounced dead. He Jie’s family, suspecting that Zhang Dong murdered her, requested a forensic investigation. Because Zhang Dong and his son refused to grant permission to examine He Jie’s stomach contents, the forensic pathologist’s tests were inconclusive. He Jie’s body was cremated.Footnote ³

Owing to failures in the Chinese civil courts, divorce cases do lead to suicides and spawn criminal domestic violence cases, including homicides. In the grand scheme of divorce litigation, however, He Jie’s tragedy is an extreme case in terms of both the number of times she filed for divorce and her ultimate fate. Other themes emerging from her case, however, are hardly aberrations from the utterly common experiences of abuse victims who file for divorce in court:

• In their divorce petitions, plaintiffs often present claims of domestic violence in gory, harrowing detail, and support them with legally admissible documentation.

• These plaintiffs commonly report their fruitless prior help-seeking efforts with the police, local government agencies, and the All-China Women’s Federation.

• Plaintiffs often face pressure from all sides to withdraw their petitions.

• In order to justify their adjudicated denials of abuse victims’ petitions for divorce, judges downplay and normalize domestic violence and underscore batterers’ contrition. In so doing, judges reinforce the gaslighting efforts of husbands, parents, parents-in-law, other family members, police, and village leaders.

• Written court decisions are rife with judges’ contorted efforts to establish mutual affection despite plaintiffs’ claims and prima facie evidence of domestic violence. Judges commonly cite defendants’ desire to stay together and remorse as proof that mutual affection has not broken down. Whereas pledge letters are supposed to be used as evidence of domestic violence, for purposes of establishing the breakdown of mutual affection, judges tend instead to use them as evidence of defendants’ repentance, for purposes of establishing the existence of mutual affection.

• When plaintiffs return to court after an unsuccessful first attempt, they often report the intensification of domestic violence in the interim and their efforts to escape it by staying with family or participating in labor migration.

• In child custody determinations, judges privilege physical possession over domestic violence allegations. The judges in He Jie’s case never had to determine child custody because they never granted any of her divorce petitions. Had they done so, they likely would have granted child custody to the defendant because He Jie, like so many abuse victims, left her son in the physical possession of her husband when she fled to safety.

I encountered other cases similar to He Jie’s. In 2014, Henan Province’s Zhongmu County People’s Court denied the petition of a woman on her fourth attempt. According to the court decision, she and her husband moved in together in 2007. Like many rural couples, they had a traditional wedding ceremony but did not officially register their marriage. Because the husband came from a poor family without the means to support the dominant rural practice of patrilocality, they moved in with her parents. Only in 2009, a year after giving birth to a son, did they retroactively register their marriage. In 2011, their twins – one boy and one girl – were born. According to the plaintiff, the defendant regularly punched and kicked her when things were not to his liking. On one occasion, during a fight, he allegedly cut her parents with a knife when they tried to calm him down. When the plaintiff filed her first petition in 2011, village leaders intervened to persuade her to reconcile. In consideration of their son and given that she was pregnant, she agreed to give him another chance. Later in the same year, after no change whatsoever, the plaintiff filed a second petition, which the court denied. She withdrew her third petition in 2012, when her in-laws persuaded her to reconcile. Her fourth trial, like many divorce trials in China, was held with her husband in absentia. To support her claims, the plaintiff submitted as evidence a police report documenting an unspecified emergency incident. In its decision, the court wrote:

Unless the defendant is AWOL, the first step of the Chinese divorce litigation process is judicial mediation for the purpose of marital reconciliation. In this case, village leaders and family members also intervened in the mediation process. They acted in concert with the court to gaslight the plaintiff by characterizing her claims of marital violence as “trifles.” Their efforts to persuade her to give her abusive husband another chance for the sake of the children and family unity succeeded when the plaintiff withdrew her first and third petitions. The court denied by adjudication her second and fourth petitions. This case illustrates not only the importance of mediation and petition withdrawals but also the unimportance of domestic violence allegations. The police report documented a visit in response to a call for help from the plaintiff, but even when they do explicitly describe the contents of police reports in their court decisions, judges tend to ignore, downplay, or negate their relevance.

Most divorce cases in Henan and Zhejiang involve couples from rural locales. A couple from Henan’s Huojia County held their marriage ceremony in 2011 and registered their marriage a year later. In her third divorce petition, filed in 2015, the plaintiff claimed she and the defendant had been separated since 2012 owing to his regular habit of late-night drinking, their incompatible personalities, and their lack of communication. In 2013, during their separation, when the defendant visited her at her workplace (a KTV club), the discussion became heated and he allegedly beat her, causing her eardrum to bleed. She filed for divorce the following month but ultimately withdrew her petition. In 2014, she withdrew her second divorce petition. In 2015, the plaintiff supported her third petition for divorce by submitting the diagnostic result of an ear endoscopy performed at the Huojia County Red Cross Hospital showing an external injury to her left ear and bleeding from – but no obvious perforation of – her left eardrum. The court refused to affirm the evidence because “the medical documentation proves only that an injury occurred but not that the defendant caused it.” With the defendant in absentia, the court denied the plaintiff’s divorce petition on the grounds that her claims of physical separation and violence lacked sufficient proof (Decision #1386750, April 2, 2015).Footnote ⁵

As a pretext for excluding admissible evidence of domestic violence, courts commonly hold that it fails to link the defendant to the plaintiff’s injury. The previous two examples also illustrate the prevalence of in absentia divorce trials. A defendant’s failure to participate in trial proceedings in no way diminishes a court’s legal authority to grant a plaintiff’s divorce petition. Nonetheless, courts can be reluctant to grant a divorce when the defendant is absent.

In her fourth divorce trial at the Xinchang County People’s Court in Zhejiang Province, a plaintiff lamented her three unsuccessful prior attempts. She supported her claim of marital strife with a copy of a pledge letter, which she said proved that her husband beat her. In his defense, the defendant stated, “It’s true that the plaintiff’s previous three attempts to divorce were unsuccessful, but it’s not true that I beat her. I believe mutual affection has not broken down and do not consent to divorce.” He challenged the plaintiff’s use of his pledge letter by saying, “I think I wrote it just to reconcile with the plaintiff.” To justify its decision to deny the plaintiff’s fourth petition, the court wrote:

Defendants in most cases deny allegations of violence made against them. Even when they admit, on the record, to beating their wives, they usually withhold their consent to divorce, which is all judges need to hold that plaintiffs’ evidence is insufficient to prove the breakdown of mutual affection.

How Marital Decoupling Informs Theories of Institutional Decoupling

Why have Chinese judges been so unwilling to apply the breakdownism standard to grant unilateral no-fault divorces? Why have they increasingly applied the breakdownism standard to deny rather than to grant divorce petitions? And why have they done so even when they were both empowered and obligated by law to apply faultism standards to grant divorce petitions on the basis of spousal wrongdoing? For decades, Chinese law has called on judges to grant divorces in cases involving spousal abuse. And yet, a Beijing court’s 2013 ruling to grant a divorce to Kim Lee has been heralded as “landmark” not because the plaintiff was American, but rather because the court granted her divorce on the grounds of domestic violence (Reference FincherFincher 2014:156; Reference JiangJ. Jiang 2019:241–42). Only exceedingly rarely have judges granted first-attempt divorce petitions on fault-based grounds.

Previous research offers clues regarding Chinese courts’ routine and egregious violations of global legal norms about the freedom of divorce, gender equality, and the protection of the physical security of women. The existing literature points in at least four possible directions of inquiry. First, we could consider the supply of China’s domestic laws that address divorce rights and domestic violence (Reference Htun and Laurel WeldonHtun and Weldon 2018; Reference Hudson, Bowen and NielsenHudson, Bowen, and Nielsen 2011; Reference Wang and SchoferWang and Schofer 2018). We would quickly strike off this explanation upon discovering China’s arsenal of laws and policies rooted in a deep ideological commitment to gender equality common to communist states (Reference WangCheng and Wang 2018; Reference Htun and Laurel WeldonHtun and Weldon 2018:297–301; Reference 493HuangHuang 2005; Reference Tang and ParishTang and Parish 2000:237). Just as the “freedom of marriage” came to symbolize the liberation of women from the oppression of arranged marriages, bigamy, and other “feudal” practices, the “freedom of divorce” too became an enshrined legal principle, particularly for purposes of providing relief to women (Reference JiangJiang 2009a; Reference 504PalmerPalmer 1995:122; Reference TsuiTsui 2001:105).

Second, we could consider China’s international legal commitments (Reference 485Englehart and MillerEnglehart and Miller 2014; Reference Htun and Laurel WeldonHtun and Weldon 2018; Reference Hudson, Bowen and NielsenHudson, Bowen, and Nielsen 2011; Reference Wang and SchoferWang and Schofer 2018). This avenue is another dead end, given that China has strongly endorsed relevant global legal norms by signing all seven (and ratifying six) core UN international human rights treaties, including the Convention on the Elimination of All Forms of Discrimination against Women (CEDAW) (Reference Runge, Goel and GoodmarkRunge 2015; Reference ZhangZhao and Zhang 2017). Numerous official reports and white papers detail China’s pledges and concrete steps to support international goals concerning the status of women in general and the protection of women against violence in particular, and its ostensible progress fulfilling these commitments (e.g., Information Office of the State Council 2015; Reference Rong and TanRong 2016; Reference Zhao, Li and TianZhao 2016).

But, of course, the law is not self-enforcing. China’s on-the-ground judicial practices that subvert its domestic laws and international legal commitments point to a third literature on “loose coupling” and “decoupling,” a gap between policies and practices, form and substance, intentions and results, appearance and reality (Reference 489Hafner-Burton and TsutsuiHafner-Burton and Tsutsui 2005; Reference Hafner-Burton, Tsutsui and MeyerHafner-Burton, Tsutsui, and Meyer 2008; Reference Meyer, Boli, Thomas and RamirezMeyer et al. 1997). It has become a sociological truism that largely ritualistic and ceremonial conformity in organizational appearance belies enormous local variation in on-the-ground organizational behavior (Reference DiMaggio and PowellDiMaggio and Powell 1983; Reference Meyer and RowanMeyer and Rowan 1977). For over two decades, scholars in the “world society” tradition have demonstrated the global ubiquity of decoupling, sometimes called “ceremony without substance” (Reference ColeCole 2013; Reference Frank, Hardinge and Wosick-CorreaFrank, Hardinge, and Wosick-Correa 2009; Reference Meyer and RowanMeyer and Rowan 1977; Reference Meyer, Boli, Thomas and RamirezMeyer et al. 1997; Reference Schofer, Hironaka, Frank, Longhofer, Amenta, Nash and ScottSchofer et al. 2012), and which can also be thought of as “empty promises” (Reference 489Hafner-Burton and TsutsuiHafner-Burton and Tsutsui 2005) and “rights without remedies.” World society scholars, also referred to collectively as the “Stanford school of sociological institutionalism” (Reference Haley and HaleyHaley and Haley 2016; Reference True and MintromTrue and Mintrom 2001), focus on the “strong commonalities in international discourses on a wide range of topics, from human rights to environmentalism” (Reference Schofer, Hironaka, Frank, Longhofer, Amenta, Nash and ScottSchofer et al. 2012:59).

The word “decoupling” in this book’s title is a double entendre. On the one hand, it refers to the “decoupling of married spouses” (Reference Mortelmans and MortelmansMortelmans 2020:2).Footnote ⁷ On the other hand, it refers to the decoupling of – or a gap between – official promises of the law and the degree to which courts fulfill them in practice. Most of the world society literature is devoted to measuring and explaining the global diffusion of stuff on the “official promises” side of the decoupling gap: exogenous norms of secular individualism, scientific rationality, universalism, equality, human rights, and the like (Reference Boli, Lechner, Smelser and BaltesBoli and Lechner 2001; Reference Boli and ThomasBoli and Thomas 1997; Reference Boyle and MeyerBoyle and Meyer 1998; Reference Meyer, Boli, Thomas and RamirezMeyer et al. 1997; Reference Wotipka, Ramirez, Simmons, Dobbin and GarrettWotipka and Ramirez 2008). More recent efforts in this tradition have sought to measure and explain the stuff on the “promise fulfillment” side of the decoupling gap. World society scholars have thus moved beyond their initial focus on the emergence and proliferation of standardized scripts governing organizational appearance and behavior to a new focus on the extent of their local implementation (Reference ColeCole 2005; Reference 489Hafner-Burton and TsutsuiHafner-Burton and Tsutsui 2005; Reference Pope and MeyerPope and Meyer 2016; Reference SwissSwiss 2009). In other words, world society research has shifted from describing superficial norm adoption to assessing its real-life impact (Reference Schofer, Hironaka, Frank, Longhofer, Amenta, Nash and ScottSchofer et al. 2012).

This third literature is reminiscent of the “gap and impact studies” of the 1960s and 1970s in the field of law and society (Reference Gould and BarclayGould and Barclay 2012:330). Gap studies can be traced even further back to the 1920s and 1930s, when legal realists sought to demonstrate that judicial decision-making can never be isolated from its social, cultural, and political contexts (Reference Gould and BarclayGould and Barclay 2012:324–25). A quip widely attributed to legal realist Jerome Frank – that a judge’s ruling has less to do with the law than what he ate for breakfast – has been dubbed the “digestive theory of law” (Reference BlackBlack 1989:5). By highlighting the gap between the law on the books and the law in action, gap studies helped define the early years of the law and society movement (Reference Gould and BarclayGould and Barclay 2012:324).

Critics of gap studies focused on a naïve and optimistic view of gaps as bugs that could be fixed. Law and society scholars subsequently came to treat gaps not only as bugs but also as features (Reference Gould and BarclayGould and Barclay 2012). A gap sometimes reflects the limits of good intentions, a will without a way: insufficient capacity to realize a well-intentioned local effort to adhere to world society norms such as human rights and gender equality (Reference 483ColeCole 2015). A gap sometimes also reflects bad intentions and hypocrisy: the adoption of laws that symbolically advance gender equality for the purpose of obscuring the perpetuation of practices that undermine gender equality (Reference Fallon, Aunio and KimFallon, Aunio, and Kim 2018). Gaps deliberately engineered by state actors as institutional features have been called “radical decoupling” (Reference 489Hafner-Burton and TsutsuiHafner-Burton and Tsutsui 2005) and “state-led decoupling” (Reference Fallon, Aunio and KimFallon, Aunio, and Kim 2018).

In some studies, domestic and international laws appear to reduce gender violence and improve gender justice (Reference Htun and Laurel WeldonHtun and Weldon 2018; Reference Hudson, Bowen and NielsenHudson, Bowen, and Nielsen 2011). Even when adopted by states with no intention of enforcing them, international treaties and conventions can, according to some scholars, shrink the gap between promises and practices (Reference ColeCole 2013; Reference Cole and RamirezCole and Ramirez 2013). Such an outcome is the “paradox of empty promises” (Reference 489Hafner-Burton and TsutsuiHafner-Burton and Tsutsui 2005), and happens because “the entire system ‘drifts’ toward legitimated models” (Reference Schofer and HironakaSchofer and Hironaka 2005:27). World society scholars similarly argue that the ratification of international treaties promoting women’s rights and the enactment of gender-equal national divorce laws have promoted the freedom of divorce and in so doing helped drive rising divorce rates around the world (Reference Wang and SchoferWang and Schofer 2018). China poses a challenge to these optimistic accounts of the impact of global legal norms. We will see later that China’s divorce explosion obscures durable local institutional forces militating against domestic laws promoting gender equality and the freedom of divorce. China’s rising divorce rates are limited to mutual-consent divorces in the Civil Affairs Administration. Meanwhile, China’s judicial clampdown on divorce reflects a widening gap between rights and protections formally provided to divorce-seekers and their practical application by courts.

Although courts contribute only a small share of all of China’s more than four million divorces processed annually in recent years (Ministry of Civil Affairs of China, various years), they are the only place where people can take contested, unilateral, ex parte divorce requests that often stem from domestic violence. Courts contribute only a small and shrinking share of divorces in part because they have become increasingly averse to granting adjudicated divorces. Between 2000 and 2018, the annual number of divorce requests courts granted through adjudication shrank by 16%, while the annual number of divorce requests denied by court adjudication more than tripled, rising by 206% (Ministry of Civil Affairs of China, various years).

Despite being far outnumbered by uncontested, voluntary, mutual consent “divorces by agreement” (协议离婚) processed by marriage registration offices in the Civil Affairs Administration, divorce cases in courts exert an outsized influence that extends into and colors the nature of divorce outside court. Divorce litigation casts a long shadow over couples’ negotiations (Reference Mnookin and KornhauserMnookin and Kornhauser 1979). Divorce-seekers’ spouses take advantage of and benefit from the divorce twofer. Courts’ tendency to deny first-attempt petitions gives spouses of divorce-initiators enormous bargaining leverage over the terms of divorce agreements processed outside court. The freedom of divorce is anything but free. Even if divorce is relatively easy to obtain outside court, divorce-seekers, the majority of whom are women, often sacrifice marital property and child custody in exchange for their husbands’ consent to divorce (Reference LiLi 2022). Divorce in China thus illuminates and obfuscates the limits and possibilities of world society’s influence on the freedom of divorce.

In contrast to world society scholars’ focus on exogenous models, templates, scripts, and blueprints (Reference Frank, Camp and BoutcherFrank, Camp, and Boutcher 2010; Reference Frank, Hironaka and SchoferFrank, Hironaka, and Schofer 2000; Reference Frank and MossFrank and Moss 2017), a fourth literature brings into high relief the less obvious endogenous forces that animate organizational behavior (Reference BartleyBartley 2018; Reference Bartley and NiklasBartley and Egels-Zandén 2016; Reference Dezalay and GarthDezalay and Garth 2010; Reference Haley and HaleyHaley and Haley 2016; Reference HallettHallett 2010; Reference Lazarus-BlackLazarus-Black 2007; Reference MerryMerry 2006; Reference Pache, Santos, Lounsbury and BoxenbaumPache and Santos 2013; Reference Raynard, Lounsbury, Greenwood, Lounsbury and BoxenbaumRaynard, Lounsbury, and Greenwood 2013; Reference WimmerWimmer 2001).Footnote ⁸ According to legal endogeneity theory, organizations interpret, give meaning to, and thus shape the application of the very laws intended to govern their behavior. Law, particularly when it contains ambiguities, is often endogenous to organizational practices (Reference EdelmanEdelman 2016). After the passage of federal equal employment opportunity laws in the United States, private corporations responded by establishing organizational policies, structures, and practices that redefined legal compliance in terms of symbolic commitment to diversity. Laws intended to combat employment discrimination have thus served to obscure and enable employment discrimination (Reference EdelmanEdelman 2016). Likewise, when hospital personnel and patients’ family members struggle to assert neonatal intensive care decision-making authority on the basis of competing legal norms and rules, organizational insiders usually prevail, owing to their greater power to define patient care routines and practices (Reference HeimerHeimer 1999).

Chinese courts, too, offer an opportunity to assess the relative importance of competing norms and practices – some consistent with and some antithetical to world society models. Judges, as organizational insiders, have redefined, reinterpreted, and applied laws in ways that advance their own professional interests, courts’ organizational interests, and the political interests of the party-state, and in so doing have undermined the lawful rights and interests of divorce-seekers. Whereas faultism is consistent with global norms about protecting female victims of marital violence and supports granting divorces, breakdownism, a competing no-fault legal standard, is consistent with local norms about protecting the institution of marriage, social stability, and the interests of judges, and supports denying divorces. To borrow the conceptual language of Reference EdelmanEdelman (2016) and Reference HeimerHeimer (1999), we will see that China’s domestic fault-based standards, consistent with world society norms, are “symbolic laws” containing “symbolic rights” that, in a twist of tragic irony, have largely failed to penetrate its own civil courts, whereas the routine application of a countervailing no-fault standard to deny the petitions of plaintiffs seeking to dissolve abusive marriages has largely stuck.

From a methodological standpoint, this fourth literature eschews efforts to draw macroscopic generalizations from superficial country-level indicators and points instead to in-depth, nuanced, contextually specific scrutiny of local processes animating organizational behavior as a more fruitful means of explaining the puzzle of decoupling (in both senses of the word) in China’s civil courts. Studies such as those in the world society literature that aim to explain variation between dozens of countries in the implementation of laws and policies intended to advance human rights and gender equality must rely on a limited set of crude measures. Owing to this inherent limitation of macro-comparative cross-national research designs, studies that adopt them would have us search in the wrong places for explanations for Chinese courts’ systematic failure to protect women seeking to divorce their abusive husbands. In our search for answers, we would consider China’s bureaucratic capacity to enforce its domestic laws and international commitments (Reference 483ColeCole 2015; Reference 485Englehart and MillerEnglehart and Miller 2014; Reference Htun and Laurel WeldonHtun and Weldon 2018; Reference True and MintromTrue and Mintrom 2001). We would consider the availability and character of monitoring mechanisms (Reference ColeCole 2005). We would consider foreign aid (Reference Dawson and SwissDawson and Swiss 2020; Reference Wei and SwissWei and Swiss 2020). We would consider the strength and autonomy of domestic feminist movements (Reference Htun and Laurel WeldonHtun and Weldon 2018). In the end, we would discover that none of these explanations helps us discern the most salient local obstacles Chinese divorce-seekers face in court.

In the Chinese context of divorce litigation, world society norms coexist with and are neutralized by orthogonal institutional logics. Chinese family law embodies world society norms of equal rights to marriage and divorce (Reference Wang and SchoferWang and Schofer 2018). At the same time, the divorce twofer – Chinese judges’ tendency to deny first-attempt divorce petitions and to grant subsequent petitions – stems from three institutional pressures unrelated to world society: a political ideology that emphasizes family preservation; heavy court dockets; and performance evaluation systems that motivate judges to maximize measures of social stability and judicial efficiency and to support other political priorities. I also argue that their unequal treatment of female and male plaintiffs stems from a fourth institutional logic incongruous with world society: patriarchy. I thus build on scholarship, some of it in the world society tradition, calling for scrutiny of local values and practices inimical to the reception of global norms, including religious doctrine and misogyny (Reference Boyle, McMorris and MayraBoyle, McMorris, and Gómez 2002; Reference Htun and Laurel WeldonHtun and Weldon 2018; Reference Inglehart and NorrisInglehart and Norris 2003; Reference Inglehart, Ponarin and InglehartInglehart, Ponarin, and Inglehart 2017; Reference 505PierottiPierotti 2013; Reference Wang and SchoferWang and Schofer 2018; Reference WelzelWelzel 2013).

Vague and contradictory guidance from the law requires judges to exercise discretion. Legal ambiguity provides space for judges to apply the law in creative ways that serve their own interests. It also invites bias. Consider French divorce judges. As street-level bureaucrats overwhelmed by heavy caseloads and under pressure to meet quantitative productivity targets, they exercise discretion by disposing swiftly of cases they deem unworthy and by approving divorce agreements they know to be unfair to one of the parties (Reference Biland and SteinmetzBiland and Steinmetz 2017:313–14). Remarkably similar dynamics are at play in China. Chinese law is ambiguous on what constitutes the breakdown of mutual affection; domestic violence; evidence sufficient to prove a legal claim; the unknown whereabouts of a defendant; an “important, complicated, and difficult” dispute requiring the application of the ordinary civil procedure by a three-member collegial panel; and the best interests of the child in custody disputes. When judges exploit legal ambiguity to cut corners and close cases quickly, they often do so at the cost of due process. Owing to pervasive patriarchal cultural beliefs, women seeking to divorce their abusive husbands have paid a disproportionate share of this cost.

I am not the first to grapple with decoupling in Chinese courts. Sida Reference LiuLiu (2006), for example, has shown that courts derive more legitimacy from their durable adherence to local practices such as mediation than from their symbolic adherence to global norms. The story that emerges from the evidence I present in this book is about (endogenous) local institutional norms and practices that serve to marginalize and even neutralize China’s domestic laws consistent with (exogenous) global legal norms protecting the freedom of divorce and the equal rights of women. Although it is a China-specific and divorce-specific story, it points to generalizable conditions of decoupling that may be found in other institutional contexts elsewhere in the world. If we are sufficiently attuned to local institutional pressures and practices, we will likely find similarly durable and even intensifying institutional decoupling in other contexts characterized by the same basic conditions present in the Chinese context of divorce litigation: close symbolic alignment to exogenous world society norms, and local agents – such as street-level bureaucrats (Reference LipskyLipsky 2010) – motivated to uphold countervailing endogenous institutional norms.

For women seeking relief from abusive husbands, courts are not the solution but rather part of the problem. I argue that the key to understanding the marginal relevance of marital violence in Chinese divorce adjudication despite its importance in official state rhetoric and black-letter law lies in countervailing legal standards, institutional norms, and practices that overwhelm China’s ceremonial commitments to protect vulnerable women. By privileging a no-fault legal standard of the “breakdown of mutual affection” over competing fault-based legal standards of spousal wrongdoing, including domestic violence, courts themselves are an obstacle to women’s freedom of divorce. Courts subvert the very legal principles of divorce rights and gender equality they symbolically embrace. In China, no-fault divorce laws consistent with legitimized global models are perversely used at best to delay and at worst to suppress divorce in general and female-initiated divorce in particular, even when plaintiffs make claims of domestic violence and support them with evidence. Even if most divorce-seekers eventually find a way to achieve their goal, justice delayed is justice denied. We will see that the delay and denial of justice are highly gendered.

In the remainder of this chapter, I set the stage for the remainder of this book. First, I delineate the empirical scope of this book by situating divorce litigation within the larger backdrop of divorce procedures. I then describe the cast of characters – both human and institutional – who star in this drama. Finally, I map out the organization of the book.

The Landscape of Divorce in China: Procedures and Trends

Litigation is the act of making, defending, and disposing of claims in court, and is handled by a judge or panel of judges. The litigation process begins when a plaintiff files a legal complaint, which I also refer to as a petition. When litigation is processed by adjudication, the decision is binding regardless of whether any or all parties agree with it. Only a minority of civil lawsuits in China are disposed of by trial, however. Most are disposed of by judicial mediation and plaintiffs’ dropping their lawsuits (Chapter 2). Judicial mediation is a Maoist legacy and remains a mainstay practice in China’s courts (Reference HuangHuang 2006). When they mediate disputes, judges apply less formal, more ad hoc, and somewhat free-flowing procedures intended to facilitate negotiation and compromise. Mediated decisions are agreements reached, in principle, voluntarily by all litigants. Likewise, a plaintiff’s “voluntary” request to withdraw her petition, another common outcome of judicial mediation, takes effect after the court approves it.

This book’s empirical focus is adjudicated outcomes in basic-level courts. I use the words “trials” and “adjudications” synonymously. I analyze mediation and petition withdrawals to a far lesser extent because written court decisions are poorly suited for their study (Chapter 4). Figure 1.1 maps out the key steps of the divorce process. It puts the general role of courts and the specific role of court adjudication in perspective.

Figure 1.1 The divorce process

Note: Black boxes and thick lines denote the empirical focus of this book.

Because the Civil Affairs Administration, shown on the left side of Figure 1.1, can only process uncontested mutual-agreement divorces, courts are the only place in China to which people can take contested or unilateral divorces. Generally speaking, divorce is readily accessible outside the court system if both sides consent and can agree on all terms. In China, a more than threefold surge in the annual volume of divorces since the year 2003 is attributable entirely to an explosion in the routine, administrative processing of uncontested, mutual-consent divorces outside the court system in local Civil Affairs Bureaus (Ministry of Civil Affairs of China, various years). Prior to the implementation of the 2003 Marriage Registration Regulations, divorces in the Civil Affairs Administration required an introduction letter from a work unit or villagers’ committee and a one-month approval period. In the first year after the 2003 Regulations took effect, the absolute number of Civil Affairs divorces rose by over 50%, and since then annual percentage growth has averaged over 10%. Between 1990 and 2018, Civil Affairs divorces as a proportion of all divorces more than doubled from 37% to 85%. Fewer than one in six divorces are processed by courts, and absolute numbers of divorces granted by courts (through both mediation and adjudication) have remained flat since 2003. Civil Affairs divorces have driven China’s rapidly rising divorce rates.

Court cases, shown on the right side of Figure 1.1, account for only a small fraction of all divorce outcomes in China. They nonetheless involved about 1.4 million couples in China in each year between 2015 and 2018 (Ministry of Civil Affairs of China, various years), many of whom were vulnerable abuse victims. Moreover, as mentioned earlier, courts’ influence is vastly disproportionate to the share of divorce cases they process. Knowing that their odds of success in court would be slim on the first attempt, divorce-seekers, often in desperation, “voluntarily” accept unfavorable divorce agreement terms as a condition of a quick and certain Civil Affairs divorce.

A Civil Affairs divorce is considerably cheaper than divorce litigation. Some provinces and municipalities had already waived the ¥9 (about US$1.50) marriage and divorce registration fee before the Civil Affairs Administration abolished it nationwide in 2017 (Xinhua 2017).Footnote ⁹ A court divorce case, by contrast, can cost thousands of yuan even when it does not result in a divorce. According to the 2007 Measures on Paying Litigation Fees, courts should charge between ¥50 and ¥300 for divorce cases that do not involve property division (Article 31, Item 1). Among the cases I analyze in this book, most litigants were charged the ¥300 regardless of whether their cases involved property claims. The following judicial practices reflect provisions in the 2007 Measures. Courts discount litigation fees by 50% when they apply the simplified civil procedure. When property claims are involved, courts may charge additional inspection, appraisal, and preservation fees according to the value of the property in dispute and which can total thousands of yuan. When a defendant’s whereabouts are unknown, courts also charge a public notice fee. Absent an agreement between litigants, judges have discretion to order one party to pay court fees or both parties to share court fees. When they denied divorce petitions, courts almost always ordered plaintiffs to assume sole responsibility for court fees. When they granted divorces, courts were somewhat more likely to order defendants to pay all or half of court fees.

On top of court fees are legal service fees charged by lawyers and legal workers (who are discussed in more detail later in the chapter). Legal advocates were involved in at least half of the cases in my samples of first-attempt divorce adjudications. Legal workers in rural areas often charge a flat fee of several thousand yuan (Reference LiLi 2015a:103, 107). Lawyers in urban areas often bill for their services according to the economic value of contested property on top of a base fee of several thousand yuan (Reference MinMin 2017:180). Means-tested legal aid is available but uncommon in the divorce cases in my samples. Legal aid is provided by government legal aid centers, nonprofit and nongovernmental legal aid clinics, and private law firms fulfilling mandatory pro bono quotas.

Uncontested divorce cases rarely enter the court system. If a spouse is unable to appear in person at the local Civil Affairs Bureau’s marriage registration office, the divorce-seeker would be forced to file for divorce in court even if her spouse agrees in writing both to the divorce and to all terms of the divorce. This is one of the few scenarios in which courts handle uncontested divorce petitions. Whereas both sides must be physically present for a Civil Affairs divorce, courts routinely proceed with divorce trials in the absence of defendants, and under special circumstances will even permit the representation of absentee plaintiffs in court proceedings.

Basic-level courts, as courts of first instance, are the first stop in the divorce litigation process. Similar to France’s mandatory conciliation hearings, which are “the first procedural step for all disputed divorces” (Reference Biland and SteinmetzBiland and Steinmetz 2017:314), judicial mediation with the aim of marital reconciliation is required by every version of China’s Marriage Law (1950, 1980, and 2001) as well as the 2020 Civil Code that replaced it on January 1, 2021. Several judicial interpretations issued by the Supreme People’s Court (SPC) echo this requirement.Footnote ¹⁰ A court can grant a divorce only if its mediation efforts have failed to achieve marital reconciliation (Reference 493HuangHuang 2005, Reference Huang2006). A divorce granted by adjudication therefore implies the court’s failure to salvage the marriage. This first step of the divorce process appears in Figure 1.1 as a choice, however, primarily because defendant absenteeism, a common occurrence, precludes the possibility of mediation. If mediation is successful, the plaintiff withdraws her petition, and the couple is considered to have reconciled. If mediation fails to bring forth this outcome, the court may redirect its mediation efforts toward helping the couple agree on the terms of divorce as amicably as possible. Mediation agreements approved by courts are final and cannot be appealed.

Courts generally do not publish approved mediation agreements because they are considered private settlements. When judges grant divorces, they sometimes indicate in their written decisions that mediation has failed to achieve marital reconciliation. Mediation also animates the adjudication process (Reference MengMeng 2012:86). In the course of trial proceedings, a judge may informally cajole a litigant into backing down from an original demand or otherwise help the litigants work out a compromise. In their written decisions granting divorces, judges sometimes refer to and formalize such informal negotiations as a way of saying that their adjudicatory rulings on divorce terms reflect the voluntary will of the litigants. When they do so, however, the information they provide pertaining to judicial mediation is sparse and cryptic (Chapter 10). Ethnographic research designs are obviously better suited for the study of micro-processes in general and judicial mediation in particular in Chinese divorce litigation (Reference HeHe 2017; Reference He and KwaiHe and Ng 2013a, Reference He and Kwai2013b; Reference LiLi 2022; Ng and Reference HeHe 2014).

If mediation is not attempted (unlikely), fails to reconcile the couple (very likely), or is unable to produce an agreement on divorce terms (quite likely), the court will adjudicate the case unless the plaintiff withdraws her petition. Sometimes the court will adjudicate immediately after a half-hearted, pro forma reconciliation attempt. A plaintiff can withdraw her petition at any stage of the process, which is why a petition withdrawal is depicted in Figure 1.1 as a possible result of either mediation or adjudication.

Adjudication and its two primary outcomes – to grant or to deny the divorce petition – are denoted in black with thick lines in Figure 1.1 because they are the focus of my empirical scrutiny in this book. The vast majority of people whose divorce cases go to trial the first time will still be married at the end of the process. The “Divorce Petition Granted” box contains a secondary outcome to which I devote two empirical chapters: child custody. A litigant who is unhappy with her first-instance trial outcome may file an appeal with the municipal intermediate court, which is the court of second instance. Appeals are uncommon. When a plaintiff or defendant does file a second-instance petition, she usually seeks a more favorable ruling on child custody or property division after a first-instance verdict to dissolve her marriage. Sometimes a defendant unwilling to divorce will pursue a second-instance reversal of a first-instance court’s decision to grant her spouse’s divorce petition. For reasons discussed in Chapter 3, plaintiffs rarely appeal adjudicated denials. When plaintiffs do return to court following an adjudicated denial, they almost always do so to file a new first-instance petition after a six-month statutory waiting period. Plaintiffs who withdraw a first-instance divorce petition have the same right to refile after waiting six months. The right to file a new first-instance petition under these circumstances is unique to divorce cases (Chapter 3), gives rise to the feedback loops in Figure 1.1, and therefore enables the divorce twofer.

In Figure 1.1, the “Divorce Petition Granted” box is populated mostly by mediations, and the “Divorce Petition Denied” box is populated mostly by adjudications. Between 2015 and 2018, the slightly more than four million divorce cases that courts nationwide closed using mediation and adjudication were divided roughly evenly between these two modes of case disposal. However, courts tended to use mediation to grant divorces and to use adjudication to deny divorces. In the same time period, courts granted 91% of all the divorce cases they closed by mediation but only 41% of all the divorce cases they closed by adjudication. As a consequence, court-mediated divorces outnumbered court-adjudicated divorces by a ratio of more than 2 to 1. Of all 17 million marital dissolutions in China processed over these four years both inside and outside courts, 11% were court-mediated, 5% were court-adjudicated, and 84% were processed in the Civil Affairs Administration. Courts’ aversion to granting divorces by adjudication intensified dramatically beginning in the mid-2000s. Adjudicated approvals of divorce petitions as a proportion of all divorce adjudications dropped precipitously from 69% in 2000 to 38% in 2018. China’s judicial clampdown on divorce simply reflects courts’ growing unwillingness over time to grant first-attempt divorce petitions.Footnote ¹¹

Dramatis Personae

The primary actors at the center of the divorce litigation stories I tell in this book include the litigants themselves, many of whom are victims of domestic violence. Courts are the stage set where judges decide litigants’ legal fates. I also describe legal advocates even though they play only a cameo role in this book.

Main Acts: Overview of the Book

This book is roughly divided into two halves: (1) causes underlying the divorce twofer, the judicial clampdown on divorce, and judges’ bias against women, and (2) gendered consequences of these problems, including women’s dimmer prospects obtaining an adjudicated divorce, their worse child custody outcomes, and their greater risk of becoming a victim of criminal battery or murder following a divorce attempt. The book is also organized empirically according to two key decisions judges make in the divorce litigation process: (1) the decision to grant or deny a plaintiff’s divorce petition and, when they do grant a divorce, (2) the decision to grant child custody to the mother, the father, or – in the case of siblinged children – both parents.

The next three chapters lay the groundwork for my empirical analyses of marital decoupling. Chapter 2 provides an overview of formal legal divorce rights, particularly as they pertain to gender equality and domestic violence.

Chapter 3 then identifies and explains countervailing institutional norms and pressures that have, at a minimum, blunted and, at most, neutralized the force of these formal legal rights, and that are therefore responsible for institutional decoupling inside China’s divorce courts.

Chapter 4 contains details about my collection of 4.5 million written court decisions from two provinces and how I studied them. China’s courts began posting their decisions online en masse for the most part in 2013 and 2014. At the time I finished writing this book, they had posted the text of over 90 million decisions and metadata of another 22 million decisions on the SPC’s China Judgements Online website (https://perma.cc/9VH9-ZMH8). This book adds to growing sources of guidance on how to exploit this gold mine of information about judicial decision-making (Reference Liebman, Margaret Roberts and WangLiebman et al. 2020).

Chapters 5 and 6 focus specifically on the causes and consequences of judges’ heavy caseloads. My two-province comparative research design reveals that spectacular growth in civil litigation gave rise to the problem of “many cases, few judges.” Judges adopted innovative coping strategies, including the divorce twofer, to deal with their crushing dockets. Divorce cases are casualties of clogged courts, and women are casualties of divorce cases. Whereas Chapter 6 shows how the divorce twofer benefits judges, subsequent chapters show how it harms women.

In Chapter 7, I begin my sustained empirical focus on gender injustice. After first quantitatively demonstrating the prevalence of domestic violence allegations in divorce petitions, I then qualitatively demonstrate their unimportance to judges.

Chapter 8 is a quantitative analysis of the decision to grant or deny a divorce petition. The content of judges’ holdings was virtually identical regardless of whether plaintiffs made allegations of domestic violence. Men enjoyed various kinds of preferential treatment in divorce trial proceedings. Consequently, an adjudicated divorce on the first attempt was considerably less likely for a female plaintiff than for a male plaintiff. Female divorce-seekers’ disadvantage, however, was limited to rural areas.

Chapter 9 explores two tragic consequences of decoupling in cases involving domestic violence. First, it has spawned a sizable population of female marital violence refugees who took flight from their abusive husbands. Second, it has spawned criminal cases. Judges’ practice of denying divorce petitions as a means of protecting themselves and abuse victims has no basis whatsoever in law. Police intervention, including public security administrative punishment and enforcement of personal protection orders, is the primary legal mechanism for protecting abuse victims but has proven to be woefully ineffective in practice. With the more effective support of public authorities, including the police, judges would undoubtedly save lives by – precisely as stipulated by Chinese law – granting the divorce petitions of abuse victims. Courts revictimized abuse victims not only by denying their divorce petitions but also, as Chapters 10 and 11 show, by granting child custody to their abusers in the process of granting their divorce petitions.

Chapter 10 shifts the empirical focus to child custody determinations. Consistent with global legal norms, Chinese laws stipulate that child custody should be determined according to the best interests of the child. No Chinese law privileges fathers with respect to the custody rights of sons. In practice, however, courts tend to formalize the status quo. In so doing, they flout the best interests of the child doctrine as well as guidelines from the SPC directing them not to grant child custody to domestic batterers.

Chapter 11 reports the results of quantitative analyses of the decision to grant or deny child custody. Abused mothers were disadvantaged in child custody determinations, and women’s child custody prospects were determined to a large extent by both the number and gender composition of children. Consistent with the logic of patriarchy, courts almost never granted child custody of only-sons to mothers. Mothers’ best chances for child custody came from multiple children and from only-daughters. When there were multiple children, courts usually split them up between the parents. When multiple children included both genders, courts usually granted custody of sons to fathers and of daughters to mothers.

Chapter 12 concludes with lessons for research on the globalization of law. I discuss the substantive implications of scholars’ methodological choices and constraints. Only by directly measuring judicial behavior and identifying the extrajudicial institutional forces that shape it – my key tasks in this book – can we properly assess the limits and possibilities of the local penetration of global legal norms.