In this chapter, we shift our attention from divorce cases to criminal cases. China’s judicial clampdown on divorce has diverted marital disputes into the criminal justice system. When judges failed to protect battered women, domestic violence sometimes escalated to criminal battery or homicide. Consequently, some women seeking relief in civil court ended up as victims in criminal court after their husbands harmed or murdered them, or as defendants when, in response to chronic abuse, they took matters into their own hands. So far, China’s criminal courts appear not to recognize domestic violence as a sufficiently mitigating factor to merit acquittal in homicide cases. Nonetheless, reforms introduced in 2015 have clearly turned the tide toward leniency in sentencing.

Women abused by their husbands often pursued help before filing for divorce. They were aware of their legal rights and did their best to advance them by seeking the help of relevant public authorities, including public security organs, villagers’ committees, residents’ committees, work units, and branches of the All-China Women’s Federation. Court decisions I present in this chapter are consistent with previous studies documenting public authorities’ reluctance to intervene in “private domestic matters” as well as families’ pressure to stay with abusive spouses (Reference FincherFincher 2014; Reference HanHan 2017; Reference Lin, Sun, Yuning and XueLin et al. 2021; Reference Liu and ChanLiu and Chan 1999). The All-China Women’s Federation gave false hope by routinely advising battered women to file for divorce without also advising them of the Sisyphean nature of Chinese divorce litigation. In their court petitions, women often reported that police failed to intervene adequately – or at all. In a representative divorce case, the female plaintiff who made a domestic violence allegation stated that “the police wouldn’t take my incident report on the grounds that it was a domestic dispute [家庭纠纷]” (Decision #1573098, Yucheng County People’s Court, Henan Province, June 25, 2015; also see Reference ZhengZheng 2015:161).Footnote ¹ Others reported that police intervention was limited to a brief mediation session that ended after their husbands expressed a requisite measure of contrition. Sometimes police officers or villagers’ committee members, in the course of carrying out mediation, likewise seemed content to have resolved the problem after making the husband apologize or write a pledge letter (Reference ZhengZheng 2015:161). Women who, for good reason, lack confidence in the commitment and ability of police to stop their husbands’ abuse may fear reporting domestic violence to the police at all. In the words of one abuse victim, “When he smashed things and verbally threatened me, I didn’t always report it to the police out of fear that he would retaliate” (Decision #4387302, Hangzhou Municipal Yuhang District People’s Court, Zhejiang Province, April 1, 2016; also see Reference FincherFincher 2014:140).Footnote ²

Many women testified that their families had also pressured them to stay with abusers. Owing to the pervasiveness of family persuasion and mediation aimed at marital reconciliation, and not wanting their parents to lose face or worry, some women reported that they feared telling their natal family members about the abuse they suffered (also see Reference ZhengZheng 2015:165, 168). In surveys of survivors of intimate partner violence in China, the majority of respondents reported seeking no help at all (Reference Hu, Xue, Kai Lin, Sun and WangHu et al. 2020; Reference Wang, Fang and LiWang, Fang, and Li 2013:35–36).

Even if they know where to seek help and want to escape, battered women often fear doing so. One plaintiff stated to the court that whenever her husband “found something even slightly against his liking, he would curse and beat the plaintiff. For this reason, the plaintiff tried to leave him many times. Under the pressure of the defendant’s threats, however, she resigned herself to continuing to live together with the defendant.” Her husband denied the allegations and expressed unwillingness to divorce. The court, asserting that husband and wife could still reconcile if they treasured marital affection, denied the plaintiff’s divorce petition (Decision #2365494, Rui’an County People’s Court, Zhejiang Province, August 5, 2010).Footnote ³ Inadequate intervention from public authorities, including courts’ reluctance to grant divorces, heightens the risk of prolonged domestic violence and thus motivates some victims to resort to desperate measures of self-help.

Recall from Chapter 2 that the drafters of the 1980 Marriage Law feared that a clampdown on divorce could result in homicides. We will see in this chapter that their fears have been realized.

What do abused women do when all their efforts to escape domestic violence fail? Some seek relief by way of “fight or flight.” At a literal level, the hormonal fight or flight response is a well-documented involuntary clinical reaction triggered by the traumatic stress of domestic violence (Reference WalkerWalker 2017:325). At another level, it represents the pragmatic choices and coping strategies of women attempting to survive (Reference ZhengZheng 2015). When they receive no protection through official channels, sometimes they take flight and escape their abusers; sometimes they fight and kill their abusers.

Marital abuse can be terminated in several ways. One way is to mobilize the force of law by petitioning for divorce or requesting intervention from other public authorities. A second way is to take flight; in China, women often finance their flight from abuse through migrant labor force participation. A third way to terminate abuse is to terminate the victim; abusers sometimes kill their victims. A fourth way is to terminate the abuser; victims sometimes kill their abusers. The evidence I present throughout this book points to the near futility of the first way. In this chapter, a mix of quantitative findings and case examples reveals the too common consequences of that failure. I will begin by demonstrating that the divorce twofer prolonged abused women’s exposure to violence when courts denied their first-attempt divorce petitions.

Leaving Abusive Men Is Dangerous

Staying with abusive men is dangerous. Leaving abusive men is also dangerous. Some women stay with their abusers when, in their assessment, the risks of leaving outweigh the risks of staying. In the United States, women’s risk of getting murdered is greatest when they leave their abusive partners (Reference WalkerWalker 2017:Chapter 12). Domestic abusers everywhere want the subjects of their abuse to get the message, “If I can’t have you, no one will” (Reference WalkerWalker 2017:114, 306). In China, too, abusive men’s threats to murder their wives – such as, “If you don’t kill me first, I’ll make sure you die a horrific death” and “If you dare try to divorce me, I’ll murder your entire family” – are pervasive (Reference WangX. Wang 2017:25). In this section I show that they sometimes put their threats into action.

A criminal case in Zhejiang documents how close a woman came to losing her life when she insisted on divorcing her husband, Li Fufa.Footnote ⁵ When discussing the practicalities of their divorce, including household debt, Li pinned his wife down on the sofa and started stabbing her face with a knife. At this moment his mother-in-law happened to enter the home. When she rushed into the living room and pulled the husband off, he stabbed her hand. His wife seized the opportunity to flee. Li chased and caught her outside a restaurant, where many people witnessed him continue to stab her face and neck. His mother-in-law once again pulled Li off, allowing the wife to escape. According to a forensic pathology report, Li’s wife sustained a traumatic tracheal rupture, a traumatic transection of the thyroid, a laryngeal nerve injury, and multiple injuries to the left and right jaw as well as to the back of the skull. Li was sentenced to 11 years in prison for attempted homicide (Decision #3236920, Yiwu Municipal People’s Court, Zhejiang Province, November 28, 2014).Footnote ⁶

In a similar case, a man with the surname Zeng nearly killed his wife who was trying to divorce him. When Zeng was stabbing his wife in the head, chest, and other areas, his mother-in-law tackled him, allowing his father-in-law to disarm him, at which point his wife escaped and notified the police. She suffered a punctured tongue, an arm laceration, and a radial nerve rupture. The court gave Zeng an eight-year prison sentence for attempted homicide. To foreshadow my discussion later in this chapter about the role of compensation in criminal sentencing, the court attributed its lenient sentence to the wife’s expression of forgiveness of Zeng offered in exchange for the compensation (of an unspecified amount) she received from his family (Decision #2638685, Ningbo Municipal, Yinzhou District People’s Court, Zhejiang Province, December 30, 2013).Footnote ⁷

Many women do not survive attacks like these. Such was the fate of one woman who returned to her natal family after what the court euphemistically described as “arguing and fighting over trifles.” In his testimony, Li Suzhen admitted hitting his wife. He made several trips to her natal home to persuade her to return home with him. On his final rebuffed effort, she declared her desire to divorce him. As Li put it, “I was furious. I thought, if we couldn’t be together, she should die, I should die, and that would be the end of it. I ran into her family’s kitchen, grabbed a cleaver, and cut her neck with great force.” She bled to death after he severed her left neck artery and vein. Li was sentenced to life in prison. In the course of the trial, Li compensated her family with ¥230,000 and received a forgiveness letter from her parents in return, which may have spared him from a death sentence. Criminal reconciliation of this nature is a topic to which I will come later in this chapter (Decision #1184232, Puyang Municipal Intermediate People’s Court, Henan Province, March 5, 2014).Footnote ⁸

Tragically, the foregoing criminal case examples are merely the tip of the iceberg. My samples contain hundreds more. According to official reports, about 10% of all homicides in China are related to domestic violence (Reference Palmer, Fu and PalmerPalmer 2017:290; Reference ZhengZheng 2015:162). A crude – and conservative – method I developed to identify criminal cases involving murders of spouses produces an identical estimate in the Henan sample (10%) and a modestly lower estimate in the Zhejiang samples (8%). The criminal cases in this section are part of a larger pool of at least several hundred cases of spousal murder and over 1,500 cases of spousal battery in my two provincial samples.Footnote ⁹ Later in this chapter I will analyze the full pool in more detail. But first I will show that the divorce twofer, by extending the divorce process, ipso facto also extends the time during which abused women are forced to remain with their husbands. When courts routinely deny their divorce petitions, the dangers of their situations are prolonged.

Divorce Denials Prolong Dangers to Women

Since courts began increasing their suppression of first-petition divorces, the population of divorce-seekers awaiting relief has swelled. The duration of time from first filing until divorce can be calculated in two ways: (1) by searching for first-attempt filing dates in the text of subsequent-attempt decisions and (2) by linking first-attempt and subsequent-attempt court decisions. In so doing, we learn that mean/median time to an adjudicated denial on the first attempt was 70/62 days and 45/35 days in the Henan and Zhejiang samples, respectively. When the divorce was granted on a subsequent attempt, total mean/median time was 410/408 days and 391/362 days in the two samples, respectively (beginning from the time when the initial petition was filed). Thus, simply subtracting the former values from the latter values, the mean/median delay to divorce caused by first-attempt adjudicated denials was 340/346 days and 346/327 days in the Henan and Zhejiang samples, respectively.Footnote ¹⁰ Given what we know from Chapter 6 about judges’ preference for applying the simplified civil procedure in adjudicated denials, the mean/median delay to a divorce might be more realistically estimated as the time from an adjudicated denial using the simplified procedure to time to granting a divorce using the ordinary procedure: 380/381 days in Henan and 434/441 days in Zhejiang. By all measures, mean and median delays to divorce caused by the divorce twofer range from almost one year to over one year in both samples. If many plaintiffs can return to their home jurisdictions to file for divorce only once per year for the Spring Festival national holiday, as we saw in Chapter 4, we should not be surprised that the statutory six-month waiting period often becomes a one-year waiting period in practice.

If courts are more likely to deny first-attempt divorce petitions filed by women, as Chapter 8 proved, then it can only be true that the delay to freedom is longer for women than for men. Another way to view gender disparities in denials of and delays to divorce is to compare the number of attempts required to obtain an adjudicated divorce. Table 9.1 contains all court decisions from both samples in which divorce petitions were granted and in which plaintiff sex is known. It shows differences between women and men in the likelihood of requiring only one attempt to do so among plaintiffs who successfully obtained an adjudicated divorce.

The first pattern to emerge from Table 9.1 is a familiar one: the divorce twofer was more prevalent in Zhejiang than in Henan. Of those who successfully divorced, the majority in Henan did so on the first attempt, whereas the majority in Zhejiang required at least two attempts. Differences between the two provinces narrowed over time as the judicial clampdown on divorce intensified in Henan and remained stable in Zhejiang (Chapter 6). Prior to 2012, 75% of divorces granted by adjudication in the Henan sample required only one attempt. By 2015, this proportion had declined to 53%; almost half of all divorces granted by adjudication in Henan had been previously denied. Second, in both provinces women required more attempts than men. The probability of success on the first attempt was 11 and 17 percentage points greater for men than for women in the Henan and Zhejiang samples, respectively. In both provinces, gender disparities were greatest in rural areas, where divorces were concentrated. In rural courts, differences between men and women in the probability of success on the first attempt were 13 and 21 percentage points in Henan and Zhejiang, respectively. Meanwhile, in urban courts, gender gaps were a substantially narrower 1 and 9 percentage points, respectively. In short, rural women were the most impacted by the divorce twofer.

Table 9.2 builds on this analysis by considering specific durations of time required to obtain an adjudicated divorce. It contains fewer court decisions than Table 9.1 owing to a large number of missing filing dates. The first thing we notice is that few divorces granted by adjudication were finalized within three months of initial filing. Among plaintiffs whose petitions were granted within three months, gender differences range from nil to relatively small, with women enjoying the advantage, particularly in urban courts. Of all divorces granted by adjudication in both provinces, the vast majority (94–96%) were finalized within two years of initial filing, regardless of plaintiff sex. Among plaintiffs whose divorce petitions were ultimately granted by adjudication, 38% and 52% were still married six months after initially filing for divorce in Henan and Zhejiang, respectively. Over time this gap shrank as the two provinces converged in their embrace of the divorce twofer; by 2015, 48% of plaintiffs in Henan were still married at the six-month mark. Gender disparities are only pronounced at this stage in rural courts, where the probability of a successful divorce within six months was 13 and 19 percentage points greater for men in Henan and Zhejiang, respectively. Gender disparities at the point of one year after initial filing are entirely consistent with but less pronounced than those at the point of six months after initial filing.

Table 9.3 brings together the previous two analyses – namely, of the number of attempts necessary to divorce and of the duration of time from initial filing to divorce – into a regression analysis. It reveals that women’s longer delays to divorce are explained almost entirely by their greater susceptibility to the divorce twofer. In Model 1, the baseline model without controls, we see that women’s mean/median time to divorce exceeded that of men by 19/6 days in Henan and 31/33 days in Zhejiang. Model 1 also shows that the gender gap was limited to rural courts, where women’s mean/median time to divorce was 24/7 days longer than men’s in Henan and 43/45 days longer than men’s in Zhejiang.

The type of civil procedure applied to the trial (ordinary vs. simplified) is added to Model 2. We know from Chapter 2 that the ordinary procedure slows down trials, and from Chapter 6 that this is the main reason why judges prefer to apply the simplified procedure. Not surprisingly, therefore, Model 2 shows that the application of the ordinary civil procedure delayed adjudicated divorce by between two and three months in both provinces in rural and urban courts alike.

Adding the civil procedure variable to Model 2 widened the gender gap in rural courts. We learned in Chapter 8 that male plaintiffs in rural courts were more than twice as likely to have their divorce petitions heard in public notice trials. For this reason, women – particularly rural women – were less likely to have the ordinary civil procedure applied to their divorce petitions. Owing to the SPC requirement that public notice trials be conducted according to the ordinary civil procedure (Chapter 4), delays associated with the ordinary civil procedure are partly a function of the 60-day public notice period (Chapter 2). Among plaintiffs, women’s relatively long delays to divorce were therefore mitigated by their relatively greater exposure to the simplified civil procedure, which, in turn, was partly a result of their relatively smaller exposure to public notice trials. In other words, the gender gap in time from initial filing to successful divorce would have been even wider had the ordinary civil procedure been applied at identical rates to the trials of female and male plaintiffs. In Model 2, among plaintiffs with identical levels of exposure to the ordinary civil procedure, women’s mean/median time to divorce was 25/8 days longer than men’s in Henan and 43/39 days longer than men’s in Zhejiang. Also in Model 2, the gender gap remains limited to rural courts: 30/10 days in Henan and 53/55 days in Zhejiang.

When my measure for the divorce twofer (“granted on subsequent attempt”) is added to Model 3, the gender gap disappears completely in Zhejiang and even reverses direction in Henan. The interpretation of this pattern is simple: rural women’s longer delays to adjudicated divorce were the direct consequence of their greater likelihood of experiencing the divorce twofer. They were far more likely than men to have refiled for divorce after an adjudicated denial. Rural women experienced relatively long delays to adjudicated divorce because rural courts were much more likely to deny their first-attempt petitions and thus to force them to refile after the statutory six-month waiting period. Many, however, appear not to have returned to court (Chapter 6). Among women whose initial petitions were denied, some may have sought to divorce outside the court system, and others may have abandoned their quests for divorce altogether.Footnote ¹¹

Because adjudicated denials delay the divorce process, the divorce twofer elevates dangers to women’s physical safety, particularly in rural areas. Courts’ routine denial of first-attempt divorce petitions fuels the expansion of a population of frustrated and often vulnerable plaintiffs awaiting divorce, among whom women are vastly overrepresented. If female plaintiffs are more likely than male plaintiffs to be victims of domestic violence, which we know is true, then it must also be true that the divorce twofer prolongs women’s exposure to domestic violence. Even if plaintiffs who return to court for another attempt are guaranteed a divorce, the mean and median delay of over one year introduced by the divorce twofer helps enable the continuation of violence. This epitomizes the principle that justice delayed is justice denied. Judges delay justice by denying divorce petitions and deny justice by delaying divorces. The next case example illustrates a homicide committed after a divorce twofer involving domestic violence.

A woman, surnamed Yu, filed her second divorce petition in May 2010. She claimed that her husband, Wang Jinya, frequently argued with and beat her over family trifles. Although online repositories of court decisions do not contain the court’s denial of Yu’s original divorce petition in 2007, her statement to the court on her subsequent attempt in 2010 – namely, that “in the time since then, the defendant has failed to rectify himself” – suggests she had made the same allegations the first time.Footnote ¹² The court granted the divorce on June 13, 2010, almost three years after denying her prior petition. On June 23, 2010, when Yu was collecting her belongings, he tried to persuade her to move back in. When she refused, he murdered her by bludgeoning her head with a wooden hammer, after which he stabbed and cut her neck with a knife, and once again bludgeoned her head with a copper rod. According to the forensic pathology report, the knife wounds severed her trachea and esophagus, as well as an artery and a vein on the right side of her neck, causing massive blood loss, and the blunt force trauma from the hammer and rod crushed her skull. When their landlord heard Yu’s scream for help and knocked on the door, Wang said nothing was going on, that he and his ex-wife were simply having a chat. Later, after the landlord heard a loud banging noise, he knocked on the door again. When no one answered, he called the police, who entered the premises and discovered Wang hanging from the ceiling. They rushed him to the hospital, where he was saved. One of their three children testified that Wang’s regular abuse was the reason why Yu had filed for divorce. The court held that Wang’s actions were consistent with domestic violence and sentenced him to death (Decision #5012675, Ninghai County People’s Court, Zhejiang Province, December 21, 2010).Footnote ¹³

Effective intervention from any number of sources might have saved Yu’s life. Yu’s divorce was needlessly – and, arguably, unlawfully – prolonged. To at least some degree, courts have blood on their hands. We have no way of knowing the extent to which the court’s adjudicated denial of Yu’s initial petition was responsible for her subsequent murder. However, had the authorities believed her allegations and taken them seriously the first time, the court may have granted her divorce years earlier than it did and reduced opportunities for Wang to kill her. After noting Wenzhou’s pioneering anti-domestic violence work (see Chapter 7), a judge from Zhejiang made the flabbergasting assertion that legal intervention to protect women like Yu is beyond the scope of public authorities: “If, after a divorce, a domestic abuser seeks to commit revenge or a violent attack, it would be difficult to prevent. This would no longer be a legal matter.”Footnote ¹⁴

As we saw in Chapter 7, in their efforts to deny divorce petitions, courts trivialized and negated women’s domestic violence allegations. In so doing, courts simultaneously evaded their responsibility to refer criminal domestic violence cases to procurators. According to both the 2015 Opinions Concerning the Handling of Criminal Domestic Violence Cases in Accordance with the Law and the 2015 Anti-Domestic Violence Law, judges are supposed to transfer to the procuracy cases in which they discover domestic violence that constitutes a criminal offense. They are also supposed to inform victims of their right to initiate private criminal prosecution (provided by Article 112 in the Criminal Procedure Law; Reference ZhangR. Zhang 2017:52). The Criminal Law also includes all kinds of provisions that could serve as the basis for criminally prosecuting domestic violence, such as maltreatment and desertion, assault and battery, rape, and homicide (e.g., Articles 17, 95, 234, 235, 236, 237, 260). Finally, the Marriage Law stipulates that public security organs, upon the request of domestic violence victims, should carry out administrative punishment of offenders (Article 43). It also stipulates that domestic violence victims have the right to initiate private criminal prosecution, at which point public security organs should conduct criminal investigations and the procuracy should initiate criminal prosecutions (Article 45). From judges’ perspective, notifying procurators of criminal wrongdoing they discover in divorce litigation would validate plaintiffs’ domestic violence claims, oblige judges to grant divorces on fault-based grounds, and thus undermine the professional benefits of the divorce twofer. Judges are reluctant to issue personal protection orders for the same reason (Reference JiangJ. Jiang 2019:235).

Police, too, appeared to take a hands-off approach to domestic violence. The court decisions in my samples show that many women sought police help, but few received it. Thousands of divorce petitions in my samples contain both allegations of domestic violence and reported instances (often documented) of calls made to police for help. According to their testimony, police often failed to provide adequate intervention – or to intervene at all – after women called 110 or reported domestic violence in other ways. Tens of thousands of court decisions – both civil and criminal – in my samples contain references to public security administrative punishment decisions. Rarely, however, do they pertain to domestic violence. Fewer than 200 divorce petitions in my samples contain both allegations of domestic violence and references to public security administrative punishment decisions, which can be and are used as evidence of wrongdoing. Even among the roughly 2,000 criminal cases I found in my samples involving intentional injury and murder between spouses, references to public security administrative punishment decisions are few and far between, also numbering fewer than 200. The most common types of criminal case involving administrative punishment are drunk driving resulting in injury or death – which is classified as the offense of “dangerous driving” (危险驾驶) – and theft (盗窃).

When courts deny divorce petitions, and in so doing prolong women’s exposure to their abusive husbands, women face difficult, high-stakes choices. Some may pursue divorce in the Civil Affairs Administration. However, the procedural requirement of mutual consent to divorce and mutual agreement on all terms of the divorce gives enormous bargaining leverage to the spouse who did not initiate the divorce. Consequently, when courts deny their divorce petitions, women often give up child custody and marital assets in exchange for freedom from their abusive husbands (Reference LiLi 2022). Other women resign themselves to staying married rather than risking destitution, the loss of their children, and the potentially deadly consequences of leaving their abusive husbands. Battered women aware of the hidden rule of the divorce twofer understand that filing for divorce will likely fail on the first attempt and result in violent retaliation from their husbands (Reference DengDeng 2017:113).

When courts fail to provide relief, some women seek protection from other public authorities. Study after study, however, shows inadequate intervention on the part of police, civil government agencies, local residents’ and villagers’ committees, and government-operated nongovernmental organizations such as the All-China Women’s Federation (Reference ChenChen 2018; Reference Cheng and GaoCheng and Gao 2019; Reference FincherFincher 2014; Reference GuoGuo 2019; Reference ZhangH. Zhang 2014:232; Reference ZhengZheng 2015). Indeed, according to the author of one study of domestic violence victims, “all the women to whom I talked have sought help from the police and All Women’s Federation [sic], but to no avail. In their words, ‘The police and the All Women’s Federation [sic] are dog’s fart (gou pi),’ meaning that they are useless” (Reference ZhengZheng 2015:172).

As a result, many women take measures to protect themselves. Prominent among women’s self-protection strategies is flight.

Flight: Fleeing Abusers

My samples show that abuse victims who returned to court after the statutory six-month waiting period often claimed worsening violence (also see Reference XuXu 2007:204). As they awaited their next opportunity to divorce, many abuse victims often became marital violence refugees. One plaintiff indicated the following in her statement to the court:

In this case, the duration of the first attempt (from case filing to adjudicated denial) was 50 days, but the total duration of time between original first-attempt case filing and adjudicated divorce was 2,492 days: almost seven years.

Another plaintiff stated to the court: “In the time since suffering a beating by the defendant in 2007, I have been in hiding, afraid to return home, for over three years. In early 2010 my divorce petition was denied by the Song County People’s Court, after which I have still not dared to return home” (Decision #562570, Song County People’s Court, Henan Province, April 1, 2011).Footnote ¹⁶ The litigation took over one year, but the overall process of divorcing lasted three or four years.

In her statement to the court, a woman claimed that after her first filing for divorce in 2009, her husband and his family prevented her from participating in the trial by physically blocking and verbally threatening her before dragging her back home, where they beat her. The court subsequently denied her second divorce petition three years later. On her third attempt in 2014, she testified that “currently I am raising our children elsewhere by myself, doing my utmost to avoid him, and living in a constant state of fear. At this point we have already been physically separated for five years.” Although the plaintiff did everything right by both satisfying the statutory physical separation requirement and submitting as evidence a copy of her husband’s “pledge letter” in which he admitted carrying out domestic violence, the court ignored her claim of physical separation and ruled the pledge letter inadmissible after the defendant recanted its contents. The basis of the court’s decision to deny her divorce petition was that “plaintiff and defendant have been married for 14 years and have a son and a daughter. Conflicts in their everyday life are difficult to avoid but not fundamentally insurmountable. Plaintiff and defendant should cherish the marital affection they have already established. They are capable of reconciling if, from this point forward, they improve their communication skills” (Decision #1168173, Changge Municipal People’s Court, Henan Province, February 8, 2014).Footnote ¹⁷

One decision contains the story of a woman who, after the Lankao County People’s Court denied her initial divorce petition in December 2011, “had no choice but to escape the reality of my situation by getting a job outside my place of residence. I never imagined my suffering would not lessen after I fled, much less that it would gradually deepen over time. My current state of mental health is on the verge of collapse” (Decision #890371, Lankao County People’s Court, Henan Province, November 7, 2012).Footnote ¹⁸ In another second-attempt divorce petition, the plaintiff submitted police records of nine requests for police help in support of her claim of intensifying violence following the court’s denial of the first-attempt petition. The court denied the petition after affirming that the evidence proved only that calls to the police were made as a consequence of “disputes,” but not that domestic violence occurred (Decision #4405727, Changxing County People’s Court, Zhejiang Province, November 24, 2015).Footnote ¹⁹

When defendants contested plaintiffs’ claims of uninterrupted separation, or if plaintiffs returned briefly for planting and harvesting crops in efforts to maintain claims to farmland, judges not infrequently held that plaintiffs failed the applicable statutory physical separation test and, on this basis, denied their second-attempt divorce petitions.

The defendant in this case denied committing any act of domestic violence. The defendant further claimed that the plaintiff had occasionally returned home during the alleged period of physical estrangement. The court, citing the plaintiff’s double failure to prove that domestic violence caused the breakdown of mutual affection and to meet the two-year separation test, denied her divorce petition (Decision #3525299, Tongxiang County People’s Court, Zhejiang Province, December 9, 2014).Footnote ²⁰

When women flee marital violence, their default destination is often their natal families. Because their husbands can so easily find them there, however, this poses risks not only to themselves but also to their family members. For this reason, many abused women go into hiding. They frequently participate in labor migration in order to escape their abusers, to support themselves, and to accumulate money necessary to finance the divorce (Reference LiK. Li 2015a:101–6). China’s migrant labor force thus includes marital violence refugees as well as women hoping to satisfy the one-year physical separation test before their next divorce attempt.

Themes shared by the following examples of women’s flight from marital violence include husbands’ jealousy and control, and wives’ efforts to free themselves and earn a livelihood through migrant labor force participation. These examples also show, however, that their husbands often find them. I will begin with the applicant’s statement in one of the very few protection orders in my samples of court decisions.

The defendant in the next example similarly hunted down his wife after she fled to her parents’ home and then to Guangdong Province, a common destination for migrant workers from Henan.

The court denied the plaintiff’s divorce request on the grounds that they had bought a house together – supposed proof of the strength of their marital foundation – and that the defendant exhibited contrition by recognizing his mistakes (Decision #338557, Xixia County People’s Court, Henan Province, June 28, 2010).Footnote ²²

In some ways, the women whose stories are captured in these court decisions are the lucky ones. Taking flight may have increased their chances of survival. The women whose stories appear in the next section were less fortunate.

Leniency in Criminal Domestic Violence Cases

We have seen that courts, in their written decisions, indicated that they extended leniency to defendants who surrendered to the police, gave full confessions, cooperated throughout the investigation, displayed sincere remorse, and so on. Most of these conditions for leniency are written into the Criminal Law’s provisions on sentencing (e.g., Articles 62, 63, 67, and 78). We have also seen that courts at most only rarely affirmed justifiable self-defense as the basis for leniency.

In the homicide case of Yao Shuangxia, the court attributed its lenient sentence to the following: the crime occurred at home, the victim bore fault for triggering the crime, Yao was a first-time offender, she reported the crime on her own accord, she surrendered to the police, and she confessed to her crime. Yao described the events culminating in her killing her husband.

Forensic evidence such as the victim’s elevated blood alcohol level and semen collected from Yao’s body corroborated her testimony. Moreover, several witnesses, including relatives and co-workers, corroborated Yao’s allegations of her husband’s domestic violence, infidelity, and alcohol abuse. Her mother also testified that, following the Spring Festival incident, Yao wanted a divorce, but that family members had intervened. Although the court did not affirm justifiable self-defense, it came close by affirming that “unbearable domestic violence and psychological torture inflicted by the victim caused the defendant to kill him.” Likewise, although the court rejected the defense lawyer’s argument that the circumstances of her crime were “relatively minor,” it accepted the argument that the homicide victim was at grave fault. For these reasons, and in light of the defendant’s cooperation, the court sentenced the defendant to eight years in prison. Because it was only a tiny fraction of the ¥821,359 her in-laws requested in the civil lawsuit they attached to the criminal prosecution, the court’s order that Yao compensate them ¥17,115 for funeral expenses was probably at most only a marginal factor in the leniency of her sentence (Decision #1059225, Zhengzhou Municipal Intermediate People’s Court, Henan Province, November 8, 2013).Footnote ³²

Beyond this specific case, however, compensation for civil damages plays an important role in criminal sentencing. Since the mid-2000s, the SPC has promoted criminal reconciliation as part of China’s broader stability maintenance agenda. The theory behind what has been dubbed “blood money” (Ng and Reference HeHe 2017b) and “cash for clemency” (Reference TrevaskesTrevaskes 2015) is that the criminal offender’s payment of compensation to the victim’s family can nip two potential sources of social unrest in the bud. In exchange for compensation, the victim’s family makes a formal expression of forgiveness and the court spares the offender’s life. Criminal reconciliation practices were thus designed to placate the anger of the victim’s family members with compensation and thus to reduce the likelihood they will protest or petition in response to what they perceive as an unjust ruling not to execute the offender. By the same token, by preempting the anger of the offender’s family that might otherwise be caused by a death sentence, leniency serves to reduce the likelihood the other side will protest or petition the court decision (Reference LiebmanLiebman 2015:214–15). Compensation is sometimes ordered by courts in their rulings on petitions for civil damages attached to criminal cases (刑事附带民事诉讼) – what Reference McConville, Choongh, Wan, Hong, Dobinson and JonesMcConville et al. (2011) call “incidental civil action” in criminal litigation. Courts also recognize compensation agreements reached privately or through mediation outside the court system.

Criminal reconciliation has been formalized through “standard cases” (典型案例) issued by the SPC as models to establish best practices in criminal sentencing (Reference TrevaskesTrevaskes 2015). Beyond standard cases, the 2010 Provisional Guiding Opinions of the SPC on Criminal Sentencing (which in 2017 became the Guiding Opinions of the SPC on Sentencing for Common Crimes) stipulates that compensating the victim’s economic losses can mitigate prison sentences by up 30% (Article 9) and that forgiveness from the victim or the victim’s family can mitigate prison sentences by up to 20% (Article 10; Reference XingXing 2013:27). The payment of compensation from the offender or the offender’s immediate family in exchange for both forgiveness from the victim’s family and sentencing leniency from the court has become an institutionalized practice in China’s criminal courts (Reference LiebmanLiebman 2015:180–85). Criminal reconciliation has even become part of judicial performance evaluation systems (Yanhong Reference WangWang 2013:33). According to one study, victims’ families submitted “forgiveness letters” (谅解书) to the court in 80% of cases of women who killed their husbands. Some of these letters were jointly signed by local residents (Reference 496Li and JiaLi and Jia 2019:64, 67). In some cases, the parents of homicide victims kneeled in front of the judges begging for leniency (Reference XingXing 2013:27).

In my samples of homicide cases involving spouses, women’s sentences were far more lenient than men’s. Whereas 13% of men convicted of murder received death sentences (most of which were suspended), not a single woman was sentenced to death. Likewise, 55% of male defendants and 39% of female defendants received life sentences. Finally, the remaining 32% of men and 61% of women received prison sentences ranging from two to 15 years. Among defendants who received fixed-term prison sentences, women (19%) were three times more likely than men (6%) to receive mitigated sentences of less than ten years for the “relatively minor circumstances” of their cases (according to Article 232 of the Criminal Law). Female and male defendants in the criminal domestic violence cases in my samples were almost equally likely to receive forgiveness (41% and 39%, respectively). Sentencing disparities between women and men therefore cannot be attributed to differences in the likelihood of receiving forgiveness. Courts’ greater leniency toward female defendants seems to have more to do with judges’ greater likelihood of finding them to be more cooperative, their risk to society to be smaller, the criminal circumstances of their cases to be less serious, and their criminal acts more likely to constitute justifiable or imperfect self-defense.

Forgiveness was associated with leniency for both female and male defendants. However, forgiveness in the absence of compensation was relatively rare for male defendants. Men convicted of murder tended to buy forgiveness from their victims’ families and thus to buy leniency from courts. Perhaps because they had greater financial wherewithal to pay compensation, men were twice as likely as women to compensate their victims’ families (35% and 18%, respectively). Among male defendants, 76% received forgiveness when they provided compensation compared to only 19% when they did not. Women’s lower incidence of paying compensation, however, did not reduce their likelihood of receiving forgiveness. Indeed, in contrast to male defendants who tended to receive forgiveness in exchange for compensation, female defendants tended to receive forgiveness without compensation.

Some husbands who murdered or attempted to murder their wives were given remarkably lenient sentences after paying compensation. A man in the city of Wenzhou named Yu Qing made a murder-suicide plan after his wife insisted on a divorce despite his numerous efforts to talk her out of it. After his final effort failed, he choked her to death before slitting his own throat and wrists. His suicide attempt failed when he was discovered and rushed to the hospital. In court, Yu’s defense lawyer argued for leniency on the basis of reasonable suspicion that his wife was having an affair and therefore bore a certain responsibility for her own murder. He further argued that the character of maliciousness in murders like this one is inherently different from that in ordinary murders and that the reason he chose to murder her and kill himself was because “he loved her too much.” In consideration of the victim’s family’s expression of forgiveness in exchange for ¥170,000 in compensation, the court gave him a lenient prison sentence of 15 years (Decision #4848656, Wenzhou Municipal Intermediate People’s Court, Zhejiang Province, September 28, 2016).Footnote ³³

Some women also compensated their victims’ families. Wu Jinrong had been subjected to over a decade of marital violence in her second marriage. On one occasion, her husband beat her so hard she permanently lost hearing in one ear. According to the court, “on a regular basis, after getting drunk, he physically injured her, insulted her dignity, psychologically threatened her, smashed and burned household objects, and committed other forms of domestic violence.” One night, after coming home drunk, he entered her daughter-in-law’s bedroom (the wife of the son from her first marriage) and smacked her grandson until he woke up. After her daughter-in-law protested, he began to break objects in the living room and, wielding a cleaver, threatened to harm family members. The defendant, recalling many similar incidents, killed him by bludgeoning his head with a hoe and a shovel. Both village authorities and the court initiated mediation between the defendant and the victim’s family. In exchange for financial compensation of ¥50,000 and a minivan, the victim’s family expressed its forgiveness. Although the court rejected the defense lawyers’ request for probation, it showed leniency by giving her a mitigated prison sentence of five years (Decision #821646, Nanyang Municipal Wancheng District People’s Court, Henan Province, July 3, 2012).Footnote ³⁴

Cases like this, however, are unusual. Relatively few women who killed their husbands provided compensation, and the payment of compensation had little effect on their sentences. Women who killed their husbands were far more likely than their male counterparts to receive forgiveness without compensation. Victims’ families and the public were naturally inclined to extend their empathy, sympathy, and forgiveness to women who killed their abusers. By contrast, forgiveness of men who killed their wives tended to be financially induced.

The case of Xue Aihua exemplifies one of the hallmarks of abusive men: “pathological jealousy” that is “a cornerstone to homicidal rage” (Reference WalkerWalker 2017:307). Xue killed her husband by chopping his face, head, and neck with an axe, resulting in hemorrhagic shock from a ruptured jugular vein. Owing to his chronic abuse, Xue had previously divorced him, after which – under the husband’s coercion – they remarried. She had started working at a cotton mill with the help of her sister-in-law, who also worked there. Prior to this she stayed mostly at home doing farm work. Her husband, owing to his jealousy, was in the habit of checking in on her at work. She stated in her testimony to the court that, on the day before the homicide, he

Xue’s ten-year-old son, mother, sister-in-law, husband’s nephew, and brother-in-law all corroborated her testimony and vouched for her good character. Her oldest daughter testified that her father

Xue’s testimony about the homicide itself was as follows:

In her testimony, the sister-in-law recounted the defendant’s answer to her question of why she hacked him to death. She said she only wanted to hack his eyes so he could no longer beat her, but once she started hacking, she lost control of herself.

In its holding, the court cited provisions in the 2015 Opinions on mitigating circumstances and justifiable self-defense in intentional homicide. Also citing the victim’s family’s forgiveness and request for leniency as well as Xue’s remorse, the court sentenced her to five years in prison (Decision #1401817, Xinye County People’s Court, Henan Province, April 24, 2015).Footnote ³⁷ This is the only court decision in my Henan sample that cites the 2015 Opinions. My Zhejiang sample likewise contains only one decision that cites the 2015 Opinions. Indeed, according to a keyword search on China Judgements Online, the 2015 Opinions have been cited in fewer than 100 court decisions (as of June 2020). Perhaps because the contents of the 2015 Opinions are largely derivative of other bodies of law, judges and lawyers have chosen to cite the original sources of their contents. Or perhaps judges and lawyers apply its provisions without citing it by name.

Abused women sometimes fear divorce owing to their economic dependence on their abusers (Reference Cheng and GaoCheng and Gao 2019:12; Reference ZhengZheng 2015:176), that is, owing to the economic control of their abusers. Over the years, Yao Rongxiang’s philandering husband had doused her with scalding water, hit her head against the floor, hit her with a steel pipe, and hit her with a beer bottle. One night he announced his decision to divorce her and split custody of their four children. In despair at the prospect of having no means to raise her children, Yao concluded that she no longer had a reason to live. She waited until he was sound asleep before bludgeoning his head with a threaded steel pipe and then cutting his neck with a cleaver. He died on the scene from a craniocerebral injury and massive blood loss. Yao stated that her original plan was to commit suicide, but, after thinking about her children, she decided instead to surrender to the police. In their testimony to the court, the victim’s parents expressed their forgiveness, requested a lenient sentence, and withdrew their original request for civil damages. In her expert testimony to the court, Chen Min stated:

In the only decision in my Zhejiang sample to cite the 2015 Opinions, the court sentenced the defendant to five years in prison (Decision #3431222, Wenzhou Municipal Intermediate People’s Court, Zhejiang Province, March 5, 2015; also cited in Chen and Reference Yang and TanYang [2016:21]).Footnote ³⁸

In addition to illuminating the influence of forgiveness on sentencing, this final example also foreshadows a finding I present in Chapter 10. When couples with minor children do get divorced, siblinged children tend to be split between their parents, often according to arrangements determined by the father.

Summary and Conclusions

As we know, a failed initial divorce petition granted on a subsequent attempt is the divorce twofer’s primary defining characteristic. Insofar as almost everyone who seeks a divorce can eventually get one, some observers might wonder what the harm is in waiting. Perhaps, skeptics might contend, the primary harm is merely inconvenience, which might be more than offset by the benefits some couples enjoy by reconciling after an adjudicated denial.

This book documents the manifold harms of the divorce twofer. If the divorce twofer provides any benefits, they are monopolized by judges in the form of gains to their work efficiency and performance evaluation scores (Chapters 3, 5, and 6). From the standpoint of families, the social and personal harms of the divorce twofer clearly outweigh their benefits. We have seen that a lot of adjudicated denials never return to court (Chapter 6). Perhaps divorce-seekers let down by courts ended up pursuing mutual-consent “divorces by agreement” in the Civil Affairs Administration. Given that abusers, simply by withholding their consent, could activate the divorce twofer in court (Chapters 8 and 9) and prevent Civil Affairs divorces outside the court system, women desperate to divorce grievously sacrificed property and child custody in exchange for their husband’s consent (Reference LiLi 2022). We will see the prevalence of precisely this trade-off in the context of child custody determinations in Chapter 10. Unable to make the financial sacrifices their husbands demand and unwilling to give up custody of their children, some women resigned themselves to staying married.

This chapter documents the mortal harms associated with the divorce twofer: its grave physical security and public health implications. An official justification for the divorce twofer is that it introduces a de facto cooling-off period designed to de-escalate conflicts (Chapter 3). Among battered women who eventually obtained an adjudicated divorce, the divorce twofer typically protracted their exposure to domestic violence by one year or more. Indeed, domestic violence often intensified between divorce attempts. Delays caused by the divorce twofer have swelled the ranks of marital violence refugees who take flight from their abusive husbands. After participating in labor migration, often as an escape route from domestic violence, some women resign themselves to remaining married but separated owing to the logistical difficulties of returning to their residential court jurisdictions to file for divorce and participate in trial proceedings (Reference LiK. Li 2015a:98). The divorce twofer thus puts many women in a sort of divorce purgatory.

This is primarily a rural story. Divorce petitions were concentrated in rural courts. Likewise, the gender gaps in delays to divorce were limited to rural courts. Finally, most of the case examples in this chapter are from rural areas.

Among women who did not or could not take flight, delays caused by the divorce twofer may have done more to prolong and intensify marital abuse than to cool it off. As a consequence, some women unable to take flight may have felt compelled to fight. Judges were delusional if they truly believed they were saving marriages by denying divorces. They professed concern that granting divorces to battered plaintiffs would generate instability and even murders. Even if their concerns were partially valid, the opposite was undoubtedly equally or more valid: denying divorces to abuse victims generated murders and suicides, the very essence of instability. Judges more faithfully fulfilled their ideological mandate to rescue marriages by denying divorces than their legal mandate to rescue abuse victims by granting divorces.

Granting divorces the first time would certainly save lives. To be sure, we cannot blame the divorce twofer for all incidents of violence that follow in the wake of adjudicated denials of divorce petitions. After all, men also beat and kill their ex-wives. Public authorities, however, are undoubtedly less reluctant to intervene in violent incidents between two people who are not married to each other. If an abuser were no longer married to his victim, public authorities would have even flimsier excuses for not intervening. If police were as willing to arrest and punish wife-beaters as they were to arrest and punish drunk drivers, the likely consequence would be fewer men killing their wives and fewer women killing their husbands.

Policies and practices that prevent divorces have done less to create harmonious families than to prolong marital misery and violence. Recall from Chapter 2 that one impetus for the introduction of the breakdownism standard in the 1980 Marriage Law was to prevent homicides that might otherwise result from forcibly preserving acrimonious marriages. By turning the breakdownism standard on its head and using it to deny divorce petitions, courts have transformed civil cases into criminal cases in a way the original drafters of the breakdownism standard feared. Recall also from Chapter 3 that Chinese government leaders have justified the divorce twofer by invoking a widespread claim that divorce contributes to juvenile crime. If juvenile crime is driven as much by marital conflict as by marital dissolution (Reference AmatoAmato 2000; Reference Amato and CheadleAmato and Cheadle 2008), then China’s judicial clampdown on divorce, by forcibly preserving unhappy marriages, may contribute more than divorce itself to juvenile crime, another source of criminal cases.

A considerable number of eminently preventable criminal cases have stemmed from the poor outcomes of women’s prior help-seeking efforts. Women are at risk of sustaining harm when they seek to divorce their abusive husbands. Women are also at risk of inflicting harm on their abusers when their help-seeking efforts are stymied. In trials of women charged with killing their abusive husbands, criminal courts have steadfastly eschewed the concept of battered woman syndrome, and have therefore been averse to acquit – or even to sentence to probation in lieu of prison time – the very women they affirmed to be victims of domestic violence. Nonetheless, I have shown in this chapter that China’s criminal courts have taken domestic violence increasingly seriously, particularly beginning in 2015. Indeed, domestic violence appears to be taken far more seriously in criminal courts than in divorce courts. China’s criminal courts experienced a watershed in 2015, when they started more fully and consistently recognizing domestic violence as a mitigating factor in criminal trials of women who killed their husbands. China’s divorce courts experienced no corresponding watershed. On the contrary, China’s judicial clampdown on divorce intensified after 2015 (Chapter 6).

When women take flight to escape their abusive husbands, they often have little choice but to leave their children at home. As we will see next, judges tend to grant custody to the parent with physical possession of the child even when the child’s living arrangement is the direct consequence of domestic violence.