How Protection Orders Became a Thorny Legal Issue in China
China’s 2015 Anti-Domestic Violence Law (ADVL) was a milestone for the country’s legal community. Even as recently as 2000, no law in China had even used the term “domestic violence.” The ADVL not only signaled a new urgency in combating gender violence; it also adopted internationally proven mechanisms such as personal safety protection orders, which authorize judges to firewall domestic violence victims from offenders.
Yet, the implementation of protection orders has been a mixed bag. On the one hand, the number of orders issued by courts has increased steadily, and the approval rate rose from 52% in 2016 to 73% in 2021, according to official data. But both the number of petitions and the number of approved orders have remained low in absolute terms: between 2016 and 2021, Chinese courts issued fewer than 11,000 protection orders.
According to China’s Supreme People’s Court, courts across the country issued 5,695 protection orders in 2023 — a significant jump from previous years, but still an average of fewer than two orders per court. By contrast, Singapore, a country of 4 million, issues around 3,000 protection orders annually.
The size of this gap made me curious: What explains the limited adoption of protection orders in China? And what factors influence a judge’s decision when handling petitions for protection orders?
These questions can be analyzed empirically, by comparing the issuance of protection orders across different regions. Indeed, there exist enormous regional differences in judicial and popular attitudes toward protection orders in China, with official statistics indicating that some regions, such as Jiangsu in the east or Chongqing in the southwest, process five times as many protection applications as others.
In 2022, I began conducting fieldwork in two courts: Court G, along China’s prosperous coast, and Court S in the country’s less economically developed interior. Between 2019 and 2022, both courts issued more protection orders than the national average, yet the number issued by Court S was still considerably lower than that of Court G. The individuals covered by Court G’s orders were also more diverse, including not only wives but also children, the elderly, and cohabitants. This was despite Court G’s higher workload: its judges handle an average of 330 cases annually, almost one per calendar day.
Through in-depth interviews with judges from both courts, I was able to explore what caused these differences. I found that, while they may be working off the same law, judges’ interpretations can vary widely, and their practice is influenced by local attitudes toward domestic violence, marriage, and social stability.
A case of contrasts
The two courts I studied differed significantly in how they processed protection order applications, including their approach to evidence, the scope of protection, and procedural support.
For Court G, the central question to be answered before issuing a protection order was simple and clear: Is the applicant at risk? Accordingly, applicants did not need to present any evidence when filing a petition for a protection order. Instead, they needed only to complete an application form specifying the most recent violent incident, the most serious one, and the first time the violence occurred.
The head of the court’s Family and Youth Division explained: “For us, the ‘danger’ of domestic violence justifies a protection order. At this stage, we do not need to ascertain whether domestic violence indeed occurred. It is enough that the applicant claims such danger.”
“As long as the information does not seem to be made up, we grant the order,” she added.
By contrast, all protection orders granted by Court S came with police documentation of prior abuse. In other words, the court regarded actual suffering, rather than a realistic threat, as the precondition for issuing protection orders. Moreover, all successful cases contained evidence of medical treatment: receipts for medications, hospitalization costs, or doctors’ diagnoses.
As for the scope of domestic violence, Court G adopted a fairly broad definition, covering most forms of violence. Acts of domestic violence listed in the application form include non-physical harm, such as humiliation and verbal abuse, as well as marital rape, which hasn’t been criminalized in China.
By contrast, Court S treated physical violence as a precondition for issuing protection orders. For example, Court S denied the application of a woman in her late 20s who applied for a protection order after her father-in-law threatened to kill her over a financial dispute. The court stated that the man didn’t know the applicant’s current whereabouts, and that the applicant had neither suffered actual violence, nor was she at risk of potential future violence warranting special protection.
Court G also takes a more proactive approach to handling protection order applications. Acknowledging that some threats are imminent and severe, the court has set up a “green channel,” allowing applications to be accepted without an appointment, even in times of high caseloads. In such cases, the application files can be transferred to the judge on the same day as they are submitted, and a decision will be made within 72 hours.
This “green channel” is absent from Court S, however, and the court rarely coordinates with other official institutions to support the applicant. Instead, the court usually focuses more on mediation until the divorce case is settled, rather than providing shelter for women immediately to prevent imminent harm.
In short, for Court G, the default attitude toward protection order applications is supportive, while Court S exhibits more hesitancy, maintaining that a protection order should only be issued when divorce is inevitable, and when there is copious evidence of prior physical violence. This hesitancy is all the more obvious in cases involving families, as I gradually came to realize.
A family matter
In my discussions with judges from both courts, I paid particular attention to how their views on families, children, and domestic violence shaped their decisions. Our conversations revealed that their different approaches toward protection order issuance were often linked to their perceptions of marriage, gender violence, and the relationship between family and social stability.
Judges at Court G held the view that gender equality and individual rights are not subordinate to the institution of marriage. An official document published by the court vowed to “scientifically and lawfully protect each family member’s legal rights.” While that may not seem extraordinary, the key phrase here is “each family member.” That is, the integrity of the family was not the top concern of the court, and marriages did not take precedence over family members’ rights.
A judge of the court echoed this view in an interview: “Of course, if the divorcing couple can reconcile, we are happy to help them. However, maintaining the integrity of marriage should be balanced against other considerations, most notably, family members’ legal rights.” Another judge from Court G highlighted the urgency of helping women escape from abusive relationships and added that she would even “encourage (women) to make compromises on other less urgent issues, such as property or custody.”
Interestingly, judges in Court G tended to view delays in issuing protection orders as a threat to social stability — long a primary concern for China’s legal system. A senior official of the court explained that “Family stability does not equal social stability. If we tolerate domestic violence, it may lead to malicious incidents, such as women who kill in response to domestic violence. That will actually threaten social stability.”
Judges at Court S put more emphasis on maintaining the integrity of the family. Influenced by traditional norms of marriage that prioritize lineage continuation, they prioritized keeping families together even in cases involving domestic abuse — as long as they believed divorce was not inevitable. In addition, Court S judges often believed that any punishment of abusive partners risked an escalation of the crisis, which it treated as the greatest threat to social stability.
This view was shared by a senior judge at Court S. A man in his mid-40s. He had been an enthusiastic advocate of the protection order mechanism but grew skeptical after some of the orders he issued not only failed to end the dispute, but also got extended family members involved. “Rather than bring an end to the disputes, the orders may add more fuel to the flames,” he told me.
The differentiated views held by judges of the two courts offer insight into how nationwide laws can translate into contrasting practices, but my question remained unsolved. Judges across China receive the same training in ADVL implementation, and the majority of judges in both courts were not locals. So what exactly was behind these localized practices around protection orders?
The price of marriage
One major factor, as it turned out, was caili, or the “bride price.” Simply put, a bride price is the money a man’s family must pay to a woman’s family in order to facilitate a marriage. According to a 2020 survey, almost three-quarters of marriages in China involve bride prices, though specific practices vary widely by region. In some areas, it’s framed as compensation to the bride’s family for losing a member; in others, it’s simply the price of getting married.
Policymakers have repeatedly attempted to stamp out the practice. But it has persisted, and in the process, become a major headache for the country’s courts. In particular, regional differences in bride price practice mean there’s no consensus as to what happens in cases when marriage ends in divorce. A search under the keyword “bride price” on the database of adjudication documents maintained by China’s Supreme People’s Court returns 100,000 hits from 2013 to 2018.
The amount paid in bride price also varies across regions. In the city where Court S is located, I was fascinated by the exorbitant bride prices being paid — and their impact on judges’ attitudes toward domestic violence. The average bride price in this not particularly wealthy agricultural region is 130,000 yuan ($18,000), or about three times the local GDP per capita, based on local media reports. Unlike some other regions, this money is generally held by the parents, rather than passed on to their daughter.
By contrast, in the wealthier City G, the average bride price was only 29,000 yuan, or about 18% of the local GDP per capita, and there was no clear evidence that judges there even took it into consideration in their deliberations.
Court files in City S show that when a high bride price is involved, judges often deny applications for protection orders, even when there is evidence of domestic violence. For example, a woman in City S had an abortion without the consent of her husband’s family, who spent 130,000 yuan on her bride price. When the woman filed for divorce, her husband beat her and several members of her family, demanding the return of the bride price.
Instead of issuing a protection order, the court mediated a settlement, with the woman returning 110,000 yuan. In my conversation with the judge who handled the case, she admitted that “beating (the woman) is extreme,” but emphasized the husband’s family’s expectation that the woman would produce an heir. “The man’s family exhausted all their savings,” the judge told me. “If the woman disappears or demands a divorce, the man’s whole family and all his relatives will certainly put up a desperate fight.”
Judges at Court S sometimes advised petitioners to keep evidence of domestic violence, not for use in getting a protection order, but to negotiate a more favorable divorce settlement. While perhaps practical, this did not eliminate the risk of domestic violence.
The contrasting practices around protection orders in the two cities I studied reflect the disparate approaches taken by judges in different parts of China as they navigate complex webs of family and social practices. It may be tempting to suggest judges with a more impressive record of protection order issuance be relocated to cities struggling with patriarchal marriage culture, but it’s doubtful they would be able to import their interpretations of the ADVL effectively. Instead, a more holistic approach is required to address the immense challenge of creating a society free from gender violence.
Editor: Cai Yineng; portrait artist: Zhou Zhen.
(Header image and in-text icons: Visuals from VCG, reedited by Sixth Tone)
