9.2.1 China’s Long History of Wrongful Convictions

Song Ci in 1247 wrote a book called Collected Cases of Justice Rectified. It is regarded as the first book on forensic science. It cautioned that careful examination is required to avoid “any regrets” in the use of the death penalty (Ci, Reference Ci1247: 10). It warned against tunnel vision or confirmation bias by stating, “the object of re-examination is not to verify against a previous verdict, but to guard against injustice” (Reference Ci1247: 14). Around the same time, Hanqin Guan wrote a play called the Injustice to Dou E That Touched Heaven and Earth. Dou E is wrongfully convicted by a corrupt official on the basis of false confessions obtained by torture and threats to torture her mother-in-law. Her innocence is confirmed after her death by three prophecies that she had predicted before her execution would confirm her innocence: blood raining from the sky; snow in June and a three-year drought. Those responsible for the wrongful conviction were eventually brought to justice. There was an early recognition in China of the dangers of wrongful convictions.

Another ancient story published in 1627 involved a man and a woman who were tortured into confessing to a murder that shocked the community. The local Governor did not take their protestations of innocence seriously. The wrongful conviction of the pair was revealed within a year after the real perpetrator confessed to the killing. This person was also executed. The Governor who presided over the wrongful conviction lost his position because he had not conducted an adequate investigation. Ming Hu and Bin Liang (Reference Hu and Liang2022) have argued that this fictional tale reveals factors such as the use of torture, influence from public opinion, lack of checks between police, prosecutors and courts and lack of defence rights that are still present in China. The ancient texts have contemporary reliance in illustrating the need to be cautious with both confessions and the death penalty. They also reflect desires not to convict those who are clearly innocent and to punish those responsible for convicting the innocent. These are deeply rooted and universal desires, but these ancient writings also reflected a sense of fatalism that wrongful convictions and wrongful executions will occur.

As in Europe (Langbein, Reference Langbein1977), torture has deep roots in Chinese society. It was allowed but regulated from the Zhou dynasty (1066–256 BC) to the Qing dynasty (1644–1912). At the same time, China also used alternatives to immediate execution during this time. Between 1744 and 1840, China had a lower execution rate than England. In both countries, the majority of death sentences were not carried out because of executive clemency (Xiong, Liu and Liang, Reference Xiong, Liu and Liang2022).

There have been estimates of 20,000,000 executions during the Cultural Revolution between 1966 and 1976. Mass trials were held during this time, and the death penalty was often tempered by a two-year delay to allow for re-education. In 1980, the Supreme People’s Court (SPC) estimated that between 17 and 39 per cent of the convictions during the Cultural Revolution may have been wrongful. After the Cultural Revolution, attempts were made to rectify over four million convictions (Zhong and Dai, Reference Zhong and Dai2019: 262). So many innocent people were released from prison camps that arrangements had to be made to allow some of them to stay and continue to work at the camps (Lin, Reference Lin2022: 427). There are ugly predecessors to China’s contemporary wrongful convictions (Muhlhahn, Reference Muhlhahn2009: 193). As in the United States, however, connections between older and newer wrongful convictions are rarely made, and much wrongful conviction discourse is ahistorical.

9.2.2 Three Back from the Dead Cases

Three cases where murder victims were discovered alive are among China’s best-known recent wrongful convictions. Such cases of obvious wrongful convictions have also occurred in the United States. Rob Warden (Reference Warden2013) has documented eleven such cases, including the 1819 wrongful convictions of the Boorn brothers for killing a relative who turned up alive to the 1973 wrongful conviction of Antonio Rivera and Merla Walpole for murdering their three-year-old daughter even though their daughter was found alive in 1975. Warden also identified two nineteenth-century cases where the wrongfully convicted man was hanged before the alleged victim of the murder was found alive (Reference Warden2013: 156–158). One of the men, William Marion, was granted a pardon by the Nebraska governor in 1987 based on innocence (Reference Warden2013: 156–158).

Back from the dead cases are the most obvious forms of wrongful convictions of the innocent. They were DNA exonerations before DNA exonerations. This is relevant to a central theme of this chapter: the highly punitive society of China, like the United States, rations justice severely so that remedies are only provided in the most obvious cases of injustice.

9.2.3 Other Well-Known Cases

Two truckers named Zhang were convicted of murdering a girl they gave a ride to in 2003 even though their DNA did not match that of DNA found under the victim’s fingernails and video showed them leaving the town where they dropped the victim before the time she was killed. The Zhangs were interrogated and tortured for seven days. They were beaten by fellow inmates at the instructions of the police before they signed false confessions. They were exonerated in 2010 after one of them provided police with the name of a serial killer and rapist whose DNA matched those found under the victim’s fingertips. That person had already been executed for a similar crime (Jiang, Reference Jiang2016: 46–48).

DNA exonerations and confessions from the true perpetrator are found in many of China’s remedied wrongful convictions. This makes them obvious cases but also underlines that wrong perpetrator wrongful convictions can, as in the United States, be portrayed as a failure of crime control (Petro and Petro, Reference Petro and Petro2010). It will be suggested in section 9.4 that, China, unlike the United States, has not remedied wrongful convictions for crimes that did not happen.

In November 2001, Chen Keyun and Wu Changlong were convicted of detonating a bomb at a Communist Party office. Both confessed, but claimed they were subject to prolonged torture including fifty-three days of interrogations that resulted in three failed suicide attempts. The prosecutor’s office twice sent the case back to the police because of concerns about insufficient evidence. Although an appeal court found that the facts were not clear and the evidence was insufficient, it also found that the accused were not tortured. They were convicted again at a new trial in 2006. Chen told western reporters that “they treated me as less than a dog. I was an old Communist Party Cadre who had been about to retire, I never thought that something like this could happen” (Associated Press, 2013). On further appeal, they were acquitted in May 2013 (Jiang, Reference Jiang2016: 53–55). The case attracted media attention because of the rarity of terrorism charges and the men’s continued claims of innocence. Some of the families of the accused were temporarily detained as a result of their petitioning. After the eventual acquittal, the Propaganda Bureau restricted media coverage, but there was eventual compensation to the two men (US-Asia Law Institute).

The more powerful are not immune from wrongful convictions in China. A police officer, Du Piew, was convicted of murdering his wife, who was also a police officer and another police officer in 1999 and was sentenced to death. Later that year, his convictions were affirmed, but because of doubt about his guilt, his sentence was reduced to a death with a two-year reprieve. He was released the next year after twenty-six months in prison when the real perpetrator confessed to the crime and ballistics evidence linked that person’s gun with the two murders (Li, Reference Li2016: 121).

In 1998, Yu Yingsheng was convicted of killing his wife on the basis of a fifty-page confession obtained through torture. He made forty-one different and often conflicting statements while in police custody. The police disregarded semen and fingerprints found at the crime scene that did not belong to him. He was only declared to be innocent on a retrial in 2013 after five law professors reviewed the evidence and declared it to be insufficient, prompting the Supreme People’s Procuratorate to reopen proceedings, which led to an acquittal after which the true perpetrator was identified through the DNA evidence (He, Reference He2014: 492–493; Jiang, Reference Jiang2016: 55–56).

Huugjilt (as known as Huge Jiletu), an eighteen-year-old Mongolian person, was executed in 1996 for a rape/murder after he found a woman murdered in a public washroom and summoned the police. He confessed to the crime after police interrogation and torture. He was executed sixty days after his conviction (He, Reference He2016: 184). A serial rapist confessed in 2005 to the crime. From 2006 to 2014, Huugjilt’s parents campaigned for his innocence, and his case received much media attention. Only in 2014 and after some official recognition of the reality of wrongful convictions in China did the High People’s Court of Inner Mongolia “declare his innocence” (Reference He2016: 189) with his parents being given less than $5,000 in compensation. Two years after the conviction was finally overturned, twenty-six officials involved in his case were given demerits or sanctions. The officials included eleven former police officers, seven former employees of the prosecutors’ office and eight former employees of the courts. He Weifang, a law professor in China, explained that “the people punished are not the real decision-makers, so the authorities can only give them light punishments” (New York Times, 2016). At the same time, the Chinese state reported in 2022 that the punishment of twenty-six officials in this case represented one of the “largest punishments for those involved in wrongful convictions” adding that over 500 prosecutors had been held accountable for their role in wrongful convictions (Sixth Tone, 2022). The punishment of those involved in wrongful convictions could be defended as an attempt at accountability, but it may also deter the recognition and remedying of wrongful convictions except in the most obvious cases.

More recently revealed wrongful convictions include the case of Wu Chunhong who was first convicted of murdering his children in 2005 and was convicted three more times on retrials before he was acquitted after a fifth retrial. As in the United States, multiple legal proceedings are often necessary before wrongful convictions are corrected. Although he confessed after being tortured outside of a detention centre, he maintained his innocence and mailed more than 600 petition letters. The charges were laid during a campaign to clear murder charges and the police did not investigate alternative suspects. He was eventually compensated 3.14 million RMB for the 5,612 days he served in prison (US-Asia Law Institute).

Chen Xiaying and two others were convicted of the kidnapping and killing of an eleven-year-old boy. They were first convicted in 1998 after the police had detained and tortured not only Chen but also his alibi witnesses. They were convicted on two more retrials that were ordered because of unclear facts and insufficient evidence. Chen was only acquitted in 2015 after his lawyer found a new alibi witness. He received 5.37 million RMB in compensation (US-Asia Law Institute).

Zhang Yuhan was interrogated and tortured for six days before making a false confession to the murder of young children in his village. The police had investigated all sixty-one households in his village and found scratches on Zhang. The police believed that Zhang had acted out of revenge for a prank the young victims had played on him and did not investigate four other alternative suspects. After extensive pre-trial detention and torture, Zhang confessed and was convicted and sentenced to death with a two-year reprieve. He won a retrial on appeal but was convicted again on a second trial in 2001. After volunteer lawyers found new evidence, an appeal court ordered a third trial in 2018, and he was finally acquitted and released in 2020 after twenty-six years in prison. He claimed over 20 million RMB and received 5 million RMB (US Asia Law Institute). His wrongful conviction was reported in the western media including the New York Times, which noted that in 2019, China had a 99.9 per cent conviction rate (Mozur, Reference Mozur2020). The British Broadcasting Company also reported his wrongful conviction and described him as China’s longest-serving wrongfully convicted inmate who has served 9,778 days in prison. The High Court judge who acquitted Zhang Yuhan explained, “After we reviewed the materials, we found there is no direct evidence that can prove Zhang’s conviction. So, we accepted the prosecutors’ suggestion and have declared Zhang innocent” (British Broadcasting Company, 2020). As in the United States (Roach, Reference Roach2013b), remedied wrongful convictions in China typically involve several rounds of litigation before the wrongful conviction is remedied. There are some similarities between Chinese rule by law and American extralegalism.

The vast majority of remedied wrongful convictions have involved male accused. One seventeen-year-old female, Qian Renfeng, was wrongfully convicted in 2002 of murdering a child who died from contaminated food at the school. She was harshly interrogated for twelve hours and made a false confession. The police did not pursue two suspects that the owner of the child care centre told them about. Qian Renfeng was acquitted on a retrial in 2015 that was only held after her lawyer publicized the case online. She received 1.72 million RMB in compensation and an apology from the court for the over 5,000 days she was imprisoned (US Asia Institute, Qian Renfeng).

9.2.4 Studies of Wrongful Convictions

The government-sponsored Chinese press has reported that between 2014 and 2019, China had sentenced 7.1 million accused while remedying forty-three wrongful convictions involving sixty-three different accused. The story stressed that all sixty-three wrongfully convicted people had received compensation. Zhou Qiang, president of China’s highest court, was quoted as declaring that Chinese courts are “learning a lesson from wrongful convictions” (Yin, Reference Yin2019a). As will be seen later, the state-controlled media in China has taken a special interest in portraying China’s record and response to wrongful convictions in a favourable light.

Yu Mou has examined a number of studies of remedied wrongful convictions in China. One 2014 study of forty wrongful convictions included four cases where the deceased victim was actually found alive, and twenty-five other cases involved confessions by the real perpetrator. In another 2015 study of thirty-four remedied wrongful convictions, twenty-one cases involved the real perpetrator’s confession (Mou, Reference Mou2020: 3–5). These studies illustrate how definitive evidence of innocence is often required in China to remedy a wrongful conviction.

In a 2021 study of forty-five wrongful convictions in capital cases, sixteen involved a confession by the real perpetrator, and in three cases, the victim came back from the dead. Only two cases involved prosecutors withdrawing charges after an appellate court ordered a re-trial. The author of the study, Professor He Jiahong of Renmin University, concluded that all forty-five wrongful convictions revealed a “shortage of evidence” in terms of its quantity and quality with most involving false confessions. As will be seen in Chapter 10.2, a similar pattern is seen in India, where remedied wrongful convictions often feature false confessions and an absence of other physical or forensic evidence linking the accused to the crime.

Zhong and Dai (Reference Zhong and Dai2019: 267) reported on 141 wrongful convictions involving 206 accused. Consistent with an emphasis in China and the United States on proven innocence, these researchers only included wrongful convictions where “factual innocence” was established (Reference Zhong and Dai2019: 265). In 115 of these cases, the victim died and the “rest of the cases concerned rape, robbery, assault, and setting a bomb causing grave injuries to victims.” In 33 per cent of the cases, the wrongly convicted accused was sentenced to immediate death, and in 28.6 per cent, they received a suspended death sentence. The death penalty has dominated remedied wrongful convictions in China. Both guilty pleas and less serious cases are generally absent from China’s remedied wrongful convictions.

Torture by the police was a factor in 122 or 86.5 per cent of the 141 remedied cases. The prevalence of police torture as revealed in remedied wrongful convictions in China and India suggests that police reforms should be a priority in those jurisdictions. At the same time, over 80 per cent of the wrongful convictions occurred between 1991 and 2005. Delays in providing remedies make it difficult to make confident determinations whether reforms enacted in response to well-publicized wrongful convictions between 2010 and 2013, including an exclusionary rule for involuntary confessions, have been effective.

Only 14 of the 141 remedied cases involved mistaken eyewitness identification, and 19 involved “inconsistent witness testimony” (Zhong and Dai, Reference Zhong and Dai2019: 270), thus suggesting that many of the remedies discussed in Chapter 3 targeting immediate causes of wrongful convictions other than those for false confessions may not be a priority in China. In addition, only 6 of the 141 wrongful convictions involved DNA, underlining the relatively “low tech” nature of Chinese wrongful convictions, something that is also the case in India. Inadequate defence representation was not even recorded as a cause even though defence lawyers were present in about two-thirds of the cases (Reference Zhong and Dai2019: 274). It is difficult to overstate the marginal and precarious role of defence lawyers in China.

Another striking finding of this study is that in the majority of cases, the accused faced three trials or more, underlining the complex and arduous process of correcting wrongful convictions and the initial reluctance of Chinese courts to recognize and remedy wrongful convictions (Reference Zhong and Dai2019: 267–268). This underlines yet another similarity with the American approach, which, as discussed in Chapter 6.5, often involves multiple post-conviction proceedings in both states and Federal courts. The extensive use of multiple re-trials in remedied wrongful convictions can be seen as a legalistic display of rule by law with courts being reluctant to recognize wrongful convictions without clear signals from those in charge. That said, Zhong and Dai (Reference Zhong and Dai2019: 272) record that the Communist Party’s Political Legal Action Committee only intervened in 9 of the 141 cases though this may underreport such interventions given their reliance on media sources. They were only able to discover court decisions in 11 of the 141 cases, underlining a lack of transparency in the Chinese judiciary. Multiple post-conviction proceedings in both the United States and China demonstrate less trust in courts than in the English system, where the accused often only has one level of appeal from a conviction (Roach, Reference Roach, Huff and Killias2013a).

A 2022 study of 103 remedied wrongful convictions in death penalty cases found that most of the wrongly convicted were of low economic status and from a rural area. Ninety per cent of the cases involved murder in the course of a rape or a robbery. Although there was only one exoneration between 1983 and 1995, there were 3.67 cases per year between 1996 and 2011 and 7.83 cases per year between 2012 and 2018 (Lu et al., Reference Lu2022: 541). The researchers found that 94 per cent of the cases involved torture by the police and only one case involving a false confession without police torture (Reference Lu2022: 543). The interrogations producing the interrogations lasted on average eleven days. After 2012, all of the confessions were recorded, but sometimes the recordings were edited (Reference Lu2022: 544). This underlines that the recording of confessions alone will not prevent false confessions. In 20 per cent of the cases, the police were sanctioned for their conduct (Reference Lu2022: 548). A 2020 study in the United States similarly found 19 per cent of American police involved with wrongful convictions had been disciplined, though often for a pattern of misconduct and not for involvement in wrongful convictions. In the United States, thirty police officers but only two prosecutors involved in exonerations were convicted of crimes (Gross et al., Reference Gross2020: 115–122). Perversely, the willingness to sanction even a minority of those involved in remedied wrongful convictions may deter recognition of wrongful convictions.

A 2024 study examined 355 remedied wrongful convictions between 1990 and 2010. It found that in 81 per cent of those cases, the wrongful convictions were not remedied until the chair of the relevant provincial political legal committee had exited office. This measures correlation and not causation and may also reflect normal turnover among governing elites. Yet the same researchers found that the chances of wrongful convictions being corrected were lower if the chair of the political legal action committee was the direct subordinate of the former chair (Li and Li, Reference Li and Li2024). This underlines the role that local politics may play in the correction of wrongful convictions in China. One American parallel may be the close to 800 exonerations by the end of 2024 that involved conviction integrity units, which are often established as a reform mechanism by a newly elected prosecutor. Na Jiang had earlier and bravely argued that political legal committees “are just a branch of CCP bodies rather than judicial institutions” and noted that they are often instrumental in remedying wrongful convictions (Jiang, Reference Jiang2018a: 23), Professor Na Jiang notes that “convicted people who insist on their innocence often seek” assistance from local Political Legal Committees and other “unprofessional political-legal institutions” (Jiang, Reference Jiang2016: 225). This suggests that remedied wrongful convictions in China may often have, below the surface of appeals and multiple retrials, a populist and political dimension that in western democracies is generally associated with executive clemency.

9.5.1 False Confessions, Recording and the Exclusionary Rule

In 2010, five judicial and prosecutorial authorities promulgated regulations providing for the exclusion of involuntary confessions. The regulations empowered not only courts but prosecutors to exclude such confessions including those obtained by torture. An exclusionary rule for involuntary confessions was included in Article 54 of the 2012 Criminal Procedure Law. The amendments were complex and affected nine articles of the new criminal procedure code (Kielsgard, Reference Kielsgard2020: 388). These reforms were influenced by research on the use of the exclusionary rule in the United States and elsewhere (Guo, Reference Guo2019).

The Chinese exclusionary rule was a top-down reform that was done without police or defence lawyering reforms. Many of the measures taken in China to prevent wrongful convictions have been directed at prosecutors and courts, not the police. Margaret Lewis (Reference Lewis2010: 636) has suggested that the Communist Party introduced the exclusionary rule mainly to preserve the integrity and stability of the government. This also ties in with arguments made below that China’s wrongful conviction reforms may have been more about legitimating the system in the wake of obvious and high-profile wrongful convictions than actually preventing them from happening.

Writing in 2014, professor Na Jiang expressed some optimism that the 2012 reforms by requiring the recording of statements and allowing for defence lawyers to challenge confession evidence, could “cast doubt on the conclusions of the police and prosecutors” (Jiang, Reference Jiang2014a: 416). Renmin University professor He Jiahong (Reference He2014) also saw the rule as progress but warned about the gap between the law in the books and the law in action, especially given that torture had been illegal in China since at least 1979 and that the new law allowed for the admission of physical evidence derived from involuntary statements. By 2018, however, Na Jiang concluded that the exclusionary reforms had not been successful, with low rates of exclusion and convictions still occurring in cases where evidence is excluded (Jiang, Reference Jiang2018b: 6–7). Even if torture was eliminated in detention centres, there were concerns that suspects would be tortured when held in temporary custody in police stations (Yin, Reference Yin2017). As will be seen in Chapter 10.2, there are similar concerns in India, and a 2023 law makes it easier for suspects to be transferred in and out of police custody.

Most studies suggest that Chinese courts have been reluctant to exclude statements, and even when they do exclude statements, the accused is still convicted (Lin, Chen and Wang, Reference Lin, Chen and Wang2019). One study has, however, found only one successful exclusion claim from 2009 to 2012 but thirty-five between 2016 and 2018. At the same time, few cases of exclusion result in complete acquittals with partial reversals and reductions of sentences being a much more frequent remedy (Kielsgard, Reference Kielsgard2020: 400–402).

The transplant of exclusion of improperly obtained evidence from western democracies to China was fraught because the police in China were part of the military. They often lacked investigative skills and capacity and so had to rely on confessions (Lin, Chen and Wang, Reference Lin, Chen and Wang2019). The absence of an independent defence bar in China was also a factor (Jiang, Reference Jiang2016: 163; Lewis, Reference Lewis2010). Gao Yong (Reference Yong2024: 222) has found that, while according to inquisitorial approaches both prosecutors and judges can exclude involuntary confessions on their own initiative, practically there must, as in adversarial systems, be an application by a defence lawyer. Another factor limiting the exclusionary rule’s effectiveness was that the Criminal Procedure Law was contradictory in providing a right against self-incrimination alongside an obligation to answer questions truthfully (Jiang, Reference Jiang2016: 112). Finally, even when applied, the Chinese exclusionary rule was not absolute. It continued to allow for the introduction of any physical evidence discovered as a result of an involuntary confession. The Chinese exclusionary rule, like the American one, is subject to many exceptions, and it emerges more as a symbol than a guarantee of due process. The American exclusionary rule has not prevented false confessions (Leo, Reference Leo2008: 280), and the same is likely true of the Chinese exclusionary rule.

9.5.2 Death Penalty Reforms

As in the United States, much wrongful conviction reform has centred around the death penalty. In 2006, the Standing Committee of the National People’s Congress restored mandatory review of death penalty cases by the Supreme Peoples’ Court (SPC). This was only a year after the first two famous back from the dead wrongful convictions were revealed. In any event, the SPC had inadequate resources to conduct such reviews. It rarely held oral hearings despite regulations requiring such hearings (Jiang, Reference Jiang2016: 153–155). As with the underfunding of the English CCRC discussed in Chapter 5.5, paper reviews reflected a lack of resources and undoubtedly missed some wrongful convictions (Reference Jiang2016: 165).

In 2013, the Chinese government announced a 50 per cent decline in executions from 2008 with an increased use of suspended death sentences (Miao, Reference Miao2024: 329). The exact number of executions, however, remained a state secret. There was a danger that even in cases of suspected wrongful convictions, the court would simply suspend the immediate execution of the death penalty while still upholding the conviction in order to avoid triggering possible accountability measures for the lower courts who may have participated in a wrongful conviction. This underlined the importance of not only examining the law in the books but also the law in action. Professor Yu Mou (Reference Mou2020, 7) of the School of Oriental and African Studies in London has described the fact that both She Xianglin and Zhao Zuohai were not executed because of suspensions of their death sentences as “an application of the implicit criminal law policy ‘extenuation for reasonable doubt’ (zuiyicongqin).”

The 2012 Criminal Procedure Law increased the role of lawyers in death penalty reviews at least on paper. This resembled in some respect provisions in the American 2004 Innocence Protection Act, which provided for increased funding for defence lawyers and prosecutors in capital cases. There is a danger in both countries that wrongful conviction reforms will help legitimate the use of the death penalty by trying to avoid the embarrassing spectacle of executions of those who have been proven innocent through inconvertible evidence.

Moulin Xiong and Michelle Miao published an analysis of “122 death-sentenced innocents, of which 109 have been exonerated and five have been wrongly executed” (Xiong and Miao, Reference Xiong and Miao2018). The authors compared this record with American cases including the wrongful execution of Cameron Todd Willingham for killing his children on the basis of flawed fire expert evidence. They also cited the English wrongful executions of Timothy Evans, Derek Bentley and Mahmood Matten examined in Chapter 5.3. They were adamant that their research was empirical only and should not be “used for evidence for criticizing the criminal justice system” (Reference Xiong and Miao2018: 290). Their data set taken from an examination of Chinese decisions and media mirrors the many death row exonerations in the United States. In both countries, there are plausible claims that death row exonerations may have had a role in reducing the number of executions, though caution is warranted given the lack of official data in China about such numbers.

As in the United States, death row exonerations in China often required “repetitive appeals,” years of incarceration and “cycles of hope and despair” (Reference Xiong and Miao2018: 312). Those who received death row exonerations in China were convicted on average 3.68 times in suspended death penalty cases and 4.40 times in immediate death penalty cases before their exoneration. They served 6 or 13.6 years respectively in prison (Reference Xiong and Miao2018: 313–315). Multiple proceedings within different court systems are a common feature of both American and Chinese legal systems and can be contrasted with other systems, including England, that often allow only one level of appeal within the court system (Roach, Reference Roach, Huff and Killias2013a).

There are some differences in the common and extensive death penalty litigation in both China and the United States. The Chinese system retains the willingness of inquisitorial systems to allow re-trials. This explains why death row exonerees were convicted about four different times before being exonerated. In the United States, the multiple legal proceedings are mainly explained by successive post-conviction litigation in both state and Federal courts and the frequent holding of errors to be harmless. Re-trials are rarer in the United States and not always successful with fifteen of the first 250 DNA exonerates being convicted on a retrial before their eventual exoneration (Garrett, Reference Garrett2011a: 197).

Suspended death penalty cases doubled in the post-2007 era. This prevented wrongful executions but likely allowed wrongful imprisonment. Moulin Xiong and Michelle Miao (Reference Xiong and Miao2018) concluded that “instead of the presumption of innocence, the sentencing reduction” was being used to remedy wrongful convictions (297, 305). They related this to an unwillingness of the courts to admit error and a traditional principle of “lighter sentences for unclear convictions” (Reference Xiong and Miao2018: 311). In this way, Chinese courts were more lenient than American Federal courts, which as discussed in Chapter 6.5 have prohibited accused from arguing residual doubt at the penalty stage of their capital cases (Oregon v. Guzek, 546 U.S. 517, 523 2006).

There are some troubling signs from the available evidence that the SPC may be more frequently approving death sentences as Xi asserted power and China was becoming more authoritarian. One study of 650 published death sentence review transcripts from 2013 to 2016 found only ten cases where the death sentence was disapproved compared to 381 offenders whose murder convictions were approved and 206 offenders whose death sentences for non-murder offences such as robbery and drugs were approved (Xiong, Liu and Liang, Reference Xiong, Liu and Liang2022). As will be examined in chapter 10.4, Indian courts have much more frequently overturned death sentences.

A legal process approach demands consideration of institutional alternatives. Professor Moulin Xiong (Reference Xiong, Pascoe and Novak2021) of China’s Southwestern University has bravely called for China to allow executive clemency from death sentences as is required under Article 6(4) of the International Covenant on Civil and Political Rights. He noted that executive clemency was used in China before the Communist Party. Similar to the United States experience with executive clemency as discussed in Chapter 6.4, amnesty laws in Communist China have been driven by political concerns as opposed to concerns about possible wrongful convictions and innocence. Professor Xiong notes that these laws and the idea of more executive clemency have not been popular in modern China in part because of a saying that “to show mercy to the enemy is to show cruelty to the masses” (Xiong, Reference Xiong, Pascoe and Novak2021: 177). He concluded, “As a result, hundreds of thousands of murderers have been executed without exception in modern China” (Reference Xiong, Pascoe and Novak2021: 181).

9.5.3 2013 Guidelines

In 2013, a variety of guidelines and directives were issued responding to the remedied wrongful convictions and laying out policies in an attempt to prevent similar ones in the future. In July, the Central Committee of the Communist Party issued a “milestone guideline” (Jiang, Reference Jiang2016: 168) on preventing wrongful convictions on the heels of recently revealed wrongful convictions and statements of the Party’s then new General Secretary Xi Jinping about the need to prevent wrongful convictions. It contained fifteen specific provisions accompanied by the following preamble:

The first directive responded to the prime role of false confessions by requiring interrogations to be recorded and only conducted in detention centres. This was backed up by the fourth and fifth directive emphasizing the role of prosecutors in supervising police arrests and investigations and the third directive stressing that “confessions obtained through torture or other illegal means, or witness testimony or victim statements obtained through violence, threats, or other illegal means may not be used as the basis for conviction.” This message against the use of torture to obtain false confessions was re-enforced in the thirteenth directive providing: “Acts such as coercion of confession through torture, use of violence to obtain evidence, and withholding or fabricating evidence shall be severely punished in accordance with the law.” China’s punitive penal culture was turned inwards towards those responsible for wrongful convictions. This ignored, however, that torture had been a crime in China since at least 1979 (He, Reference He2014). The criminal law itself was not applied to state actors who used torture as an investigative method until the Central Political-Legal Committee so instructed.

Other parts of the Central Committee’s directive told courts not to convict based on evidence that was not “presented, examined, and cross-examined in court and verified through these court investigation procedures” and not to convict on the basis of uncorroborated confessions. The directive recognized the use of suspended death sentences in cases of doubt and instructed prosecutors and courts to respect the presumption of innocence and stated, “standards cannot be lowered to issue a verdict with a ‘margin of error.’” On its face, this respected the reasonable doubt standard but raised questions about the differences between the law as articulated in the directive and the law in action.

The eighth directive warned both prosecutors and courts not to act illegally “due to pressure caused by public-opinion hype, disruptive petitioning by a party to the case or their relatives, ‘case-solving deadlines,’ or local ‘stability preservation.’ This constructed public pressure as a cause of wrongful conviction even though public criticism, especially in high-profile cases, could also be a factor in remedying wrongful convictions.

Only the ninth directive addressed defence lawyers. It instructed prosecutors and courts to hear defence counsel “according to law” and explain why they either reject or accept defence arguments. The 2012 Criminal Procedure Law provided more robust rights for defence lawyers with Article 35 providing that their duties “are to present materials and opinions regarding the innocence of a criminal suspect or defendant….” Article 35 is consistent with the idea that China and the United States shared a focus on proven innocence. At the same time, Article 42 of the 2012 law criminalized the concealing, destruction or forging of evidence by defence lawyers as well as any other interference “with the procedures of judicial authorities.” This combined with the retention of the infamous Article 306 of the Penal Code with respect to obstruction of justice allowed for “retaliatory prosecution” of defence lawyers and meant that their “actual status is still vulnerable in China’s practice” (Jiang, Reference Jiang2016: 166–167). As always, there is a need to be attentive to differences between law in the books and law in action.

The final directive recognized the centrality of the Political Legal Committees by instructing them not to coordinate authorities in cases where “the facts are unclear or the evidence is insufficient” and “generally refrain from making specific opinions regarding the determination or substantive handling of a case.” This directive has been re-enforced by subsequent directives. It was, however, a double-edged sword because intervention by such committees in individual cases could both cause and remedy wrongful convictions.

In 2013, the Supreme People’s Court issued an opinion or general directive “for preventing unjust, false and wrongly decided cases” (Jiang, Reference Jiang2016; SPC, 2013). The first directive stressed the importance of not convicting the innocent. The second directive echoed the Central Committee’s directive in instructing courts not to respond to “to stirred-up public opinion, parties petitioning and causing a commotion, local ‘stability maintenance’ or other such pressures.” Both directives were partly a response to public outrage at well-publicized convictions of the clearly innocent, but also condemnation of public outrage at crimes that may have played a role in the wrongful conviction. This ambivalent response to public opinion – something that Susan Trevaskes (Reference Trevaskes, Hood and Deva2014) has aptly called “fear and loathing (Party fears of the masses’ loathing)” – underlines how the directive itself was an instrument to assert the state’s control of the public.

Article 6 repudiated the oft-acknowledged practice of using doubts about guilt to mitigate a punishment by stating, “a judgment that ‘leaves room for error’ must not be handed down.” At the same time, the increased use of suspended death penalty sentences and declines in immediate death sentences provided some support for the idea that this practice continued and helped reduce the use of the death penalty even while condoning and partially legitimating wrongful convictions. The fourth and eleventh directives were also denials of the reality of the law in action by stating that the Supreme People’s Court should “adhere to the principle of open trials” and “courtroom hearings shall be the centre of the trial.” The idea of open hearing was used to legitimate what often remained a closed inquisitorial process where even judges who did not hear the evidence might still have a role in the judicial decision (Jiang, Reference Jiang2016).

Other articles in the SPS guidelines responded to the facts of the remedied wrongful convictions. Article 8 stressed the need to exclude not only confessions obtained by torture but also those that were not recorded or conducted in designated detention centres.

Article 9 stressed the importance of using DNA and fingerprints and not just witness accounts to verify the truth including the identity of the deceased. In an allusion to the back-from-the-dead cases, it underlined the need to confirm the victim’s identity, if possible, by fingerprints or DNA. Yu Mou’s (Reference Mou2020: 117) empirical study of the processing of cases in China revealed a very limited use of forensic evidence with one prosecutor stating, “Our forensic techniques are really weak. Ordinary evidence such as fingerprinting is rarely used in the court. We rely too heavily on the suspect’s statements” and another prosecutor stated that even in murder cases, “we rarely use DNA evidence” (Reference Mou2020: 142). As will be seen in Chapter 10.5, India enacted a law in 2023 that in an attempt to improve both the efficiency and accuracy of prosecutions requires forensic investigations in serious cases. At the same time, American innocence projects have recognized from the start that the idea of scientific certainty was an illusion (Scheck, Neufeld and Dwyer, Reference Scheck, Neufeld and Dwyer2000). China has attempted to improve the regulation of forensic science, but many such services are located with the police and suffer some of the same oversight and validity challenges as forensic science in western democracies (Jiang, Reference Jiang2014b).

Finally, the 2013 directive contained its own version of the 1996 American law that tried to speed up death penalty cases by restricting successive uses of the federal habeas corpus claim. Article 18 urged the SPC not to send cases back for retrial “where the facts of an initial judgment are unclear and the evidence insufficient.” If followed, this could result in speedier exonerations with less use of repetitive litigation. At the same time, such litigation may have been necessary to provide time for political legal committees to decide if they were prepared to remedy wrongful convictions.

Some of the background to the 2013 directive was provided by an article by Shen Deyong, the Executive Vice-President of the SPC. He quoted General Secretary Xi Jinping’s concerns about wrongful convictions and argued “that reliance should be placed on the leadership of the party in efforts to prevent unjust, false and wrongly decided cases….the Party’s leadership is an important political guarantee that judicial trial work will be done well” (Deyong, Reference Deyong2013). He warned that “a single unjust, false or wrongfully decided case” could “greatly shake the public’s faith in the rule of law” (Reference Deyong2013). In his view, wrongful convictions “tend to be the consequence of following orders, abandoning principles or sloppy work and dereliction of duty” (Reference Deyong2013). He then seemed to concede that the Chinese judiciary fared poorly in comparative perspective by stating, “In the west, judges are synonymous with fairness, and we also feel that judges are the embodiment of fairness, are the guardians of fairness and justice, and if the guardian become the assailant, the professional stigma cannot be washed off in a lifetime” (Reference Deyong2013). He called DNA testing “truly a blessing and a curse” because of its ability to convict the guilty but also reveal wrongful convictions.

Shen Deyong praised defence lawyers as “an important member of a joint corps of legal professionals, are allies of the people’s courts, are an irreplaceable and important force in bringing about fair trials and effectively preventing unjust, false and wrongly decided cases.” This statement, however, ignored the marginal role of defence lawyers and their vulnerability to prosecution. He also seemed to absolve prosecutors of their responsibility by stating, “they will instinctively pay more attention to evidence of guilt and aggravating sentences, and this is understandable” (Reference Deyong2013).

After the various changes between 2006 and 2013, wrongful conviction reforms in China seem to have lost momentum. Shen Deyong’s (Reference Deyong2013) speech was greeted by a speech by a deputy chief prosecutor Zhu Xiaoqing, who disagreed with “comrades” who suggested that preventing wrongful convictions may justify allowing “some cases to slip through the cracks.” He also defined wrongful convictions narrowly as cases of factual innocence where authorities “got the wrong guy” and seemed to equate the prosecutorial standard of “the facts are clear and the evidence is credible and sufficient” with proof of guilt beyond a reasonable doubt (Zhu, Reference Zhu2013).

In 2018, the SPC stressed “three rules” involving pre-trial conferences, investigations and exclusion of illegally obtained evidence. It mentioned the need for both prosecutors and defence lawyers to disclose evidence that could prove the accused’s innocence. It also re-enforced the idea that political-legal committees should not intervene in individual cases, but as suggested earlier, this could also inhibit remedies for wrongful convictions.

The only follow-up directives by Chinese officials on wrongful convictions that I have discovered was one issued in August 2021 by the Communist Party’s Central Committee. Others have detected less activism in China with respect to wrongful conviction reforms (Said et al., Reference Said2025: 135). The Reference He2021 directive repeated the calls in the 2013 directive “to promptly detect and correct illegal acts during investigations,” such as extracting confessions through torture and illegally obtaining evidence, to prevent wrongful convictions (Zekun, Reference Zekun2021). It was accompanied by a story in the English language China Daily, noting that the annual report of the Supreme People’s Court reported 656 acquittals in the last year in public prosecution cases and 384 acquittals in private prosecutions including the reversal of the wrongful conviction of Zhang Yuhuan after twenty-seven years in jail. It also reported that the man had “received about 4.96 million yuan ($764,000) in compensation from the State” (Reference Zekun2021). This report conflated acquittals at trial with the correction of wrongful convictions. Reporting acquittals as remedied wrongful convictions may demonstrate sensitivity to China’s conviction rate remaining very close to 100 per cent, with one recent study reporting a 0.005 per cent acquittal rate (Miao, Reference Miao2024: 306 fn 52). In any event, some important recent developments in China, namely the encouragement of guilty pleas, have increased the risk of wrongful convictions in the name of increasing the efficiency of its criminal justice system.

9.11.1 Taiwan

Some remedied wrongful convictions in Taiwan are similar to those remedied in China. Five of thirteen involved false confessions but three of these were obtained before Taiwan adopted, as part of its transition to democracy, Miranda-style protections of the right to counsel and recordings of interrogations (Chin and Lin, Reference Chin, Lin and Robins2023). That said, police in Taiwan avoid recording and Miranda warnings by conducting pre-interrogation interviews or Fantan which supposedly do not involve custodial interrogation. This follows American studies demonstrating that Miranda has been circumvented and does not provide significant guarantees against false confessions (Leo, Reference Leo2008). Despite this, one recent commentator has concluded that the reforms “have successfully suppressed the use of physical coercion during police interrogation … the movement from coercion to deception represents a triumph of the rule of law since police in Taiwan have become oriented to the legal norms of due process and human right protections. Police generally no longer resort to physical violence or other highly coercive interrogation tactics” (Chien, Reference Chien2022: 35, 43).

Some of Taiwan’s remedied wrongful convictions have not been nearly as obvious or clear-cut as those remedied in China. In Lu Chieh-Min’s case, a wrongful conviction was overturned because of concerns about the accuracy of teeth marks found on the victim (Chin and Lin, Reference Chin, Lin and Robins2023: 219). Wrongful convictions have also been reversed because of concerns about mixed DNA samples, drug tests, the use of polygraphs and errors in cell phone tracking (Reference Chin, Lin and Robins2023: 220–222). This is much closer to the full array of remedied wrongful convictions in democracies than China’s remedied wrongful convictions based almost entirely on false confessions obtained by police torture.

At the same time, concerns have been expressed that ten of sixty-two capital convictions in Taiwan were “seriously flawed.” In almost half of the cases, the court made reference to the accused’s prior convictions and courts did not fully explore the accused’s claims of self-defence, mental disorder or police coercion(Chang and Hoyle, Reference Chang and Hoyle2019: 14). Taiwan executed its first person in 2025 since 2020 despite concerns that it was not clear that his conviction had been unanimously affirmed in all courts as required by a 2024 decision of its Constitutional Court that stopped short of prohibiting the death penalty while prohibiting the execution of those with mental health conditions (Lehrfreund and Hoyle, Reference Lehrfreund and Hoyle2025).

Taiwan has amended its Criminal Procedure Law to expand new evidence that can be considered in challenging a conviction and to require a new trial to be granted if there was a “reasonable probability” of a different result. This is more similar to the standards used by the Scottish and Canadian commissions examined in Chapter 5.6 than a standard of proven innocence often used in both China and the United States. In 2016, the Taiwan Innocence Project also had provisions for new DNA testing added to the Code. In 2020, the law was broadened to allow a person petitioning for a new trial to point to areas where the court can investigate in an attempt to find new evidence (Code of Crim. Proc. art. 429–433). These are signs of bottom-up democratic law reform not found in mainland China.

In 2016, a group of senior prosecutors at the request of the Ministry of Justice was formed to review death penalty cases in part to prevent wrongful convictions. Like similar prosecutorial committees that vet the use of jailhouse informants in Canada, there is no avenue for defence participation in such reviews. A year later, however, forensic experts, scholars, lawyers and retired judges were added to the Review Committee with provisions for receiving opinions including from human rights and bar organizations. The decisions made by the Committee are submitted to prosecutors and are final with no appeal (Rong Gong-Li, Reference Rong and Cohen2019). This stops short of a Criminal Cases Review Commission but may be a step in the right direction that is not found on the mainland.

As discussed in Chapter 4.3, plea bargaining has largely been rejected in Taiwan since it was introduced in 1997 in large part because of an inquisitorial-based resistance (Lin, Reference Lin, Li, Yuan and Zhang2024: 172). This should limit false guilty pleas as a type of wrongful conviction. As seen earlier, the mainland has no such professional loyalty to inquisitorial ideals raising concerns that 2018 efficiency-based reforms have produced American-like high guilty plea rates and false guilty pleas, that in the absence of democratic freedoms that have led to mass exonerations in the United States and England, will never be corrected.

9.11.2 Hong Kong

Hong Kong has so far kept much of the colonial system inherited from Britain. The most serious cases are tried by juries. They typically consist of only seven people, and a five-person majority can convict the accused. A 2015 study found that Hong Kong residents placed the lowest value on testimonial evidence such as confessions and concluded that there was no evidence of the Hong Kong system moving towards reliance on confession evidence as on the Mainland. As Professor Simon Young (Reference Young, Roberts and Hunter2012) has pointed out the Hong Kong courts inherited a generous common law approach to the voluntariness of confessions as applied by the Judicial Committee of the Privy Council (Lam Chi-ming v. R. [1991] A.C. 212). Moreover, they have decided that, unlike in England where there has been statutory abrogation of the right to silence, that no adverse inferences should be drawn from pre-trial silence (Lee Fuk Hing v. HKSAR [2004] 7 HKCFA 600). In 1984, a Hong Kong court noted in reviewing testimony by a witness with an incentive to lie that “the courts have always been jealous in protecting the innocent against wrongful conviction and as such put in place measures to minimise wrongful conviction.”(Queen v LukYiuKwong [1984] HKEC 85, 85). The Hong Kong Bill of Rights is based on the International Covenant on Civil and Political Rights, and courts can award remedies for the breach of rights including against unlawful arrest or detention or various fair trial rights.

Hong Kong courts have also expressed concerns about mistaken eyewitness identifications and the danger of convicting simply because the accused raised a false alibi (R. v. Hui Kwok Hung [1991] HKLY 175). In 2010, the Court of Final Appeal explained that many evidentiary exclusionary rules are “designed to avoid wrongful convictions” (Kissel v. HKSAR [2000] 2 HKLRD 435 at para 200, 2000; Mujuzi, Reference Mujuzi2021). As in England, the accused can appeal on the grounds that the conviction is unsafe or unsatisfactory. In a leading 2021 case, the Hong Kong Court of Appeal held that “there is no limitation or restriction on the type of matter that might cause the Court of Appeal to conclude that the conviction is unsafe.” It elaborated that a conviction may not be safe both if the evidence is not sufficient or the trial has been unfair (HKSAR v. Lam Hing Chit 2021 HKCA 1297). In other words, Hong Kong law maintains a broad approach to miscarriages of justice that does not require proven innocence. There has even been a recent application by the Hong Kong Court of Appeal of a lurking doubt standard for overturning appeals that as discussed in Chapter 5.2 still exists in England but is rarely applied (HKSAR v. Ma Ka Kin [2021] HKCA 1188 [54]).

There are also statutory provisions allowing the Chief Executive to refer convictions back to the court for a second or subsequent appeal “if he thinks fit” (Criminal Procedure Ordinance s.83P) In 2016 the Hong Kong Court of Appeal held that while this provision granted the executive “unfettered discretion” (HKSAR v. Chang Wai Hang Alab[2016] HKCA 45 at para 48), it would not rule out the possibility that in a compelling and exceptional case, the court could review the Chief Executive’s exercise of discretion. Moreover, the court could ask the Chief Executive to refer a conviction for a second or subsequent appeal in cases where an applicant “put forward cogent and forceful important points to convince the court that his conviction is clearly so unsafe and unsatisfactory that disallowing him to appeal is clearly unfair to him” (HKSAR v. Chang Wai Hang Alab at para 64). This still compares favourably to the absence of executive clemency on the mainland (Xiong, Reference Xiong, Pascoe and Novak2021).

Plea bargaining both with respect to sentences and charges is encouraged in Hong Kong by customary sentencing discounts of a third of a sentence that would have been given if the accused was convicted after trial. Concerns have been raised about false guilty pleas in Hong Kong with one defence lawyer telling researchers “a pressure to plead guilty just to save years of your life is wrong” (Cheng and Chui, Reference Cheng and Chui2015: 408). In 2018, the Hong Kong Court of Final Appeal stressed that no court should accept an “equivocal” plea of “guilty but” (HKSAR v. Chan Chi Ho Lincoln [2018] HKCFA 64). In a 2021 case, the Court of Appeal quashed a drug conviction of a twenty-year old who had no criminal record and had been sentenced to twenty-one years’ imprisonment for importing cocaine. The Court expressed concern that the prosecution had brought lesser charges against a co-accused and that the young accused’s lawyers may have had preferred the older accused’s interest to that of their younger client. The Court of Appeal also applied a lurking doubt standard in quashing the conviction (HKSAR v. Ma Ka Kin [2021] HKCA 1188).

He and Liu (Reference He, Liu and Li2024) have argued that both Hong Kong and China are increasingly encouraging guilty pleas for efficiency reasons despite their distinct legal systems. There is truth to this, but the above-mentioned cases suggest that Hong Kong courts have been alive to the dangers of false guilty pleas. Enshin Li (Reference Li, Li, Yuan and Zhang2024) has stressed that on the mainland, plea bargaining reflects a long-standing policy of “leniency to those who confess, severity to those who resist” and allows prosecutors to continue to dominate the system (Reference Li, Li, Yuan and Zhang2024: 255, 270). In contrast, plea bargaining emerges more from negotiation between prosecutors and a Hong Kong defence bar that is more empowered than on the mainland (Reference Li, Li, Yuan and Zhang2024: 261). Different penal cultures also make a difference. As He and Liu note, the mainland is becoming increasingly punitive with cases that result in sentences of three years or more rising from 53.9 per cent in 2000 to 77.4 per cent in 2020 with a conviction rate that exceeds 99.9 per cent (He and Liu, Reference He, Liu and Li2024: 201,205, 210).

The Hong Kong Bill of Rights based on the International Covenant on Civil and Political Rights provides for compensation for some wrongful convictions. As will be examined more fully in Chapter 11.5, Article 14(6) is quite restrictive and only requires compensation for those acquitted on the basis of new exonerating evidence after their ordinary appeals have been exhausted. It is thus not surprising that Hong Kong courts have rejected claims of compensation by a person who spent thirty-three months in pre-trial detention but was acquitted on his first appeal because of a lack of disclosure (A v. Secretary for Justice [2020] HKCFI 427 [31]). This is a rare example of the mainland system being more generous because Article 15 of China’s Law on State Compensation allows claims for “wrong detention of a person without incriminating facts or proof substantiating a strong suspicion of a crime.” As will be suggested in Chapter 11.6, however, after-the-fact compensation is not a substitute for the prevention of miscarriages of justice.

Unlike in England, the Hong Kong executive has maintained the alternative of granting ex gratia compensation. In the above-mentioned case, the courts also upheld the executive’s decision not to grant compensation on the basis of its concerns that the applicant might not be innocent (A v. Secretary for Justice [2020] HKCFI 427: para 42). As in England, compensation in Hong Kong often requires something close to proven innocence despite arguments that will be examined more fully in Chapter 11.5 that a more generous approach is possible and desirable (Mujuzi, Reference Mujuzi2021). Nevertheless, this standard is not applied to criminal appeals where Hong Kong courts are concerned about the safety of convictions and even lingering doubts about guilt, much less restrictive standards than used on the mainland.