Rethinking the Factors Affecting the Prohibition and Prevention of Torture in China—A Qualitative Comparative Analysis ^†

Abstract

As China continues to intensify its judicial reform efforts, there are multiple factors that influence the prohibition and prevention of torture in China. This issue leads to the main research question of the article: what factors influence the protection of human rights in China to achieve the prohibition and prevention of torture? This paper uses fuzzy set (fs)/qualitative comparative analysis (QCA) to explore how the configurations of multiple factors influence the establishment of a strong human rights protection system, to achieve effective torture prevention and prohibition in China. This article will focus on the cases of torture that have been made public (1998–2021). When considering how to build a strong human rights protection system to achieve the prohibition and prevention of torture, it is necessary to identify the most effective combination of cooperation with international monitoring bodies, transparency, legislation and judicial practice, attention allocation on torture cases and independent monitoring bodies for torture prevention. The analysis concludes that: (1) there are three ways of configuration to establish effective torture prevention and prohibition, (2) active engagement with international monitoring bodies is key to establishing a high level of human rights protection, particularly concerning China’s current mechanisms, and (3) the three types of configurations also provide solutions for future policy and decision-makers in China. This study contributes to demonstrates the benefits of fs/QCA in testing the complementarity of judicial practice and provides conceptual and empirical evidence for previously understudied factors.

1. Introduction

The prohibition of torture is jus cogens in international law, and it is imperative for China, as a state party, to fulfil its international obligations to Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (UNCAT) (United Nations Treaty Collection 1984). China is a State Party to the UNCAT and shall submit its periodic report under Article 19 of the Convention1, but until now (2022), China has not presented its sixth periodic report to the Committee Against Torture, which is overdue since December 20192 (UN Treaty Body Database 2022). China became increasingly unwilling to facilitate the Committee’s examination of human rights practices of the government. The core question of this article is to examine which combinations of factors affect the achievement of a high level of human rights protection in China by exploring the multiple factors that influence the prohibition and prevention of torture through the results of different pathways of fuzzy set (fs)/qualitative comparative analysis (QCA)’s configurational analysis? Previous literature has explored the factors affecting the anti-torture in China, although many factors affecting the prohibition and prevention of torture, such as a lack of transparency in the way laws operate and inadequate legal texts, have also been raised in the literature. However, it ignores how the configuration between these potential factors affects anti-torture in China. If a factor is absent, does it affect the prevention and prohibition of torture? What are the factors present and absent in the different pathways? Does the presence or absence of these factors affect the achievement of a high level of human rights protection as far as the prevention and prohibition of torture is concerned? The aim of this article is to analyze the above questions in the light of the equivalence and asymmetrical character of QCA. This research on the factors affecting anti-torture in China is based on the results of previous scholarship on these factors and employs QCA to reconstruct these factors into different pathways. The different pathways may provide flexibility for future policy makers. The challenges to achieving this outcome includes the misapplication of the law or unexplained legal reasoning in judicial practice, the official concealment of cases of torture (Ahl 2017) and evading scrutiny from international monitoring bodies, cross policy or law-making, judicial practice and relevant institutions and international cooperation. The research method of QCA is promoted by experts in the field of international law (Castillo-Ortiz 2021). Currently, qualitative comparative analysis is employed in this study and explore human rights protections through the prohibition and prevention of torture issues. Thus, using configurational methods, such as QCA as elaborated by Ragin (1997), to establish causal studies between these conditions and outcomes, is emerging in the field of social science. First, QCA allows for cross-case and cross-sectoral comparisons to be made (Ragin 1997). Second, it facilitates the synthesis of information and allows alternative explanatory model testing that leads to favorable or unfavorable condition variables through the identification of different outcome paths (equals) and is based on multiple concurrent causal relationships. Third, it allows academics and practitioners to assess the combination of conditions under which a particular policy is achieved. Finally, QCA’s configurational approach is considered a core asset for policy-oriented analysis, producing results that are applicable to a specific set of cases, while still maintaining some generalization. This study aims to explore the links between legislation and judicial justice, transparency, and the interaction between internal and external prevention and prohibition bodies. In addition, it proposes a methodological contribution to the traditional doctrinal legal method through a comparative analysis of the QCA. fs/QCA could indeed handle sample sizes ranging from very small (<50 cases) to extremely large (thousands of cases) (Greckhamer et al. 2013, p. 5). fs/QCA is useful for a variety of data formats and case-oriented qualitative analysis (for example, Likert-scale, clickstreams and multimodal data), the researcher translates them into fuzzy sets, which would cover all values in the [0, 1] range in such instances. (Pappas and Woodside 2021). QCA focuses on the study of the aggregated, nonlinear, and asymmetric relationships between multiple elements and outcomes, whereas regression analysis is correlational and linearly symmetric. in QCA, the conditional elements are considered as a set. In actual judicial practice, each conditional element cannot be strictly independent of each other, and there will be multiple complex intersections between the sets of elements, such as X (conditional variable) leads to Y (outcome variable), but the absence of X may also lead to Y (Furnari et al. 2018). Therefore, QCA enables to realistically reflect the different judgment results of cases caused by multiple factors in judicial practice. QCA studies consider that the composition of outcomes can have multiple paths, each of which is a complex combination between different elements. In the set perspective, when different subsets of elements are superimposed on each other to form a concurrent set, also known as a “configuration”. The superposition of different subsets of elements may produce the same result (Schneider and Wagemann 2010). The QCA method explains the necessity and sufficiency of each conditional element and outcome of a case, measured by the degree of coverage between the set of outcomes and the subset of conditions, while the regression analysis explains an overall average trend, which is not generalizable to a specific case (Ordanini et al. 2014).

In the process, the QCA’s depth is highlighted, as is the additional benefit of having case-based and theory-rooted knowledge to strengthen the analysis. Therefore, the following research question was derived: which core configurations of conditions of torture legislation and judicial practice, as well as the interaction between internal and external prevention and prohibition bodies, are conducive to building strong human rights protection?

The structure of this article is as follows: Section 2 presents the theoretical background; Section 3 discusses how to deal with data transformation for the QCA; Section 4 presents a selection of cases and conditions; Section 5 analyses the results and discussion of the fs/QCA; Section 6 describes the limitations of the article, steps for future research and suggestions for practitioners and policy makers.

2. Literature Review and Analysis Framework

2.1. Exploring Torture Prohibition for Human Rights Protection: Practical Experience and Research Progress

Torture prohibition is jus cogens in international law, and its judicial practice is pervasive at the international level and in Human Rights Courts, such as the European Court of Human Rights (ECtHR). Many treaty-based human rights protection mechanisms are therein established. For instance, in the context of torture prohibition and prevention, there is a specific Body: Human Rights Committee, which monitors the implementation of the International Covenant on Civil and Political Rights (ICCPR), and its Optional Protocols3. The Human Rights Committee, in accordance with ICCPR Articles 7(5) and 10(1), provides written Views urging the State party to grant effective remedies to torture victims. In addition, the Committee Against Torture (CAT), established under the Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (UNCAT), monitors its implementation. The Subcommittee on Prevention of Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (SPT) established based on the Optional Protocol to the Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment4 (OPCAT), visits State Party detention areas.

Torture prohibition has also attracted a wide range of attention in academic circles. Scholarly analyses and research on torture prohibition could be categorized into four perspectives.

Judicial practice and legislation prohibit torture. In Saadi v Italy5 (2008) and Ben Khemais v Italy6 (2009), the ECtHR confirmed the principle that no circumstances, including the threat of terrorism, can justify exposing individuals to a serious risk of human rights violations. There is an absolute guarantee against torture and inhuman or degrading treatment or punishment (Gentili 2010). However, arguments regarding torture in Ireland v United Kingdom (1978)7, emphasize the tension between strict adherence to the law by the British judiciary, and the executive interest of maintaining national security in the United Kingdom (Gearty 2020). These articles suggest that perpetrators of torture can be brought to justice, while maintaining the lawful punishment of torture (Gearty 2020). Furthermore, in German legislative practice, the absolute prohibition of torture is explicitly stated in German Police Law and Constitutional Law (Brugger 2000). In China, the legislature was ratified by the UNCAT in 1988, but according to Gong (2016), torture still exists due to the lack of a complete definition of torture and an independent investigation mechanism in the legislation. Simultaneously, a torture victim’s right to sue is not guaranteed and a climate of impunity (Gong 2016) still prevails. The prevalence of torture and prison abuse in China is due to the traditional Chinese culture of “torture” and the weak rule of law, as well as the lack of independent monitoring mechanisms (Thelle 2006). The case study and qualitative interview research methods were used for an in-depth study of the Chinese criminal justice system, and interviews with offenders provided insight into the real dynamics of justice. (Lubman 2012). However, the fact that the interviewees were limited to offenders only, in fact, resulted in limited research findings. Since the actors of public power in judicial practice such as judges, prosecutors, and police officers should also be included in the scope of the investigation. Although China has introduced the exclusionary rule in its criminal procedure law, its definition is too vague causing judges and prosecutors to not apply the rule and police to selectively record confessions in judicial practice (Jiang 2018). In other words, the reliability of the evidence is insufficient (Guo 2019). Similarly, it is also noted that the provisions on torture in the Chinese Criminal Law are limited, setting restricted scenarios and the identity of the perpetrator (Ahl 2017). Electronic recordings of interrogations are de facto controlled in their entirety by prosecutors and police, and without the mandatory presence of a lawyer during custodial interrogations, recordings will not be effective as a deterrent to torture and police misconduct (Zhu and Siegel 2015). The difficulty in applying the exclusionary rule also exists in Chinese judicial practice, where the trade-off between fighting crime and protecting human rights and between substantive justice and procedural justice is overly weighted in favor of the former (Guo 2016). Enforcement of the exclusionary rule is required to the procuratorial authorities should prove the legality of the evidence and the courts should apply the exclusionary rule when deciding cases (Guo 2019).

Independent monitoring bodies exist for torture prevention, and fact-finding and technical assistance from the CAT to States parties, particularly regarding implementation mechanisms (Gaer 2020). European systems of torture prohibition are an organic combination of three organizations: the Council of Europe (CoE), the European Union (EU) and the Organization for Security and Cooperation in Europe (OSCE), as well as the advanced torture prevention system in the European Convention for the Prevention of Torture and Inhuman or Degrading Treatment or Punishment (ECPT) and the ECtHR’s legal framework to implement the binding European Convention on Human Rights and Fundamental Freedoms (ECHR) concerning torture (Bicknell 2020) The Human Rights Committee monitors State parties’ compliance with, and violations of the ICCPR, through their periodic reports, examining complaints (i.e., “communications”), revealing views and “general comments” (Heyns et al. 2020). The Inter-American System of Human Rights, based on the Organization of American States (OAS) is structured by regional human rights conventions. It aims to extend the perpetration of torture, in an official capacity, to a private one (i.e., private armed groups and violence against women in private or domestic settings) (Rodríguez-Pinzón 2020). The African Commission on Human and Peoples’ Rights (ACHPR), the African Court on Human and Peoples’ Rights (ACHPR), and the African Committee on the Rights and Welfare of the Child (ACRWC), combined with the African Union Commission (AUC) and the African Union (AU), constitute the African human rights system. It prohibits torture and is concerned about the number of accessions or ratifications to human rights treaties and laws criminalizing torture and remedies for victims (Murugu Mute 2020). This system focuses on victims’ prompt reparation and rehabilitation, and on community-based and localized anti-torture institutions (Rodley 2009). In Sweden, a highly transparent institution responds to situations, where acts of torture occur. This institution reports the acts perpetrated by individuals and society and analyses available data on torture and the development of a human rights measurement model (Eck and Fariss 2018). China enacted the Oversight Law of the People’s Republic of China (Oversight Law) in 2018 and established the National Supervision Commission as the highest supervisory authority8, as well as supervisory commissions at various levels at the local level. In addition, cases where the causes of action were corruption and torture were also selected for assignment analysis. It is possible to follow the verdicts of cases in early 2018 and beyond and find cases where the Supervisory Commission implemented detention measures under Article 229 of the Oversight Law. Its function is to exercise independent powers of supervision and investigation10.The State Supervisory Commission is not an independent torture monitoring body de facto that States parties are obligated to establish under the Convention against Torture. The variable of an independent monitoring body (National Supervision Commission) is mentioned here because the Oversight Law stipulates that the National Supervision Commission performs the task of monitoring and investigating crimes committed by public officials in their positions, including torture to extract confessions and other ill treatment, as well as corruption of public officials. In the written legal texts and judicial practice, the National Supervisory Commission appears to be the highest supervisory organ of the state independent of the system of public security organs, courts and procuratorates.

Measures are implemented for torture prohibition and prevention. Legal reform is one of the effective ways to combat torture, and human rights monitoring also contributes to the elimination of violence, particularly regarding torture in extra-custodial violence and anti-corruption of torture prohibition (Kelly 2019). Torture risk reduction is made possible by identifying and addressing the factors that increase the risk of torture and ill treatment. These concrete and practical measures include the prohibition of incommunicado detention; the immediate notification of family members after an arrest; access to a lawyer and their presence during interrogation; medical examination by an independent doctor; and video recording interrogations (European Union 2012). In addition, “letters of rights” can be used to inform persons of the rights to relief available to them (European Union 2012). Finally, moving away from an over-reliance on confessions in criminal proceedings (APT 2019) will also reduce the possibility of torture. In both domestic and international law, evidence obtained by means of torture and other cruel, inhuman, or degrading treatment is categorically excluded from criminal and administrative proceedings, following the exclusionary rule (Pollard 2020).

Although there are several studies on the factors affecting torture prohibition, and the necessary measures to prohibit it, their lens is predominantly doctrinal, rather than empirical. Further, literature that uses empirical analysis has mainly explored the single value of many potential elements, neglecting the interaction between the multiple factors that influence torture prohibition. To address these limitations, this article introduces the fs/QCA, that is Ragin’s (2009) approach, to explore linkage effects of factors affecting torture prevention and prohibition in China.

2.2. Human Rights Protection Regarding Torture Prohibition

2.2.1. Legislation and Judicial Practice

China has incorporated provisions for the rules from the UNCAT into its domestic legal frameworks, such as the Criminal Law provides for the extortion of confessions by torture, and the Criminal Procedure Law of the Evidence Rules and introduced exclusionary rules as an effective means of reducing torture (Ahl 2017). However, there are inconsistencies with the Convention regarding the prohibition of torture law-making practices. China has clear provisions against torture, in both domestic statutory law and policy, but they are contradicted in judicial practice (Lewis 2019).

In judicial practice, case verdicts show a tendency towards impunity, and lighter penalties (Lewis 2019). Similarly, concerning confessions extracted under torture, where the key evidence was not obtained before the indictment, courts often admitted the illegally obtained fake confessions for “efficiency”. Further, a lack of clear legal reasoning and appropriate applicable laws have been seen in Chinese domestic courts (Gong 2016).

The UNCAT requires States parties to enact laws on the prohibition and punishment of torture in domestic law, which can be found in the Criminal Law of the People’s Republic of China (Criminal Law), the Law of the People’s Republic of China on State Compensation (State Compensation Law), the Law of the People’s Republic of China on Prisons (Prison Law). In judicial practice involving Article 24711 of the Criminal Law, there has been confusion over the legal application of the article in cases of confessions extracted under torture, death by torture and injury by torture. There has been a lack of clarity as to the limits of the legal application of the distinguishing provision. In the case of Article 24812 of the Criminal Law, a lighter penalty or immunity is granted in cases of torture. Furthermore, an exploration of the case guidance system of the Supreme People’s Court (SPC) showed that there are only three typical cases and no guiding cases on torture; this situation increases the obstacles encountered by Chinese domestic courts in following the decisions of previous cases. Even in the typical case involving article 248 of the Criminal Law, there is little record of the specific circumstances of the case in the judgement, and no statement from the parties or legal basis for the court can be found. Although the State Compensation Law also provides access to remedies in cases where torture was used to extract confessions and ill treatment of prisoners in China, victims, in most cases, did not receive compensation from the State when they were eligible for it.

The punishment for what is called “torture” is regulated in Chinese domestic law. There are two forms of incrimination in the Criminal Law, which are the crime of extorting a confession under torture or the crime of obtaining evidence through violent means and the crime of maltreatment of prisoners. Article 247, which covers the crime of extorting a confession under torture or the crime of obtaining evidence through violent means, provides for the punishment of judicial personnel who extort confessions from criminal suspects or defendants by torture or use violence to obtain the testimony of witnesses. Meanwhile, Article 248 stipulates that if a supervisor of a prison, detention center, guardhouse or other custodial institution beats or physically abuses a person under supervision, and the circumstances are serious, he shall be sentenced to fixed-term imprisonment of not more than three years or detention; if the circumstances are particularly serious, he shall be sentenced to fixed-term imprisonment of not less than three years but not more than ten years. If a person causes injury, disability or death, he shall be convicted and punished severely in accordance with the provisions of Articles 23413 and 23214 of Criminal Law. A supervisor who instructs a supervisee to beat or physically punish and abuse another supervisee shall be punished in accordance with the provisions of the preceding paragraph. Meanwhile, the provision is made for the compensation of victims of torture and other inhuman treatment. Specifically, Article 1715 of the State Compensation Law provides state compensation for victims who have been subjected to torture, beatings, ill treatment and the unlawful use of weapons or police weapons.

2.2.2. Transparency in the Justice System and Emphasis on Torture Cases

China’s judicial reform focused on the establishment of an open legal database (Finder 2019) and smart courts (Li 2020), to promote judicial openness, fairness and effectively enhance judicial credibility. It also builds publicity on four main platforms: trial processes and procedures, court activities, judgment documentation and information on court actions (OHCHR 2016). However, the publicly available legal database has limited access groups (Ahl and Sprick 2017), and some cases involving public officials or maladministration are categorized as unsuitable for public release (Chen 2016). The criteria for online disclosure are determined by the authorities and there is not much judicial transparency (Tang and Liu 2019). Judicial reform for improved transparency is also seen as the centralization of the judiciary system and decision-making (Chen et al. 2021). In addition, regarding China’s Case Guidance System16, the Supreme People’s Court (SPC) has not published any guiding cases on cases of action relating to torture to extract confessions, prison torture or ill-treatment. Instead, there is only one typical case of a complaint against the State compensation regarding confessions extracted under torture (Pkulaw.cn 2022)17. The definition of “transparency” in the article is based on the selection of the “four platforms of openness” referred to in the periodic report submitted by China to the Committee Against Torture (UN Treaty Body Database 2022), where the court’s activities are made available to the public at different stages of the court’s judicial activity. The highest value of transparency discussed in the article is met when a particular case is simultaneously available to the public on all four platforms. Likewise, the assignment of values is explained in detail below.

2.2.3. The Interaction between Internal and External Prevention and Prohibition Bodies

China has established a National Supervisory Commission as part of its progress in judicial reform (Wei 2019). Although the supervision of various complaints against public officials, for offences including confessional and prison torture or ill treatment, could be found in legal texts, in practice the Commission is primarily aimed at punishing and preventing corruption among public officials (Oversight Law of the People’s Republic of China 2018)18. In other words, in the Chinese legal context, however, the National Supervisory Commission is a supervisory body independent of the courts, procuratorates and public security organs. Nonetheless, its functions are mainly directed towards the investigation of corruption and torture cases and the retention of public officials involved in corruption and torture. In addition, China has submitted five periodic reports to the UN Committee Against Torture in fulfilment of its State Party obligations since becoming a State Party, but has not presented its sixth periodic report, which has been overdue since December 2019 (UN Treaty Body Database 2022). In its periodic report, China describes initiatives relating to judicial reform. For example, the number of cases of torture in criminal scenarios has decreased following the implementation of the relevant legislation (Criminal Law and rules on the exclusion of illegal evidence in the Criminal Procedure Law) prohibiting torture.

In fs/QCA, the relationship between different matching patterns and results is identified through cross-case comparisons. Condition configurations with equivalent outcomes could also be identified to facilitate the recognition of the differential driving mechanisms that lead to varied outcomes in different case scenarios. In addition, further discussion of the matching and substitution relationships between conditions could be identified. From the configuration perspective, the aforementioned factors are not isolated from each other and could have simultaneous synergistic effects. Furthermore, they could be mutually reinforcing through adaptation or mutually offsetting through substitution.

3. Approaching Data Transformation for QCA

This study uses fuzzy set qualitative comparative analysis and follows a four-valued fuzzy for affiliation taking; that is, each antecedent condition variable is valued from [0, 0.33, 0.67, 1], according to the specific situation. To implement the QCA, any initial data, whether numeric or non-numeric, must be transformed into crisp or fuzzy set membership values. This data transformation is more than just a technical exercise; it corresponds to a deeper epistemological viewpoint (Kröger 2021). In this paper, the four-value fuzzy-set assigns each variable to a value (Oana et al. 2021). Then, the fuzzy set degree of accuracy employs and specifies each of their values before matching the qualitative categories to fuzzy set values.

For this study, the fuzzy set’s accuracy was calculated based on the qualitative data information level. Six conditional variables were selected and assigned: (1) competence in torture law-making, (2) legal reasoning and applicable law, (3) transparency, (4) attention allocation on torture cases, (5) independent monitoring bodies for torture prevention and (6) cooperation with international monitors. The variables are based on purely domestic legal provisions and international human rights treaties that have been ratified and judicial practice, and all variables in the article are derived from judgments that have entered into force.

For the legislation and judicial practice of torture which includes (1) competence in torture law-making and (2) legal reasoning and applicable law. The assignment of values is based on the circumstances of different cases, and the value of a single case does not represent the full range of circumstances in that case. The assignment of legislation as a conditional variable is based on the application of the law and the circumstances of the legal reasoning. A complete, comprehensive and practicable legislation was assigned a value of [1]. A partly comprehensive and practicable legislation was assigned a value of [0.67]. An incomplete and “empty shell” comprehensive and practicable legislation was assigned a value of 0.33. An entirely incomprehensive and practicable legislation was assigned a value of [0]. For legal reasoning and applicable law, a correct and clear applicable law and logical legal reasoning were assigned a value of [1]. A partly correct and clear applicable law and logical legal reasoning were assigned a value of [0.67]. A barely correct and clear applicable law and logical legal reasoning were assigned a value of [0.33]. An entirely incorrect applicable law and logical legal reasoning were assigned a value of [0]. Transparency includes four major open platforms of trial processes and procedures, court activities, judgement documentation, and information regarding court actions. Complete and high transparency (four open platforms) is assigned a value of [1]. Higher transparency (three open platforms) is assigned a value of [0.67]. Almost no transparency (one to two open platforms) is assigned a value of [0.33]. Totally out transparency (no open platform) is assigned a value of [0]. For attention allocation of torture cases (China SPC of Case Guidance system), both guiding and typical cases were assigned a value of [1]; only guidance cases were assigned a value of [0.67]; only typical cases were assigned a value of [0.33]; and no case guidance or case examples were assigned a value of [0]. The interaction between internal and external prevention and prohibition bodies includes independent monitoring bodies for torture prevention and cooperation with international monitors. Cases, where the cause of action is corruption and torture, were considered to be under this condition (National Supervisory Commission). Cases without oversight were assigned a value of [0], and cases with oversight were assigned a value of [1].

International monitoring bodies’ cooperation refers to China’s disclosure of issues related to the prevention and prohibition of torture in Universal Periodic Review and its response to the relevant recommendations made by the reviewing states. Before 2019 (UN Treaty Body Database 2022), strong international cooperation and a high level of compliance with international obligations were assigned a value of [1]. Partial international cooperation, and compliance with international obligations, before 2019 were assigned a value of [0.67]. Hardly any international cooperation and compliance with international obligations, after 2019, were assigned a value of [0.33]. No international cooperation and compliance with international obligations, after 2019, were assigned a value of [0] (Appendix A.1).

The outcome variable is that the occurrence of a lawsuit in which the judgement enforced and the perpetrator is punished, as well as the State compensation is obtained by victim a value of [1]. The value of [0.67] is assigned to the occurrence of a lawsuit in which the judgement enforced and the perpetrator is punished. A value of [0.33] is assigned to the occurrence of a lawsuit with a verdict turned out to be an inconsistent or flawed application of the law and legal reasoning. Such as impunity or lighter penalties for serious crimes A value of [0] is assigned if a lawsuit is brought but the verdict is inconsistent or flawed application of the law and legal reasoning of impunity or misdemeanor and a retrial or appeal is dismissed and remains pending. The outcome variables are assigned based on the judgments that have been validated and publicly available in the professional legal database PKULAW, in which the cause of action in the cases are torture to extract confessions (刑讯逼供), torture and ill-treatment in detention areas (拘留场所内的酷刑和虐待) and prison maltreatment or ill treatment (监狱酷刑), torture and state compensation (酷刑与国家赔偿) as well as torture and corruption (酷刑与腐败). The facts of these cases are based on acts of torture and other ill-treatment that were actually committed by the perpetrators against the victims. Those cases in which the defendants were acquitted without committing acts of torture, in fact, are not within the scope of the article’s research.

4. Cases and Conditions Selection

QCA is fundamentally an empirical data-driven model developed through a conversation between theory and data. This is often an iterative procedure that is repeated several times. Researchers testing models through QCA analysis often encounter contradictory results (homozygous causes and effects). One way to deal with contradictory problems is to introduce new variables. However, the problem of limited diversity is raised when the number of cases is insufficient. An overabundance of conditions may lead to one or a few special cases, as there are few case scenarios reflecting individual configurations (Schneider and Wagemann 2012). For example, if eight conditions are included in the model, there will be 256 possible combinations of these conditions. However, if only 50 cases are investigated, then at least 206 combinations become logical residuals (no case coverage), and the number of logical residuals will increase further if multiple cases are categorized into the same combination. Currently, there is no clear rule on the upper limit of the number of conditions, but several studies provide recommendations. For example, Berg-Schlosser et al. (2009) suggest that small samples of QCA studies (10–40 cases) should limit the model to seven conditions. Marx and Dusa (2011) provided probability distributions for finding consistent solutions to QCA based on simulations with different sample sizes and number of conditions. In this case, a model containing four conditions should contain at least 12 case numbers (the number of cases in the figure corresponds to the case where the column value is 0), and a five-condition model contains at least 15 cases by analogy and has enabled limited diversity (Marx and Dusa 2011). In summary, this article was conducted by searching the PKULAW specialist legal databases for cases where the causes of action were torture to extract confessions, prison abuse, state compensation and corruption. The number of cases meeting all these causes of action is 463 cases (until 2022), but the number of available data (with clear applicable law and legal reasoning) is only 79 cases. In such case, the six conditional variables and 32 case studies selected for this article are operationally compliant with the QCA.

Furthermore, a QCA-based analysis may be utilized to analyze sample situations (Thomann and Maggetti 2017). The 32 cases included for this article are from China’s hierarchical courts. Cases from primary people’s courts, intermediate people’s courts, higher people’s courts, the Supreme People’s Court and the Supreme People’s Procuratorate Gazette, and those cases where only the title of the case is disclosed but where it is not appropriate to disclose the specific circumstances in accordance with the principles of online disclosure. These instances represent, to some extent, the contemporary judicial practice of China’s hierarchical courts. This article compares and analyses 32 cases in criminal and administrative scenarios (Appendix A.3) from a public platform PKULAW database, where the causes of action includes torture to extract confessions (刑讯逼供), torture and ill treatment in detention areas (拘留场所内的酷刑和虐待) and prison maltreatment or ill treatment (监狱酷刑) and cases relating to torture or ill treatment but where the cause of action is state compensation (酷刑和国家赔偿) as well as torture and corruption (酷刑与腐败) (1998–2021). These cases were chosen for analysis, because they mostly cover the current state of Chinese legislation and judicial practice. These cases all involved misconduct by public officials such as torture and ill treatment and range from current scenarios of torture occurring within the context of Chinese Criminal Law and Administrative Law. These cases involve police and other public officials being tortured and other ill treatment of suspects or detainees or prisoners, during interrogation or detention measures or while serving sentences, for the purpose of obtaining confessions or undue advantage. Simultaneously, the accessibility of these cases was also a major factor, when considering transparency. In addition, they reflect the implementation status of China’s international obligations and the interaction between internal and external prevention and prohibition bodies.

5. fs/QCA Results and Discussion

The fundamental principle of the QCA is to calibrate causal conditions and outcomes into sets, and to determine the necessity and sufficiency of each condition and its combination for the outcome by calculating the subset of relationships between the sets. Consistency and coverage are two important metrics for assessing the strength of the subset relationships in a set. For fuzzy sets, cases have a set affiliation ranging from 0.0 to 1.0, i.e., partial affiliation. A fuzzy subset relationship exists when the affiliation of a case in one set is always less than or equal to its affiliation in another set (Ragin 2008).

Consistency with respect to sufficient conditions. In an earlier work, Ragin (2008) argued that, although the range of consistency is between 0 and 1, values between 0.0 and 0.75 indicate substantial inconsistency, and therefore, a critical value of consistency should be above 0.75. Subsequently, Fiss (2011) used 0.8 as a recommended criterion for adequacy consistency and one that is currently followed by most studies (Misangyi and Acharya 2014; Bell et al. 2014; Campbell et al. 2016). In addition, where conditions allow, there are studies that use higher consistency criteria, such as 0.85 (Gupta et al. 2020), 0.87 (Jacqueminet and Durand 2020), 0.9 (Park et al. 2020) and 1.00 (McKnight and Zietsma 2018), capable of filtering out subsets of more strongly related histories and streamlining the number of histories. In contrast, the consistency judgement criteria for the necessary conditions are more uniform, using a higher threshold of 0.9 (Fiss 2011; Misangyi and Acharya 2014; Bell et al. 2014; Gupta et al. 2020).

The QCA investigation in this article follows the reporting idea of the earlier Ragin and Fiss’ solution, i.e., combining the intermediate and parsimonious solutions to derive the core and edge conditions. The details could be found in

Table 1. Necessary conditions analysis.

Condition Variables	China’s Human Rights Protection Regarding Torture Prohibition and Prevention		~China’s Human Rights Protection Regarding Torture Prohibition and Prevention
	Consistency	Coverage	Consistency	Coverage
Competence	0.857245	0.719593	0.469688	0.478131
~Competence	0.378301	0.37037	0.724544	0.860238
Legal	0.761599	0.969119	0.25309	0.390554
~Legal	0.521056	0.365183	0.979988	0.832916
Transparency	1	0.656514	0.547969	0.43627
~Transparency	0.141328	0.204969	0.56857	1
Allocation	0.571021	0.960384	0.174809	0.356543
~Allocation	0.617416	0.381562	0.980577	0.734892
Independent	0.190578	0.534	0.137139	0.466
~Independent	0.809422	0.436154	0.862861	0.563846
Cooperation	0.810849	1	0.271925	0.40669
~Cooperation	0.518915	0.370163	1	0.865071

Notes: ~ indicates not.

. In principle, there are three main types of solutions (parsimonious, complex and intermediate). The conditional configurations of intermediate solutions are the main source of the article’s findings, as shown in the Table 1 above, the high-level human rights protection regarding prohibition and prevention of torture in China configuration influence paths include a total of three. Coincidentally, the intermediate solution paths in this paper are consistent with the complex solution paths, and the consistency of each path is higher than 0.9, while the coverage of the results is 0.810849 and the consistency is 1. The running results are true and valid.

As could be seen from the Table 1, there are three paths of influence of high human rights protection configuration in the complex solution, of which the consistency of the solutions is higher than 0.8, and the coverage of the overall operation result is 0.810849 with a consistency of 1. Greater than 0.9 means that the conclusion is valid. However, from the theoretical point of view, the conclusions of complex solutions that do not incorporate logical residuals are not reported. The conditional configuration path of the simple solution, such as the simple solution with high human rights protection in the table, has a total of 1 path for Cooperation. The simple solution will be reported in combination with the intermediate solution as the final conditional configuration path. The conditional configuration of the intermediate solution is the main source of the article’s findings, as shown in the table above, the path of high human rights protection Configuration influence includes a total of three. Coincidentally, the intermediate solution path in this article is consistent with the complex solution path, and the consistency of each path is higher than 0.9. While the coverage of the results is 0.810849, and the consistency is 1, and the operation results are true and valid. At the same time, the three paths of the intermediate solution are combined with one path of the parsimonious solution to distinguish the core condition and the edge condition of the final configuration pathway. If a condition variable appears in both the parsimonious solution and the intermediate solution, it is the core condition and plays an important role in the configuration path; if it only appears in the intermediate solution, it is the edge condition and plays an auxiliary role. The same is observed for the situation of low human rights protection, and the two situations are synthesized in Table 1.

5.1. Necessary Conditions

The analysis of the necessary conditions for human rights protection, regarding China’s torture prohibition and prevention (Table 1), shows that the consistency of each condition is below the critical value of 0.9, except for transparency. Therefore, the condition variable of transparency becomes a necessary condition for the prohibition of torture in order to establish a high level of human rights protection. In the Chinese context, transparency is analyzed in the context of the different scenarios of the disclosure of judgment documents in the four major platforms in this text chapter. This suggests that high transparency (the four platforms) may be an explanation for the high level of human rights protection in China’s prohibition and prevention of torture. Moreover, the low level indicates that only the weak cooperation with international consistency is greater than the critical value of 0.9. Significantly, the figures in this table show the complexity of the issue—that is, legislation and judicial practice, justice system transparency, the attention allocation of torture cases torture and the interaction between internal and external prevention and prohibition bodies—are major influencing factors in China’s human rights protection regarding torture prohibition and prevention.

The coverage-degree itself has no referential significance but is used to judge the empirical explanatory strength of this relationship after the subset relationship has been confirmed. Further, for multidimensional equivalent configurations (i.e., multiple combinations of sufficient conditions satisfying the consistency criterion), the coverage of different configurations could be used to compare the strength of interpretation of the results by different configurations. Thus, although there is no minimum threshold for coverage, lower coverage indicates that a given configuration may be rarer (Leppänen et al. 2019).

5.2. Configurations

This article presents results based on Ragin’s QCA analysis (Ragin 2009). This analysis method provides a clear indication of the relative importance of each condition in the configuration. Configuration 1 (

Table 2. Configurations for achieving strong human rights protection regarding prohibition and prevention of torture in China.

	1	2	3
Competence in torture law-making	•	⊗	•
Legal reasoning and applicable law		•	•
Transparency (four major platforms: trial processes and procedures, court activities, judgement documentation, and information regarding court actions)	•	•	•
Attention allocation to torture cases torture (China SPC of Case Guidance system)		⊗	•
Independent monitoring bodies for torture prevention	⊗
Cooperation with international monitoring bodies	$●$	$●$	$●$
Consistency	1	1	1
Raw coverage	0.643112	0.213419	0.428266
Unique coverage	0.286938	0.0485368	0.0720915
Solution coverage		0.810849
Solution consistency		1

Notes: $●$ indicates core conditions; • indicates the presence of a condition; ⊗ indicates the absence of a condition; empty space indicates that the condition does not matter.

; Figure 1) shows that a core factor in prohibiting and preventing torture and building strong human rights protection in China, is the need to cooperate with international monitoring bodies. This factor alone constitutes the sufficient condition for explaining the outcome. In other words, China needs to undertake its obligations, as UNCAT’s state party, submit its periodic reports on time and actively participate in the Committee Against Torture’s periodic sessions after considering various countries’ recommendations to China on torture prevention and prohibition. Simultaneously, a high level of transparency and competency in torture law-making are complementary conditions for achieving torture prohibition and prevention in China. In the absence of independent monitoring bodies for torture prevention, these conditions could also contribute to the establishment of a strong human rights protection system against torture. However, the lack of independent monitoring bodies hinders China from building a strong human rights protection system for the prohibition and prevention of torture. As far as the cases 31, 1, 26 and 27 are concerned, their causes of action relate to cases of torture to extract confessions and prison maltreatment or ill treatment) in the criminal context. They apply articles 247 and 248 of the Criminal Law, and the judgements resulted in a final trial, in which the victim received compensation from the State. The proceedings were fully concluded, without appeal or other circumstances. However, those causes of action do not relate to corruption and torture. Therefore, it falls under Configuration 1, which is not subject to the oversight of the National Supervisory Commission.

The core conditions of Configuration 2 (Table 2; Figure 2) are the same as Configuration 1. The difference is that legal reasoning, applicable law and transparency play a supporting role in this path, because the crucial factors in the outcome of a case are correct legal reasoning and applicable law. Currently, the cases available on China’s public platforms contain incorrect legal reasoning and applicable law with blurred boundaries. Regarding incompetence in torture law-making and attention allocation on torture cases (China SPC of Case Guidance system), legal reasoning, applicable law and transparency could play a complementary role in achieving a high level of human rights protection. Furthermore, transparency (disclosure on the four major platforms) remains an important method to support interested parties in accessing cases and grasping Chinese judicial practice. With regard to cases 14 and 15, their causes of action relate to cases of torture to extract confessions and prison abuse in criminal scenarios are not covered by the guiding cases and typical cases issued by the SPC. There is also blurring of the boundaries of the application of Article 247 and Article 234 of the Criminal Law to cases where the cause of action is torture to extract a confession and torture to cause injury. These verdicts in the case were that the perpetrators were brought to justice and the case was concluded. However, the cause of action was not state compensation, so the victim did not obtain state compensation. Deficiencies are highlighted because the cause of action is in cases involving torture or ill treatment, if the applicant does not take the initiative to lodge a complaint with the obligatory compensation authority, then the victim is denied the remedy of State compensation.

In Configuration 3 (Table 2; Figure 3), the attention allocation to torture cases and China’s current case guidance system, which acts as a de facto “precedent” in judicial practice (Klöhn et al. 2021) for cases with similar causes of action, differs from the two previous configurations. Accordingly, there are only typical cases, but no guiding cases regarding the cause of action of torture (Pkulaw.cn 2022). To build a strong human rights protection system, the influence of the former cases on later ones cannot be ignored. As far as the cases (25, 26, 27, 31) are concerned, their causes of action relate to ill treatment in both criminal and administrative scenarios, and contain typical cases published by the SPC that also relate to existing Chinese legislation on torture, such as the Criminal Law and the State Compensation Law. At the same time, these are also available in the four major platforms. This allows other entities or individuals to gain an insight into the existing measures to prevent and prohibit torture and human rights protection in China. The decisions made in these cases were implemented and the victims obtained state compensation. These three configurations are examined so that policymakers and decision-makers should take multiple ways toward establishing robust human rights protections in the future judicial system by preventing and preventing torture in China.

These judgments highlight a reflection on the applicable law and legal reasoning in Chinese domestic court practice. In some perplexing cases involving torture under Article 247, the person was sentenced for multiple offenses based on the findings of intentional injury or intentional homicide and torture to extract a confession, whereas, in others, the person was sentenced for the offence of torture with the goal of obtaining a confession. There have regularly been occurrences of “lenient felony sentences” and “non-convictions” in circumstances where Article 248 has been applied. It is remarkable to come across a case in which mistakes in the application of the law or legal reasoning resulted in significantly different conclusions in the initial trial court and the Court of Appeal. In other words, upon considering a matter, the Court of Appeal has seldom reversed the initial judgment. It has been claimed that the judgement in such situations implies leniency towards torturers. The conclusion of these instances demonstrated that the courts implemented the law incorrectly and that there were holes in the court’s causation argument.

Many more comparable examples demonstrate how police officers tortured the victim in order to elicit a confession. The law was applied inconsistently in these circumstances. Furthermore, these cases can be broadly divided into cases of judicial officers torturing the victim to extract a confession (under Article 247 of the Criminal Law) and cases of judicial officers torturing the victim or the victim’s family (under Article 247 of the Criminal Law and Articles 232 and 234 of the Criminal Law). Nevertheless, under Article 247 of the Criminal Law and Article 17(4) of the State Compensation Law, state criminal compensation was not always paid out for the victims’ injuries or death as a result of being tortured into a confession. The Court of First Instance committed factual and relevant law errors during the trials in the instances described above. Furthermore, appeals have not been successful; the Court of Appeal did not review the errors committed and affirmed the initial verdict. Similarly, in other torture instances when the Court of First Instance delivered the original judgement based on ambiguous facts and a legal error, the Court of Appeal has not revisited the decision. In respect to Article 247 of the Criminal Law, it has been rather unusual for a case to have radically different conclusions in the Court of First Instance and the Court of Appeal.

With concerns of torture, Articles 247 and 248 of the Chinese Criminal Law are congruent with relevant requirements in international treaties. According to Chinese court practice, legal requirements are handled differently in situations with comparable facts. However, in order to achieve a result that is compatible with the case, the real facts of each instance must be considered. However, even when the facts are largely the same, various courts may reach conflicting conclusions and make the same incorrect inferences of causation. As demonstrated in these examples, the judgment’s causative inference was based on conclusions regarding the seriousness of the circumstances, and the decision contained the sentence suspension. The incorrect conclusion that the seriousness of the offence in the case justified the imposition of a suspended sentence is manifestly contradictory to the application of the Suspension of Sentence. In these situations, the court relied on mitigating circumstances to make its judgement. These examples have been examined in earlier chapters as well as in Criminal Law Article 248. In similar circumstances, legal errors have continued to occur. The scarcity of publicly available evidence on torture-related cases under Article 248 of the Criminal Legislation has made it more difficult to investigate the relevant law and conform to precedent.

5.3. An Analysis of the Alternative Effects of Condition Variables

An internal comparison of each type of path reveals that different marginal conditions can lead to the same results, when the core conditions are consistent. For achieving strong human rights protection in the case of prohibition and prevention of torture in China, a comparative analysis of Configurations 1 and 2 leads to the conclusion that both belong to fractionation under the combination of conditions of cooperation with international monitoring bodies and transparency (four platforms). The only difference is that competence in torture law-making conditions is not present in Configuration 2, whereas it is present in Configuration 1. The other conditions’ variables are in an insignificant state.

This also illustrates that improving the transparency of the judicial system, strengthening international and domestic information exchanges and cooperation are crucial when it comes to fulfilling international obligations against torture and building a strong human rights protection system. Configuration 3, which is fully evolved, has all the conditions present, except independent monitoring bodies for torture prevention. As such, this is the most appropriate path for China to implement measures against torture and strengthen human rights protections, specifically because there are no independent monitoring bodies for torture prevention in China, currently.

5.4. In-Depth Case Studies

In Prosecutor v Liu, the court determined that defendant Liu, a correctional officer, abused a monitored person and conspired to have detainees abuse a supervised person named Du with whom they were imprisoned in Prosecutor v Liu. The victim had been physically harmed, and the circumstances were dire. Liu’s actions constituted ill treatment of a person in prison under Criminal Law article 248. The victim suffered substantial and exacerbated injury as a result of the case’s conclusion. Despite the fact that the perpetrators were convicted under Article 248, they were free from criminal penalty and were not convicted or sentenced for the offence of intentional injury under the aggravating circumstances specified in Article 248. Article 17 of the State Compensation Law, in particular, offers state compensation for victims of torture, beatings, ill treatment, and the unlawful use of firearms or police weapons. Court’s experience shows that the typical case decided by the Supreme People’s Court, Huang caihua and others v Public Security Bureau of Lianping County, occurred in 1997. This marks the first judgment in which the compensation committee of the People’s Court ordered state compensation for the death of a person caused by the extortion of a confession by torture. The case was released as a typical case by the Supreme People’s Court. As a result, it was especially significant in the discussion that the case filled a gap in the legal jurisprudence regarding state compensation in cases of torture, and so the case has had ramifications on similar cases. It was determined in the case of Prosecutor v Lou that defendant Lou, a prison official, had beaten, physically punished and abused a prisoner in custody. The circumstance was severe, and Lou’s actions equaled the crime of prisoner abuse. The defendant was charged with the offense by the Public Prosecutor’s Office, and the charge was upheld by the court. Lou was found guilty of prisoner abuse and was granted immunity from criminal punishment on the grounds that the court did not find any relevant grounds or basis for punishment. However, the wording used in the verdict included conflicting language about the “serious circumstances” of the crime and the “minor harmful consequences” of the crime. The perpetrator was sentenced to a penalty and punishment appropriate for the offense in the case of Sun Shanwen v Public Security Bureau of Da’an City. The majority of the decisions relating to Article 247 of the Criminal Law were in accordance with the jus cogens prohibition of torture under international law and constituted a modification of international law. The Dandong Intermediate People’s Court concluded that Wang Chang and Liu Chenyang, as judicial officers, had forced suspects to confess their guilt by using torture and that their conduct constituted the crime of extorting confessions by torture. The court also concluded that the victims’ death was directly related to the perpetrators but that the two perpetrators could be exempted from criminal punishment. However, the judgement did not explain the court’s reasoning for the finding of mitigating circumstances and lack of social harm. The perpetrators of the case were exonerated from criminal punishment and the victims did not receive state compensation, and contradictory legal reasoning sequentially could be drawn from the judgment. In the cases of the Prosecutor v Zhuang Hanzhong and others, the Court of First Instance and the Court of Appeal unanimously concluded that the perpetrator’s conduct, a police officer who beaten the victim for no reason while questioning him as a suspect in a public security case, resulting in minor injuries, constituted the crime of intentional injury under Articles 247 and 234 of the Criminal Law. In contrast, in the case of Xiong Dongsheng v Public Security Bureau of Yunmeng County, the Court of First Instance condemned Dongsheng, who caused injuries by torturing the victim to obtain a confession under Article 247. All these cases show that the police tortured the victims to extract confessions resulting in minor physical injuries to the victims. However, the application of the law is inconsistent, with some cases applying only Article 247 of the Criminal Law, and others involving both Article 247 and Article 234. This situation could lead to sentences with different consequences, but the court did not explain the circumstances under which Article 234 could be applied.

6. Conclusions

China has made some efforts to fulfill its obligations to prohibit and prevent torture. However, most of the existing studies are based on doctrinal discussions and lack empirical analysis. Thus, this article finds a new method for qualitative comparative analysis to explore the impact of multiple factors on torture prevention and prohibition how to achieve strong human rights protection. This article demonstrates how human rights go local and how local organs should enact a strong prohibition of torture through three pathways, capturing the core elements of the policy-making process in order to better guide judicial decision-making in the future. One of the most significant findings to emerge from this study is that cooperation with international monitoring bodies is crucial in establishing a high level of human rights protection against torture. Significantly, China has not presented its sixth periodic report to the Committee Against Torture, and it has been overdue since December 2019. Three types of configuration conditions will constitute the driving path for a high-level human rights protection system that could combat torture. The potential substitution relationship between the identified conditions suggests that, given specific objective endowments, competence in torture law-making, legal reasoning, and applicable law could build strong anti-torture and human rights protection objectives through equivalent substitutions in a “homogeneous” manner. The findings reported here shed new light on legal reasoning and applicable law as subjective and controllable conditions that are viable effective options for this cause.

6.1. Core Methodological Contribution

Concerning the methodological contributions related to the QCA, several points should be highlighted. First, this article is based on existing theoretical foundations that frame the research model. A rooted theory approach is used to select the conditional variables. Second, this paper empirically explores the concurrent synergies and linkage patterns of legislation and judicial practice, transparency in the justice system and attention allocation to torture cases. In addition, it explores the interaction between internal and external prevention and prohibition bodies in promoting torture prohibition and prevention through a Configuration Framework. Based on the existing theory, it explains the “causal complexity” that impacts China’s torture prohibition and prevention to achieve a high level of human rights protection. Finally, in this paper, the fs/QCA approach allows us to clarify the equivalent driving mechanisms and substitution relationships between conditions that promote this cause. It also helps in exploring the factors that contribute to low levels of human rights protection from a “causal asymmetry” perspective.

6.2. Limitations and Steps for Further Research

There are some limitations to this study. First, the cases selected for this article (n = 32) were a small sample. Although this reflects “limited diversity”, when the number of cases analyzed is small, the conditions included in the QCA model are restrictive. Second, this study is influenced by the inherent limitations of qualitative research methods and some of the data handling tools used in the research process, such as rooted theoretical coding and variable calibration. Simultaneously, these cases are only from open platforms that are accessible to the public. There are many unpublished cases that have not been accessed and investigated.

In addition, the factors (conditional variables) selected for this study were derived from the textual content of previous scholarly literature that was searched for specific keywords such as “torture” and “China” and then extracted factors from these literatures related to the case studied in the article. As these factors are subjective and selective, they could only reflect how the configuration relationship between the particular case under study and the particular factor selected influences the outcome variable. The definition of the term “torture” discussed in the article is limited and is confined to the context of the cases and factors selected for the article. The analysis does not cover the whole of the current situation of torture prevention and prohibition in China, nor does it analyze torture from the perspective of other disciplines, such as political science or legal philosophy. In Chinese judicial practice, only an extremely small number of the cases selected for this article have provided victims of torture with a fair trial and remedies, while most of them have some degree of error and logical flaws in the application of the law and legal reasoning. Trying to comprehensively analyze the situation of torture prohibition and prevention in China is a complex and wide-ranging issue. Understanding the full dimensions of “torture” is affected by the imbalance between the public power and citizens’ rights as well as the level of the rule of law in China.

The QCA approach is effective in understanding the particular set of cases and factors related to torture in the selected contexts but is limited in analyzing the broad definition of “torture” in China. Based on the equivalence and asymmetrical nature of the QCA, some factors are absent in the resulting pathway does not mean that policy makers are unable to make a decision that could potentially contribute to the establishment of a high level of human rights protection in the context of anti-torture in China. Enhanced cooperation with international monitoring bodies and increased transparency in the judicial process remain the core condition and a necessity condition in the configuration pathways. This analysis also suggests that China should continue to publicly disclose judgments and relevant information on torture cases in its future judicial practice, as well as actively, implement the UNCAT obligation for States parties. Enshrining fair trials and promoting the rule of law as a crucial task in judicial practice.

With these limitations in mind, future research should consider the following steps. First, there should be an assessment of whether the three paths can indeed feasibly achieve a high level of human rights protection for the prohibition and prevention of torture in China. Second, the number of torture and ill treatment cases must be expanded beyond the criminal and administrative contexts to broaden our understanding of torture and ill treatment in non-custodial settings. In other words, by increasing the number of cases in different parts of China or the specific time periods, a wider range of factors affecting human rights protection could be detected. Third, the objectivity of rooted theory coding should be enhanced by exploring more appropriate methods of assigning fuzzy sets of affiliation to variables. Finally, by following a more qualitative and case-oriented approach, future research could also reveal configurations in different legal systems, since what works in the common law system may not be effective in China. This is certainly relevant and useful for stakeholders and policymakers when exploring and implementing good practices.