Public security forces in the PRC push for an ‘informatization (信息化)’ of their work, and increasingly apply analytical techniques for not only solving past crimes but also preventing future crimes. Such measures are far-reaching and part of a highly integrated framework which, according to Xi Jinping, is part of a larger policy approach for which his administration coined the phrase ‘social governance (社会治理)’. More and more different policy fields are drawn into what has been described as a ‘pluralization of security work’.
While there is still no nationwide program or concept for predictive policing in China, there are however local projects that make use of analytical data technology (e.g. in Zhongshan, Guangzhou and Hangzhou, Zhejiang) and several specific crimes (e.g. drug-related crimes and telecommunication fraud) are targeted by predictive policing measures in China. Additionally, the pervasive surveillance of the entire (Muslim) population in Xinjiang produces high volumes of data that are used for operational purposes in China’s ‘People’s War against Terror’. Daniel Sprick‘s latest paper in the Nordic Journal for Law and Social Research (available for free here) asks: Can China overcome the problems and drawbacks frequently discussed in the context of predictive policing in general?
In his paper, Daniel Sprick gives an overview of ongoing predictive policing programs and related technology- and data-driven undertakings of the Chinese security apparatus in the context of China’s comprehensive approach towards maintaining order. Based on these observations, he analyzes China’s potential for an effective predictive policing by scrutinizing the availability of possible solutions for the inherent flaws of predictive policing that are frequently conceptualized in the existing English-language literature on the matter.
He finds that it is the propagandistic value of predictive policing, promising fairness and justice to be the single most important end possibly even unsurmountable obstacle in establishing an effective crime reducing system of this kind in China. The Chinese security apparatus appears systemically unfit to critically evaluate, acknowledge error, re-adjust methodologies and adapt responses, which is an indispensable process in making predictive policing work. If predictive policing is seen as an instrument to further target specific (dissident) groups however, China may be able to successfully employ Big Data technology for this particular objective, “It is however not conceivable that this technology will substantially change police operation and police culture in China, it will rather amplify pervasiveness and bias of its practices.”
Daniel Sprick is an Associate Researcher and Lecturer at University of Cologne’s Chair for Chinese Legal Culture. He has publicized widely on law and criminal justice in China. Find out more about his work and get in touch with him here or on LinkedIn.
The automation and digitisation of justice (司法信息化 ‘judicial informatisation’) in China has been ongoing for two decades. The latest development is the emergence of “smart courts” (智慧法院), which are part of the Chinese party-state’s efforts to reform and modernise its administration of justice and governance capacity. The advent of Smart Courts is an example of the willingness of the party-state to harness new technologies for its governance reform goals. However, the academic reaction has not been uniformly enthusiastic; there is scepticism about the benefits of increased automation and digitisation. Straton Papagianneas explores the phenomenon in the course of his PhD at Leiden University. In this post he sets out to map the academic reaction to some of the smart courts’ implications among Chinese scholars.
Chinese scholarly work has traditionally been a neglected group in academic discussions, including in the English-language literature on automated and algorithmic justice, whereas the latter is thoroughly cited and discussed by Chinese scholars. However, China is at the vanguard of judicial automatization and digitisation. The implications of its development can certainly be useful for other jurisdictions. Therefore, their academic discussion deserves attention.
A Brief Introduction to Smart Courts
The definition of a “smart court” is difficult to capture, partly because different courts use different technologies. Among the different official definitions, the clearest one, from the 2017 New Generation AI Development Plan (2017 AI Plan, translation here), states that a court can be considered ‘smart’ if it has a:
“[…] courtroom data platform that integrates trials, personnel, data applications, judicial disclosure, and dynamic monitoring, and promotes the application of artificial intelligence in evidence collection, case analysis, legal document reading and analysis; realising a smart court trial system and smart trial capacity.”
A smart court is not necessarily a court where everything is completely automated, with a self-learning ‘robot judge’ adjudicating over cases independently from any human interference. It is a court where judges use software applications to conduct the judicial process in a digital environment. ‘Intelligent legal applications,’ that is, applications that can render expert legal advice or decision making based on big-data analytics and without human interference, are still limited (Sourdin, 2018).
Central to the smart court is the human-computer interaction that results from integrating different technological applications supported by algorithms and big-data analytics into the judicial process. These applications range from systems that can automatically prompt similar cases as a reference for judges, to systems that can process and cross-examine all collected evidence, to ones that can automatically detect contradictions or relevant information for the judge to review (Cui, 2020). Ultimately, it is still the judge that adjudicates, albeit with the aid of technology.
Following the 2017 AI Plan, there are different degrees of smart courts. Some are more ‘intelligent’ than others. For example, there are three types of Internet Court, in Beijing, Guangzhou, and Hangzhou. These courts provide full online dispute resolution for limited types of e-commerce disputes (Xu, 2017). All activities, from the filing of a case through to the enforcement of a judgment, can be conducted online, with litigating parties and the judge all connecting remotely.
It is only later that these Internet Courts progressed to a ‘higher’ level of intelligence. Recently, the Hangzhou Internet Court introduced an AI judge that can take over simple functions during online court trial hearings, thereby assisting human judges, who still monitor the proceedings and make the final decision (Mei, 2019). According to its 2019 White Paper, the Beijing Internet Court developed an intelligent judgment generation system that is able to automatically generate standard instruments, as well as judgments, rulings, and settlements.
Hangzhou, where the PRC’s first smart court was established in 2017
Their Purpose
The 2017 SPC Opinion on Accelerating the Building of Smart Courts (translation here) explains that the purpose of judicial informatization and smart courts is to achieve the following judicial reform goals of making the judiciary more efficient and improving its transparency, consistency, and even autonomy from unwanted internal and external interference.
Remarkable is that technologies are treated almost like a ‘cheat code’ to bypass genuine, structural reforms, which requires an internalisation of norms and changes in behaviour that take a lot more time and effort. Introducing applications that force behavioural change is easier.
However, Chinese scholars are less confident. Smart courts and judicial informatisation are primarily framed as set to improve judicial efficiency and consistency. The advantages can be considered as evident, yet a review of the literature shows that there are doubts they can help achieve the ultimate aim: namely restoring the faith in and credibility of justice.
Efficiency
Efficiency (more output for less input) is low-hanging fruit and is therefore often mentioned as one of the advantages of these smart systems. Automation and digitisation will make the judicial process run smoother and faster, at a lower cost. In general, the efficiency benefits of judicial informatisation are left unquestioned (Guo 2017; Pan 2017; Qian 2018).
Chen and Sun (2019) show that digitisation has only gone so far, and that many judicial institutions have developed isolated data-silos. Many judicial departments have their own databases, but due to secrecy requirements, this data barely moves around between judicial organs. Additionally, for the data to be useful it still needs to be manually selected, cleaned, interpreted, and then finally labelled; increasing the workload of judicial officers after a case is done.
However, Wang (2019) notes that this efficiency discussion is only relevant for ‘traditional technology’, whereas smart technology driven by algorithms and big-data analytics are aimed to achieve much more, such as more accountability, more consistent adjudication, better monitoring and supervision of cases etc. The implications go far beyond an expedited judicial process.
By equivalating efficiency with “a more just and fairer judiciary”, reform goals of a more abstract level are implicitly achieved despite not being explicitly addressed. Technology is not a ‘magic weapon’ that will suddenly help achieve, for example, judicial credibility and fairness. An efficiently automated judiciary, does not, in and of itself, constitute a credible and just judiciary.
Consistency
A major issue plaguing the Chinese judiciary has been inconsistent adjudication, caused by the relative vagueness of laws and different interests trying to influence the judicial decision-making process to the detriment of consistent application of law (Ahl, 2019; Ng & He, 2017). Alongside previous judicial reforms (Ahl, 2014; Ahl & Sprick, 2017), smart courts are expected to enhance consistent adjudication or “similar judgments in similar cases” (同案同判).
Judicial databases feed into applications that conduct big-data analyses to provide adjudicating judges relevant references, or warn them that their judgment is deviating too much from the average judgment of previous, similar cases. Consistency is thus achieved, partly through automation, but also through the supervision of adjudication judges by these applications.
This has worried scholars. Technology should not be more authoritative than the human judges themselves (Y. Liu, 2019). Substantive justice is related to considering the unique circumstance of a case. Automated systems cannot maintain this balance between staying consistent while also considering unique factors. This is only something that human judges with sufficient judicial discretion can achieve. ‘Prefab’ judgments via nearly automatized decision-making would severely damage this (Pan, 2018)
Sun (2019) and Wang (2019) foresee the end of judicial discretion by this fully technologically embedded judicial process that minimises human interference. Judges would become screening bureaucrats that only concern themselves with inputting the right information in the automated system and reviewing its output.
The judicial system risks surrendering its power to technology, shifting the nexus of decision-making power to technical expertise. Judicial pluralism will be endangered by an exaggerated focus on uniformity and automation (P. Liu & Chen, 2019).
This can lead to ‘technological alienation’. Litigating parties can become frustrated by rigid automated system deciding over their cases, subverting the reform goals of restoring judicial credibility and faith (Jiang, 2019; Y. Liu, 2019).
Technology is being heralded as the bringer of a modern, efficient, and consistent judiciary. While this might be the case in most instances, a review of the literature shows the other side of the medal: The instrumental gains of judicial informatisation are no guarantee for a fairer and more credible judiciary.
What is clear from the literature is that judicial automation and digitisation needs to be accompanied by genuine reforms. ‘Cheating’ only gets one so far.
Straton Papagianneas explores China’s smart courts in the course of his PhD at Leiden University. Under the supervision of Dr. Rogier Creemers, he is part of the project called “The Smart State: Law, Governance and Technology in China”. Find him on LinkedIn here and follow him on Twitter.
Ahl, B. (2014). Retaining Judicial Professionalism: The New Guiding Cases Mechanism of the Supreme People’s Court. The China Quarterly, 217, 121-139. doi:10.1017/S0305741013001471
Ahl, B. (2019). Judicialization in authoritarian regimes: The expansion of powers of the Chinese Supreme People’s Court. International Journal of Constitutional Law, 17(1), 252-277.
Ahl, B., & Sprick, D. (2017). Towards judicial transparency in China: The new public access database for court decisions. China Information, 32(1), 3-22. doi:10.1177/0920203X17744544
Cui, Y. (2020). Artificial Intelligence and Judicial Modernization. New York: Springer Publishing.
Guo, S. (2017). Informatisation of the Judicial Process – Preliminary Study of Building Courts for the Internet Age (司法过程的信息化应对———互联网时代法院建设的初步研究). Jinan Journal (暨南学报)(10), 25-32.
Jiang, Q. (2019). The Scope and Limits of using AI in Judicial Adjudication ( 论司法裁判人工智能化的空间及限度). Academic Exchange (学 术 交 流)(2), 92-104.
Liu, P., & Chen, L. (2019). The Datafied and Unified Evidence Standard (数据化的统一证据标准). Journal of the National Prosecutors College (国家检察官学院学报)(2), 129-143.
Liu, Y. (2019). The Theory and Practice of Modernization of Trial System and Trial Capacity in the Era of Big Data (大数据时代审判体系和审判能力现代化的理论基础与实践展开). Journal of Anhui University (安徽大学学报)(3), 96-107.
Ng, K. H., & He, X. (2017). Embedded Courts: Judicial Decision-Making in China. Cambridge: Cambridge University Press.
Pan, Y. (2017). The Value and Position of AI Application in the Judicial Field(人工智能介入司法领域的价值与定位). Current Affairs Observations (时事观察)(10), 101-106.
Pan, Y. (2018). Analysis of Integrating AI into the Judicial Field (人工智能介入司法领域路径分析). Eastern Legal Studies (东方法学)(3), 109-118.
Qian, D. (2018). China’s Process of Judicial AI: Function Replacement and Structural Enhancement (司法人工智能的中国进程：功能替代与结构强化). Legal Review (法学评论)(5), 138-152.
Sourdin, T. (2018). Judge v. Robot: Artificial Intelligence and Judicial Decision-Making. UNSWLJ, 41, 1114.
Sun, D. (2019). Knowledge Deconstruction and Corresponding Logic of China’s Criminal Judicial Intelligence (我国刑事司法智能化的知识解构与应对逻辑). Contemporary Law (当代法学)(3), 15-26.
Wang, L. (2019). The Dangers and Ethical Regulation of Using Judicial big data and AI Technology (司法大数据与人工智能技术应用的风险及伦理规制). Law and Business Research(2), 101-112.
Xu, A. (2017). Chinese judicial justice on the cloud: a future call or a Pandora’s box? An analysis of the ‘intelligent court system’of China. Information & Communications Technology Law, 26(1), 59-71.
