ENQUIRE PROJECT DETAILS BY GENERAL PUBLIC

Project Details
Funding Scheme : Early Career Scheme
Project Number : 27603721
Project Title(English) : The Socialist Reception of Precedent: A Computational Analysis of Guiding Cases in the People’s Republic of China 
Project Title(Chinese) : 先例制度在社会主义法律体系的发展:中国指导性案例的量化计算分析 
Principal Investigator(English) : Prof Chen, Minhao Benjamin 
Principal Investigator(Chinese) :  
Department : Faculty of Law
Institution : The University of Hong Kong
E-mail Address : benched@hku.hk 
Tel :  
Co - Investigator(s) :
Panel : Humanities, Social Sciences
Subject Area : Social and Behavioural Sciences
Exercise Year : 2021 / 22
Fund Approved : 696,320
Project Status : Completed
Completion Date : 31-8-2024
Project Objectives :
To computationally analyse implicit references by lower courts to the Guiding Cases by using an optimized text reuse algorithm to detect unattributed copying from the Guiding Cases. Methods and early findings will be presented at a conference on empirical studies of the Chinese legal system to be hosted by HKU. Preliminary results will be reported as part of the conferencing proceedings. The final product will be disseminated in the first article targeted for journals like Journal of Empirical Legal Studies or Journal of Law and Courts.
To computationally analyse the impact of Guiding Cases on the type of disputes litigated in lower courts by using an optimized language model to evaluate the similarity between a lower court judgment and a Guiding Case. Methods and early findings will be presented at a conference on empirical studies of the Chinese legal system to be hosted by HKU. Preliminary results will be reported as part of the conferencing proceedings. The final product will be disseminated in the first article targeted for journals like Journal of Empirical Legal Studies or Journal of Law and Courts.
To examine the theoretical implications of empirical findings for legal development and judicial politics in China. Findings and arguments will be presented as part of a research trip to Durham University or ETH Zurich to audiences in the United Kingdom and Continental Europe. They will also be advanced in a second article targeted for journals like American Journal of Comparative Law or the Columbia Journal of Asian Law.
To examine the theoretical implications of empirical findings for general theories of comparative law and judicial decision-making. Findings and arguments will be presented as part of a research trip to Durham University or ETH Zurich to audiences in the United Kingdom and Continental Europe. They will also be advanced in a second article targeted for journals like American Journal of Comparative Law or the Columbia Journal of Asian Law.
To analyse commonalities between the language models that perform best in retrieving similar cases to derive generalizable insights into machine processing of text in a specialized domain. Depending on circumstances, conclusions draw from the experiments may be presented at a forum like the International Conference on Artificial Intelligence and Law.
Abstract as per original application
(English/Chinese):
Is the idea of precedent anathema to socialist legality? If so, is the aversion to case law merely superficial or is it instantiated by judicial practice and legal outcomes? And what do the answers to the questions tell us about how legal ideas morph and spread throughout the globe? This project explores the receptivity of socialist legal systems to the common law notion of precedent. In contrast to common law jurisdictions like the United States, the United Kingdom, and Hong Kong, socialist and civil law regimes do not recognize judicial opinions as establishing a legal rule for future cases. Under socialist legal ideology, judges administer the rules laid down by legislatures but do not pretend to expound them. Because judicial opinions are binding only on the disputing parties, cases do not alter the legal fabric. Recent legal developments in the Socialist Republic of Vietnam and the People’s Republic of China are blurring the divide between legal systems that are built on precedent and those that are not. Over the last decade, the supreme courts of the two socialist regimes have sought to interpret and unify national law by promulging cases that all judges must take reference from. In contrast to the Vietnam, however, reception of this innovation in China has been lukewarm. According to empirical studies conducted by researchers at the City University of Hong Kong, Peking University and Stanford University, Guiding Cases are very rarely cited in Chinese judgments. The divergence between China and Vietnam might be due to the former’s commitment to socialist legality and the latter’s estrangement from it. But geo-political stature and ambition may also furnish an explanation. We revisit conventional wisdom about the Chinese resistance to precedent by applying computational methods to study how the Guiding Cases shape legal disputes and discourse in the lower courts. This project traces the influence of Guiding Cases by looking beyond citations—overt references—to search for echoes of their reasoning—or implicit references—in the decisions of lower courts. It also explores the impact of a Guiding Case on the kinds of disputes that are channeled into the judicial system. The novel approach taken here promises not only to illuminate the practical operation of the Chinese legal system but also to contribute to a more sophisticated understanding of how legal concepts and values travel and how they are transformed.
先例的概念与社会主义法治相抵触吗?如果是这样,对判例法的排斥仅仅是表面上的呢?还是由司法实践和结果所导致的?这些问题的答案能告诉我们法律概念是如何在全球范围内演变和传播的吗?这个课题探讨和洞察社会主义法律体系对普通法先例概念的接受程度。与美国、英国和香港这样的普通法系辖区不同,社会主义和大陆法系不将司法判决看作是为未来的案件建立法律规则的一种正式的法律渊源。在社会主义法律意识形态下,法官执行运用立法机关制定的规则,但并不佯装对其进行延展。由于司法判决仅对争议双方具有约束力,因此案件并不会改变法律原有的结构和规则。 越南和中国最近的法律发展正在模糊建立在先例基础上的法律体系和没有先例的法律体系之间的界限。在过去的十年中,这两个社会主义国家的最高法院通过颁布指导性案例来解释和统一国家的法律,而这些案例是要求所有法官在类似案件中必须参照的。然而,与越南相比,中国法院对这一改革的采纳程度不高。从香港城市大学、北京大学和斯坦福大学的学者的实证研究结果来看,中国的判决中很少引用指导性案例。中国和越南之间的分歧可能是由于前者对社会主义法制的坚持,而后者则与之逐渐失和。但地缘相关的政治地位和追求也可以阐释这一差异的体现。 我们通过计算机科学来探讨最高法院公布的指导性案例是如何影响下级法院所处理的法律纠纷,从而重新审视之前有关地方法院抵触先例的研究结果的可靠性。该项目通过跟踪下级法院的判决中对指导性案例明示的引用或默示的参照,来量化指导性案例在地方法院审判实践中的影响。我们的课题也将探讨指导性案例对今后在法院起诉的纠纷类型所产生的影响。课题中所采用的机器学习这一前沿的研究方法不仅有望阐明中国的司法实践过程,而且有助于我们更深入地理解法律概念和价值是如何传播以及如何转化的。
Realisation of objectives: 1. The first objective was fully accomplished. By adapting the Basic Local Alignment Search Tool (or BLAST) algorithm, typically used to compare genetic sequences, for Chinese characters, the investigators were able to identify unattributed references to the Supreme People’s Court’s Guiding Cases in the judgments of Beijing and Shanghai courts. These unattributed references were then validated by human labellers and analyzed. Quantitative findings at various stages of the study were presented at the Conference on Empirical Legal Studies in Asia, hosed by Academica Sinica in Taipei and Computational Legal Studies, hosted by Singapore Management University. The methodology and results are published in the article “Detecting the Influence of the Chinese Guiding Cases: A Text Reuse Approach” which appeared in Artificial Intelligence and Law in May 2023. 2. The second objective was partially accomplished. By estimating document embeddings from a corpus of Beijing and Shanghai court judgments, investigators were able to generate clusters of similar cases. However, these clusters were not sufficiently discriminating to be reliable for social scientific causal inference. Moreover, the drawing of causal inferences, even from big data, requires identifying assumptions that may or may not obtain. Some of the challenges are described and illustrated in the article “Data Still Needs Theory: Collider Bias in Empirical Legal Research” which appeared in the Hong Kong Law Journal in December 2023. 3. The third objective was fully accomplished. By using regular expressions to count the number of attributed references in the same corpus of Beijing and Shanghai judgments, the investigators were able to give some indication of the extent to which the previous studies underestimate the significance of Guiding Cases. Contrary to the literature, the results of this project suggest that the Guiding Case System is neither marginal nor dysfunctional. These arguments and their implications for our understanding of the PRC legal system were explored in a seminar given at the Southwestern University of Finance and Economics and summarized in a blogpost by the European Chinese Law Research Hub, housed in the University of Cologne. 4. The fourth objective was fully accomplished. Through a close reading of the matches generated by the BLAST algorithm, the investigators discerned two techniques that PRC judges employ to implement a Guiding Case without citing to it: statutory ventriloquy and jurisprudential retracing. These phenomena indicate that the doctrinal rejection of cases as precedent may be more form than substance. These arguments and their implications for the comparative study of law and legal institutions were explored in seminars given at ETH Zurich and Durham University. 5. The fifth objective was not substantially accomplished. Given (1) the limited utility of document embeddings in making fine distinctions between cases, as related above, (2) the overarching purpose of the project and (3) the timeframe of the project, the investigators determined that effort and resources would be better allocated to more promising avenues of inquiry illuminated by the successful application of text reuse methods to judicial documents.
Summary of objectives addressed:
Objectives Addressed Percentage achieved
1.To computationally analyse implicit references by lower courts to the Guiding Cases by using an optimized text reuse algorithm to detect unattributed copying from the Guiding Cases. Methods and early findings will be presented at a conference on empirical studies of the Chinese legal system to be hosted by HKU. Preliminary results will be reported as part of the conferencing proceedings. The final product will be disseminated in the first article targeted for journals like Journal of Empirical Legal Studies or Journal of Law and Courts. Yes100%
2.To computationally analyse the impact of Guiding Cases on the type of disputes litigated in lower courts by using an optimized language model to evaluate the similarity between a lower court judgment and a Guiding Case. Methods and early findings will be presented at a conference on empirical studies of the Chinese legal system to be hosted by HKU. Preliminary results will be reported as part of the conferencing proceedings. The final product will be disseminated in the first article targeted for journals like Journal of Empirical Legal Studies or Journal of Law and Courts. Yes75%
3.To examine the theoretical implications of empirical findings for legal development and judicial politics in China. Findings and arguments will be presented as part of a research trip to Durham University or ETH Zurich to audiences in the United Kingdom and Continental Europe. They will also be advanced in a second article targeted for journals like American Journal of Comparative Law or the Columbia Journal of Asian Law. Yes100%
4.To examine the theoretical implications of empirical findings for general theories of comparative law and judicial decision-making. Findings and arguments will be presented as part of a research trip to Durham University or ETH Zurich to audiences in the United Kingdom and Continental Europe. They will also be advanced in a second article targeted for journals like American Journal of Comparative Law or the Columbia Journal of Asian Law. Yes100%
5.To analyse commonalities between the language models that perform best in retrieving similar cases to derive generalizable insights into machine processing of text in a specialized domain. Depending on circumstances, conclusions draw from the experiments may be presented at a forum like the International Conference on Artificial Intelligence and Law. Yes10%
Research Outcome
Major findings and research outcome: Jurisprudentially, cases are not a source of law in socialist legal systems, including that of the People’s Republic of China. In 2011, however, the Supreme People’s Court introduced the Guiding Case system. Guiding Cases are selected by the Supreme People’s Court and must be referred to by courts at all levels when adjudicating similar controversies. The practical impact of Guiding Cases on judicial practice has however been called into question by many legal scholars and commentators. Previous studies almost uniformly find that citations to Guiding Cases are sparse, and many Guiding Cases are not cited at all (Daum 2017; Zhang 2018). The ‘extremely low’ incidence of citations is taken as symptomatic of ‘the dysfunction of the [G]uiding [C]ase system as a type of case law’ (Wang 2019) and as proof of the incongruity of case-based adjudication in China (Ahl 2014; Zuo & Chen 2015; Finder 2017; Jia 2016). However, citations might not be an accurate measure of the influence of cases. The premise of our research is that reuse of text uniquely ascribable to the Main Points of Adjudication is a strong indicium of the influence of Guiding Cases on judicial decision-making. The Basic Local Alignment Search Tool or BLAST algorithm is employed to search for text reuse in local court decisions. Ultimately, we find more instances of unattributed text reuse than citations for the Guiding Cases examined. A close reading of local court decisions reveals that it is not uncommon for them to cite statutory or regulatory instruments for propositions reproduced from a Guiding Case (“statutory ventriloquy”). In addition, rather than quote the Main Points of Adjudication as directed by the Supreme People’s Court’s Detailed Rules, many judgments simply repeated the various legal arguments adduced in the Guiding Case (“jurisprudential retracing”). To conclude, previous studies report that Guiding Cases are rarely cited in judicial decisions, suggesting that their practical effect is negligible. The anaemic impact of Guiding Cases has sometimes been taken as demonstrating the fundamental incompatibility between socialist legality and judicial precedent. We demonstrate, however, that Guiding Cases are more influential than conventionally thought. By uncovering how the Guiding Cases are referred to other than by name, we not only illuminate the operation of Chinese law but also inform debates about legal traditions, cultures, and transplants.
Potential for further development of the research
and the proposed course of action:
Most immediately, the research can be developed in two major directions. The first direction is to broaden the scope of the study to include other subnational jurisdictions beyond Beijing and Shanghai. Variation between jurisdictions might uncover the social, economic, and cultural factors that animate the refusal to cite Guiding Cases even when they are being adhered to. The second direction is to test text reuse algorithms other than BLAST, for instance, Smith-Waterman algorithm. The Smith-Waterman algorithm is supposed to be more accurate but also more computationally expensive than the BLAST algorithm. A comparison between BLAST and Smith-Waterman could generate insights into the trade-offs between the two algorithms in the context of PRC legal text. Efforts have been taken to further the research along these dimensions. Recently, the investigators have obtained access to a database that contains decisions from local courts in all provinces. The reliability and completeness of this database is still being accessed and lies outside the parameters of the funded project. In the longer term, researchers may consider applying the techniques developed in this research to investigate the influence of cases in other jurisdictions that do not recognize precedent as a source of law.
Layman's Summary of
Completion Report:
In socialist legal systems like that of the People’s Republic of China, cases are not traditionally considered a source of law. However, in 2011, the Supreme People’s Court introduced the Guiding Cases system. Guiding Cases are selected by the Supreme People’s Court and must be referred to by lower courts where relevant. The practical impact of the Guiding Cases has been widely questioned. Previous studies report that they are rarely cited, suggesting their lack of importance. We adopt a novel approach to analyse the influence of Guiding Cases. Instead of counting citations, we examine local court decisions for the reuse of text from the Guiding Cases. We find many instances of such reuse, indicating a subtle but distinct influence of Guiding Cases on how local courts decide disputes. We also find that local courts often apply Guiding Cases by copying their legal reasoning or mixing-and-matching language from other sources. In so doing, local courts avoid citing Guiding Cases even when they are following them. In conclusion, the Guiding Cases are more significant than they might seem at first glance. This and other discoveries enrich our understanding of diverse legal traditions and cultures.
Research Output
Peer-reviewed journal publication(s)
arising directly from this research project :
(* denotes the corresponding author)
Year of
Publication
Author(s) Title and Journal/Book Accessible from Institution Repository
2023 Benjamin M Chen, Zhiyu Li,* Daniel Cai, Elliott Ash  Detecting the Influence of the Chinese Guiding Cases: A Text Reuse Approach  No 
2023 Benjamin M Chen* and Xiaohan Yin  Data Still Needs Theory: Collider Bias in Empirical Legal Research  No 
Recognized international conference(s)
in which paper(s) related to this research
project was/were delivered :
Month/Year/City Title Conference Name
Taipei Precedent and Socialist Legality: The Quiet Influence of the Chinese Guiding Cases  Conference on Empirical Legal Studies in Asia 
Singapore The Silent Influence of the Chinese Guiding Cases: A Text Reuse Approach  Computational Legal Studies 
Other impact
(e.g. award of patents or prizes,
collaboration with other research institutions,
technology transfer, etc.):

  SCREEN ID: SCRRM00542