This research uses critical discourse analysis (CDA), specifically socio-cognitive discourse analysis (SCA), also known as the "Discourse-cognitive-society Triangle," to examine Chinese court judgments (CCJ). SCA, developed by van Dijk, analyzes discourse considering psychological representation, processing, shared knowledge, and ideology. Unlike traditional sociolinguistics, SCA emphasizes the cognitive processes behind social interactions within discourse. The study focuses on Chinese court judgments because they represent a unique legal discourse, combining logical reasoning with tangible evidence. The researchers aim to explore the deep relationship between discourse components, cognitive sources, and social functions within these judgments. A corpus of 1.74 million words, comprising 54 civil, 57 criminal, and 5 administrative judgments from the Supreme People’s Court (SPC) and local people’s courts, forms the basis of the study. The importance lies in understanding how these judgments reflect both social structures and cognitive processes in legal decision-making.

Existing research on legal discourse often focuses on the wording, syntax, and style of legal texts, with comparative studies highlighting differences in discourse styles across countries. While critical discourse analysis (CDA) explores language's role in social functions and power dynamics, less attention has been given to legal language itself. Prior CDA studies of court judgments examine intertextuality, interpersonal relationships, and communicative purposes, highlighting the influence of past judgments and interactions between various participants. Researchers have also looked at specific linguistic features such as rhetorical preferences, reflexive pronouns, epistemic modality, reporting verbs, and the use of evidentiality in judgments, revealing how these elements contribute to legal fact-construction and judicial reasoning. However, this study differentiates itself by integrating corpus linguistics and socio-cognitive discourse analysis to provide a more comprehensive understanding of the interplay of discourse, cognition, and social factors within Chinese court judgments.

The study employs a corpus-based approach within CDA, using a corpus of 1.74 million words of Chinese court judgments. The methodology involves several steps: Firstly, the researchers use the corpus tools Wmatrix3.0 and ConcGram1.0 for identifying expressions indicating cognitive sources and social functions. Wmatrix3.0 performs automatic semantic analysis based on the UCREL Semantic Analysis System (USAS) tagset, while ConcGram1.0 generates word frequency lists and analyzes co-text. Secondly, they manually label and categorize discourse components (vocabulary, phrases, sentences), cognitive sources (faith, induction, paraphrase, inference), and social functions (citation, depiction, distance, summary). Thirdly, linear regression models (LRM) using Stata16.0 are constructed to explore quantitative relationships between these variables. The models investigate how cognitive sources and discourse components influence social functions. Fourthly, a variance inflation factor (VIF) test using SPSS23.0 is conducted to address multi-collinearity issues among variables. Finally, to ensure the robustness of the results, the researchers re-run the analysis using the Tobit method. The study uses various models (Equations 1-18) to analyze the relationships between different variables, including the impact of various discourse components (e.g., first-person vs. third-person subjects, adverbs indicating confidence levels, prepositional phrases, verb-object phrases, conditional clauses, and causal clauses) on cognitive sources and social functions.

The study yields three key findings: First, the discourse dimension of CCJs is both society-oriented and cognition-oriented. Discourse components serve as markers of cognitive sources. Second, the cognitive source of faith is rooted in social cognition of law and regulations. Induction and paraphrase reflect personal cognition applied to evidence, while inference transforms personal cognition into social consensus through reasoning. Third, the social function of CCJs aligns with cognitive sources and surface structures formed by various discourse components. Specifically, the regression analysis reveals that: * The cognitive source of "faith" (Fai) positively influences the social function of "citation" (Cit), with phrases and sentences also playing a significant role. * "Induction" (Ind) and "paraphrase" (Par) have positive effects on "depiction" (Dep). * Paraphrase strongly influences "distance" (Dis). * "Inference" (Inf) significantly affects "summary" (Sum), along with phrases and sentences. Further analysis of the citation function shows that law, regulations, and evidence (as sources of faith) strongly affect citation, alongside phrases and sentences. Additional models examining the influence of various discourse components on specific social functions and cognitive sources reveal that particular vocabulary, phrases, and sentence structures play distinct roles in shaping these elements within CCJs. Robustness checks using the Tobit method confirm the validity of the findings.

The findings demonstrate the intricate relationship between discourse, cognition, and social context in Chinese court judgments. The study's approach, integrating corpus linguistics and SCA, successfully reveals the subtle ways in which cognitive processes and social factors shape the language of legal judgments. The identification of specific linguistic features associated with different cognitive sources and social functions provides a valuable tool for understanding how legal meaning is constructed and communicated. The results highlight the importance of considering both social cognition (reflected in the reliance on established laws and regulations) and personal cognition (evident in the use of induction and paraphrase) in legal reasoning. The transformation of personal cognition into social consensus through inference demonstrates the dynamic nature of legal interpretation. This study contributes to a deeper understanding of how legal discourse functions, offering practical implications for legal text interpretation and judicial decision-making in the Chinese context.

This research offers a novel corpus-based analysis of Chinese court judgments using socio-cognitive discourse analysis, revealing a complex interplay between discourse, cognition, and social context. The study highlights the dual society- and cognition-oriented nature of these judgments and clarifies the connections between cognitive sources, social functions, and specific linguistic features. Future research should focus on expanding the corpus size, investigating the impact of different audiences' understanding, and exploring the role of inter-semiotic operations in legal interpretation more comprehensively. The creation of a large, shared corpus would enable a more thorough examination of these complex interactions and enhance the generalizability of the findings.

The study acknowledges limitations stemming from the use of a specific corpus of Chinese court judgments. While the corpus is large (1.74 million words), it may not fully represent the diversity of cases and styles across all Chinese courts. The study primarily focuses on the language of the judgments themselves, neglecting the broader communicative contexts surrounding the production and reception of these texts, such as the interactions between judges, lawyers, and litigants. Additionally, the quantitative approach, while valuable, does not capture the full nuance and complexity of legal interpretation.