Abstract The Climate Change Advisory Opinion (AO) by the International Court of Justice (ICJ) demonstrates the growing prominence of general principles of law in international law. The Climate Change AO was handed down at the end of the International Law Commission's project on general principles of law with the adoption of its Draft Conclusions. In the Climate Change AO, the ICJ accords general principles of law particular importance in environmental protection. This article documents how States identified general principles of law as the bedrock of the international climate change regime, and how the ICJ employed a systematic approach to ‘thicken’ climate change law, both in terms of normative content, obligations and consequences of breach. It then examines the general principles of law affirmed by the ICJ, in particular, the principles of common but differentiated responsibilities and intergenerational equity, both extracted from the broader general principle of equity. These principles guide the interpretation of ‘how far’ or ‘how much’, operating as balancing tools in relation to other obligations. The broader significance of this development lies in the ICJ's growing recognition of general principles of law as a means of supporting and structuring its legal reasoning. The article further argues that the normative development of these principles has been reinforced by reports of the Intergovernmental Panel on Climate Change (IPCC), and that Article 38(1)(c) of the ICJ Statute provides a broader gateway for taking account of normative contributions by actors such as the IPCC. The identification of customary law and peremptory norms ( jus cogens ) is more narrowly defined than general principles of law. The article concludes by examining the IPCC's role in underpinning the normative character of certain general principles of law, building on the interaction of law and science, and suggests that strengthening these principles may facilitate their more robust incorporation into future treaty‐design mechanisms.

Abstract In recent years, models of a fortiori argumentation from the field of artificial intelligence and law, developed to describe legal case-based reasoning based on precedent, have been successfully applied to improve interpretability of data-driven decision systems. To aid with these applications, we further develop the theory of a fortiori case-based reasoning by extending the knowledge representations on which these models operate. More specifically, we modify the representations to accommodate incomplete information, as well as to incorporate both dimensional (as opposed to binary) and hierarchical (as opposed to unstructured) information. This results in four models—one for each combination of accommodating dimensional or hierarchical information. We investigate their formal properties, and find they are monotonic with respect to the addition of new precedents and of new facts, and that some are conservative extensions of other models. In addition, we exemplify each through a running example from the penitentiary law domain.

Abstract Crime scene analysis requires the evaluation of multiple factors to determine a suspect’s guilt, a process that can be lengthy and costly. The integration of Artificial Intelligence (AI) into the judicial system is emerging as an opportunity to improve the efficiency of investigations and legal decision-making. In this study, we propose a Machine Learning (ML)-based methodology to support the assessment of road homicide cases under Italian law. Our approach employs a Large Language Model (LLM) to extract 51 features from crime scene descriptions automatically. Four ML models are then analyzed: (RF), (GBM), (DT), and (LR). We evaluated the performance of these models on a dataset of 100 road homicide court rulings in Italy, achieving 95% accuracy in crime classification. The validation was conducted by comparing the model outputs with a legal ranking established by four legal experts, allowing us to verify the consistency of the algorithmic predictions with human legal reasoning. The results indicate that the shows the highest correlation with legal evaluations ( $$\rho = 0.857$$ ρ = 0.857 , $$\tau = 0.714$$ τ = 0.714 , p values $$< 0.05$$ < 0.05 ), while performed the worst. This study highlights the potential of AI in legal decision support, emphasizing the need to ensure transparency and bias mitigation to comply with European Union Regulations while maintaining human judgment as the central authority in legal proceedings.

This research article examines the concept of Istihsan (juristic preference) within Islamic jurisprudence and highlights its significance in deriving legal rulings, particularly within the Hanafi school of thought. The study begins with a conceptual introduction to Istihsan and discusses its juristic definition and methodological foundations. It then analyzes the legal status of Istihsan in the context of the four major Sunni schools of law, presenting the perspectives of the Hanafi, Maliki, Shafi‘i, and Hanbali jurists with references from primary classical sources. Furthermore, the article explores various juristic applications of Istihsan in practical legal matters, demonstrating how this principle helps address complex social realities and provides flexibility within Islamic law. Through selected examples from fiqh literature, the study illustrates the role of Istihsan in facilitating ease, preventing hardship, and ensuring justice in legal rulings. Finally, a comparative analysis of the four schools is presented, showing that although the terminology and methodological acceptance of Istihsan differ among the jurists, its practical spirit is reflected in their legal reasoning in various forms. The research concludes that Istihsan serves as an important juristic tool that strengthens the adaptability and dynamism of Islamic law in changing circumstances.

This study examines Decision No. 8/Pdt.G.S/2024 of the Mataram District Court concerning the execution of fiduciary security following Constitutional Court Decision No. 18/PUU-XVII/2019. It focuses on the prohibition of unilateral execution by creditors and the legal reasoning behind the rejection of the debtor’s claims for material, immaterial, and loss-of-use damages. Using a normative juridical methodology, the study finds that the judges consistently applied the Constitutional Court’s interpretation, declaring the unilateral repossession of the vehicle without a court order as an unlawful act. However, the rejection of all compensation claims reflects insufficient protection for the debtor’s actual losses, particularly loss of use. The decision also highlights a normative gap regarding proof of non-economic damages in fiduciary execution cases. Therefore, more operational judicial guidelines are needed to ensure that civil judgments provide not only formal justice but also substantive justice for debtors.

Abstract Legal reasoning is complex and multi-faceted, requiring a broad set of skills. By employing domain knowledge from legal experts, we design five elements that can be included in prompts for large language models that could aid in legal reasoning tasks. We use additional legal guidelines, 1-shot prompting, dictionary definitions, knowledge representations of legal articles, and IRAC-style prompting. We investigate the effect of each prompt element on the model’s performance on a legal entailment task. Certain prompt elements can improve performance, depending on the context and the model. For the smaller models, increasing the number of prompt elements improves performance on average. For any particular combination of model and sub-task, only using a subset of the prompt elements seems to work best. For the most advanced reasoning model we evaluate, using a selection of prompt elements increases average performance across all evaluated sub-tasks. Results indicate that the problem space of the legal entailment task may be too large for a single model and prompt. In future research, we therefore aim to investigate the capabilities of an ensemble of specialized models.

The epistemological foundation of legal education has long been dominated by positivistic thought, strongly influenced by thinkers such as Hans Kelsen and H.L.A. Hart, emphasizing legal certainty, formal logic, and the mechanical application of rules. While this paradigm has been effective in cultivating technical competence and doctrinal precision, it has simultaneously marginalized the moral, philosophical, and social dimensions of law as a living and dynamic system. Consequently, legal education often produces graduates who are procedurally skilled yet insufficiently responsive to substantive justice and humanitarian concerns. This study aims to reconstruct the epistemology of legal education by shifting from a rigid positivist framework toward a humanistic paradigm that integrates ethical consciousness, critical reflection, and social responsibility. Using a normative-philosophical approach supported by conceptual analysis of classical and contemporary legal theories, including critiques advanced by Ronald Dworkin and Jürgen Habermas, this research examines how the dominance of positivism has constrained legal reasoning and distanced law from its emancipatory purpose. The findings reveal that formalistic and text-centered approaches in legal education limit students’ capacity to engage with justice in its broader social and moral contexts. Therefore, this study proposes an epistemic reconstruction centered on dialogical learning, intersubjective understanding, and contextual engagement with social realities. By situating law within ethical, cultural, and societal frameworks, this humanistic model envisions legal education not merely as a mechanism for producing technically proficient jurists, but as a transformative process that nurtures reflective, compassionate, and socially responsible legal thinkers committed to the realization of substantive justice.

This review paper examines how bail and pretrial detention are applied in the United States and Canada, with attention to fairness, equality before the law, and the protection of individual rights. Bail decisions determine whether a person accused of a crime is released or held in custody while waiting for trial, even though they are legally presumed innocent. In the United States, the bail system relies solely on cash bail, meaning that many people remain in jail simply because they cannot afford to pay for their release. However, research shows that this approach affects low-income individuals and underrepresented groups, leading to unnecessary detention, loss of employment, family disruption, and a higher likelihood of conviction before trial. In contrast, Canada’s bail system is based on constitutional principles that emphasize the right to reasonable bail and the presumption of release. Canadian law requires that detention be justified by clear legal reasons, such as ensuring court attendance, protecting public safety, or maintaining public confidence in the justice system. Monetary bail plays a limited role, and courts are encouraged to use the least restrictive conditions possible. By comparing these two approaches, this paper highlights how legal frameworks shape real-world results in pretrial justice and discusses lessons that can inform ongoing reforms aimed at improving fairness and reducing unnecessary detention.

Abstract In the digital information age, artificial intelligence is increasingly being applied to national governance and judicial decision-making assistance. Existing studies lack case studies and empirical analyses of the effectiveness of large models in aiding judicial decisions. To address this research gap, this study designs a comprehensive evaluation framework encompassing five core task dimensions: Task-oriented Information Extraction, Legal Article Citation, Event Extraction, Judicial Decision Generation, and Legal Opinion Generation. By using carefully crafted prompts to activate the legal reasoning capabilities of the models, we conducted extensive testing on 13 mainstream large language models (LLMs). The experimental results demonstrate that large models perform excellently in processing legal texts and providing preliminary legal opinions, but still exhibit shortcomings in complex legal reasoning and precise decision-making. On this basis, we applied a weakly supervised learning strategy to fine-tune the LLMs for targeted improvements. The results indicate that introducing a small amount of task-specific learning can significantly enhance the performance of LLMs in judicial tasks. This further underscores the critical role of data and the acquisition of domain-specific knowledge in applying AI technology to judicial tasks. Additionally, this study briefly discusses the issue of the boundaries of AI’s involvement in judicial activities, aiming to provide theoretical foundations and practical guidance for the deep integration of AI technology with legal practice.

On 12 November 2024, The Hague Court of Appeal in Shell v. Milieudefensie overturned the preceding 2021 judgment which obliged Shell to reduce its CO2 emissions resulting from its global operations by 45%, relative to 2019 levels, by 2030. Indeed, the Court found that Shell's violation of legal obligations had not been sufficiently demonstrated in relation to its Scope 1 and 2 emissions and refused to affirm a concrete reduction target by 2030 of Shell's Scope 3 emissions. While the verdict cannot be celebrated as a success from a climate justice's perspective, the legal reasoning of the Court of Appeal provides some important takeaways concerning the doctrine of the indirect horizontal effect of human rights, the relevance attributed to EU climate legislation and the consequences of projected investments in new oil and gas fields. The present contribution focuses on these aspects, with a view to highlighting the major implications and the positive impact they may have for future climate litigation action.

Artificial intelligence is increasingly deployed in high-stakes decision-making, raising critical questions about accuracy, fairness, and transparency in regulated domains. This study evaluates the use of machine learning models to generate sentencing recommendations within the U.S. criminal justice system, examining whether such models can reliably support judicial decision-making without amplifying existing inequities. Using a comprehensive dataset of sentencing records enriched with engineered features reflecting criminal history, offense severity, demographics, and jurisdictional context, we develop and compare a range of predictive models, including Logistic Regression, tree‑based ensembles (Random Forest, XGBoost, LightGBM), deep learning architectures (MLP, LSTM, Bi‑LSTM), and hybrid ensemble frameworks. Models are assessed on both continuous sentence length prediction and classification of above‑median sentencing, using metrics such as mean absolute error, R‑squared, AUC‑ROC, and F1‑score. Fairness metrics are computed across gender, age, and jurisdictional groups, and interpretability analyses employ feature importance, attention weights, and SHAP values to ensure transparency in decision logic. Results indicate that hybrid and stacked ensembles achieve the best balance of accuracy and fairness improvements over baselines, with interpretability tools confirming alignment with legal reasoning and risk factors. These findings suggest that responsibly governed AI systems can augment sentencing decisions as decision‑support tools, provided continuous bias monitoring and ethical oversight are integrated into deployment practices. The study contributes empirical evidence and methodological guidance for integrating machine learning into judicial contexts.

This paper discusses a multimodal AI system applied to legal reasoning for tax law. The results given here are very general and apply to systems developed for other areas besides tax law. A central goal of this work is to gain a better understanding of the relationships between LLMs (Large Language Models) and automated theorem-proving methodologies. To do this, we suppose (1) two cases for the theorem-proving system: one where it has a countable number of total meanings for its countable number of atoms and the other is where it has an uncountable number of total meanings for its countable number of atoms, and (2) LLMs can have an uncountable number of token meanings. With this in mind, the results given in this paper use the downward and upward Löwenheim–Skolem theorems and logical model theory to contrast these two AI modalities. One modality focuses on syntactic proofs and the other focuses on logical semantics based on LLMs. Particularly, one modality uses a rule-based first-order logic theorem-proving system to perform legal reasoning. The objective of this theorem-proving system is to provide proofs as evidence of valid legal reasoning when enacted laws are applied to particular situations. These proofs are syntactic structures that can be presented in the form of narrative explanations of how the answer to the legal question was determined. The second modality uses LLMs to analyze and transform a user’s tax query so this query can be sent to a first-order logic theorem-proving system to perform its legal reasoning function. The main goal of our application of LLMs is to enhance and simplify user input and output for the theorem-proving system. Using logical model theory, we show how there can exist an equivalence between laws represented in logic of the theorem-proving system, fixed in time when the theorem-proving system was set up, and new semantics given by LLMs. These results are based on logical model theory and Löwenheim–Skolem theorems.

The rapid integration of artificial intelligence into legal practice raises fundamental questions about its compatibility with criminal justice, a field that has traditionally been based on human judgment and discretion. This relevance becomes particularly acute with regard to evaluative concepts, which are indispensable for context-sensitive decision-making but at the same time create risks of inconsistency and unpredictability. Against this backdrop, this article aims to assess whether artificial intelligence can function as an auxiliary tool for the application of evaluative concepts in criminal proceedings and whether such use is legally and economically justified. The object of the study is the application of evaluative concepts in criminal justice, and the subject is the economic and legal consequences of applying artificial intelligence to evaluative concepts. The study is based on doctrinal legal analysis, comparative legal reasoning, and the methodology of law and economics as a theoretical and methodological basis. By synthesising legal theory and economic analysis, the article considers artificial intelligence as a normative problem and as a tool for optimising economic efficiency. The article demonstrates that artificial intelligence can enhance analytical capabilities in criminal proceedings by systematising large volumes of case law, identifying patterns in the application of evaluative concepts, and highlighting deviations from established trends in decision-making. As a result, artificial intelligence can contribute to greater consistency and predictability in judicial practice. At the same time, the study reveals structural limitations of algorithmic approaches, in particular reduced sensitivity to unique contextual factors, difficulties in providing normative justification, and the risk of reinforcing existing interpretative patterns. From an economic perspective, the analysis shows that artificial intelligence has the potential to reduce transaction costs, optimise the allocation of judicial resources and speed up procedural decision-making, provided that its use remains auxiliary rather than substitutive. The practical value of the study lies in substantiating a balanced model for integrating artificial intelligence into criminal justice, in which algorithmic tools serve as analytical aids, while final decisions remain under human control, ensuring both efficiency and compliance with fundamental legal guarantees.

Judicial review of regulations below statutes by the Supreme Court of Indonesia plays a crucial role in safeguarding the hierarchy of laws and controlling executive regulations that have broad public impact. However, the procedural design of such judicial review has long been characterized by closed, document-based mechanisms with limited participation and transparency, raising concerns regarding the fulfillment of open justice and due process of law as minimum requirements for the legitimacy of erga omnes decisions. This study aims to examine whether the procedural design and practice of judicial review at the Supreme Court have met these principles and to formulate rational and constitutional procedural reforms without altering the Court’s authority. This research employs normative legal research methods using statutory, conceptual, and case-based approaches, supported by content analysis of selected judicial review decisions. The findings reveal that although procedural requirements are formally satisfied, judicial review at the Supreme Court remains procedurally minimalistic, non-deliberative, and insufficiently transparent, resulting in limited procedural justice and weakened normative legitimacy. To address this deficit, the study proposes procedural reforms consisting of limited open hearings, mandatory written and selective oral hearings, and minimum standards of legal reasoning, which can be implemented through amendments to Supreme Court Regulations and internal institutional policies. This study contributes to constitutional law scholarship by shifting the focus of judicial review discourse from authority-based debates to procedural legitimacy and demonstrates that strengthening due process and procedural openness is essential to enhancing the accountability and rationality of judicial review without expanding judicial power. Future research is encouraged to integrate empirical approaches to assess the effectiveness of procedural reforms and their impact on public trust in the judiciary.

The study is the first empirical evaluation of euthanasia requests in the terminally ill patients in Nigeria. A sample of 312 patients found that 18.9 per cent. had contemplated or actually requested euthanasia. The main predictors of use were severe uncontrolled pain (adjusted odds ratio=3.89, p=0.001) extreme financial burden (adjusted odds ratio= 2.76, p=0.003), poor access to palliative care (adjusted odds ratio= 3.21, p=0.001), and depression (adjusted odds ratio= 2.98, p=0.002). Among 178 interviewed physicians, 72.6 per cent of them said that euthanasia must never be permitted in any situation, and gave the reasons of moral and religious and legal reasons. Out of the 245 family caregivers interviewed, 34.7 per cent had a conditional supportive attitude to euthanasia with unremediated suffering. These results reveal some major gaps in the end of life care system of Nigeria among which are the absence of palliative facilities, insufficiency in pain management, and inadequacy in the provision of psychosocial care. The findings thus highlight the urgent need to increase the services of palliative care, streamline the process of pain management, encompass mental-health care, and carry out systematic policy debates to minimise preventable pain and encourage ethical practice.

Modeling legal reasoning with artificial intelligence and machine learning presents formidable challenges. Legal decisions emerge from a complex interplay of factual circumstances, statutory interpretation, case precedent, jurisdictional variation, and human judgment—including the behavioral characteristics of judges and juries. This paper takes an exploratory approach to investigating how contemporary ML techniques might capture aspects of this complexity. Using pharmaceutical patent litigation as an illustrative domain, we develop a multi-layer analytical pipeline integrating text mining, clustering, topic modeling, and classification to analyze 698 U.S. federal district court decisions spanning January 2016 through December 2018, comprising substantive validity and infringement rulings under the Hatch-Waxman regulatory framework. Results demonstrate that the pipeline achieves 85–89% prediction accuracy—substantially exceeding the 42% baseline majority-class rate and comparing favorably with prior legal prediction studies—while producing interpretable intermediate outputs: clusters that correspond to recognized doctrinal categories (Abbreviated New Drug Application—ANDA litigation, obviousness, written description, claim construction) and topics that capture recurring legal themes. We discuss what these findings reveal about both the possibilities and limitations of computational approaches to legal reasoning, acknowledging the significant gap between statistical prediction and genuine legal understanding.

Abstract to be supplied.

This article examines the dialectical integration of fiqh (Islamic jurisprudence) and Sufism (taṣawwuf) within Wahidiyah teachings in Indonesia and critically explores its implications for contemporary Islamic legal thought. Employing a qualitative normative-legal approach combined with philosophical analysis, this study analyzes Wahidiyah doctrinal texts alongside classical fiqh literature and authoritative Sufi works through deductive legal reasoning, maqāṣid al-sharīʿah analysis, and ethical–spiritual interpretation. The findings demonstrate that Wahidiyah articulates a systematic and operational integrative framework in which fiqh provides normative structure and legal certainty, while taṣawwuf functions as an ethical–spiritual orientation that animates legal reasoning and practice. This integration is concretely structured through a hierarchical sequence of principles—lillāh–billāh, li al-rasūl–bi al-rasūl, li al-ghawth–bi al-ghawth, yuʿṭī kulla dhī ḥaqqin ḥaqqah, and taqdīm al-aham fa al-aham thumma al-anfaʿ fa al-anfaʿ—aligning intention, authority, obligation, and priority within Sharīʿah boundaries. Departing from previous studies that predominantly frame Wahidiyah as a devotional or cultural Sufi movement, this article conceptualizes Wahidiyah as an integrative model of Islamic legal philosophy. It contributes theoretically by demonstrating how Sufi ethical consciousness can be methodologically incorporated into jurisprudential reasoning without undermining legal certainty. Methodologically, the study offers a maqāṣid-oriented analytical framework for examining spiritually grounded legal practices. Practically, it provides an alternative paradigm to rigid legalism and unstructured mysticism by showing how spirituality, legal normativity, and ethical responsibility can be systematically harmonized in contemporary Muslim life. By positioning Wahidiyah as a transferable model of fiqh–tasawwuf dialectics, this research advances contemporary Islamic legal theory and offers a contextually grounded approach to developing spiritually informed yet normatively coherent jurisprudence in plural Muslim societies

This study discusses the judge’s considerations in granting child custody of a non-mumayyiz (underage) child to the biological father in the Religious Court Decision of Tanjung Karang Number: 1379/Pdt.G/2024/PA.Tnk. Generally, according to Islamic law and the Compilation of Islamic Law (KHI), custody of a non-mumayyiz child is granted to the biological mother. However, in this case, the panel of judges decided to grant custody to the biological father. The purpose of this research is to identify the legal reasoning behind the judge’s decision and the legal implications of such a ruling. The study employs a normative juridical method with a descriptive-analytical approach, supported by primary data obtained through interviews and secondary data from legislation and legal literature. The results show that the judge’s consideration was based on the best interest of the child principle, as the mother had remarried a non-Muslim man and converted to another religion, which made her ineligible to be the custodian under Islamic law. The legal implication of this decision reinforces the authority of judges to interpret hadhanah provisions contextually, ensuring the protection of the child’s rights and supporting their growth in accordance with religious teachings and prevailing social norms.

This article examines how judges in Italy deal with sociocultural diversity in family law cases. The aim is to investigate how understandings about culture are shaped and constructed in legal reasoning and what institutional dynamics influence this process. Based on qualitative research – semi-structured interviews with judges and critical discourse analysis of judicial decisions – the study explores the tension between individually held understandings of culture and the institutional conditions under which legal decisions are produced. Some judges show awareness of the dynamic nature of culture, but such perspectives rarely translate into the rulings. On the contrary, culture is often treated as a fixed attribute, made legible through essentialist classifications shaped by procedural, bureaucratic constraints and reliance on external assessments. Combining Practice-Based Theory and Critical Discourse Analysis, the article shows how judicial reasoning is constructed in routines that reinforce dominant cultural assumptions, suggesting the need for structural change to support more context-sensitive, pluralistic forms of legal interpretation.