Moral judgment is an integral part of judicial adjudication. If courts disregard the moral concerns of the public, their judgments can easily spark moral controversy. This article examines the moral considerations underlying judicial adjudication, analyzing the moral pressures on justice, the ways justice responds, and the boundaries between moral considerations and legal rules. In China, moral disputes are particularly pronounced in cases involving justifiable self-defense. This article conducts a quasinatural experiment using such cases as samples. The findings are as follows: First, there is a phenomenon known as the “result-only theory” in the trials of justifiable self-defense in China, which significantly contributes to the controversy in these cases. Second, in response to the moral pressures surrounding self-defense, the judiciary has begun to address moral claims, often employing a consequentialist approach. Third, while legal rules fundamentally determine the nature of legal reasoning in justifiable self-defense trials, moral considerations mainly guide the severity of the sentencing outcomes. Additionally, we must be aware of the potential risks associated with judicial responses to moral demands and seek ways to mediate these risks throughout the process.

Probation is a crucial institution in modern criminal law, embodying the principles of fairness and justice while contributing to the harmonious development of society. Despite its importance, the current Intelligent Judicial Assistant System (IJAS) lacks dedicated methods for probation prediction, and research on the underlying factors influencing probation eligibility remains limited. In addition, probation eligibility requires a comprehensive analysis of both criminal circumstances and remorse. Much of the existing research in IJAS relies primarily on data-driven methodologies, which often overlooks the legal logic underpinning judicial decision-making. To address this gap, we propose a novel approach that integrates legal logic into deep learning models for probation prediction, implemented in three distinct stages. First, we construct a specialized probation dataset that includes fact descriptions and probation legal elements (PLEs). Second, we design a distinct probation prediction model named the Multi-Task Dual-Theory Probation Prediction Model (MT-DT), which is grounded in the legal logic of probation and the Dual-Track Theory of Punishment. Finally, our experiments on the probation dataset demonstrate that the MT-DT model outperforms baseline models, and an analysis of the underlying legal logic further validates the effectiveness of the proposed approach.