关键词: 道德化, 去道德化, 放大模型, 推拉模型, 中国文化

Abstract:

Moralization refers to the process through which individuals ascribe moral meanings to previously neutral behaviors, beliefs, or objects. This paper systematically reviews the psychological mechanisms, theoretical models, and consequences of moralization and proposes an integrative “Cognitive-Affective-Social (CAS)” model that incorporates macro-level social factors overlooked by existing frameworks. Drawing on multidisciplinary evidence, the study advances a comprehensive understanding of how moral beliefs emerge, intensify, and influence individual and collective behavior.

Previous research has primarily conceptualized moralization through two models: the moral amplification model (Rhee et al., 2019) and the push-and-pull model (Feinberg et al., 2019). The former delineates moralization as a two-stage process, moral recognition and moral amplification, manifested in moral judgments, attitudes, and the expansion of moral concern. The latter emphasizes the dynamic interplay between moralization (push) and de-moralization (pull) across emotional and cognitive domains. Despite their contributions, both models treat moralization as an individual-level phenomenon, neglecting the broader sociocultural contexts in which it unfolds. To address this gap, this paper introduces the “cognitive-affective-social (CAS)” integrative model of moralization, positing that moralization arises from the interactive effects of cognitive (such as perceived harm, cognitive reflection, moral identity, ideology), affective (such as disgust, anger), and social (such as inequality, cultural tightness, social crises) antecedents.

At the cognitive level, individuals who are more sensitive to perceived harm or who view morality as central to their self-concept are more likely to moralize neutral issues. Those who endorse binding moral foundations such as loyalty, authority, and purity also tend to see neutral behaviors as moral issues. This tendency is further strengthened among individuals with conservative ideology. Affective processes, especially emotions of disgust and anger, trigger rapid moral intuitions and amplify moral condemnation. At the macro-social level, social threats such as inequality, cultural tightness, and crises (for example, pandemics or terrorism) enhance the tendency to moralize everyday behaviors. The proposed CAS model integrates these multilevel factors into a dynamic framework. The CAS model further illustrates the dynamic interactions between these three domains: social factors can trigger specific cognitive appraisals (e.g., of harm) and emotional responses (e.g., moral outrage), which in turn cause moralization. Furthermore, cognitive and emotional factors can drive moralization on their own, or function interactively as both cause and effect, ultimately working in tandem to influence the process of moralization. This integrative approach bridges micro and macro perspectives, offering a more comprehensive understanding of moralization processes.

The paper further elaborates on the double-edged consequences of moralization. On the one hand, moralization enhances social cohesion, motivates prosocial behavior, and legitimizes collective moral norms. It sustains moral identity, strengthens goal commitment, such as healthy living and hard work, and provides psychological standing for actions. On the other hand, however, moralization can foster cognitive biases, stigmatization, polarization, and intergroup hostility. Highly moralized beliefs promote dehumanization of outgroups, the spread of misinformation, and moral dogmatism, often amplified by social media algorithms that reward moral and emotional content. Thus, moralization simultaneously stabilizes social order and fuels social conflict.

In response to these ambivalent outcomes, the paper outlines several future research directions. First, it calls for a systematic exploration of de-moralization, the process through which individuals or societies detach moral meaning from issues once moralized. Understanding its mechanisms and potential interventions, such as intellectual humility, shared moral values, and meta-cognitive training, is essential for mitigating moral conviction. Second, it highlights the role of social media technologies in accelerating moral amplification and moral contagion, creating moral echo chambers that reinforce ideological divisions. The paper argues that future research should develop mechanistic models explaining how digital affordances such as anonymity and virality transform moral cognition and expression. Third, the paper extends the moralization framework to the age of artificial intelligence. As AI systems increasingly assume decision-making roles, humans begin to moralize AI agents themselves, treating them as moral agents, patients, or proxies depending on perceived mind attributes. This raises novel questions about the moral responsibility of AI and AI users. Empirical findings reveal that resistance to AI often stems from moral rather than instrumental objections, and that AI users are perceived as having less morality. These emerging phenomena demonstrate that moralization is not confined to human relations but extends to human-machine interaction, signaling a new frontier for moralization studies. Finally, the paper advocates developing indigenous perspectives on moralization within Chinese cultural contexts. Traditional Chinese thought, characterized by pan-moralization, infuses moral meanings into natural, material, and ritual domains. For instance, the moral symbolism embedded in jade, archery, and landscape appreciation exemplifies this tendency to moralize material and social practices as reflections of morality. Investigating these indigenous moralization phenomena will provide invaluable cultural diversity and novel insights into the global understanding of moral psychology.

Key words: moralization, demoralization, amplification model, push-and-pull model, Chinese culture