This is a preprint version of a theoretical article intended for journal submission. This article examines a semantic problem in rational-action explanation: the compression of materially, evaluatively, normatively and reputationally different consequences into a single utility vocabulary. It argues that many disagreements between rational choice theory, bounded rationality, prospect theory, social exchange theory and norm-based accounts arise not simply because these theories make incompatible claims, but because they specify different consequence dimensions with different degrees of explicitness. The article proposes a minimal benefit–harm semantic decomposition that distinguishes material benefit, evaluative benefit, material cost and evaluative cost. It interprets action through situated subjective rationality: choice under the actor’s represented consequences, probabilities, constraints, identities and normative expectations. The article is not offered as a replacement for rational-action theory or as a general theory of social life. Its aim is more limited: to clarify what utility contains, which consequence dimension is doing the explanatory work, and under what conditions material incentives, shame, recognition, commitment, moral self-regard and institutional sanctions should generate different behavioural predictions. The framework is illustrated through public goods contribution and morally costly action, and it specifies failure conditions to avoid post hoc rationalization. This manuscript has not yet undergone peer review. The preprint is made available for scholarly discussion, citation and timestamping prior to journal submission.
Yoshiaki Ikematsu (Wed,) studied this question.