Describing emotions in terms of valence is second nature, in everyday life and in science alike. Yet in the paper, van Kleef and Sauter argue that the concept does surprisingly little to explain how primary emotions actually work. Reviewing conceptual issues as well as empirical findings at both the intrapersonal and interpersonal levels of analysis, they conclude that the effects of emotional experiences and expressions are better understood in terms of specific appraisals, action tendencies, and social functions, and that the notion of valence may hamper rather than help scientific understanding and progress.

Rather than discarding valence altogether, the authors propose repurposing it at the level of meta-emotions: how people feel about how they feel. They suggest that the valence of meta-emotions is constructed through assessments of the momentary goal congruence and sociomoral evaluation of primary emotions, and that people subsequently represent emotional episodes in terms of how they felt at the time relative to how they wanted to feel. At this level, valence does real work: it predicts emotion regulation strategies, enables economical representation of experiences in memory, and characterises how people communicate about their emotional experiences. This may help explain why valence is so pervasive in human thinking about emotions, despite the conceptual and empirical problems that beset the construct.