James, Lange, Damasio, Feldman Barrett, and others regard the labeling of sensations as the core of emotions. However, this chapter shows that, as Wierzbicka points out, different languages partition the domain of ‘emotions’ differently. Moreover, a given language may have no name for a given emotion, and languages typically give different names to a given emotion in different domains or contexts. Within any one language, people use any label inconsistently and ambiguously. Moreover, most emotion labels refer to more than one emotion. For all of these reasons, it is essential to avoid the lexical fallacy of reifying vernacular lexemes as scientific constructs. Among other implications, this poses a methodological challenge: we cannot identify emotions by their labels alone. We must use multiple criteria for discerning the emotion a person experiences on a given occasion.