Language limits the experience of emotions: Comment on “The quartet theory of human emotions: An integrative and neurofunctional model” by S. Koelsch et al.

Abstract

Emotions are phylogenetically ancient and involve complex interactions of neural, behavioral, and physiological processes. A complete theory of emotions must incorporate, or at least be informed by, current knowledge from neurobiology and comparative psychology [1]. The Quartet Theory of Human Emotions by Koelsch and colleagues [2] is therefore a welcome step towards a more integrative affective science. A notable feature of the Quartet Theory is the clear proposed relationship between language and emotion systems, which diverges from current trends in affective science. Many theories of emotion that specify a role for language in affective processes suggest that language and concepts either create, differentiate, facilitate awareness of, and/or express emotional states [e.g., 3–5]. These theories are based in part on the assumption that emotion words can accurately and precisely encapsulate internal feeling states. Some even suggest that verbally categorizing diffuse affective states coheres them into what we then experience as discrete and differentiated emotions [4,5]. In contrast, the Quartet Theory suggests that linguistic categorization of emotion percepts—the summed “felt” components of affect—is neither an automatic nor essential step in an emotion experience. Language facilitates regulation (e.g., through reappraisal), and sometimes, expression of emotions. However, the authors argue that translating a pre-verbal emotion percept into language results in the loss of important diagnostic information [6], which is problematic (as the philosopher Ludwig Wittgenstein argued) because speakers cannot be confident that their understanding of an emotion word is the same as another person’s.1 Thus it is unlikely that language and concepts play a fundamental role in the emotion itself because categorizing an emotion percept reconfigures it. We agree that emotion percepts are pre-verbal internal states and emotion labels therefore do a poor job of communicating them. We also agree that language is a powerful tool for conscious emotion regulation. We would like to push the argument even further and suggest that the simple act of labeling an ongoing emotion percept automatically