Emotions Embodied

In one of the most frequently quoted passages in the history of emotion research, William James (1884: 189f) announces that emotions occur when the perception of an exciting fact causes a collection of bodily changes, and “our feeling of the same changes as they occur IS the emotion.” The same idea occurred to Carl Lange (1984) around the same time. These authors were not the first to draw a link between the emotions and the body. Indeed, this had been a central theme of Descartes’ exquisite opus, The Passions of the Soul. But James and Lange wanted to push things farther than most, suggesting that emotions are exhausted by bodily changes or perceptions thereof. Other kinds of mental episodes might co-occur when we have an emotion state. For James, an emotion follows an exciting perception. But the exciting perception is not a part of the emotion it excited (Ellsworth, 1994, reads James differently, but see Reisenzein et al.’s 1995 convincing response). The majority of contemporary emotion researchers, especially those in philosophy, find this suggestion completely untenable. Surely, emotions involve something more. At their core, emotions are more like judgments or thoughts, than perceptions. They evaluate, assess, or appraise. Emotions are amendable to rational assessment; they report, correctly or incorrectly, on how we are faring in the world. Within this general consensus, there is a further debate about whether the body should figure into a theory of emotions at all. Perhaps James and Lange offer a theory that is not merely incomplete, but entirely off base. Where they view judgments as contingent and non-constitutive concomitants of emotions, it is actually bodily perceptions that deserve this demotion. Perhaps emotions can be, and often are, disembodied in some fundamental sense. I propose to defend James and Lange, though not completely. They should be criticized for their failure to reckon with what can broadly be regarded as the rationality of emotions. That failure, however, has a remedy that does not depart from the spirit of the James-Lange approach. Emotions are somatic, but they are also fundamentally semantic: meaningful commodities in our mental economies. I will not be especially concerned with presenting the somatic theory exactly as it appears in James and Lange. I will not assume that emotions are always consciously felt, as James sometimes implies, nor that the relevant bodily changes must have the origins in the vasomotor system, as suggested by Lange. The core idea that I will defend is that emotions are perceptions (conscious or unconscious) of patterned changes in the body (construed inclusively). I begin by briefly presenting some of the positive evidence for this core idea. Then I discuss six stubborn objections. I argue that the objects can be answered without abandoning the core idea but forward by James and Lange, but they do demand an important amendment.