Emotion

States combining feeling, judgement and motivation - and the question of which of the three is doing the defining.

Theme: Mind, self and consciousness
Tradition: Cross-tradition
Term: πάθος · pathos
Term fixed: c. 340 BC

In plain terms

Fear has three parts that usually arrive together: a feeling, a thought about danger, and a push toward getting away. Any theory of emotion has to say which of those is the emotion and which are its trappings.

Say the feeling, and you have to explain how emotions can be reasonable or unreasonable - a fear can be mistaken, and mere sensations cannot be. Say the thought, and you have to explain why knowing a spider is harmless does not stop the fear. Say the motivation, and you leave out everything it feels like.

And there is a prior question that the three-part picture assumes an answer to: whether emotions are natural kinds at all, the same in everyone, or categories a culture assembles and teaches.

As a philosopher would put it

Cognitive theories make emotions constituted by evaluative judgements - the Stoic version is that anger just is an assent to the proposition that one has been wronged, which is why they held emotions could be eliminated by correcting beliefs. Nussbaum's contemporary version keeps the judgement and drops the elimination.

Feeling theories run the other way. James and Lange argued that the bodily response precedes the emotion and constitutes it: we do not tremble because we are afraid, we are afraid because we tremble. The standing objection is recalcitrance in reverse - identical bodily states accompany different emotions, so the body cannot be doing the individuating.

The empirical dispute is now about kinds. Basic-emotion theory, following Ekman, holds that a small set is universal, discretely realised and expressed in recognisable facial configurations. Constructionist accounts, principally Barrett's, hold that emotions are conceptual categories the brain applies to undifferentiated affect, learned like other concepts, and that the universality findings are artefacts of method.

Where you meet it

Paul Ekman's mid-century studies asked people in many societies, including one in Papua New Guinea with minimal outside contact, to match photographs of facial expressions to short stories. Agreement was high, and the result became the standard evidence that a handful of emotions are universally recognised.

In 2014 Maria Gendron, Lisa Feldman Barrett and colleagues ran a different version with the Himba of northern Namibia. Instead of supplying the emotion words as options, they gave participants the photographs and asked them to sort them into piles however they liked.

The Himba sorted by action - smiling, gazing, and so on - rather than into the Western emotion categories, and free-sorting by American participants reproduced the expected groupings. The finding is that the original method supplied the answer along with the question. The dispute is unresolved and both camps are active, which is a useful thing for a reference page to say plainly: one of the best-known results in psychology turns out to depend on how the question was asked.

What is actually disputed

Whether emotions are natural kinds.

If they are, then anger in one culture is the same state as anger in another, differing in expression and occasion but not in nature, and a science of emotion can proceed as a science of six or seven things. If they are constructed, then emotion words carve up affect differently in different languages, the categories are learned, and cross-cultural comparison is comparing concepts rather than states.

The evidence is genuinely contested and the stakes are practical. Automated emotion recognition, courtroom demeanour evidence and clinical assessment all assume the first answer. There is also a much older dispute running underneath: whether emotions are impediments to good judgement, as the Stoics held, or constitutive of it, as Aristotle held and as most contemporary work assumes - a person who felt nothing about anything would not be an ideal reasoner but an impaired one.

Defined against

Reason The contest between them is the oldest structural claim about the mind, and almost every tradition rules on it

Philosophers

Aristotle Emotions as involving judgement, and as educable
Epictetus Emotions as assents that can be withheld
Baruch Spinoza The affects as increases and decreases in a thing's power of acting
David Hume The passions as the source of all motivation
William James The bodily change first, and the emotion as the feeling of it

Works

Nicomachean Ethics Emotion as part of good character rather than an obstacle to it
Discourses and Enchiridion The Stoic discipline of assent
Ethics The systematic account of the affects
A Treatise of Human Nature The passions as primary
Pragmatism The psychological reversal of the usual order

Schools

Peripatetic school (Aristotelianism) Emotions as educable and part of virtue
Early Stoicism Emotions as mistaken judgements
Roman Stoicism The practical discipline built on that
Spinozism The affects as changes in power
Classical Pragmatism The bodily-first account

Related concepts

Desire c. 380 BC · Suffering c. 450 BC · Virtue c. 340 BC · Compassion c. 450 BC · Imagination c. 330 BC

Aristotle's account in the Rhetoric and the Ethics treats emotions as involving beliefs and as educable; the Stoics make them judgements outright, which is where the cognitive theory begins.