Do we cry because we're sad, or are we sad because we cry?

Do we cry because we are sad, or are we sad because we cry? Common sense says the first. In 1884 the philosopher and psychologist William James argued for something close to the second. His claim was that the bodily changes come first, and that our feeling of those changes as they occur is the emotion.

James liked a bear for his example. Most people would say you see a bear, become afraid and run. James thought you see the bear, your body reacts, and you tremble, and only then do you become afraid and run. As his view is often summed up, we feel sad because we cry, angry because we strike and afraid because we tremble. The physician Carl Lange had developed similar ideas independently in 1885, which is why the idea carries both names.

James backed this up with a thought experiment. Take a strong fear and subtract every feeling in the body: the quickened heartbeat, the shallow breathing, the trembling lips and weak limbs, the goose-flesh, the stirrings in the gut. What fear would be left? James found it quite impossible to think.

The idea did not go unchallenged. In the 1920s Walter Cannon and Philip Bard offered a rival in which bodily changes arise independently of emotion. Later studies found cases where physiological patterns could not tell emotions apart, and a 2009 study of brain-damaged patients found that bodily responses and emotional experience could come apart.

Yet the idea has survived in a softer form. Modern theorists propose that emotion is shaped by physiological feedback and by other information, and one psychologist says most modern affective neuroscientists would support that view.