7 - A Sceptical Look at Faces as Emotion Signals

Summary

As the authors of other chapters in this volume elaborate, it is common sense that we express our emotions in our face and that onlookers can read the emotions so expressed. These seemingly obvious premises stem from a folk theory that dates back at least to Aristotle. They were articulated into a simple and heuristic scientific research program known as Basic Emotion Theory (BET) and that has dominated the psychological study of emotion for decades. The theory was readily accepted both within and outside science. Researchers, clinicians and security guards accepted its advice on how to detect emotions by watching for tell-tale facial signs. Philosophers relied on its claims in analyzing emotion and expression (see, e.g., Green 2007 and this volume). Alas, as often happens in science, the evidence gathered indicates that BET must be revised or, as I suggest, abandoned altogether. I first review that evidence and then turn, briefly, to consider a way to think about facial changes that does not assume that emotions cause them.

The Prototype Version of Basic Emotion Theory

In the early decades of the twentieth century, experimental psychologists sought to bring into the laboratory the age-old idea that all human beings express their emotions by the same facial movements and that onlookers can accurately read those emotions from those expressions. Indeed, Langfeld (1918a: 183) called this idea ‘a well-known fact’. The experimentalists were therefore presumably surprised when the evidence was mixed. A person's emotion did not reliably result in a specific facial movement (Landis 1924), and onlookers were often inaccurate in reading emotions even from highly selected facial expressions (Buzby 1924; Feleky 1914; Guilford 1929; Langfeld 1918a, 1918b; Sherman 1927). Anthropologists observed cultural differences in how emotions were expressed (Bateson and Mead 1942).

In 1962, Tomkins published the first of four volumes that proposed what I shall call BET. BET was a return to the millennia-old common-sense folk theory of human facial expressions, although BET enthusiasts sometimes date this theory to a much more recent event, the publication of Charles Darwin's (1872) book on the expression of emotion. BET captures our taken-for-granted presuppositions about facial expressions.