Giving Darwin His Due

Darwin’s Contributions to Our Understanding of Emotion and Expression

The Expression of the Emotions in Man and Animals by Charles Darwin was published in 1872, a year after The Descent of Man. It is, without a doubt, a brilliant book, forecasting many of the fundamentals of not just facial expression but emotion itself. It is the first, pioneering study of emotion and, in my view, should be considered the book that began the science of psychology.

Darwin charted a course for the field of emotional expressions through five major contributions.  The first of his contributions was to  treat the emotions as separate discrete entities, or modules, such as anger, fear, disgust, etc. The German physician Wilhelm Wundt proposed an alternative view of emotion about a decade later. Wundt wrote about variations in dimensions or continua of pleasantness and activity or intensity. This very different conceptualization enjoyed popularity in 20th century psychology, with Schlosberg (1941) as the major mid-century proponent, followed by Russell at the end of the century.

Many different kinds of research — neuroscience, perception, and cross cultural evidence — show that Darwin’s conceptualization of emotions as separate discrete entities has proven to be very useful. Of course, each emotion varies on attributes such as intensity or acceptability, which can be considered as dimensions that describe differences within each discrete emotion.

The second major contribution was his focus primarily on the face, although he did give some attention to vocalizations, tears, and posture. To date, facial expression has been the richest source of information about the emotions. The voice has yet to be shown to be a source for as many discrete emotional states as the face, although it is harder to fabricate or regulate than facial expressions.

Darwin took for granted that it is the morphology of facial expression that conveys information about which emotion is occurring. The timing of an expression carries information as well, but not about which emotion is occurring. My own research has found that facial expressions reach an apex of the maximum muscular contraction that is going to occur, which is held typically for a few seconds with little noticeable variation during the apex. Any slice of time within that apex carries the information about which emotion is being signaled. For that reason, I call these snapshot expressions to distinguish them from aggregate signals, which incorporate a sequence of expressions.

Darwin’s third major insight was that facial expressions of emotion are universal. In the last few decades, the preponderance of evidence from Western and Eastern, literate and preliterate cultures strongly supports Darwin’s claim (based on sparse evidence, but in all likelihood demonstrated to him by his experience traveling around the world during his five year journey on the Beagle). Universality did not support his evolutionary theory — if we all descended from Adam and Eve, expressions would be universal. But it did support Darwin’s challenge to the racists of his time — who claimed Europeans had descended from a more advanced progenitor than Africans — by showing common descent, allowing Darwin to proclaim the unity of mankind.

Darwin proposed that facial expressions of emotion are universal, although he also proposed that gestures are culture-specific conventions. This has proven to be correct. The same hand movement, for example the first finger touching the thumb to form a circle in the North American “A-OK” gesture, has a radically different meaning in other countries.*

The fourth insight was that emotions are not unique to humans, but are found in many other species. His examples range from bees to roosters, dogs, cats, and horses, as well as other primates. For much of the last century, that view was considered an example of bad science, of anthropomorphism. Underlying that view was a reification of language and verbal self-report. If we can’t examine a species’ report of their experience how can we know if emotion is occurring? That stance would require that we regard infants as not having emotions prior to their acquiring speech!  Words are used to describe or reflect our emotional experience, but the words are representations of emotion not the sine qua non of emotion.

A fifth contribution was Darwin’s explanation of why particular movements signal a particular emotion. Why is the raised upper lip included in one of the anger expressions, for example? Darwin attributed this to it having been a “serviceable habit,” exposing the canine teeth, which threaten harm to come, as well as preparing for the attack. Stripped of its Lamarckian baggage, this explanation is consistent with contemporary ethological accounts of how signals evolved from intention movements, providing the foundation for current formulations of how signals become ritualized or formalized. Darwin also proposed a principle of antithesis, whereby a signal has a certain form because it is the opposite of another signal. For example, the dog (and many other animals) puffs itself up to appear larger in a potentially antagonistic encounter, which Darwin explained as based on the principle of serviceable habits. But the antithesis of that movement is the submissive slinking and lowering of the body.

Why, we might ask, was Darwin right about so many aspects of emotion and expression? One answer is that it was the product of his evolutionary perspective — a perspective that would suggest much of what he proposed when it is focused on emotion. Another related answer is that Darwin turned to the biology of emotion, noting what he could about the physiology of emotion and the anatomy of facial expression (where in his time much more was known). Darwin’s other important source was the French neurologist Duchenne De Boulogne. Amazingly Darwin’s publisher omitted three of Duchenne’s photographs that Darwin discussed at length in Expression. Those photographs never appeared in any subsequent edition of Expression, until the recent third edition that I edited (Darwin, 1998).

Darwin’s keen observational skills were applied to more different data sources than anyone before or since has included in an article or book about emotion: infants (his own), children (likewise), adults, animals in the zoo, the mentally ill, and reports he obtained from many people he wrote to or who wrote to him about what they had observed in other cultures.

Another methodological contribution was Darwin’s focus not just on changes in appearance but the musculature that generated those changes. In most of the 20th century scientists instead described expressions in terms that mixed inference about underlying state with description (e.g., smile, frown) and were imprecise to boot. Another mistaken path was to describe changes in the appearance of the features or wrinkles without considering what muscular actions produced those changes. Building on Duchenne, Wally Friesen and I published a comprehensive, anatomically based tool for describing/measuring any facial movement: the Facial Action Coding System (FACS; 1978).

Still another of Darwin’s methodological contributions was to show photographs of facial expressions to observers and note what emotions they attributed to each expression. This is still the most widely and easily used method for studying facial expression, referred to currently as a judgment study. It is a useful method, but there are many questions that it cannot answer, that must be addressed by measuring facial movement itself (see Chapter 2 in Ekman, 1982, for a comparison of the different methods for studying facial expression).

Issues Not Considered by Darwin

Darwin did not attempt to provide a method for measuring facial movement. And he did not consider how to define the boundaries of each emotion family. There is little doubt that there are many variations on the expression of any emotion. We do not yet know how many variations, nor do we know how many of those variations are linked to differences in social context or subjective experience. This is, in my judgment, the most serious gap remaining in our understanding of facial expressions, and it is a very large one. FACS provides the means for describing all the variations, but we have yet to map them completely for any emotion, nor do we have an empirical basis for knowing how many of the possible distinctions or variations merit consideration because they provide different information.

Another issue that Darwin did not consider but that needs to be addressed with vigor is the distinction we described (Ekman & Friesen, 1969) between an indicator and a communicative signal. When writing that paper, I didn’t know how to apply this distinction to facial expressions, but Duchenne’s observations about the differences between a voluntary smile and an involuntary smile of enjoyment provide an excellent illustration of the value of this distinction between indicators and signals. The action of zygomatic major, a muscle extending from the corner of the mouth to the cheekbone (AU 12 in FACS terms), provides a very strong smile signal, even when the action of that muscle is weak. But as Duchenne suggested, the absence of orbicularis oculi, a muscle controlling contraction around the eyes (AU 6 — Duchenne failed to exclude AU7, the inner part of that muscle which we found is not relevant to distinguishing enjoyment) “unmasks the false friend.” However, neither Duchenne or Darwin noted that the difference in appearance is very subtle and hard to recognize without precise measurement. The difference between a spontaneous enjoyment signal and a voluntarily or habitually produced facsimile is an indicator, not a signal. It is rarely recognized by members of the same species.

Darwin was not interested in how to distinguish deceptive from genuine facial expressions of emotion. The conceptualization of the role of emotion in perpetrating and betraying a lie was clearly not of much interest Darwin, being one of the very few topics he left to others to chart. (See Ekman, 2003, for an extended discussion of Darwin’s few contributions about deception.)

*For discussion of symbolic gestures see Ekman (1976, and Chapter 4 in Ekman, 1985/2009).


A longer version of this article, with more detail in particular about the issues Darwin did not consider will be in the upcoming December 2009 Transactions of the Royal Society.

References and Further Reading:

Darwin, C. (1998). The expression of the emotions in man and animals (3rd ed., P. Ekman, Ed.). New York: Oxford University Press. Ekman, P. (1976). Movements with precise meanings. Journal of Communication, 26, 14-26. Ekman, P. (1982). Emotion in the human face (2nd ed.). Cambridge, United Kingdom: Cambridge University Press. Ekman, P. (2003). Darwin, deception, and facial expression. Annals of the New York Academy of Sciences, 1000, 205-221. Ekman, P. (2009) Telling lies. New York: WW Norton. (Original work published 1985) Ekman, P., & Friesen, W.V. (1969). The repertoire of nonverbal behavior: Categories, origins, usage, and coding. Semiotica, 1, 49-98. Ekman, P., & Friesen, W.V. (1978). Facial action coding system: A technique for the measurement of facial movement. Palo Alto, CA: Consulting Psychologists Press. Schlosberg, H. (1941). A scale for the judgment of facial expression. Journal of Experimental Psychology, 29, 497-510.
Observer Vol.22, No.8 October, 2009

Leave a comment below and continue the conversation.


Leave a comment.

Comments go live after a short delay. Thank you for contributing.