The (Provisional) Truth About Emotions

The March 2 New York Times guest editorial is here.

[By virtue of a perhaps over-sensitivity to much abused copyright law, we have chosen to provide a link to the original guest editorial rather than reproduce it on this page.]

This is Drs. Ekman and Keltner's response shortened by its authors to less than 200 words per The New York Times' requirements:

Feldman Barrett (New York Times, March 2, 2014) failed to undermine the science showing universality in the interpretation of facial expressions. She doesn’t understand what Darwin said and left out most of the evidence. Darwin never claimed that all facial expressions are universal, only a specific set of expressions that he had observed and studied. In the late sixties, Izard and Ekman each found high agreement about the emotion shown when many of the expressions proposed by Darwin were shown to people in literate cultures. Ekman obtained similar findings when he studied members of an isolated Stone Age culture in New Guinea.

 

Feldman Barrett not only doesn’t understand the difference between unselected and theoretically selected facial expressions, she totally ignored published studies supporting Darwin in which entirely different methods were used: (1) spontaneous facial expressions were measured in different cultures; and (2) physiology differed when the various universal facial expressions occurred.

Darwin emphasized the importance of some universal facial expressions in establishing the unity of mankind, challenging the racist assertions of his time that Europeans had descended from a more advanced progenitor that Africans. Those findings and his conclusions remain unchallenged. Further, government agencies using those findings are not mistaken to do so.

 

Paul Ekman, Emeritus Professor, University of California, San Francisco
Dacher Keltner, Professor, University of California, Berkeley

 

This is Drs. Ekman and Keltner's response, edited and published by The New York Times on March 12:

To the Editor:

 

“What Faces Can’t Tell Us,” by Lisa Feldman Barrett (Sunday Review, March 2), seeks to undermine the science showing universality in the interpretation of facial expressions. In her eyes, recent evidence challenges “the theory, attributed to Charles Darwin, that facial movements might be evolved behaviors for expressing emotion.”

 

In the late 1960s, Carroll Izard and one of us, Paul Ekman, showed people in literate cultures many of the expressions proposed by Darwin and found high agreement on the emotions recognized. I (Paul Ekman) obtained similar findings when I studied Stone Age cultures in New Guinea.

 

Some published studies supporting Darwin used entirely different methods, such as measuring spontaneous facial expressions in different cultures and physiological activity that differed when various universal facial expressions occurred.

 

Darwin emphasized the importance of some universal facial expressions in establishing the unity of mankind, challenging the racist assertions of his time that Europeans had descended from a more advanced progenitor than Africans. Those findings and his conclusions remain unchallenged.

 

PAUL EKMAN
DACHER KELTNER
San Francisco, March 5, 2014

Mr. Ekman is emeritus professor of psychology at the University of California, San Francisco. Mr. Keltner is a professor of psychology at the University of California, Berkeley.

 

A version of this letter appears in print on March 12, 2014, on page A26 of the New York edition with the headline: Reading Facial Expressions

This is Drs. Ekman and Keltner's full response:

Darwin’s Claim of Universals in Facial Expression Not Challenged

 

Lisa Feldman Barrett’s recent contribution (New York Times, March 2, 2014) seeks to undermine the science showing universality in the interpretation of facial expressions. In her eyes, recent evidence “challenges[ing] the theory, attributed to Charles Darwin, that facial movements might be evolved behaviors for expressing emotion.” Such a disagreement really belongs in exchanges of findings and theory in a scientific journal, evaluated by colleagues as evidence accumulates, not the public press. This is not the first time that Feldman-Barrett publicized her views in the press. We didn’t respond then, but feel compelled to do so now so that the public is not misled, and is apprised of the broader, Darwin-inspired science of emotional expression many scientists are working on today.

 

First, let’s get the science right. Darwin never claimed in his great book The Expression of Emotions in Man and Animals (1872) that all facial expressions are universal, only a specific set of expressions that he had observed and studied. Nearly one-hundred years later Silvan Tomkins helped Ekman and Carrol Izard refine and add to Darwin’s list. In the late sixties, Izard and Ekman in separate studies each showed photographs from Tomkins’ own collection, to people in various literate cultures, Western and Non-Western. They found strong cross cultural agreement in the labeling of those expressions. Ekman closed the loophole that observing mass media might account for cross cultural agreement by studying people in a Stone Age culture in New Guinea who had seen few if any outsiders and no media portrayals of emotion. These preliterate people also recognized the same emotions when shown the Darwin-Tomkins set. The capacity for humans in radically different cultures to label facial expressions with terms from a list of emotion terms has replicated nearly 200 hundred times.

 

Feldman Barrett is right to ask whether individuals in radically different cultures provide similar interpretations of facial expressions if allowed to describe the expressions on their own terms, rather than a list of emotion terms. Haidt and Keltner did such a study comparing the free responses to the Darwin-Tomkins set of expressions and some other expressions, with people in rural India and the U.S. Once again the findings of universality were clear cut, and evidence of universality in the expression of embarrassment was also found. The evidence on the judgment of the Darwin-Tomkins facial expressions is robust; so we suppose is Feldman-Barrett’s evidence for the expressions not covered in the Darwin-Tomkins set. She has missed that point, not understanding the difference between unselected and theoretically selected facial expressions.

 

Feldman Barrett also ignores two other very powerful data sets that don’t involve showing portrayals of facial expressions to people. Instead what people actually do, spontaneous facial expressions, is measured in numerous, different emotional contexts. Ekman and Friesen published what might be the first such study comparing the spontaneous facial expressions shown by Japanese and American subjects in a private and public setting, finding universal facial expressions – the Darwin-Tomkins set—in private, and different expressions in public. Since then over a hundred studies have been published measuring spontaneous facial expressions, enough to justify two volumes reprinting the articles of dozens of scientists by Oxford University Press.

 

Another large body of research has established different patterns of physiology – in bodily changes generated by Autonomic Nervous System (ANS) activity and in brain activity – coinciding with the appearance of the Darwin-Tomkins set of facial expressions. Separate, well-replicated studies, have also shown that voluntarily generating the Darwin-Tomkins set of facial expressions produced distinct changes in ANS and brain activity! Still other studies have related the Darwin-Tomkins set of expressions to distinct responses, including cortisol, oxytocin, dopamine, and the cytokine response that is part of the immune system. This work, ignored in Feldman-Barrett’s critiques, suggests that facial expressions not only are informative about individuals’ feelings, but patterns of neurophysiological activation in their bodies. Darwin emphasized the importance of some universal facial expressions in establishing the unity of mankind, challenging the racist assertions of his time that Europeans had descended from a more advanced progenitor that Africans. Those findings and the conclusion that all human beings have a shared set of facial expressions remains unchallenged.

 

Paul Ekman, Emeritus Professor, University of California, San Francisco
Dacher Keltner, Professor, University of California, Berkeley

This is our contribution on the subject to the Huffington Post:

The (Provisional) Truth About Emotions

 

I am not a fan of The New York Times. I rarely use my quota of free-to-the-internet articles. My view is that, institutionally, they are not nearly so clever as they appear to believe. They are slow to acknowledge or correct errors. Although I also recognize that, over the years, their reporters have produced some important and, occasionally, pivotal investigative pieces, the Times is a relatively minor source in my efforts to keep up with what is called news.

I continuously have to remind myself that the world the media describes is only tangentially connected to the one most of us inhabit. They will tell you that it is the important part. Perhaps. Or maybe it’s an echo chamber.

On March 2, the Times published a guest editorial attacking Paul Ekman and his (and many others’) decades of work on the facial expression of emotions.

First, you should know that Paul Ekman is a friend. One characteristic of a true friend is a person with whom one can disagree and from whom one thereby learns and grows.

That’s also the way that science is presumed to work. The history of science provides ample evidence that scientific knowledge is provisional. It is evolving in ways which have consistently foiled our attempts to predict.

As we collect more data and make sense of that data, we learn more. As we develop new techniques and technologies for collecting data, we learn more. As we develop new models and theories for organizing the data, we learn more.

Years ago, Melanie Mitchell suggested to us that pattern recognition may well be the key to understanding intelligence, human and otherwise. In a conference my husband attended about the same time, Nobel laureate Murray Gell-Mann presented a simplified taxonomy of recognized patterns:

• Patterns which we see which exist
• Patterns which we don’t see which exist
• Patterns which we see but don’t exist
• Patterns which we don’t see which don’t exist

We presume that the proper focus of science (and media and politics) is on the first category, with the second category pushing us to do better thinking and more research.

Different people with distinctive experiences and divergent objectives will map the same data to different patterns. When the subject of our pattern-making is the human animal (or, more generally, life), certainty about the correct, immutable pattern is perilous, at best, foolish in the middle ground or, at its worst, destructive.

The March 2 editorial states that:

“Hundreds of scientific studies support the idea that the face is a kind of emotional beacon, clearly and universally signaling the full array of human sentiments, from fear and anger to joy and surprise.”

 

The opinion piece goes on to say that companies and government agencies are using these studies to evaluate individuals. The piece then issues a blanket assertion: “(T)his assumption is wrong.”

Without being clear precisely what the assumption is, the author cites as evidence the work of the lab she directs. Such evidence hardly exhausts the subject or warrants the blanket assertion of error on Dr. Ekman’s part.

As is the case with, say, Jesus, attacking the prophet for failures of the disciples is misguided. I doubt that Dr. Ekman would defend every application of his work, particularly those motivated by economics rather than a commitment to truthfulness.

In a version of his response (which the Times refused to publish deeming it too long), Dr. Ekman says:

 

"Such a disagreement really belongs in exchanges of findings and theory in a scientific journal, evaluated by colleagues as evidence accumulates, not the public press."

Indeed.

These kinds of discussions are sufficiently difficult that playing to the crowd or stoking one’s career almost certainly will derail the search for the (currently) correct pattern. Additionally, readers of the Times do not, in this context, constitute a credible jury. Much as in might appeal to fundamentalists-of-any-tribe, a vote on the efficacy of, say, gravity is wasted.

In addition to the pattern problem, there is also the inevitable language problem: are we talking about the same thing? is my use of a word identical to your use of that word? is your jargon compatible with my jargon? do we even speak the same language?

Jim Carse has suggested (The Religious Case Against Belief, 2009) that only true-believers know the truth. Scientist and the genuinely religious seek the truth. The more we know, the more we don’t know; as the frontiers of knowledge expand, we confront ever more that we don’t know (Plato was wrong). Walter Truett Anderson took on the challenge in The Truth About the Truth (1995). He argues that we are much more complex and diverse (individually and collectively) than can be subsumed in simple definitions or categorizations of who and what we are.

So, has the Times done well by its readers? Was this “news” “fit to print”? You can read the entire, published exchange here. On this question, we’re all qualified to judge.

In my view, Dr. Ekman’s science is good as of today. I am unfamiliar with the editorialist’s science. I hope that she is appropriately diligent. If there are disagreements, there are scientific protocols available to resolve those differences. The New York Times is not among those protocols, nor are my opinions.

We would be well served by more humility from the media as well as from scientists. Politicians, too.

Someday science may provide a vaccine for hubris or a palliative which will reduce its more noxious attributes. Meanwhile, a healthy, broadly informed skepticism augmented by multiple, competing information sources works as an antidote. Right now my emotional state is fair to partly cloudy, with change on the horizon.

° ° °

A few additional thoughts (in no meaningful order):

 

•  The original guest editorial was ± 770 words. The Ekman/Keltner response was ± 170 words. The guest editorial was published in the SundayReview; the response on a Wednesday. These asymmetries are intrinsically unfair, whatever the editorial policies of the publication. Comparing the submitted version of the response to the published version further erodes a sense of fairness. Indeed, I doubt that Ekman/Keltner would even recognize their letter until they reached the last paragraph. Even that was detuned.

 

•  The Times' Op-Ed and Sunday Review editor states how the page works:

 

We're interested in everything, if it's opinionated and we believe our readers will find it worth reading.

 

These criteria suit both fiction and non-fiction. Generally, opinions do not correlate well with truthfulness. Clearly, the Times has opted for entertainment over information in this section.

 

•  The New York Times benefits from what we've come to call the as-compared-to fallacy. It's better than many of its truly awful rivals. With care, caution and liberal application of one's critical faculties, the internet now provides many good sources of information. However, good today is not necessarily good tomorrow.

 

•  Psychology describes a phenomenon named the Einstelung effect which refers to the human tendency to solve a problem using familiar tools rather than more appropriate ones. Years ago, the late historian John Keegan introduced us to the notion of "professional deformation" — e.g., professors of the history of war (which he was) favor war-historical descriptions for people and events regardless of appropriateness to the particular circumstances. For the rest of us, this is the hammer theory: if your tool is a hammer, every problem looks like a nail. Scientists (jounalists, politicians et al) are not immune to this malady.

 

•  According to the National Institutes of Health:

 

What is the BRAIN Initiative?

The Brain Research through Advancing Innovative Neurotechnologies (BRAIN) Initiative is part of a new Presidential focus aimed at revolutionizing our understanding of the human brain. By accelerating the development and application of innovative technologies, researchers will be able to produce a revolutionary new dynamic picture of the brain that, for the first time, shows how individual cells and complex neural circuits interact in both time and space. Long desired by researchers seeking new ways to treat, cure, and even prevent brain disorders, this picture will fill major gaps in our current knowledge and provide unprecedented opportunities for exploring exactly how the brain enables the human body to record, process, utilize, store, and retrieve vast quantities of information, all at the speed of thought.

 

This is the page from which the description is taken:

 

 

Some have likened this project to the Manhattan Project or the Human Genome Project or The Apollo Program. We hope it's not true. The objectives are much less clear; the subject many orders of magnitude more complex.

 

For our purposes here, those concerns are tangential. The problem in the current context is the circled graphic: "BRAIN Funding Opportunities". Research universities survive on government funding. Those scientists who have a different approach, perhaps a better approach, will not get funded, their careers dead-ended. "Emergent properties", of which many consider the mind to be a prime example, may not arise on the chosen funding path. Of course, the mind and the brain are distinct. We just don't know how. Most of us, I think, are interested in the brain because of the mind. I suspect many would-be researchers share this interest.

 

Additionally, as noted in our post, it is exceedingly difficult to predict scientific discoveries. There is a danger of finding what one is looking for, Gell-Mann's third category: seeing patterns that don't exist. It is much easier to drink the kool-aid when it's laced with money.

 

The Brain Initiative is probably good as a stimulative economic policy. As science policy, the benefits are (or should be) opened to debate.

 

•  Years ago, a prominent scientist at Tufts University told us that he had all but given up on talking to the media ("reporters" was his word). He had simply dispaired of accurate reporting because his careful and complete explanations would inevitably be simplified, distorted, caricatured and extrapolated well beyond what was scientifically plausible. Even scientific journals have to be read carefully and with a healthy skepticism.