The Ascent of Affect: Emotions Research and the History of Emotions – Interview with Ruth Leys (Part II)

Ruth Leys is Professor Emerita of the Humanities at Johns Hopkins University. Her work has focused on aspects of the history and theory of psychoanalysis; the history of psychiatry; the history of the neurosciences; trauma theory; the mind-body problem; and the history of approaches to the emotions.

Jonas Knatz & Nuala Caomhánach: In the first part of our interview, we traced the emergence of the concept of trauma, the tension between mimetic and anti-mimetic theories, and the history of the shift from guilt to shame in theorizations of trauma. In your recent book, The Ascent of Affect: Genealogy and Critique (2017), you focus especially on the post-World War Two shift you have just described from an intentionalist approach to the emotions to a non-intentionalist approach. You also note that fifty years on, the emotion research field is today in a condition of stasis: as you put it, opposing positions “cling to their contested positions and research strategies, leaving fundamental questions unresolved.” Could you outline for us what those respective positions are, what their stakes are, and why you think the emotion research field has failed to function as a cumulative science?

Ruth Leys: I start my book The Ascent of Affect by observing the lack of consensus regarding the most fundamental assumptions in the field of emotion research. As I see it, those fundamental differences concern the question of meaning, which is to say, they concern whether we should understand the emotions as intentional states or as nonintentional processes.

An intentionalist treats the emotions as “about” something. The fact that an emotion has an object at which it is directed is considered a defining feature of the experience. For the intentionalist, meaning inheres in an emotion, it is constitutive. Accordingly, the relation an emotion bears to its object is not an accidental or contingent attribute but an essential feature: the essence of an emotion is not the mere consciousness of an affective state but an awareness of something– of a real or imagined object. (It could be an ill-defined or vague object, such as when I feel pride in my country, or fear the future.) The general view that emotions are intentional states was implicit in earlier Freudian approaches to the affects. Under the influence of Jean-Paul Sartre and others, it was also adopted in America in the 1960s by Magda Arnold, Richard Lazarus, and other so-called “appraisal” theorists in the emotion field at a time when for various reasons the emotions, which had almost disappeared as a topic in the psychological sciences, became a focus of interest again. The intentionalist position has also been endorsed by many philosophers, notably Robert Solomon and, more recently, Martha Nussbaum.

An alternative position argues that emotions are non-intentional states. On evolutionary and other grounds, in the 1960s American psychologists Silvan Tomkins and Paul Ekman adopted a non-intentionalist view of the affects. They argued that there exists a limited number of reflex-like, discrete primary affects or “basic emotions” that are evolved, pan-cultural, natural “kinds” characterized by distinct, prototypical facial expressions and patterns of behavioral and autonomic responses. Tomkins included eight affects in his original list: interest, surprise, joy, anger, fear, distress, disgust, and shame, to which he later added a ninth, contempt. Ekman expanded the list of what he called the “basic emotions” to about fifteen, and his followers have now expanded the list again to include even more discrete affects.

According to Tomkins and Ekman, the primary or basic emotions are triggered by stimuli in the environment and discharged by putative, subpersonal” affect programs” in the brain. Most important, they believed that the discharge of the emotions occurs independently of the subject’s cognition or apprehension of the meaning of the triggering stimuli. We might put it that on the Tomkins-Ekman account, the object of the emotion is turned into the causal trigger or “releaser” of the affective response. The result for Tomkins and Ekman is that people can feel sad or happy without knowing why– not because, as Freudians might argue, the intentional object has been repressed, but because affect and cognition are two separate systems. On this model, the affect system can become attached to a multiplicity of objects but it is itself intrinsically independent of all objects. In other words, for Tomkins and Ekman the affects operate blindly: they have no predetermined objects of their own.

Ekman also held (and still holds) that the basic emotions may be modulated or disguised by cultural norms or conventions, which he termed “display rules,” governing how we control and manage our emotions in public, or they may be masked by attempts at deliberate deception. But outside of these cases, the face involuntarily “expresses” the affects, with the expressions acting as authentic “readouts” of discrete, internal emotional states; as such, they are context-free signals that automatically and unintentionally express affects to others. Ekman also provided an explanation of lying according to which what is hidden in deception is detectable through the “leakage” of unmanaged micromovements of the facial muscles and related behavioral-autonomic discharges.

One of the main arguments of my book is that the basic emotion theory (BET) is unsupported by the evidence and incoherent at the theoretical level. In fact, it has been repeatedly demonstrated that the emotions are not organized into discrete categories with signature facial expressions or autonomic processes in the way the basic emotion theory assumes. Ekman’s claims about the universality of the basic emotion categories based on cross-cultural judgment studies have been decisively refuted, and the empirical evidence allegedly proving the correlation between the basic emotions and specific facial expressions or behavioral-autonomic reactions has not stood up to scrutiny. Nevertheless, the Tomkins-Ekman paradigm continues to dominate the affective neurosciences.

The stakes of these opposed views are very large. If you accept the idea that emotions are intentional states you place the question of meaning at the center of your analysis. But if you believe that emotions are best understood as discrete entities that can be triggered without regard to the meaning of an object or the role of the so-called higher cognitions, you dislodge the question of meaning from your analysis of the affects. Instead, you aim to identify the exact number of discrete emotions by measuring their component elements, including especially their putative characteristic facial expressions and accompanying physiological-behavioral features.

The history of emotion research in the post-World War Two period to the present is the history of the oscillation and confrontation between these competing views. The story is one in which, after an initial moment in the 1960s when the intentionalist position seemed to prevail, the non-intentionalist position, as represented especially by the work of Tomkins and Ekman, eventually succeeded in dominating the research field and has continued to do so up until the present day.

The reason why the emotion research field is not a cumulative enterprise is that these two approaches are not negotiable, they can’t be combined. You can’t be both an intentionalist and a non-intentionalist: the competing positions require different understandings of the importance and nature of intentionality, what research questions need to be asked, and what experimental methodologies need to be pursued.

JK & NC: If you are right that the non-intentionalist position, as exemplified by the work of Silvan Tomkins and Paul Ekman, is erroneous, what explains its continuing popularity?

Ruth Leys: In part the story of the success of the non-intentionalist position concerns the several challenges facing the intentionalist. The topic of intentionality is an intrinsically difficult one even for the best philosophers to handle, so it is no wonder that many busy experimental psychologists without a deep immersion in philosophy have failed to master the issues involved in operationalizing intentionality for scientific purposes. By contrast, the non-intentionalists Tomkins and Ekman developed effective laboratory methods that, for good or ill, have turned out to have an enduring impact.

In particular, both psychologists made use of photographs of posed facial expressions, held to be characteristic of each discrete emotion, as stimuli in their research enterprises. In 1976 Ekman systematized the deployment of such posed expressions by making available in slide form a set of black and white photographs defined as prototypical expressions of six discrete affect categories (this was the result of arduous research). Two years later, he and a colleague published a coding system, the Facial Action Coding System (FACS), for measuring and analyzing facial movements. This coding system was designed to provide an anatomically based, standardized scoring method for measuring the movements of the muscles of the face that researchers could use to test their hypotheses about the relationship between emotion and facial expression.

The use of posed photographs of facial expressions raises many problems that tend to be ignored by advocates of Ekman’s position. Nevertheless, Ekman’s methods have been so helpful in laboratory research that they, or similar approaches, are still widely employed– even by his critics. They have been used to assess people’s ability to recognize emotional expressions across different cultures; to evaluate which parts of the brain are activated when subjects view such posed expressions or to assess associated physiological reactions to develop surveillance techniques designed to expose potential terrorists at the airport; and so on.

Moreover, the entire non-imntentionalist framework is compatible with many other presuppositions. Thus an important feature of Ekman’s non-intentionalist approach is the assumption that affect programs are fundamentally involuntary discharges that instantaneously communicate and spread emotions among members of the species or group. According to Ekman, facial expressions and other nonverbal signs of emotion have evolved precisely for that purpose. According to this view, the transmission of emotions occurs without any cognitive mediation at all by the receiver. As long as the display is spontaneous, the receiver has access to the emotional state of the sender that is immediate and unthinking.

The idea that the emotional states of individuals can so readily ricochet off each other is appealing to many researchers because it seems to explain social cooperation. Moreover, the picture of the emotions is one in which eruptions of our inner passions under the certain conditions are regarded as ineluctable; as such it is compatible with the idea that a clear distinction can be drawn between the authentic affects we experience and express when we are overcome by them or when we are entirely alone and allegedly free from social contamination, and those expression we try to control when we are in public by adapting to social conventions. It is a picture that conforms nicely with “Rousseauean” assumption that faces that are held to be the product of automatically-generated internal emotional states are regarded as signifying authenticity, whereas convention-governed “public” social faces are held to denote a loss of innocence “imposed” on us by cultural conventions.

Another way of putting this point is to say that Ekman attempts to secure facial signaling from the threat of lying and dishonesty by assuming that beneath conventional displays the individual’s genuine feelings are bound to leak out. Whereas an alternative intentionalist account of the kind offered by the American psychologist Alan Fridlund suggests that it is not in the evolutionary interests of animals to display their intentions at all times and that therefore the threat of dissimulation or dishonesty cannot be dissolved in this way.

JK & NC: What alternatives to the Ekman and affect position are available today and why do you think it has been difficult for researches to adopt them?

RL: In my book, I pay special attention to the contribution of the work of Fridlund because in my view his book Human Facial Expression: An Evolutionary View (1994) offers the only viable alternative to the Ekman position. I was fortunate in that, in 2005 during the course of research for my book From Guilt to Shame, I stumbled across Fridlund’s book. My exchanges with him gave me confidence that my skepticism about Tomkins’ and Ekman’s views was justified, and he has been a terrific sounding-board for me ever since.

As a former student and follower of Paul Ekman who became a strong critic of the latter’s Basic Emotion Theory, Fridlund had the credentials and expertise to make his counter-arguments stick. The fact that, nevertheless, there has been such resistance to his views tells us a great deal about how entrenched the opposing non-intentionalist, basic emotion theory is and how difficult it has been to dislodge it.

Fridlund’s basic insight is that facial movements are forms of intentional signaling: they are essentially social displays that are sensitive to their audiences and hence to the context in which they are performed. Influenced by the “new ethology” of the 1970s, Frudlund argued that it would not be adaptive for signalers to signal his or her intentions at all times. According to the new ethology position, in order to manage and control others and to maximize their own interests, signalers attempt to “mind-read” their opponents by looking for the fine clues by which they can predict how their victims might act. And of course “victims” also evolve mind-reading skills, too. The new ethology therefore emphasized the tendency of animals to manipulate and deceive others.

Based in part on these developments, in his 1994 book Fridlund rethought the phrase “the expression of the emotions” that had been inherited from Darwin. Instead of positing the existence of a set of discrete internal emotions that under the right conditions involuntarily leak out in distinct facial expressions, he emphasized the strategic-social nature of animal displays. He argued that facial displays could not be regarded as automatic spill-overs of internal affective states, but as intentional movements serving an animal’s various social motives and sensitive to its audiences and the context of its interactions.

An immediate result of Fridlund’s re-thinking was that the “read-out” theory of the emotions and the one-to-one correspondence that the Tomkins-Ekman model had posited between internal states and specific facial expression were called into question. He also questioned the fundamental distinction posited by Ekman between the innate, reflex-like, sincere faces that involuntarily leak ongoing emotions, such as our genuine smiles, and the learned, instrumental faces that we perform when, for reasons of politeness or deception or social conformity, we deliberately disguise our true feelings, as when we produce “false” smiles. Fridlund dismantled this dualism between authentic versus inauthentic displays, because according to him all displays are strategic, they all arise from social interactions. Another consequence of Fridlund’s arguments was the suggestion that deception is omnipresent in nature and potentially highly advantageous for the organism, even as he also proposed that the relationship between organisms may often be cooperative.

Among many other notable features of his well-researched and brilliantly-argued book was Fridlund’s critique of one of Ekman’s most iconic experiments, a comparative study of the responses of Japanese and American students to certain stress films, a study that purported to prove the dissociation between cultural conventions or display rules and the basic emotions. For the discerning reader, his criticisms of this experiment were so devastating as to completely undermine the validity of Ekman’s claims.

Since the appearance of his book, Fridlund’s views have received strong support from several psychologists who in my opinion have persuasively shown that the evidence in favor of the existence of discrete emotions located in the brain and characterized by signature facial expressions is indeed inadequate and that the theoretical rationale for Ekman’s position cannot be sustained. Nevertheless, the majority of emotion scientists have continued to resist Fridlund’s views. As he admits, his position has been a “hard sell,” for reasons I discuss in detail in my book.

JK & NC: In the field of History of Emotions there have been other attempts to bridge the gap between experience and expression: a historicization of the body itself, either in Monique Scheer’s ideas about emotions as part of a Bourdieuian habitus or in Rob Boddice’s attempt to write a history of emotional experience by combining Daniel Lord Smail’s Deep History with neurobiological studies of brain plasticity. Do you see these attempts as alternatives to Fridlund’s theory?

RL: By stating that in my view there is no intellectually viable alternative to Fridlund’s Behavioral Ecology View I was referring to scientific alternatives. I don’t view the historian Monique Scheer’s approach as an alternative to Fridlund’s approach, because she shares with him a commitment to studying human behavior in intentionalist terms as forms of life that are shaped by context and situation. Like Fridlund, though for different reasons, she regards emotional behaviors as socially-situated, historically-influenced forms of embodied intentional actions.

However, I would draw attention to one of the topics where in my view in her arguments she goes too fast. This is when, without discussion, she appears to accept the idea that human cognition (or perception) can be non-conceptual, as opposed to the claim that it is conceptual through and through, as the philosopher John McDowell argues. It’s as if she thinks that when learned behaviors of the kind that constitute a “habitus” in the sociologist Pierre Bourdieu’s sense of the term, become forms of skilled coping, not only can they be performed automatically without conscious awareness, but they lose all conceptual content. This issue takes us into deep philosophical waters. It’s difficult even for serious philosophers to get a handle on the topic, and surely even harder for us historians who for the most part are not trained in philosophy. There is an important debate on this issue in the philosophy of mind, as evidenced by the well-known dispute between Hubert Dreyfus and John McDowell in 2005 over the nature of skilled coping, and it would have been useful and interesting if Scheer had at least touched on its implications. I say this with all due sympathy, because I recognize that the issues involved are so difficult to master.

More broadly, I confess to being somewhat allergic to methodological recommendations as to how historians ought to proceed. It strikes me that such recommendations tend to come too late, in the sense that they end up simply describing what historians already do. In the case of Scheer’s suggestion that we embrace a methodology based on Bourdieu’s practice theory, it’s not clear to me that the actual practices of historians need to change much at all. Scheer seems to recognize this when at the end of her article she sketches “four overlapping categories of emotional practices that can be studied as part of a history of emotions,” while recognizing that “[t]hese categories pick up on existing approaches with the intention to infuse them with a new vocabulary, that of practice theory.” What her proposed methodology thus appears to amount to is the idea that historians should now use Bourdieu’s language to describe what they are already doing.

As for Rob Boddice’s attempt to combine Smail’s neurohistory with an interest in the topic of brain plasticity: it does offer an alternative to Fridlund’s approach but in my opinion not a useful one. In Ascent of Affect I critique Smail for proposing that we should understand the past, including prehistory, as the product of the traces that are unintentionally sedimented in material artefacts of various kinds– such as material marks found in archeological remains, stone tools, fossils, and other deposits—artifacts that he considers more reliable records of what happened in the past because, as a matter of “information” rather than the meaning revealed in unreliable documents and texts (since humans can lie), those artifacts can be interpreted in much the same way a population geneticist reads or decodes a strand of DNA. This leads Smail to propose that historians should focus on all those non-intentional processes that have influenced human behavior, including neurobiological changes in the brain, from the “deep time” of the prehistoric ages to the present.

Among the motors of change Smail is interested in are various “autotropic devices,” including stimulants such as coffee, tea, alcohol, opium, and psychotropic drugs, as well as new cultural practices such as novel-reading that, he argues, in the 18^th century actively shaped human emotional predispositions through their effects on the neurochemistry of the brain. In this way, he suggests that ideas and political change can be thought of less as the product of human agency than as the unintended consequences of impersonal shifts in cultural practices interacting with subpersonal, neurohormonal brain processes. In other words, Smail embraces many of the same assumptions held by the new affect theorists with their emphasis on the importance of the role of unintentional, non-signifying affective corporeal forces in causing our political, ethical, and aesthetic beliefs, and does so in terms drawn from the basic emotions approach and related assumptions that I argue are erroneous.

Boddice adopts Smail’s neurohistory framework but is critical of its implicit biological determinism. He therefore proposes to revise Smail’s position by re-emphasizing what Smail seems to deny, the crucial importance of human intentions and agency in history, even as Boddice retains the neurohistorical idea that what matters in this regard is the dynamic relationship between those intentions and the plastic changes in the brain that have occurred in specific historical contexts. He thus argues against Smail that “if experience is historical because the brain is bioculturally made, then we must be prepared to take historical actors at their word.” At the same time, and with Smail, Boddice believes that brain changes are indeed at the heart of historical experience and that accordingly it’s the task of the historian to understand those changes alongside the human actions with which they are connected. But even if Smail and Boddice are correct in characterizing the human brain as so plastic as to undergo constant alterations, it’s not clear how we can ever learn anything about those alterations in the brains of historical actors because they are dead and gone. So we seem to be left with a proposal to the effect that historians need to pay attention to what people have said and done in the past, which is what they’ve always tried to do. In short, it doesn’t seem to me that the “neuro” in “neurohistory” adds anything.

JK & NC: In Ascent of Affect you also pay attention to what has been called the “turn to affect” in cultural and social studies. Under the name of “Affect Studies,” scholars in the humanities and social sciences have adopted pre-conceptual and pre-linguistic notions of affect and feeling. In your book you suggest that such affect theorists are fundamentally mistaken. Why?

RL: The new affect theorists and the BET psychologists such as Tomkins and Ekman share one fundamental premise– that the affects or emotions are non-intentional states. Believing that the role of reason and belief has been overvalued in politics, ethics, and aesthetics, the affect theorists emphasize instead the corporeality of human existence and especially the importance of putative subliminal, bodily-affective intensities they regard as decisively influencing and indeed causing our political and other beliefs. They treat those affective intensities as prior to and independent of ideas and cognition– which is to say, as prior to intentions –because they are held to be nonsignifying forces that act in and on bodies below the threshold of consciousness and meaning. You can see right away the overlap between such ideas about affect and the non-intentionalism of the Basic Emotion theorists, and indeed several of the new affect theorists make use of Tomkins’ and Ekman’s scientific claims to support their views.

My criticism of the affect theorists is quite straightforward. First, in arguing their case, they often appeal to various scientific findings by Tomkins, Ekman, and others that, when examined carefully, turn out to be dubious, or are misinterpreted, or aren’t supportive of their claims. Second and more theoretically, I suggest that, by replacing an interest in the role of reason and belief in human life with an interest in the role of affects acting independently of reason and belief, the new affect theorists make disputes about the meaning of our beliefs and judgments irrelevant. This is because according to them, what matters in our political, ethical, and aesthetic lives is not our disagreements over what we think and believe, but the role of corporeal affects operating independently of issues of meaning. We might put it that for the new affect theorists an interest in disagreement over beliefs has given way to an interest in differences of feeling about which it makes no sense to disagree: when people have different affective responses, they don’t disagree, they just are different.

The net effect of the affect theorists’ arguments is to emphasize differences ion subjectivity and identity at the expense of disagreement, which is to say that ontology becomes the focus of attention at the expense of argument and debate. (This is a point that Walter Benn Michaels has made more than once.) For the new affect theorists, then, political views are nothing but the expression of personal preferences, and political action is a matter not of trying to convince others of the truth of one’s ideas but of attempting, through the use of various kinds of affective manipulations, to alter people’s feelings and inclinations. Who is doing the manipulations? Master manipulators who are themselves immune to the influence of the hypothesized vital affects that sub-personally manipulate all the rest of us? Who are these manipulators? And how do they justify their normative convictions and beliefs regarding the superiority of one sort of political arrangement over another, democracy over fascism, say, if their own theoretical arguments preclude them from providing reasons for such convictions and beliefs? That in a nutshell is my argument against the new affect theorists.  

***