Theories of Emotion and What They Mean for the Social Sciences
Friday 11 February 2022
A correspondent recently brought my attention to the work of Lisa Feldman Barrett, who has spent her career studying emotions. I hadn’t previously heard of her work so I made a superficial survey and was interested to find that she argues for an understanding of emotions as being constructed rather than as being human universals. The idea that emotions are human universals is familiar from the (reductionist) idea that there are six basic emotions — happiness, sadness, disgust, fear, surprise, and anger — that all human beings experience. This idea is due to Paul Ekman, but the approach is so typical of how science works that it is more or less inevitable that if Ekman hadn’t compiled his list of six basic emotions, someone else would have made a similar list that largely overlapped that of Ekman.
Ekman also explicitly identified these basic emotions with particular facial expressions, and Barrett has made a point in her books, articles, and popular presentations (she gave a widely viewed TED talk) that facial expressions are not necessarily tied to specific emotional states. So clearly Barrett is positioning herself contrary to Ekman, and one can look forward to psychology working through a dialectic of the emotions in which Ekman is the thesis, Barrett is the antithesis, and some future researcher arrives at a synthesis that takes what is best from existing theories and moves beyond the antithetical positions represented by Ekman and Barrett. Both schools of thought make good points, both have empirical research to back up their arguments, and a future theory of emotion needs to do justice to what is accurate in both theories without being merely eclectic.
There is also a school of thought about emotion that stems from cognitive science, which can be strangely disconnected from psychology, but is often found closely adjacent to computer science and the study of artificial intelligence. Allow me to share an anecdote about this. Some years ago I attended a talk in Astoria by Margaret Boden, who is well known in the philosophy of AI. The Portland Arts & Lectures series would invite speakers to Portland, and it would cost 20 or 30 bucks to go to the event scheduled in Portland, but then sometimes the same speakers would also go to other venues while they were in Oregon, so the Astoria event was free, and I went to it.
After the talk, during the Q&A, Boden was asked a question about the place of emotion in human life, and it was obvious from the way the question was asked that the individual asking the question thought that emotion was a central feature of human life and that Boden had given it short shrift. Boden responded by more-or-less dismissing emotion as a regulatory mechanism for cognition that would be necessary to the function of cognitive agency in any intelligent being, and so would inevitably evolve by some mechanism or another, but played no crucial or constitutive role in the mind. When she delivered herself of this view the person who asked the questions made some confused sounds, repeated part of the answer under their breath, and a murmur went through the room. Clearly, the audience thought Boden had said something radical and unexpected. So the audience of this Boden talk was working from an implicit paradigm of the importance of emotion in human life, and Boden made it clear that she did not share this paradigm.
I was struck by this exchange at the time, and obviously the memory has stayed with me, occasionally surfacing when I think about emotion, so my recent correspondence about Barrett’s work dredged this out of my preconscious. While I haven’t made an effort to read about theories of emotion in cognitive science and AI research, this exchange certainly gave the impression of an approach even more philosophically naïve than that of conventional academic psychology.
Thinking about emotion, I realized that there hasn’t been much productive exchange between philosophical theories of emotion and psychological theories of emotion, which represents an opportunity. This is curious given that, at least in American psychology, William James, a philosopher, wrote the classic textbook on psychology, so one would have expected a robust tradition of philosophical psychology to have followed, but this does not appear to have been the case. A careful philosophical reading of psychology, or a psychological research program into a philosophically formulated theory of emotion, would be places to start. Although I am not up on the literature of philosophical psychology, there doesn’t seem to be much sign of this. There are increasingly sophisticated philosophical engagements with physics, cosmology, biology, and archaeology, but psychology does not seem to have received the same attention, although evolutionary psychology, through philosophical work on evolutionary biology, comes closest.
Does this state-of-affairs require an explanation? Is there a reason for the relative philosophical neglect of psychological theories of emotion, and psychological neglect of philosophical theories of emotion? I can think of at least one claim of this kind. The final section of Wittgenstein’s Philosophical Investigations calls psychology into question in a radical way:
“The confusion and barrenness of psychology is not to be explained by its being a ‘young science’; its state is not comparable with that of physics, for instance, in its beginnings. (Rather, with that of certain branches of mathematics. Set theory.) For in psychology, there are experimental methods and conceptual confusion. (As in the other case, conceptual confusion and methods of proof.)”
Wittgenstein in his later years wrote a lot on the philosophy of psychology, much of which has been published posthumously. Clearly he believed there was a problem, but he didn’t point the way to the solution of the problem, nor did he attempt to write a systematic treatise that would put psychology on a better footing by addressing the conceptual confusions he saw in the discipline. As is evident from the above, Wittgenstein also thought there was a problem with set theory, and, again, his posthumously published works develop his critique at some length, but, again, he did not formulate an alternative. Other philosophers have built on Wittgenstein’s work and formulated a philosophy of mathematics sometimes called strict finitism, but I am unaware of the Wittgensteinian challenge to psychology being taken up.
We already know about the replication crisis in psychology, which has particularly affected social psychology, and which we could take as a concrete manifestation of the conceptual confusions inherent in the very way that scientific psychology as it is practiced today. Whether or not the academic institutions of psychology are taking the replication crisis any more seriously than Wittgenstein’s critique I do not know, but if we take a radical interpretation of the later Wittgenstein, the question then becomes whether, (1) scientific psychology can be revised and reformed so that it will become scientifically rigorous, or whether (2) scientific psychology must start over from new beginnings, from a theoretical blank slate, as it were, or lastly whether (3) scientific psychology is an impossible discipline that cannot be made scientifically rigorous no matter how it is formulated.
This last claim is philosophically interesting: might there be aspects or forms of human experience that are intrinsically resistant to scientific schematization? If there are, it would mean that there are recognizable, identifiable areas of human experience for which no scientific concepts can be a valid representation. This is more radical that being ineffable; it is being unconceptualizable. Or it could be argued that ineffability means that we cannot formulate concepts adequate to the representation of a region of human experience. Can we name objects — objects of knowledge? objects of experience? — without conceptualizing them?
In the case of ostensive definition — pointing at something — we identify an object without necessarily conceptualizing it. However, it could be argued that we bring a number of assumptions and presuppositions to our identification of an object, any object (a potential object of knowledge?), such as the very idea of an object, something that Gödel mentioned in his paper on the continuum hypothesis:
“…it seems that, as in the case of physical experience, we form our ideas also of those objects on the basis of something else which is immediately given. Only this something else here is not, or not primarily, the sensations. That something besides the sensations actually is immediately given follows (independently of mathematics) from the fact that even our ideas referring to physical objects contain constituents qualitatively different from sensations or mere combinations of sensations, e.g., the idea of object itself, whereas, on the other hand, by our thinking we cannot create any qualitatively new elements, but only reproduce and combine those that are given.” (Collected Works Vol. II, p. 268)
Gödel here was talking about the mathematical intuitions underlying set theory (precisely what Wittgenstein was criticizing at the same time), but what he says here is true for other intuitions, and it has applications to ostensive definition, and I have never read any philosopher who has tried to work this out in any kind of schematic or adequate way. (Therein lies an opportunity.)
Allow me another anecdote, this one relevant to ostensive definition: in 1989, on my second trip to Europe, I visited the Basilica of San Marco in Venice. Ordinarily I do not go on any kind of organized tours, but in order to see the treasures of the cathedral one had to go along with a tour, which lasted about an hour I think. As I recall, the young man who conducted the tour was from New Zealand, and was studying theology in Italy.
When the guide was talking about icons (Venice was sometimes a go-between for the Catholic and Orthodox churches, so issues concerning icons would come up more often in Venice than in other parts of Western Europe), one of the other tourists following along asked the question, “What’s an icon?” Instead of attempting to explain what an icon is, the guide pointed to a mosaic of Christ on the cross and said, “That’s an icon.” Then he went on to point out several other mosaics that served as icons. I think the response didn’t really do a good job of answering the question, and by the puzzled look on the face of the person who asked the question (notwithstanding’s Barrett’s claim that emotions can’t be read off facial expressions), she seemed more confused after the answer than before. I thought about this afterward, and have thought about it ever since as an example of the ambiguity of ostensive definition. When we point at something, we have several assumptions in mind, and if our assumptions are reasonably close to those of another person, pointing something out might be sufficient for an effective ostensive definition, but the fewer assumptions we share, the more difficult it will be to make effective use of ostensive definition.
To what extent can we detach our experiences from the assumptions and presuppositions we bring to the first step of knowing, or investigation, or exploration, or whatever we can to call it when we engage with the world? Or, if we cannot realistically detach ourselves from these assumptions and presuppositions, can we agree on some basal framework that nevertheless allows enough of the world to be as it is without our forcing a template upon it by our thinking (i.e., without going full Kant)?
With Gödel’s example of the presupposition of the idea of an object, this is a pretty minimalist framework. If we bring a similarly bare bones ontology to our exploration of the world, serving as a minimal framework, within which not all objects are alike, some of the objects that we encounter may approximate a pure identification without conceptualization. If this is the case, then the philosophically interesting scenario I identified above, whether there are aspects of human experience that cannot be conceptualized, could be approximated.
Suppose, then, as Wittgenstein suggested, that psychology is well and truly broken, but further suppose that it cannot be fixed. That is to say, suppose there is a fairly well defined region of human experience that cannot be turned into a science. If this were true, it would be very interesting, and it would entail a dichotomy that drew a line through human experience, with that which can be the object of a science, sensu stricto on the one side, and, on the other side, that which cannot properly be an object of science.
Since “pseudo-science” is already taken I could use “quasi-science,” “ersatz science,” or even “failed science” to describe an aspirational scientific discipline that is other than pseudo-science but which is also other than science in the strict sense of the term. Whereas pseudo-science apes the appearance of science, in order to give non-scientific claims the appearance of being actual scientific claims, the other form of non-science — quasi-science, or whatever we call it — apes the methodology of science, and so could be said to mimic the reality of science while still falling short of actually being science, because the purported objects of the science cannot be conceptually schematized.
Organized pseudo-science can construct for itself all of the institutional correlates of legitimate science, by creating its own institutions, its own journals, its own hierarchy, its own classics of frequently cited papers, and so on — essentially, a professional echo chamber in which only approved statements acceptable to the research community are allowed to stand. There is a sense in which this is institutionalized scientific method (or perhaps we could call it the institutionalized correlate of scientific method), but it is not to be confused with scientific method that confronts an hypothesis with empirical evidence in order to test how that hypothesis will hold up under this scrutiny.
Organized quasi-science is not merely the implementation of institutionalized scientific method, but is (or would be, if in fact it exists) a genuine process of formulating hypotheses, constructing experiments to test these hypotheses, and then confronting the hypothesis with actual empirical data. However, because the concepts are so confused at a fundamental level, the hypothesis, the experiment, and the evidence all miss the mark. The result is both an hypothesis and empirical experimental results that are sufficiently ambiguous that they can be interpreted in such a way as to keep the enterprise alive. (This would nicely explain the replication crisis.)
Suppose that this is true of psychology, as suggested by Wittgenstein, and further suppose that if this is true of a paradigmatic social science like psychology, the same problem affects other social sciences to some degree, so that anthropology, sociology, political science, and so on are all compromised. This would be one way to explain the gulf between the natural sciences and the social sciences.
Rather than the differentness of the social sciences being due to a distinctive methodology, it is simply due to conceptual confusion. I can’t imagine anyone in the social sciences accepting this argument, but a more rigorous formulation of the argument would be a good way to confront the social sciences with their failings, and would also be an opportunity to start over, if (2) above holds; but if (3) above holds, we would simply have to surrender the social sciences and accept that they are not appreciably different from magic, astrology, and divination.
