Emotion as Appraisal

On <Upheavals of Thought> by Martha Nussbaum

I first purchased my copy of Upheavals of Thought: The Intelligence of the Emotions from a Montreal bookstore in 2018, while tagging along for the CogSci conference that year. I don’t remember what made me pick it up that day, but whatever aspirations I may have nursed at the time wilted when met with the reality of the book’s 700-or-so pages. The book was dutifully transferred from apartment to apartment with each of our successive moves, remaining unopened until 2024, when a friend was starting a research project involving emotion words across languages. I like to know about my friends’ domains of expertise, and this was motivation enough to finally give it a shot. I gave myself permission ahead of time to skip chapters entirely if they didn’t seem interesting to me, and then I settled in to chip away at it little by little while commuting on the trolley; waiting in medical offices; killing time at 3am while in the grip of third trimester insomnia until—to my surprise—I finished the book six months later, after having read it without skips from cover to cover.

I’m not used to reading philosophy, so I thought I was in for very dense and convoluted writing. Instead, Martha Nussbaum’s clear and forceful prose carried me patiently, determinedly, and virtuosically along, persuading me at each breakpoint that I really mustn’t miss out on the next chapter: how could I otherwise hope to do justice to the argument should I, at some future date, have need to respond to sensible questions posed in the current chapter?

As an example, I was fairly certain I was going to skip the chapter on animal emotion, anticipating that I would find its discussion too vague to merit the page count. But: it emerged that a particular variety of Stoic considered language a precondition to emotion, and so only language-users could experience emotion, to the exclusion of animals and pre/nonverbal humans¹. Reasonable critics’ reply shared a family resemblance with the so-called duck test²: insofar as we can tell that anyone experiences emotion (e.g., by observing the way they behave in response to an event), it is self-evident that animals and babies experience emotion³. It is indefensible to invest the human adult’s expression of emotion with legitimacy while dismissing the animals’ and the baby’s as a simulacrum, given that they are often indistinguishable from each other (e.g. in the case of the emotions that we learn early on, like FEAR and ANGER). The critics maintained (a) that animals do experience emotion but (b) that judgment relies on linguistically expressible propositions, and (c) animals do not have a capacity for judgment as such. Therefore, emotion is unlikely to involve language and “they concluded, must instead consist of ‘nonreasoning movements’ (aloga pathê), housed in a separate nonreasoning part of the soul.” (p. 91) . Nussbaum says: everyone is a little bit wrong and a little bit right, and I’ll show you why in Chapters 2-4; stay with me!

So I had to take a look…and then I had to take more of a look because I was then given an account of the genesis of the emotions (a wonderfully narrated—yet much darker—version of something like Inside Out, with ANGER, SHAME, and GUILT as powerfully formative emotions). The tempo flagged only through what ended up being the relative doldrums of the book (mostly consisting of a very technical musical analysis that I may revisit later in life⁴), and then picked back up when I was promised a program (in Part III: “Ascents of Love”) to “reform” erotic LOVE and cast it as the energizing foundation of the good life. And the program very much delivered.

Forward momentum was perhaps also sustained by the sheer variety in knowledge sources that Nussbaum brought to bear in each chapter. Again, unfamiliar as I am with philosophy writ large, I am not sure how common it is for philosophers to draw on all of: personal anecdote and observations (Chapters 1-3), physiology experiments (Chapter 2), psychoanalysis (Chapter 4), musical analysis (Chapters 5 & 14), religious writings (Chapter 11), as well as narrative prose (including Virgil, Dante, Brontë, Proust, and James Joyce, among others; Chapters 9-16), and poetry (Whitman; Chapter 15). Seeing the arguments refracted through each of these genres certainly raises questions about how reliably each genre offers evidence to support the argument. But one does feel that the genre-hopping is enriching and offers fitting lenses with which to report on a construct as kaleidoscopic as emotion.

With that, I turn to choice bits from the early parts of the book, leaving to a future post the parts concerned with the role of emotion in structuring ethical social life. In particular, I want to think here about the highest-level frame shift I have since adopted from the book: the notion that emotion is as much a rational tool as logic is.

P.S. - Though it seems unlikely that anyone is looking for CliffNotes on the book in this post, in case you are…be warned: this is not the place to look! It has been a little over a year since I finished the book, and so the thoughts below consist of some combination of my notes, my (very unreliable) memory, and analogies that may or may not have come from the book (I really don’t remember) but with which I came to understand the argument.

Emotions are Evaluative & Eudaimonistic

The book seeks to justify a revision of the Greek Stoic conception of emotion. Along with the Stoics, Nussbaum agrees that emotions form part and parcel with our reasoning faculties, rather than serving as their foil or reflecting irrationality. According to Nussbaum, the Stoics argued that emotions are value judgements, consisting of an appraisal that an entity or event that is outside one’s control is somehow important to one’s own projects and personal flourishing. Emotion, in summary, is an evaluative tool in a sentience’s⁵ arsenal; one honed to serve eudamonistic aims by continually describing where we think we stand in relation to agents and events in our environments.

Core to the Stoic view was the notion that the value judgment (the emotion) implies neediness: an assessment that we need others in order to flourish, or that we need events to align with our plans to avoid becoming destabilized. We will consider ANGER, to first understand what it would mean for it to comprise an evaluation:

In order to have anger, I must have [a] complex set of beliefs: that some damage has occurred to me or to something or someone close to me; that the damage is not trivial but significant; that it was done by someone; probably, that it was done willingly. It seems plausible to suppose that every member of this family of beliefs is necessary in order for anger to be present. If I should discover that not A but B had done the damage, or that it was not done willingly, or that it was not serious, we could expect my anger to modify itself accordingly, or go away. […T]hese beliefs are essential to the identity of the emotion: the feeling of agitation all by itself will not reveal to me whether what I am feeling is fear or grief or pity. Only an inspection of the thoughts discriminates.

p. 28-29 (emphasis mine)

Beliefs & the identity of the emotion

We will test this out: If the damage to us were trivial, then we are left with the belief that it was caused by someone who did it willingly. In this case, “we could expect my anger to modify itself accordingly” into something more akin to IRRITATION or ANNOYANCE. A more challenging modification is a belief that an unwilling perpetrator caused us significant damage. Now, it seems, we might feel the need to seek out more details about the situation. How pressed was the unwilling perpetrator? How much did they resist? Were they threatened with the loss of someone precious to them if they did not comply? If we judge it to be the case that we too would have caved had we been as coerced as our perpetrator, we might not find it in ourselves to be angry anymore. On the other hand, if we believe that they had much more room to resist, we might be just as angry as we would be at the willing perpetrator. The key here is that the existence of ANGER depends on information, and how we choose to interpret it—cornerstones of rational thinking.

There is also another way that this modified scenario could have gone. We could still have believed that the perpetrator was unwilling. We could even entertain the possibility that we would have behaved identically to them were our roles reversed. We might even understand them to REGRET what they have done to us, and wish to undo it or to make amends. And we might still blame them and feel as much ANGER as if the base-case belief conditions for it had been met. I don’t remember Nussbaum’s theory of emotion really dealing with this very well, particularly because she could always rejoin with: well, if we are still angry, we must not really believe in our heart of hearts all of the extenuating assertions we have just made. Or, maybe we believe them, but our belief about the sizable nature of the damage done to us simply outweighs a true belief in the extenuating assertions.

But, what she does do well is signpost right away that emotions “may be no more reliable than any other set of entrenched beliefs.” (p. 2) Their status as rational constructs does not protect them from criticism: they need not be immediately validated just because they are felt with “heat and urgency”, and they could in fact be terribly misdirected. Any given judgment constitutes either an acceptance or rejection of an appearance. The sun, for example, may appear to me to be small enough to pinch between my index finger and my thumb, but I judge this appearance to be false, because I have learned something about the appearance of large objects when observed from a great distance. I move through the world as though it were a flat plane, yet a host of others assure me this perception is unreliable; the planet is more like a sphere. The way things appear to us may differ from what we judge them to be, and our rejection or acceptance of appearances may prove mistaken. In the most recent scenario we considered above, our attribution of blame—and the attending ANGER—is could just be misplaced, or we have decided that despite our beliefs, we need someone to blame, and the person most proximal to the damage will serve⁶.

Feeling != Emotion

The “feeling of agitation” that we might experience when under the influence of ANGER is not, according to the Stoics, equivalent to ANGER itself. To them, the feeling is a symptom of the emotion, which in turn is more like the underlying source of illness. Agitation is analogous to vomiting. The vomiting itself is not the pathogen that has invaded the host. Moreover, it is an entirely ambiguous signal. Many viruses, bacteria, and parasites can provoke vomiting in the same way that many emotions, like GRIEF or PITY—each comprising their own uniquely identifying constellations of beliefs—can provoke agitation in the “host”.

The disease analogy very much illustrates why the Stoic strategy was the complete extirpation of emotion, via belief modification. Change your beliefs about the situation: force yourself to believe the damage was trivial; to believe that you are above pain and nothing and no one around you matters very much; to think of the perpetrator as a puppet of fate, just another actor playing a role in a grand and meaningless stage. Now, they say, you cannot really be angry; you are cured!

Assuming that everyone is willing to take at face value that beliefs are modifiable—I think they possibly can be, for the most part, through a lot of patient effort; though whether one actually wishes to or not is a different matter—someone may still object: why be so extreme and get rid of all emotion? Why not just the “bad” ones? And the (probable?⁷) response: beliefs overlap across emotions like stars shared between constellations. For instance, the belief that your child is essential to your life and your flourishing gives you LOVE and JOY and perhaps PRIDE when that child is safe and happy and thriving. Were some event or person to threaten your child, or take your child from you, or cause harm to that child, you will have to experience FEAR, GRIEF, ANGER, and perhaps HATRED. If the source of harm to your child is yourself, SHAME and GUILT will haunt you. In the case where the child is safe but becomes more attached to someone else, you might suffer JEALOUSY and ENVY. The Stoics: …you would expose yourself to eight wellsprings of evil just to preserve a few bright spots of good?

In Parts II and III—which I will not discuss in detail here—Nussbaum begins in earnest to make the case that living the good life requires emotion, with all its attendant recognition of our neediness with respect to ungovernable others. This is her essential point of departure from the original Stoics, and she spends just over half of the book motivating COMPASSION (“a painful emotion occasioned by the awareness of another person’s [sizable and] undeserved misfortune”, p. 301) as a cornerstone of ethical civic life, and later (erotic) LOVE as the energizing force that trains us to look beyond the limits of our own self-centered projects and imagine the interiority of others. It is this emotion above all others, her argument goes, that equips us to value even strangers as unique, particular individuals with their own special weirdnesses and tendernesses. My impression is that Nussbaum contends that the o.g. Stoic insistence on attaining absolute self-sufficiency (the absence of emotion), is sustained not by some placid impassivity (a delusion, if there ever was one), but by another lurking emotion, one posing the most danger to her civic project: DISGUST.

DISGUST is the emotion that strives to preserve the boundaries of the self, keeping them unsullied by anything we would construe as corruption. I will skip all the arguments here for now, because they are super extensive, but suffice it to say that Nussbaum believes LOVE to be the emotion that can—when deployed appropriately—overcome the civic and interpersonal limits that DISGUST would impose on us.

Here, I will end rather abruptly with the guiding imagery that, I think, paints Nussbaum’s insistence on retaining emotion in vivid strokes, drawn from Marcel Proust’s suddenly-in-love character, M. de Charlus:

The world of Charlus in love is compared to a landscape full of mountains and valleys, produced as if by “geological upheavals of thought”; and this differentiated landscape is contrasted with the “uniform plain” of his previous unattached life, where no idea stood out as urgent or salient, no evaluation jutted up above any other. His self-sufficient world was, we might say, very much like the world seen from the point of view of a far-distant sun, a world not yet humanized by the earthquakes of human love and limitation, which are at once comic and tragic. His new world of twisted jealousy and towering love is a more agitated world, alive as it is at every moment to small movements of thought and action in a person whom he in no way controls (and who is, besides, especially inscrutable and unreliable). And yet the narrator tells us that this world is a world “enriched” — and enriched by the agitation itself […]. This normative conclusion remains to be examined. For now, we are beginning to have some idea of what it is to understand emotions as a certain sort of vision or recognition, as value-laden ways of understanding the world.

p. 88 (emphasis mine)

1

Since Chrysippus understood emotions to involve the acceptance of letka, proposition-like entities corresponding to the sentences in a language, it seemed to him obvious that creatures not endowed with linguistic capabilities could not have them; he concluded that popular ascriptions of emotion to animals (and young children) were based on a kind of loose and illegitimate antrohpomorphizing. p. 91

2. Which Sean has discussed elsewhere, in the context of determining whether large language models (LLMs) understand language.

3. “Critics such as the Stoic Posidonius and the Platonist Galen insisted that animals obviously have fear and anger and grief. Since they agreed with Chrysippus about the capacity of animals for learning and judgment, they believed this a knock-down objection to Chrysippus’ analysis of emotion as judgment. Emotions, they concluded, must be ‘nonreasoning movements’ (aloga pathê), housed in a separate nonreasoning part of the soul.” p. 91

4. Preferably to be discussed with friends who know a lot about music theory.

5. This is an unusual usage of “a sentience”, but I like it better than the alternatives. Here, I use it in place of “a sentient being” or “a consciousness”.

6. But I think, in this latter case, that Nussbaum would no longer necessarily call the emotion ANGER.

7. I actually can’t remember whether this is an explicit part of the argument, but it is a thought that was prompted by following through the implications of the theory that emotions are collections of belief, rather than the feelings and physiological consequences they provoke.