Stoicism and Emotion, V: brutishness and insanity

Orestes pursued by the Furies

The Stoic way of understanding emotional experience emphasizes that what we properly call anger, fear, love, or any other emotion actually depends on a judgment by the mind, that this or that has happened or is about to happen, and that it makes sense to react in some feeling-laden way. So what about animals? And what about people who aren’t capable of reasoning, either because they are very young children or because their mental faculties are impaired in some way? Should the strong feelings they undoubtedly have be considered emotions? Are they morally responsible for things they do in moments of strong feeling? And what about the fact that none of us is perfectly rational, unless we happen to be a Stoic sage? Is there really a difference between the actions of a typical imperfect adult in a fit of anger and what might be done by someone who is mentally ill? Interestingly, the ancient Stoics had answers to these questions.

Famously, for the Stoics everyone who is not a Sage is a fool, and in a sense “insane,” because he lacks knowledge of what is important, and consequently too readily assents to impressions from which he should recoil. This includes people who get angry, which is why Seneca calls anger a “temporary madness.” This class of individuals can certainly be held morally responsible for their actions, since they are perfectly capable of reason, they just don’t use it well. This is the set up for the fifth chapter of Margaret Graver’s book on Stoicism and Emotion, which I have been commenting upon with her help (she has kindly agreed to check my posts before publication).

In order to keep confusion at bay, I will follow Margaret’s qualification of the two types of “insanity”: paradoxical insanity (where “paradoxical” in ancient Greek just meant contra to common opinion) is the way in which we are all fools because we are not Sages (we are, literally, unsound, as one translates the Latin word insanus). The other type is “melancholic” insanity, from the Greek term literally referring to “black bile disease,” but which broadly speaking indicates a condition close to our modern conception of mental illness. Again: moral responsibility requires functional agency, so the paradoxically insane is ethically responsible for the use of his impressions, while the melancholically insane is not.

A classic example of melancholic insanity is that of Orestes, the son of Agamemnon and Clytemnestra, who at one point hallucinates that his sister is a Fury, pursuing him for his (just, arguably) matricide. According to the ancients, this condition may be the result of what we would today call extreme psychological stress, or may even be caused by drugs, including alcohol. Graver points out, for instance, the existence of a fragment in which Stobaeus refers to drunkeness as a “little insanity.” Interestingly, one of the disagreements between Cleanthes (the second head of the Stoa) and Chrysippus (the third head) is that the former thought that once acquired virtue cannot be lost, while Chrysippus points out that even the Sage may suffer the effects of melancholia or drunkenness (though she shouldn’t get drunk of her own volition in the first place!), and thus at least temporarily lose virtue. The difference between the two philosophers is not actually very big, according to Margaret, but it is indicative of the fact that the Stoics modified their positions in response to external criticism, in this case from Academic Skeptics.

Again, though — and contra popular mischaracterization — there is a clear distinction between actual insanity and the Stoic “paradoxical” insanity, as Graver forcefully reminds us:

“No bona fide Stoic text ever asserts that the mental state of every human is just the same as that of an Orestes or Alcmaeon, and none ever refers to melancholia or any other medicalized notion of insanity when speaking of the madness of humans in general. … ‘All fools are mad’ is a paradox, one of the counterintuitive teachings for which the school was renowned. Like others of its kind — ‘only the wise person is rich,’ ‘all fools are slaves’ — it runs contrary to popular opinion (para doxan) but becomes plausible when restated in other terms. It is a conversation opener, a deliberately provocative formulation meant to arouse the curiosity of the audience, later to be cashed out in a way that renders it acceptable.” (p. 117)

This isn’t quite as strange as one might at first think. Chrysippus points out that even lay language refers to people in the throng of strong emotions (anger, love, etc.) as “besides themselves,” i.e., mad, if temporarily. When those people wish to do whatever is on their mind to do “no matter what,” this is a clear indication that they are not in full possession of their rational faculties. The Stoics just pushed the point a little further: for them, anyone who sets their heart on external things — as almost all of us do — is primed for emotional turmoil at any moment. We might be fine now, but any chance event can throw us into a tailspin. Cicero too uses a similar approach, when in the third Tusculan Disputation he uses the Latin word “insania,” meaning without health, i.e., a mental condition that falls short of full health, or sanity.

Graver brings up the intriguing question of whether it is compatible with Stoic psychology to consider the possibility of a strictly emotional cause of derangement, given the Stoic emphasis on rational assent. She reminds us that the Stoics were strict materialists, for whom anything happening in our mind is the result of one sort or another of physical change in the pneuma, the substance that pervades everything. So yes, it is perfectly conceivable, within Stoicism, that either physical substances (drugs, alcohol) or repeated, strong emotional experiences will cause some permanent alteration of our psychic condition, mediated by physical changes. Nowadays, we don’t believe in pneuma, but we think — like the Stoics — that there is no separation between the mental and the physical, so the general idea still applies.

The last two sections of the chapter are devoted to Seneca’s treatment of anger, and especially of a particularly dangerous form of insanity that Seneca calls “brutishness.” They make for fascinating and insightful reading. Brutishness in the Senecan sense is no longer anger, but it has its roots in the latter condition, and it should serve as a warning for the dire, ultimate consequences of indulging in anger. The difference between anger and brutishness is that the first is motivated by a (mistaken, in Stoic philosophy) belief that one has been hurt. The sort of behavior Seneca calls brutishness, by contrast, is cruel and results in inflicting pain for fun, without even a plausible reason for it. Indeed, Margaret points out that although the term brutishness refers to animal-like behavior, this is misleading, since animals don’t attack out of cruelty, but in response to natural urges like hunger, self-defense, or defense of their offspring. Animals are not morally responsible for their actions, human beings in possess of their rational faculty are.

Seneca explains the difference between anger and cruelty also in his On Clemency, where he says that the cruel tyrant has a tendency to punish beyond what is actually required by the situation, indulging his own lust for blood and the infliction of pain. And in his Letter CXXXIII to Lucilius he describes the excesses of Mark Anthony, an example of the cruelty developed by the far gone alcoholic. We call this anti-social personality disorder, or psychopathy. This loss of rationality affects us profoundly, of course, because for the Stoics rationality is the best and most characteristic of human attributes.

The last bit of this chapter is an in-depth discussion of Seneca’s so-called three movements in On Anger, and again is well worth a detailed look. Consider first the following extended quote from Seneca:

“Let me tell you how the emotions begin, or grow, or get carried away. The first movement is nonvolitional, a kind of preparation for emotion, a warning, as it were. The second is volitional but not contumacious, like this, ‘It is appropriate for me to take revenge, since I have been injured,’ or ‘It is appropriate for this person to be punished, since he has committed a crime.’ The third movement is already beyond control. It wants to take revenge not if it is appropriate, but no matter what; it has overthrown reason. That first impact on the mind is one we cannot escape by reason, just as we cannot escape those things which I said happen to the body, such as being stimulated by another person’s yawn, or blinking when fingers are thrust suddenly toward one’s eyes. That second movement, the one that comes about through judgment, is also eliminated by judgment. And we must still inquire concerning those people who rage about at random and delight in human blood, whether they are angry when they kill people from whom they have not received any injury and do not believe that they have — people like Apollodorus or Phalaris. This is not anger but brutishness. For it does not do harm because it has received an injury; rather, it is willing even to receive an injury so long as it may do harm. It goes after whippings and lacerations not for punishment but for pleasure. What then? The origin of this evil is from anger, which, once it has been exercised and satiated so often that it has forgotten about clemency and has cast out every human contract from the mind, passes in the end into cruelty.” (On Anger II.4-5)

The standard scholarly interpretation of this is that Seneca brakes with the bit that begins “and we must still inquire,” where he starts talking about brutishness. According to some interpreters, anger is not present until the third movement, and brutishness is a separate topic entirely. If this is true, then – observes Graver – Seneca is committed to say that there is a half-way point at which one assents to the full content of anger as a judgment but still doesn’t get carried away. So the sequence of three movements would be: pre-emotion > anger not at odds with reason > full-fledged anger. But this is contrary not only to the entire corpus of Stoic doctrine (sounding suspiciously Aristotelian), but also to everything Seneca himself has been saying in On Anger up to that point.

Margaret’s interpretation, by contrast, seems to me (admittedly, as a simple Stoic practitioner, not a scholar of ancient philosophy) to make much more sense. She takes Seneca’s sequence to present anger in the middle, flanked by a pre-emotion that precedes it, and by a runaway brutishness that follows it (if one indulges one’s anger). Neither the first nor the third movement are rational. The first one because it takes place without assent, the third one because it happens to an individual who is no longer rationally competent. Only the second movement is actual anger, because it is caused by a mistaken assent given by reason to the impression of injury. Remember that in Stoicism assent to an impulsory impression does not take place before the impulse; rather, it is the impulse, analyzed at the intentional level. That analysis, of course, may be on target (the agent withholds assent and anger winds down) or off (the agent mistakenly gives assent and full anger results).

I find this analysis — and Seneca’s presentation of the issue — beautiful and clarifying. In Stoic psychology, anger is a cognitive emotion, which is why it is under our control. But neither the pre-emotion nor the descent into brutishness is under our control. The first because it is naturally inevitable, since it comes before reason kicks in. The latter because we have lost control of things and are overpowered, thus losing competence to arrive at rational judgment. This is a terrifying prospect, which has enormous practical consequences, and that’s precisely why Seneca makes a big deal of it, using appropriately horrific language to describe it. As Graver concludes at the end of the chapter:

“One thus has a powerful motive to learn ways of eliminating or at least decreasing the frequency of anger by the methods Seneca goes on to suggest, like examining one’s conscience, correcting one’s values, asking friends for help. For these are the means of preserving one’s humanity.” (p. 132)

Advertisements

Stoicism and Emotion, IV: feelings without assent

Ship in a storm, by Daniel Eskridge

There is a famous story told by Aulus Gellius, of a Stoic philosopher who finds himself aboard a ship in the midst of a severe storm. While he does not scream or cries, he shows all the outwardly signs of being emotionally disturbed: he is pale, he trembles, and his expression is one of alarm. Once the storm passes, Aulus asks the philosopher how come his Stoicism did not preclude such reactions? Isn’t that what his school teaches? Not at all, responds the philosopher, and gets out his copy of Epictetus’ Discourses, from which he quotes, in part:

“When some terrifying sound occurs, either from the sky or from the collapse of a building or as the sudden herald of some danger, even the wise person’s mind necessarily responds and is contracted and grows pale for a little while, not because he opines that something evil is at hand, but by certain rapid and unplanned movements antecedent to the office of intellect and reason. Shortly, however, the wise person in that situation ‘withholds assent’ from those terrifying mental impressions; he spurns and rejects them and does not think that there is anything in them which he should fear.” (Fragments, IX)

Margaret Graver begins in this way the fourth chapter of her Stoicism and Emotion, which I am commenting on in this series of posts. The idea expressed by Epictetus is crucial to understand the Stoic theory of emotions. The initial mental impression (phantasia, in Greek) is an inevitable component of human affective responses, and the point of Stoic training is certainly not to try to achieve the impossible. The bit that is morally significant is not whether one experiences the impression, but how one reacts to it, and specifically whether one actually judges that there is an evil is at hand. Before that point one has what Graver calls a pre-emotion, a good rendition of the Greek propatheia.

A similar treatment of the emotions is actually found in Aristotle, and — among the early Stoics — in Chrysippus:

“People who are weeping stop, and people weep when they do not want to, when the impressions created by underlying facts are similar, and there is either some impediment or no impediment. For it is reasonable that in such cases [i.e., those of involuntary weeping] something happens similar to the way that the cessation of weeping and lamentation come to pass, but rather in the beginnings of the circumstances bringing about the movement.” (cited in Galen, The Doctrines of Hippocrates and Plato IV.7.16-17)

The point is that the simple fact that people are weeping (or laughing, for that matter) should not be taken to be evidence of a fully formed emotion, because people may start weeping suddenly, in response to an external stimulus, even though they are not actually grieving, or stop weeping when they are distracted by something else, even though they are, in fact, grieving. Proper distress, for the Stoics, is volitional, because it is dependent on judgment.

Margaret observes that this view that feelings sometimes occur in the absence of assent is most fully developed, in the Stoic literature, by Seneca, particularly in On Anger. According to Seneca, anger is a high-level response, which requires not just the impression that an injury has been received, but the additional assent to such impression:

“We hold that anger dares nothing on its own; rather, it comes about with the mind giving its approval. For to gain an impression of injury received, and conceive a desire for revenge, and to link together the two ideas that one ought not to have been wronged and that one ought to take revenge-none of this is characteristic of that impulse that is stirred involuntarily.” (On Anger II.1.4)

Why is this? Because anger, fear, and other emotions depend on our believing certain things. For instance, a person isn’t really angry unless they believe they have been injured in some way: to be angry is to believe that a particular person knowingly did something bad to you which you didn’t deserve. That sort of belief is complicated; it relies on concepts like personhood, intention, and fairness. While some reactions really are involuntary — a rush of adrenalin, as we say, or jumping at a loud noise — it wouldn’t be plausible to say that a truly involuntary reaction can include all of that belief-content. So the Stoics need a lesser term for the kind of reaction that might have some of the physiological characteristics of emotion but doesn’t include the characteristic beliefs. Their word for the lesser reaction is “pre-emotion.” As Graver points out later on, in the course of her lengthy and careful exegesis of Seneca’s On Anger, the crucial distinction to the Stoics is between assent, which is the most crucial function of the rational mind (see here) and the ability to receive impressions, which we share with the rest of the animal world. Indeed, Seneca provides a long and at first sight haphazard list of examples of situations that appear to be anger but are not. What the items on the list have in common is that they are all feelings that do not depend on assent, and for which, therefore, the individual experiencing is not morally responsible. But he is human, so he will experience them just the same:

“For natural faults of body or mind are not removed by any amount of wisdom: what is innate and implanted may be mitigated by treatment, but not overcome. … Some people have lively, energetic blood that rises swiftly to their faces. This is not cast out by any amount of wisdom; if it were, if wisdom could erase all a person’s faults, then wisdom would have nature itself in charge.” (Letters to Lucilius, XI)

Moreover:

“Lest it should seem that what we call virtue strays outside the natural order, the wise person will tremble and feel pain and grow pale, for all these things are feelings of the body.” (Letters to Lucilius, LXXV)

In Letter IC, Seneca even explains that sometimes even the wise person weeps involuntarily, for instance suddenly at a funeral, and at other times voluntarily, when remembering a loved one who passed away. Contra to common misunderstanding in modern times — not to mention popular caricatures of Stoicism — our philosophy is not a magic bullet that will make you invincible, or impervious to the basic feelings associated with the human condition. It is, rather, an attempt to analyze that human condition and to derive best practices for living it in the best way possible that is actually accessible by members of our species.

Margaret’s chapter ends with an interesting discussion of the same concepts as found in the Jewish scholar Philo of Alexandria, as well as in the exegetical writings of the Christian Origen. Philo is concern with the interpretation of Genesis and Exodus in the Old Testament, for instance presenting Abraham as not really mourning Sarah — in the standard sense of the term — but rather behaving a bit like a Stoic sage. His concern, like that of the Stoics, is to explain what Scripture says about Abraham (“he came there to mourn”) in a way that allows Abraham to still be the wisest of all human beings. Similarly, Origen — who was familiar both with Philo and with actual Stoic sources — is concerned with interpreting some apparent emotional reactions displayed by Jesus as propatheia, or pre-emotions. In other words, it won’t do for these two authors to allow a possible interpretation of either Abraham or Jesus as affected by what the Stoics would characterize as unhealthy emotions, hard to reconcile with the wisdom of a prophet and the character of a god. It seems, then, the the concept of pre-emotion was instrumental in allowing the Stoic theory of emotions to be taken seriously by many in the ancient world, including their later rival, the Christians.

Stoicism and Emotion, III: vigor and responsibility

Chrysippus, Trajan Markets in Rome, photo by the Author

There is a sense in which emotions are something that happens to us regardless, or indeed in spite of, our will. As Margaret Graver reminds us in the third chapter of her Stoicism and Emotion, this idea of passivity is embedded in the very word the Ancient Greeks used to refer to emotions: pathos, the noun form of the verb paschein, which means to suffer, or to undergo.

Of course, as we have seen so far in the course of our discussion of Graver’s book, the Stoics made quite a big deal of reminding us that full fledged emotions actually include a cognitive component, which means that they are, in a sense, “up to us.” That said, we know that even Chrysippus, the third head of the Stoa, was well aware of the fact that people usually have a hard time controlling their feelings:

“Often, through the same blindness, we bite keys, and beat at doors when they do not open quickly, and if we stumble over a stone we take revenge on it by breaking it or throwing it somewhere, and we say very odd things on all such occasions.” (Chrysippus, cited in Galen, PHP 46.43-45)

Remember that two millennia-old quote the next time you shout at your computer, some things never change… According to Graver, despite the standard Stoic account of emotions which began with Zeno of Citium, Chrysippus was pretty pessimistic about the possibility of controlling them, thinking that once they begin they are for all effective purposes unstoppable. This is an important point to make, because it clearly shows that the Stoics were serious and sophisticated thinkers, and did not espouse the “stiff upper lip” caricature of philosophy that is often nowadays associated with their name.

Nevertheless, there seems to be a serious problem in the contrast between the Zenonian idea that emotions are up to us because they result, in part, from a cognitive judgment, and Chrysippus’ realistic appreciation that it ain’t quite that easy to overcome unhealthy passions. The resolution of this tension lies in the words “in part”: recall that for the Stoics fully formed emotions are the result of an impression, which is not under our control, and a cognitive judgment, which is under our control. These are co-causes of the emotions. The problem is that on the spur of the moment the impression easily overruns the cognitive judgment, as Seneca repeatedly points out in his On Anger:

“For once the mind is stirred into motion, it is a slave to that which is driving it. With some things, the beginnings are in our power, but after that they carry us on by their own force, not allowing a return. Bodies allowed to fall from a height have no control of themselves: they cannot resist or delay their downward course, for the irrevocable fall has cut off all deliberation, all repentance; they cannot help but arrive where they are going, though they could have avoided going there at all.” (On Anger, I.7)

Which means, argues Chrysippus, that the answer lies in a sort of cognitive preemptive action: we need to work on our character, our overall collection of judgments, before a specific impression hits us. As in medicine so in practical philosophy: an ounce of prevention is worth a pound of cure after the fact.

“Stoic thought does not consider the impulsory impression to be the principal cause of an action. It is indeed a cause, but the real cause or reason for the assent is to be found in the agent’s own mental character, where by ‘mental character’ is meant simply the structure and content of one’s own belief set. … Whether assent is given on any particular occasion will thus depend on the nature of the existing mental contents and the sort of standard one uses for recognizing logical fit.”

Margaret at this point discusses Chrysippus’ famous analogy with the combination of internal and external causes that allow a cylinder to roll. I refer the reader to my own summary of that famous Stoic discussion of volition. The bottom line is that two people may be subject to the same impression (say, lust for an attractive potential sexual partner) and arrive at a different fully formed emotion because their characters differ enough that one is inclined to assent to and the other to withhold assent from the initial impression.

Graver then “tests” Stoic, and especially Chrysippean, ideas against several well known examples from Greek mythology. This may seem odd to modern readers, but is, in fact, not very different from the way in which contemporary moral philosophers confront their intuitions about ethical dilemmas by way of constructing thought experiments, sometimes inspired by fictional characters drawn from modern literature. Here, for instance, is how Chrysippus analyzes the famous instance of Menelaus’ reaction to seeing again his wife, Helen, during the final sack of Troy:

“An example is Menelaus as depicted by Euripides. Drawing his sword, he moves toward Helen to slay her, but then, struck by the sight of her beauty, he casts away his sword and is no longer able to control even that. Hence this reproach is spoken to him: ‘You, when you saw her breast, cast down your blade / and took her kiss, fondling the traitor dog.’” (Cited in Galen, PHP 4.6.7-9)

Margaret comments that Menelaus’ passivity is the result of a weakness of reason, not an instance of yielding to a passion that is independent of cognitive assent. Menelaus behaves that way because he suffers from a structural weakness in his belief system, i.e., in his character.

I want to emphasize just how important this is for practical purposes, as far as modern followers of Stoicism are concerned. Too often nowadays Stoicism is brandished as a magic wand, as if one decides to “be” a Stoic and this, ipso facto, guarantees immunity from unhealthy emotions. It doesn’t, and Chrysippus, Seneca, and Epictetus would be astounded that anyone would think so. Stoic training is like training for the Olympics (a metaphor often used by Epictetus): you don’t just decide to be an athlete, start running, and win the race. You have to train, patiently, for years, improving gradually, and suffering setbacks. We are talking real life here, not wishful thinking.

Medea, of course, is another example of a weak structural belief system, often used by the Stoics, who however tended to see that tragic character with compassion, as in Epictetus. Yet another example, again discussed by Chrysippus, is more positive: it refers to an episode of the Odyssey where Odysseus (a standard Stoic role model) has returned home to Ithaca, but has adopted a disguise to study the situation. He is angered by the brazen behavior of the Suitors and the treachery of his own maidservants. But he strikes his own breast and addresses his heart, reminding it to tolerate the offense for now, so not to spoil any realistic chance at justice later on.

Graver devotes a section of chapter 3 of her book to an interesting discussion of how Galen (a known critic of the Stoics, despite being the personal physician of none other than Marcus Aurelius) unfairly characterizes Chrysippus’ ideas, even accusing the third head of the Stoa of misunderstanding Plato’s treatment of the relationship between emotions and reason. She concludes:

“The charge is implausible. What seems more likely is that Galen and his contemporaries did not stop to reflect that their own interpretation of Plato might not be the only viable able reading … [and that] Chrysippus’ approach to mental conflict avoids these difficulties [faced by Plato’s account] while preserving Plato’s central insights about the importance of moral reflection and inner harmony.”

Chapter 3 continues with a fascinating section on the so-called Posidonian objections. Posidonius was a major figure in the middle Stoa, but he is quoted approvingly by Galen against Chrysippus. It’s easy to see why. Posidonius considers a number of objections to the standard Stoic account of emotions. For instance, tears caused by an emotional response to music, where there clearly is no role played by any belief structure, since no words are involved at all. He also takes up the issue of feelings experienced by animals and young children, again situations in which belief systems seem to play no role (animals) or a very limited one (depending on how young a child is). Or consider grief, which decreases over time, even though, presumably, the belief that co-causes it in the first place, remains intact (it is still a bad thing that my grandfather died, even though it happened decades ago).

Margaret argues that it is implausible that Posidonius meant these objections as insurmountable for the Stoics — as Galen believes — or else he would simply not have been regarded as a major exponent of the Stoic school. Rather, Posidonius considered the objections in questions out of intellectual honesty, as they ought to be confronted by any system that puts cognition and volition at the center of its account of emotions:

“It will help to clarify the import of Chrysippus’ position, then, if we pause to consider how a philosopher of his commitments might reply to the objections raised by Posidonius. This is more than a thought exercise, for on several points there is evidence suggesting that the older philosopher had already advanced at least the beginnings of a position, and on others we know how later admirers of Chrysippus handled similar problems.”

The first line of defense, then, is the realization that Stoic philosophy does not need to deny the existence of feelings in the absence of judgment. In fact, the Stoics recognized that both animals and young children do have affective responses, arguing however that these resemble, but are not the same, as a human adult’s affective response (which is always mediated by cognitive judgment). Indeed, it would actually be problematic to propose too sharp a distinction between young and adult human beings, or between humans and other animals, because then one would not have any account of how character develops over time.

Another way to respond to the Posidonian objections is that phenomena like the tears caused by a particularly moving piece of music are propatheiai, not fully formed emotions (“pre-emotions,” as Graver calls them), precisely because they lack a cognitive component. In this case, they lack it because it simply cannot be developed, given that the impression is caused by a stimulus that has no cognitive content.

What about the observation that grief diminishes over time? Recall that grief, like any fully formed passion, is the result of two co-causes: the impression and our assent to it. Margaret points out that we change our assessment of the import of certain events over time, to which one could add that the strength of the impression will also naturally change with time. Since both co-causes of grief are liable to change, then it is no surprise that grief itself does too.

Moreover, the “pathetic syllogism,” which we have discussed last time, should not be taken in the spirit of a modern scientific account of emotions, but rather as describing the ethical aspects of emotional responses, which is what Stoic philosophy is concerned with in the first place. The Stoics are interested in exploring the nature of character and the boundaries of ethical responsibility, in a way similar to a notion proposed by Robert Solomon in his “On the Passivity of the Passions,” quoted by Graver:

“The truth is, we are adults. We must take responsibility for what we do and what we feel. … Arguing as I have amounts to nothing less than insisting that we think of ourselves as adults instead of children, who are indeed the passive victims of their passions.”

Margaret further comments to the effect that Stoic psychological ethics is even more realistic and nuanced than Solomon’s:

“On the same [i.e., Stoic] school’s realistic understanding of ordinary mental capacities, however, it would be truer to say we are in the process of becoming adults: our intellectual and moral characteristics are always to be compared with that normative conception of human nature which is the endpoint of personal growth and development.”

We are all prokoptontes and prokoptousai, i.e., we are — hopefully — making progress, but we are not quite there yet.

Which leads me to the final section of chapter 3, on the Stoic meaning of “freedom.” For the Stoics freedom means to be able to do what we want to do, and not to have to do what we don’t want to do — thus distancing their treatment from esoteric, and ultimately sterile, discussions about the metaphysics of “free will,” grounding their approach instead in what Wilfrid Sellars, two millennia later called the “manifest image” of the world. It follows, then, that someone is not free if his affective responses get in the way of what he wants to do, as was the case for Medea and Menelaus, but not Odysseus.

Graver reminds us that even the Stoic Sage is likely to experience strong feelings. The difference between a Sage and the rest of us is that she always experiences the right feelings:

“An awareness of having done the right thing should evoke not just a mild satisfaction but real, deep joy. The thought of abusing a child should be met with more than unwillingness: aversion should go off like an air-raid siren that arrests one’s very being.”

In a sense, then, only the Sage is truly free, but, as Margaret puts it at the very end of the chapter, “we are unfree only because we are at variance with ourselves.”

Stoicism, stoicism, and mental health

You can tell Stoicism is getting popular when anyone who disagrees with me on Twitter resorts to the “argument” (I’m using the word very loosely): “but what you said is un-Stoic!,” pretty much regardless of what I actually said, or of its logical connection to Stoicism. Anyway, another sign of popularity is the fact that mental health professionals are beginning to take an interest, wishing to empirically assess the effects of practicing Stoicism on people’s psychology.

Since Stoicism itself has from the beginning been a philosophy that included the study of psychology (under the field of “logic”), and moreover has always been explicitly open to revision, this is welcome news. What is not so welcome is when people, predictably, claim that they are studying Stoicism, while in fact they are studying stoicism (for this crucial, and not even that subtle distinction, see this article by Don Robertson).

Which is why a paper published last year in BMJ Open and entitled “Stoic beliefs and health: development and preliminary validation of the Pathak-Wieten Stoicism ideology scale” was a missed opportunity. Authored by Elizabeth Pathak, Sarah Wieten, and Christopher Wheldon, it purports to develop a scale with which to measure, as the title says, Stoic “ideology,” with the purpose of beginning to explore the mental health effects of practicing such ideology. Unfortunately, the paper hopelessly mixes Stoicism and stoicism, with a strong lean toward the latter.

The authors identify four “key domains” of Stoicism: imperviousness to strong emotions, indifference to death, taciturnity, and self-sufficiency. It is on the basis of these domains that they build their scale, which they then test on 390 subjects, most of whom are young students (aged less than 25), mostly white (though somewhat gender balanced), almost all American born.

Setting aside the usual problems with the sampling of subjects used in this sort of study, which there is very little reason to think is representative even of the American population, let alone beyond, let me begin with the four domains just mentioned. I will then move to a brief examination of additional problematic statements made by Pathak and collaborators in the paper.

I. Imperviousness to strong emotions: it is absolutely not the case that this is a Stoic value (although it certainly is a stoic one). Consider, for instance, what Seneca says to his friend Marcia in his letter of consolation to her:

“‘But,’ say you, ‘sorrow for the loss of one’s own children is natural.’ Who denies it? Provided it be reasonable? For we cannot help feeling a pang, and the stoutest-hearted of us are cast down not only at the death of those dearest to us, but even when they leave us on a journey.” (VII)

A bit earlier he writes to her:

“I am not soothing you or making light of your misfortune: if fate can be overcome by tears, let us bring tears to bear upon it: let every day be passed in mourning, every night be spent in sorrow instead of sleep.” (VI)

Does that sound to you like someone who is trying to be impervious to emotions? Or consider just how explicit on the subject Epictetus, notoriously the most stern of the ancient Stoics, really is:

“I must not be without feeling like a statue, but must maintain my natural and acquired relations, as a religious man, as son, brother, father, citizen.” (Discourses III.2)

And here is what modern scholar Margaret Graver, who wrote a whole book on Stoicism and Emotion, says: “If the psychic sensations [i.e., feelings] we experience in emotion are not simply identical with the pathē [i.e., negative emotions], then the norm of apatheia [i.e., lack of negative emotions] does not have to be cashed out as an injunction against every human feeling. One might be impassive in the Stoic sense and still remain subject to other categories of affective experience.”

So, no, the Stoics do not seek to be impervious to emotions. Rather, they work toward improving their judgments about externals, in order to re-align their emotional spectrum, de-emphasizing unhealthy emotions and nurturing and developing healthy ones.

II. Indifference to Death: one can see how people may develop a misconception here, for instance while reading what Seneca writes to Marcia:

“Reflect that the dead suffer no evils, that all those stories which make us dread the nether world are mere fables, that he who dies need fear no darkness, no prison, no blazing streams of fire, no river of Lethe, no judgment seat before which he must appear, and that death is such utter freedom that he need fear no more despots. All that is a phantasy of the poets, who have terrified us without a cause.” (XIX)

This, however, is an explanation of why we should not be afraid of what will happen after we die (because there won’t be any “us” to be concerned by things), it is hardly a council not to care about dying.

True, death — like everything that is not virtue — is categorized within the “indifferents,” either preferred or, in this case, dispreferred. But that word has a very clear technical meaning in Stoic philosophy: it doesn’t mean that one shouldn’t care about dying, but rather that death itself is irrelevant to virtue, in the sense that dying or staying alive, per se, doesn’t make you a better person.

Not convinced? Here is how Epictetus reacts to the news that a friend of his has decided to commit suicide for the hell of it:

“If your decision is justified, look, here we are at your side and ready to help you on your way; but if your decision is unreasonable, you ought to change it. ‘We ought to hold to our decisions.’ — What are you up to, man?” (Discourses II.15.6-7)

Again, does this sound like indifference (in the ordinary sense of the word) to death? I should think not.

III. Taciturnity: stoicism appeals to men, according to the authors, and goes well with the ideal (myth, really) of the solitary man who speaks by his actions. Think Clint Eastwood in The Good, the Bad and the Ugly, or in any of his other movies, for that matter. But Stoicism (the philosophy) advocates nothing of the sort. True, Epictetus famously advises his students to talk little and of important things:

“When you’re called upon to speak, then speak, but never about banalities like gladiators, horses, sports, food and drink — common-place stuff. Above all don’t gossip about people, praising, blaming or comparing them.” (Enchiridion 33.2)

Immediately below, Epictetus adds:

“In your conversation, don’t dwell at excessive length on your own deeds or adventures. Just because you enjoy recounting your exploits doesn’t mean that others derive the same pleasure from hearing about them.” (Enchiridion 33.14)

It’s obvious that this an injunction not to annoy people, and to engage others in meaningful, as opposed to idle, conversation. After all, the Stoics we know of were teachers, senators, generals, and emperors. Hardly the kind of individual who spends his life in a taciturn mood.

IV. Self-sufficiency: here too, stoicism (the attitude) seems to rely on the myth of the solitary hero who depends on no one. But Stoicism (the philosophy) is quintessentially cosmopolitan, and teaches that our primary concern should be to do good on behalf of the human polis. Sure, Marcus is often cited as saying:

“Begin the morning by saying to yourself, I shall meet with the busybody, the ungrateful, arrogant, deceitful, envious, unsocial. All these things happen to them by reason of their ignorance of what is good and evil. … I can neither be injured by any of them, for no one can fix on me what is ugly, nor can I be angry with my kinsman, nor hate him.” (Meditations II.1)

But that’s simply an honest analysis of his expectations about other people qua emperor. He also tells himself in his personal diary:

“Have I been made for this, to lie under the blankets and keep myself warm? But this is more pleasant. Do you exist then to take your pleasure, and not at all for action or exertion?” (Meditations V.1)

And what action and exertion is he referring to, other than the obviously military one (he was encamped on the Danube fighting the Marcomanni tribes when he wrote the above)?

“Do what is necessary, and whatever the reason of a social animal naturally requires, and as it requires.” (Meditations IV.24)

Seneca says something interesting about this matter:

“The wise man is self-sufficient. Nevertheless, he desires friends, neighbours, and associates, no matter how much he is sufficient unto himself.” (IX. On Philosophy and Friendship, 3)

It is clear in context that by “self-sufficient” he means capable of maintaining virtue. And at any rate he is talking about the Sage, which is as rare as the mythical phoenix. He immediately adds that even the Sage desires friends, neighbors, and associates. So, no, there is no basis in Stoicism for the idea of self-sufficiency as understood within the context of stoicism.

This brief analysis should make clear that there is a huge difference between Stoicism and stoicism. Unfortunately, Pathak et al.’s article completely mixes the two, with a far stronger dose of stoicism than Stoicism. The result is a scale that measures something, but definitely not the effects of adopting a Stoic philosophy.

Let us take a look at some quotes from the paper itself, to make sure I am not misinterpreting what the authors are doing. “Stoicism has also been invoked as a defining characteristic of masculinity and as a key explanatory factor for certain health behaviors and outcomes among men” (p. 2 of the online version). This very clearly refers to the stiff upper lip attitude, not the philosophy. Indeed, on the very same page there is mention of the Liverpool Stoicism Scale, which includes “three items that are ideological, for example, ‘one should keep a stiff upper lip.’” Precisely, but no Stoic ever advocated that, so the scale should be renamed the Liverpool stoicism Scale…

Pathak et al. “attempt to articulate an explicitly theory of stoicism [the fact that they are using the lowercase here and elsewhere is not indicative of a distinction being made between the philosophy and common parlance] and its potential impact on health. … stoicism is an ideology … we theorize that people who strongly endorse a personal ideology of stoicism may be more likely to avoid or delay seeking professional medical intervention for serious signs and symptoms of disease” (p. 3)

This can be extremely misleading, if it will lead to adoption of the scale developed in the study for social health research that focuses on stoicism but makes claims about the medical unsoundness of Stoicism.

The results are also difficult to interpret, again because of the complete confounding of stoicism and Stoicism. For instance: “men were more than two times as likely as women to fall into the top quartile of responses” (p. 5), meaning that men agreed more readily with a self-description as “stoic.” “Scores for stoic taciturnity were strongly correlated with scores for both stoic endurance and stoic serenity, but stoic endurance and stoic serenity were not highly correlated with each other. Stoic death indifference … was least correlated with the other three domains” (p. 4). Moreover: “In this study population, respondents were least likely to endorse stoic serenity and most likely to endorse stoic death indifference” (p. 5).

It is hard to know what to make of these findings, since it isn’t clear at all to what mixture of stoicism and Stoicism they refer. Please note, of course, that so far as we know likely none of the participants to the study actually had any training in, or exposure to, Stoic philosophy. Based on that, I’m inclined to say that the relevance of the study to Stoicism is close to zero, while it may tell us a lot about stoicism.

The authors draw some conclusions that seem to me to be rather unsubstantiated, or at least, again, ambiguous as to their referent: “Ironically, a personal ideology of stoicism almost guarantees failure to live up to one’s ideals. … An ideology of stoicism creates an internal resistance to external objective needs, which can lead to negative consequences. … a study of major strain among family caretakers of elderly patients with dementia found those who used stoicism as a coping strategy suffered burnout, while those who sought social support did not” (pp. 6-7).

Right, but that sounds a lot like stoicism, not Stoicism. Now one could perhaps argue that Pathak et al. never actually intended to address Stoic philosophy, only the stiff upper lip modern attitude. But that is clearly not the case from the opening line of the paper: “Stoicism is a school of philosophy which originated in ancient Greece” (p. 1).

Which makes some of their conclusions particularly troublesome, especially for people who are trying to practice the philosophy: “We hypothesise that illness behaviors may become ‘noncompliant’ or ‘irrational’ or ‘self-harming’ when specific courses of action would create an internal conflict with patients’ ideas of who they are. … This internal conflict will lead to delays in or avoidance of help seeking, with potentially life-threatening consequences. For example, empirical studies of male suicide in rural Australia have identified hegemonic masculine norms of stoicism as an important causal factor in the context of severe economic stress” (p. 8).

The situation of rural men in Australia subject to economic stress and committing suicide at high rates is horrible, but it does not seem to have anything to do with Seneca, Epictetus, and Marcus Aurelius, names that are very likely entirely unknown to the people in question. It is therefore highly misleading, dangerous, and — I’m sorry to say — irresponsible to so casually mix stoicism and Stoicism.

Of course, as Don Robertson points out in his article on the matter, the same problem applies to other philosophies. I can easily see a study of “epicureanism” showing that if people indulge frequently in large meals, a lot of drinking, and unprotected sex with multiple patterns, they will incur health risks. But to blame Epicurus — who argued for restrain in or abstention from all the just mentioned activities — would be bizarre.

So I urge Elizabeth Pathak, Sarah Wieten, and Christopher Wheldon to scrap the whole exercise, ideally retracting their paper, and writing up their findings again while making crystal clear what they mean by “stoicism.” Better yet, since there is a philosophy that has carried that name for the past 23 centuries, avoid to use the term altogether and write a paper on how to measure the potentially negative health consequences of trying to live like Clint Eastwood in a Western movie. The underlying problem — that an image of extreme self-reliance is bad for one’s health — is likely real and deserves attention. But the relevant research ought to be carried out properly.

Stoicism and Emotion, II: the “pathetic” syllogism

Stoicism is not about suppressing emotions. At least, not exactly. Last time we have seen that Margaret Graver, in her Stoicism and Emotion, makes the point that for the Stoics (as in modern cognitive science) there is a fundamental distinction between feelings and emotions. Feelings are raw materials of our subjective awareness, and they can evolve into cognitively informed emotions of different types, depending on the (implicit or explicit) judgment that accompanies them. A rush of adrenaline, for instance, may cause the feeling of fear or dread, but that feeling becomes actual fear (of a specific something) only after I have given it assent: “yes, I really should be afraid, after hearing that noise in my house in the middle of the night, because it is highly likely that someone is after me.” But I can also withhold assent, if I think the feeling is not justified: “no, there is nothing to worry about, it was, once again, the damn cat making noises.”

That is why we need to be careful whenever we talk about a major goal of Stoic training: achieving the state of apatheia, literally the lack of pathē, which are the disruptive or unhealthy emotions, not to be confused with the eupatheiai, the positive or healthy emotions. As Graver puts it:

“If the psychic sensations [i.e., the feelings] we experience in emotion are not simply identical with the pathē, then the norm of apatheia does not have to be cashed out as an injunction against every human feeling. One might be impassive in the Stoic sense and still remain subject to other categories of affective experience.”

But aren’t emotions natural? And don’t the Stoics advise us to “live according to nature”? If so, where does this talk of healthy vs unhealthy come from? It comes from the fact that even though according to the Stoics nature endowed us with feelings for good reasons, it pays to rationally scrutinize what such feelings actually are in any particular circumstance. It is Cicero — quoted by Margaret — that makes the connection clear:

“By nature, all people pursue those things which they think to be good and avoid their opposites. Therefore, as soon as a person receives an impression of some thing which he thinks is good, nature itself urges him to reach out after it.” (Tusculan Disputations, IV.12)

So the inclination to follow our feelings is natural, but how we do that depends on our judgment of whether the feeling refers to a good or a bad situation. Nature yes, but not unaided by reason…

Emotions, for the Stoics, not only come with a (either implicit or explicitly articulated) content, they are associated with a normative aspect. Graver stresses that this is a crucial characteristic of the Stoic approach: the psychological aspect of emotions must be integrated with an ethical judgment as to the appropriateness (or not) of that emotion. The ability to do that, fundamentally, is what distinguishes human beings from other animals. This position is easily traceable back to the early Stoa:

“They [the Stoics] think that the pathē are judgments, as Chrysippus says in his work On Emotions. For [he says that] fondness for money is a supposition that money is a fine thing, and similarly with drunkenness, stubbornness, and so forth.” (Diogenes Laertius, Lives and Opinions of the Eminent Philosophers, VII.3)

Cicero, in book III of the Tusculan Disputations, presents a two-step processes for analyzing emotions, informed by Stoic philosophy (recall that Cicero himself was not a Stoic, though he was clearly sympathetic to the Stoic approach):

“When our belief in the seriousness of our misfortune is combined with the further belief that it is right, and an appropriate and proper thing, to be upset by what has happened, then, and not before, there comes about that deep emotion which is distress.” (III.61)

In other words, distress is what happens when we believe that we should be distressed about whatever it is going on (or we perceive as going on). It is not the event in itself that carries the distress embedded into it, but our cognitive analysis (which, once more, to pre-empt lazy criticism, does not have to be consciously taking place at that specific moment).

Margaret calls this the “pathetic syllogism” (from pathos, nothing to do with the modern word), and she spells out the general form in this fashion:

P1: Objects of type T are evil
P2: Object O belongs to type T
P3: Object O is in prospect
C: An evil is in prospect

Take the famous example of Agamemnon, the commander in chief of the Achaean expedition against Troy, who felt fear at the idea that he was about to be defeated. The fear comes from his belief that defeat is a bad thing, plus the additional belief that one ought to have certain feelings when a bad thing is in prospect. Of course, just like in any syllogism, if one wishes to deny the conclusion (assuming that the reasoning is valid, which in this case it is), then one needs to find a premise that can be rejected. And that is a major objective of Stoic training, of course.

Another way of looking at the issue, also discussed by Graver, is in terms of externals vs. what she calls “integral” things. Both Plato and Aristotle referred to things that are determined by us (i.e., they are not externals) as good (or evils) “of the psyche.” The Stoics made this distinction — commonly referred to as the dichotomy of control — central to their philosophy, often the “integrals” as ways of handling externals, or as dispositions to use externals one way or another. Seneca, for instance, talks about an ambassadorship as an external, where the true good lies in handling it with honor. In the case of Agamemnon, the true good (or evil) lies in how the king would handle defeat, if it really did come. As it turns out, the Achaeans were not defeated, though we also know that Agamemnon was pretty bad in general at handling difficult situations, especially for a commander in chief. Graver summarizes the Stoic ethical stance in this way:

“The chief insight of Stoic axiology could very well be expressed this way: that in a rational being, external objects never merit uncompromising evaluation but integral objects always do. … The claim often appears in the form ‘virtue is the only good, vice the only evil.’”

Because the above mentioned dispositions to use externals are, of course, the virtues. The oft-neglected other side of this famous Stoic coin is that indifferents are not so in the sense that they don’t matter. Indeed, as Margaret says, they may be pursued strenuously, at times, but only on the basis of a restricted evaluation, applicable to local circumstances. The only thing that is always good, under all circumstances, is virtue, and that is why it is often referred to as the chief good, or the only (intrinsic, unqualified) good. Virtue is the good by means of which one is able to properly handle every external, including ambassadorships and defeats in war.

According to Stoic philosophy, it is possible — though very rare — for a person to align all her beliefs with each other, yielding a full and consistent evaluation of herself and her surrounding. That person would be in harmony with herself and with the cosmos at large, and of course it is referred to in Stoic lore as a Sage. Sagehood, as Seneca says in Letter XLII.1 is as rare as the mythical Ethiopian phoenix, a bird who comes back to life from its ashes (every 500 years, according to legend). Why bother with such a concept then? Graver puts it aptly:

“Alongside the dissatisfaction with our actual moral condition [for the Stoic] goes an extraordinary optimism about what we might achieve. … Becoming like the Sage would be becoming more human, not less; it would be recognizable as human maturation.”

What about the famous eupatheiai, the positive, of healthy emotions? Following Graver’s analysis, we should think of them as normative affect, i.e., as the ethically proper responses of individuals who have been practicing their virtue, and have therefore developed the right dispositions toward externals:

“A wise person who meets with an opportunity to perform some generous or courageous action might feel a kind of yearning toward that action; conversely, she may be expected to experience a horrified aversion from anything shameful or wrong.”

It is commonly assumed that the eupatheiai are somehow less intense than the pathē, leading to the stereotype of Stoics as (nearly) emotionless, or at least characterized by flat emotions. But this is nowhere to be found in the actual literature, and there is no reason, based on Stoic philosophy, to believe that to be the case. Margaret stresses that eupatheiai are “corrected” (by way of ethical training), not diminished, versions of human emotions. She makes the analogy to the seamless movements of a trained athlete: forceful but without strain.

“Preeminent among eupathic responses is the one called chara or joy. Joy is ‘well-reasoned elevation,’ corresponding on a feeling level to the happy excitement the ordinary person experiences on winning a raffle or leaving on vacation. But joy differs from those feelings in being directed at genuine goods: a generous action, for instance, would be an occasion for joy, and the proper object of the feeling would be the generosity itself, as exercised on that occasion.”

I trust you can see just how grating Stoicism can be for some modern sensibilities. I’m thinking of the sort of people who say things like “who are you to tell me I should / should not feel this way?” The answer is clear, from a Stoic perspective: emotions, as characterized above, come in bad and good varieties, and we should, indeed, work toward feeling in certain ways and avoid to feel in certain other ways. While we cannot avoid raw feelings, we are not at the mercy of our fully formed emotions, pace David Hume. (See also my recent commentary on Letter XXIII, where Seneca tells Lucilius that joy is a serious matter.)

Graver ends the chapter with a nice discussion of the Stoic classification of positive and negative emotions, first at what she calls the “genus” (i.e., broad categories) level, then at the “species” (i.e., on the basis of more detailed examples) level. The generic classification is nice and neat, while there is no consensus among the available sources about the specific classification. Graver suggests that this isn’t a reflection of disagreement among the Stoics, but rather stems from the fact that the specific examples were meant as illustrative of the generic categories, not as an exhaustive list. Figures 2, 3, 4, and 5 in this chapter are very useful summaries of both levels of classification. I am reproducing the first two tables here:

Notice the lack of a “present evil” category among the eupatheiai. The explanation is along these lines:

“We can see why one would want to claim that the person of perfect understanding has no genus of affective response for present evils. Having perfect understanding entails that one regards as evil only those things that really are evil; that is, integral evils such as personal failings, errors, and other events or situations whose causes lie within oneself. In order to believe that this sort of evil is present in the relevant sense, one would have to believe that a proposition concerning one’s own shortcomings has just become true, something like ‘I act unjustly’ or ‘I am ungenerous.’ But the person of perfect understanding is exempt by definition from everything of that kind. The situation simply never arises.”

From the third table (not shown here), consider as an illustrative example some of the “species” listed under the “genus” Desire, a pathos: anger (desire to punish someone who is thought to have armed us unjustly); hatred (anger stored up to age); rancor (anger biting its time for revenge); exasperation (anger that breaks out suddenly). And from the fourth table (not shown), here are some examples of the species listed under the genus Joy, a eupatheia: enjoyment (joy befitting the surrounding advantages); cheerfulness (joy in the sensible person’s deeds); and good spirits (joy about the self-sufficiency of the universe).

Let me conclude with one important note. Stoicism is often accused of being a self-centered philosophy, focused only on self-improvement. But as Margaret writes:

“The genus concerned with prospective goods includes some affective responses that are directly concerned with the goods of other people. … The rich affective life of the wise is being said to include some concern for other human beings that goes beyond disinterested service to the level of genuine affective involvement.”

Stoicism and Emotion, I: a science of the mind

If there is one complex, and often misunderstood, topic in Stoicism is the role played by emotions in the philosophy. You know, stiff upper lip and all that nonsense. That is why I decided to begin a multi-part series devoted to an extended commentary of Margaret Graver’s excellent book, Stoicism and Emotion.

Margaret was the keynote speaker at Stoicon 2017 in Toronto (you can read an interview with her here), and she is a serious scholar of ancient Stoicism. Her book is accessible, but not aimed at a general public, which is why I am going to do with it something similar to what I did recently with Larry Becker’s must read, A New Stoicism. As in the latter case, I have asked the author to take an advance look at my posts and, whenever possible and useful, to comment on the published version during the discussion window. Margaret has graciously agreed to it, which I’m sure will enhance the value of this series. Without further do, then, let us get started!

Stoicism and Emotion is organized in nine chapters, and from the look of it, I will have to devote a post to each, since Graver’s treatment is in-depth and requires some time to unpack. The first chapter is entitled “A science of the mind,” and it sets the stage for an understanding of Stoic psychology in general, and their treatment of emotions in particular.

The Stoics, Margaret begins, thought about emotions in what turns out to be a very modern fashion, as at least in part having propositional content. That is, they adopted toward emotions what today’s philosophers call an intentional stance: emotional reactions are certainly physiological in nature, but they also contain a judgment, say that something is threatening, or valuable. There is no contradiction between thinking this way about emotions and taking on board what modern neuroscience tells us about the underlying neurophysiology:

“The recognition of a threat [say], is analyzable on two different levels, a physiological level as investigated by the neuroscientist and an intentional level as investigated by the cognitive psychologist.”

Moreover, the Stoic approach is also very much like our own in the sense that the Stoics were materialists, so they thought of mental events in terms of physical changes effected by material substances. These two aspects are important to keep in mind throughout our discussion, because they account for why — despite getting some important details wrong, as we shall see — Stoic psychology is still very much useful today, especially in terms of its practical ethical implications. Indeed, Graver draws a direct analogy between Stoic thought on emotions and William James’ circa 1884, as well as with the more recent work by modern neuroscientists like Antonio Damasio (see, for instance, his Self Comes to Mind: Constructing the Conscious Brain).

The Stoic rejection of dualism is based on the same sort of cogent arguments accepted by most contemporary philosophers (even though, amazingly, dualism hasn’t completely died out even in 21st century philosophy). First off, the objection that famously stumped Descartes: if mental phenomena are not physical, then how on earth can we account for the causally efficacious interaction between non-physical and physical aspects of human mentation?

Moreover, the Stoics were familiar with empirical examples of mind-body interactions that, again, clearly point to a physical-to-physical connection. Consider for instance that a cut to your finger (physical) causes pain (mental), or that when you are angry (mental) your face becomes red (physical). It’s a two way street, and one does not need to invoke magical or metaphysically suspect non-physical properties to account for it.

Margaret carefully explains the Stoic theory that there is a single substance permeating the universe, the pneuma (literally, breath), a mixture of fire and air, two of the classical four primordial elements. That mixture can take different specific forms, which account for the differences between non-living things and living ones, as well as for those among plants, animals, and humans. The pneuma can take various forms because of the tension (tonos) produced by the balance of the two elements, sort of like the different types of vibrations one gets with a string musical instrument:

“It is variations in tension, and not the properties of air and fire alone, that explain differences in the qualities imparted by pneuma to things: hardness to stones, whiteness to silver, and at higher levels the sophisticated properties of plants and animals. Living things differ across the board from the nonliving in that they have much greater complexity in structure and function, and animals also differ from plants in that their more elaborate body structures and life functions require a higher level of tension to support them. The special characteristics that set humans apart have their physical explanation in yet another level. Indeed the pneuma in a human being at his or her optimal level of functioning is characterized by such a high level of tension that it is capable of maintaining its cohesion [for a time] after the body’s death.”

Of course, all of this has been superseded by modern science. But the relevant kernel of truth is nonetheless crucial: everything in the universe is made of the same stuff (we call it quarks, strings, or whatever, depending on the fundamental physical theory du jour), and yet this elemental stuff is arranged in different, and varyingly complex patterns, accounting for the variety of non-living and living matter. The implication is that the differences we observe at the macroscopic level, and that seem to be qualitative to us, are in reality the result of an underlying quantitative continuum.

What about the Stoic reference to the soul? The Greek word is psuché, and it has none of the non-physical characteristics that Christian theology attaches to the word. Psuché, for the Stoics, is material and subject to the same laws of cause and effect as anything else. It can be studied scientifically, just like everything else. And interestingly, Graver points out, does not correspond to the modern concept of mind, but rather to the entire nervous system.

What does correspond to the modern idea of mind is the hêgemonikon, the central directive faculty that combines our sensations with our judgments, and which initiates action. As I have argued in another post, the hêgemonikon is very much akin to the frontal lobes of the human brain.

Chrysippus located the hêgemonikon in the chest, and was chastised for that by Galen (Marcus Aurelius’ personal physician), who correctly thought that it was located in the brain. Once again, an example of the Stoics being wrong in the details and correct about the general picture. Lucky for us, it is the latter that matters. (Incidentally, as Margaret explains, Chrysippus’ choice was not crazy at all, but actually fit very well with Ancient Greek knowledge of human physiology.) Therefore:

“As a theoretical construct … their account of psychic function did not depend on any particular physiology. Given a more detailed knowledge of the workings of the central nervous system, a Stoic theorist should have had no difficulty in transferring to the brain the role that Chrysippus in fact gave to the heart.”

Graver then moves to a detailed explanation of the relations among thought, belief, and action in Stoic psychology, and we need to grasp at least the basics in order to make sense of their treatment of emotions. To begin with, the simplest kind of mental event is an “impression” (phantasia). This is an alteration of the psuché that tells us that something seems to be present or to be the case. Notice that animals too are capable of impressions, but not of a rational kind, since they are unable to conceptualize their phantasia.

Margaret makes the interesting point that the word “rational” (logikos) here does not have a prescriptive meaning, but rather a descriptive one: it just says that human beings are capable of complex thought, not that they get it right from the standpoint of formal logic.

The Stoics thought of impressions, again, as physical events. Zeno, for instance, used the analogy of a wax tablet that is “impressed” with something. Apparently, Cleanthes (the second head of the Stoa) took this quasi literally, so he was corrected by Chrysippus, who said that one should simply think of impressions as some (unspecified) kind of alteration in the psychic material. No need to be committed to a particular theory of human neurophysiology:

“The impression is made, i.e., caused, by some material thing, which, by impinging upon the sense organs, brings about an alteration in the material psyche, and that alteration ‘reveals itself’ together with its object through the psyche’s awareness of its own movements. But impressions may also be of that kind for which the object is more properly described as an actual or hypothetical state of affairs, i.e., a proposition.”

The impression, then, is a linguistically formulable thought. It gets translated into a more complex mental event that the Stoics referred to by a variety of terms, including “assent,” “judgment,” and “forming an opinion.” (See the book for the corresponding Greek terms. I will limit their use here to the essential ones, for ease of exposition.) This is crucial: assent is conceived of in intentional terms: by way of assent one either accepts or rejects the apparent truth of a given impression. It follows that the difference between an ordinary mind and a (Stoically) trained one is that the former has a tendency to accept impressions at face value, while the latter more wisely exercises its faculty of judgment. (As in: “That is a beautiful woman over there, I must sleep with her!” As opposed to: “That is an aesthetically pleasing human being of the female gender. Nothing else follows from such observation.”)

Margaret presents the example of the simple act of walking. If we are walking, then we have assented to the impression that, right now, it is good for us to walk (say, because we need to get to the grocery store to buy some foodstuff for dinner). The assent does not need to be conscious, but for the Stoics the fact that we are walking is either the result of a conscious judgment of the hêgemonikon, or it implies an unstated judgment of that kind, which can be articulated if need be. If someone stops you in the street and asks you why you are walking, presumably you will be able to tell him that you need to get to the grocery store and why.

What about emotions? From the beginning of the school they have been thought of in a particular way. Zeno defined them as “excessive impulses,” by which he meant a powerful kind of tendency to act. Since the cognitive mature emotions are the result of an assent, they then depend on ratifying (again, subconsciously or consciously) certain propositions about ourselves and how we think of our surroundings. Here is how the commentator Stobaeus puts it:

“Distress is a contraction of psyche which is disobedient to reason, and its cause is a fresh believing that some evil is present toward which it is appropriate to be contracted. Delight is an elevation of psyche which is disobedient to reason, and its cause is a fresh believing that some good is present toward which it is appropriate to be elevated.”

The above, it should be noted, refers to the unhealthy emotions, of which the Stoics produced a detailed taxonomy. In fact, Graver points this out immediately, mentioning that they also recognized “well reasoned” occurrences of “elevation,” “withdrawing,” and “reaching.” Moreover:

“[In] both the Zenonian and the Chrysippan definitions, there is a distinction to be made between the emotions or pathe understood as judgments (i.e., strictly for their intentional content, which may be either true or false), and the feeling one gets from a certain emotion. … Feelings which are phenomenologically similar will not necessarily represent the same kind of affective response.”

For instance, I may be sexually aroused by the sight of my partner, or by the sight of a stranger. The raw feeling is similar, but if I act on it (following my judgment that it is desirable for me to do so), the first case has a very different import from the second. There are crucial ethical implications of assenting, or withdrawing assent, from the very same emotions.

The philosophy and science of (Stoic) free will

Time to tackle again the debate that never goes away: how is the Stoic idea that we can work to improve our character, or — which is the same — Epictetus’ contention that some things are up to us and other things are not up to us, compatible with the (also Stoic) contention that we live in a universe determined by a universal web of cause and effect? In other words, how is Stoic “free will” going to square with their metaphysical position of physicalism? And, moreover, hasn’t modern neuroscience “demonstrated” that there is no such thing as free will?

Let me summarize the answers upfront, and then we’ll dig into the details, with the help both of scholarly work on Stoic philosophy and of scientific research on the human mind. In a nutshell: “free will” is an awfully confusing term that should never, ever, be used again. (Yes, I know, it’s in the title of this post. That’s because I wanted to grab your attention.) Instead, we should use the word “volition,” which — as we shall see in a moment — is both the best translation of what Epictetus means by the Greek term prohairesis as well as the appropriate technical term used by modern cognitive scientists. Moreover, the Stoics had a sophisticated causal model of volition (it may be good to re-read this post on Chrysippus’ metaphor of the rolling cylinder, before proceeding further), which turns out to be in very good agreement with the findings of modern science (yes, including the (in)famous Libet experiments popularly, but mistakenly, assumed to “demonstrated scientifically” that there is no such thing as free will; more on that below). Okay, let’s get started.

To begin with, the Stoics used two terms within the context of this discussion: hêgemonikon and prohairesis. The two are often treated as interchangeable, but Anthony Long, in his excellent Epictetus: A Stoic and Socratic Guide to Life (see my two commentaries here and here) draws an interesting and useful distinction between them.

Hêgemonikon is the most common Stoic term, and broadly speaking it is a faculty of the mind that has four “powers,” according to this helpful article in the Internet Encyclopedia of Philosophy: “presentation [phantasia], impulse [hormê], assent [sugkatathesis], and reason [logos].” The ancient Stoics thought that the hêgemonikon resided in the heart, a notion for which they were ridiculed by Galen, Marcus Aurelius’ personal doctor. Galen was right, of course, and I suggest that modern neuroanatomy locates the hêgemonikon in the frontal lobes.

The frontal lobes are areas of the brain that are particularly developed in both humans and other great apes (but, interestingly, not so in lesser apes and monkeys). The frontal lobes (one per hemisphere) are the largest of the four lobes of the mammalian brain, and experimental research has associated them with the following functions: reward, attention, short-term memory tasks, planning, and motivation. They also allow us to project the future consequences of our intended actions, to choose between what seem to us as good or bad actions, to override and suppress socially unacceptable responses, and do assess similarities and differences between things and events. That sounds to me very much like what the Stoics were talking about whenever they used the term hêgemonikon, usually translated as ruling faculty. Here is Marcus, for instance:

“Things stand outside of us, themselves by themselves, neither knowing anything of themselves nor expressing any judgment. What is it, then, that passes judgment on them? The ruling faculty.” (Meditations, IX.15)

Modern science tells us that the frontal lobes do not fully mature in human beings until our late ‘20s, and of course a variety of situations can impair their proper function, from accidents such as the famous one that occurred in 1848 to the American railroad construction foreman Phineas Gage to strokes, Alzheimer’s disease, and various forms of dementia.

Interestingly, according to Long:

“Hêmonikon does not mean rationality; it is a term that applies to the souls of animals who lack rationality as well as to human beings“ (p. 211)

Which is in broad agreement with the above mentioned scientific fact that all mammals have this structure, albeit in a less developed form than in the great apes. The Stoics were also clearly aware that there are conditions in which one’s hêmonikon is impaired, by age or disease, or not yet fully formed, in children.

Long insists, and it seems to me that he is correct, that there is a good reason why Epictetus — uniquely among the Stoics — switches to prohairesis, instead of hêmonikon. Prohairesis refers to a faculty that is unique to humans, and in a sense is a specific sub-component of the hêgemonikon:

“We should take prohairesis to refer to the human mind in just those capacities or dispositions that Epictetus constantly maintains to be completely ‘up to us’ and free from external constraint.” (p. 211)

What’s the difference? One of the characteristics of the hêgemonikon, as we have seen, is that it is presented with phantasia, or impressions, both external ones (from our sensory apparatus) and internal ones (from memory). Such presentation, according to Epictetus, is automatic, i.e., it is not “up to us.” The remaining three powers of the hêgemonikon, however, impulses, assent, and reason are up to us. So we can think of prohairesis as the ability to reason and to give or withdraw assent to our impressions — thereby controlling our impulses — that is part of the hêgemonikon. I will leave it to neuroscientists to work out exactly which structures within the frontal lobes are more concerned with reasoning and decision making, and which are devoted to sensorial inputs and memory. The answers are empirical in nature, but do not affect the philosophy at all.

[Incidentally, as Long points out, Epictetus is not alone here: Aristotle also adopted the term prohairesis to mean “the deliberated desire of things that are up to us.” Moreover, Aristotle too judges what he calls “prohairetic disposition” to be a better assessment of moral character than action, presumably because actions are not entirely up to us, only judgments are. (p. 212)]

We are now in a better position to make sense of one of the most famous passages in the Discourses:

“For a start, don’t be carried away by [the impression’s] vividness, but say: Wait for me a bit, impression; let me take a look at you and what you are about, let me test you. Next, don’t let it lead you on by painting a picture of what comes next. Otherwise, it is off and away, taking you wherever it wishes. Instead, confront it with another impression, a fine and noble impression, and dismiss this foul one.” (II.18)

It is the hêgemonikon that both presents us with the impression (say, of an attractive member of the other sex) and with memories to draw from (of pleasurable previous sexual encounters). But it is our prohairesis that applies reason to the situation and denies assent to the impression (I am in a committed relationship, and I will therefore not pursue sex with another person on the side).

That also makes sense of why the Stoics, and Epictetus in particular, insisted that the only things that are truly good or bad are our judgments: the impression itself is neutral, because it is a fact about the world, and moreover is not under our control. It is us, through the faculty of prohairesis, who attach a value judgment to things and events. That attaching of a judgment is under our control, and so the only truly good thing is to have a well functioning prohairesis, and the only truly bad thing is to have a malfunctioning one (which leads to the condition often describe as amathia, or un-wisdom).

Long provides his readers with a very good discussion of why he ends up translating prohairesis with “volition,” and it is well worth attending to his reasoning. He considers three other options: moral purpose (or moral character), will, and agency. He discards the first possibility, because prohairesis can be good or bad, while “moral” has a positive meaning only; moreover, the word is tainted by later connotations that would distract and lead us to misunderstand the Stoic meaning.

What about will? He has actually translated prohairesis that way, occasionally, particularly because Epictetus does use phrases like “if you will, you are free.” But, again, the term has become loaded with especially Christian-influenced meaning, and — more importantly — it may lead us to erroneously conclude that Epictetus is thinking of a faculty of the will distinct from assent and impulse, which he is definitely not.

Agency, says Long, is a better alternative, and indeed it is the one used by modern Stoic Lawrence Becker throughout his A New Stoicism, about which I have recently published a ten-part commentary. But Long isn’t happy with agency either. Consider this passage:

“And who told you; it is your function to walk unimpeded? What I was telling you is that the only unimpeded thing is the impulse. Wherever there is a need for the body and the body’s cooperation, you have long ago heard that nothing is your own.” (Discourses IV.1.72-73)

What Epictetus is reminding his student of, here, is that even our bodily actions are not entirely up to us: we may decide to start walking, but we could be paralyzed by disease, or chained to a wall by a tyrant. So the action is not (entirely) up to us, only the impulse to perform the action. Since “agency” typically carries a meaning of actually doing things, not just willing them, then I would have to agree with Long’s take.

Hence the final preference for volition, which in modern psychology is “the cognitive process by which an individual decides on and commits to a particular course of action. It is defined as purposive striving and is one of the primary human psychological functions.”

As Long correctly points out, volition is not contradictory to a physicalist view of the world as determined by cause and effect:

“We take people to have volitions irrespective of whether these are predetermined or independent of antecedent causation.” (p. 220)

And, moreover, modern psychological science considers volition to be “a process of conscious action control which becomes automatized” (see link just above). But wait a minute! Isn’t it a fact of modern science that free will, however one wishes to call it, is an “illusion”? Specifically, didn’t the famous experiments by Benjamin Libet back in the 1980s conclusively show that to be the case? Does that not mean, therefore, that the entire Stoic philosophy of mind, and hence the crucial idea of the dichotomy of control as expressed by Epictetus at the very beginning of the Enchiridion, crumble under the pounding of modern science? Nope, not at all. On the contrary, Libet’s experiments, and subsequent others carried out since, spectacularly confirm the ancient Stoic intuition about prohairesis.

Libet performed some fascinating experiments on conscious vs unconscious decision making, beginning back in 1983. Briefly, he asked subjects to follow the movements of a dot on the screen of an oscilloscope. The dot moved like the hands of a clock, but faster. Libet told his subjects to move a finger at a moment of their choice during the experiment, noting the position of the dot when they became aware of their decision to act. The experiment showed that the decision to move the finger entered conscious awareness about 200 milliseconds before the actual movement. But, stunningly, there was a rise in the so-called “readiness potential,” which is thought to be associated with the preparation for action, about 550 milliseconds before movement. So the subjects appeared to get ready to move the finger a full 350 milliseconds before they became conscious of their decision to do so. (Indeed, in later experiments, the readiness potential has been shown to build up even as long as 1.5 seconds before movement.)

Taken at face value, Libet’s results seem to show that we decide our actions unconsciously, and that what we call consciousness is simply a (late) awareness of a decision that has been made. There are several well known criticisms of such conclusion, beginning with the obvious one, that the experimental conditions have precious little to do with the recursive, complex behavior that we normally label “conscious decision making,” and which is understood as a continuous feedback loop between what Daniel Kahneman calls System I (fast, subconscious) and System II (slow, deliberate) brain processing systems.

But in fact it was Libet himself who rejected the facile “free will is an illusion” interpretation of his own research. Here is part of his commentary:

“The finding that the volitional process is initiated unconsciously leads to the question: is there then any role for conscious will in the performance of a voluntary act? The conscious will does appear 150 msec before the motor act, even though it follows the onset of the cerebral action by at least 400 msec. That allows it, potentially, to affect or control the final outcome of the volitional process. An interval msec before a muscle is activated is the time for the primary motor cortex to activate the spinal motor nerve cells, and through them, the muscles. During this final 50 msec, the act goes to completion with no possibility of its being stopped by the rest of the cerebral cortex. The conscious will could decide to allow the volitional process to go to completion, resulting in the motor act itself. Or, the conscious will could block or ‘veto’ the process, so that no motor act occurs.” (B. Libet, Mind Time: The Temporal Factor in Consciousness, 2004, p. 137)

Note that the motor cortex in question is part, surprise surprise!, of the frontal lobes (specifically, part of the posterior border of the lobe, called the precentral gyrus), which I have suggested is the anatomical counterpart of both the hêgemonikon and enables the faculty of prohairesis.

Also, more recent research (summarized here), has led to a re-interpretation of Libet’s original findings that aligns them even more with the intuitions of the Stoics. For instance, a group of researchers in Germany has modified the original protocol to test Libet’s idea of a veto power exercised by conscious thought. Bear with me for a minute, because some of the details are important.

Subjects were asked to hit a foot pedal as quickly as possible after seeing a green light on a screen, but also to stop themselves from doing so (i.e., cancel their own movements) whenever a red light appeared. Researchers then put the red light under the control of a computer monitoring the participants’ brain waves. The twist was that whenever the computer detected the above mentioned readiness potential building up it would make a red light appear. In agreement with Libet’s veto hypothesis, participants were, in fact, able to stop themselves from pushing the pedal, reversing the build up of the action potential. This was possible up until a point of no return: if the red light was too close after the green one (about 0.25 seconds) then the foot movement could not be completely inhibited.

And there is more. A French team of neuroscientists published a paper in 2012 in the prestigious Proceedings of the National Academy of Sciences in which they argued for a different interpretation of Libet’s original experiments. They suggest that the readiness potential does not, in fact, signal the brain’s preparation for a specific action. Rather, the potential goes up and down randomly, but movement can only occur when a certain threshold in the potential is reached. Sure enough, they conducted an experiment in which they asked subjects to press a button, either at moments of their own choosing, or when they heard a random click. The results show that the response to the random clicks were much faster when they happened to coincide with a (again, random) surge in the readiness potential then when the potential happened to be low. So the potential is not really a sign of an already made unconscious decision, but rather one of a number of co-occurring causes that facilitate the movement.

All of the above seems to me eminently compatible with the Stoic take on volition. Please understand that I am not suggesting that the ancient Stoics somehow anticipated modern neuroscience. That would be preposterous. They knew nothing about action potentials, and they even got spectacularly wrong the anatomical location of the hêgemonikon. But their intuitive understanding of human psychology — on which they built their moral philosophy of action — was right on target, which makes their philosophy perfectly compatible with modern cognitive science. (For another example of such compatibility, this one concerning the Stoic treatment of emotions as cognitively informed, see here.) There is no magic here, the Stoics were simply astute observes of human nature.

To summarize then: the hêgemonikon, our ruling faculty, is roughly equivalent to the functions performed by the mammalian frontal lobes, which are particularly developed in the great apes, and that in the human species reach full maturity in our late ‘20s. Prohairesis is our special faculty of deliberate, rational judgment, and is made possible by sub-components of the frontal lobes. Our volition is compatible with the fact that the cosmos is characterized by a universal web of cause and effect, because some of these causes are internally generated (see Chrysippus’ cylinder), which makes Stoic philosophy of mind a type of compatibilism about “free will” (and please, let’s no longer use that term!). Finally, not only there is no contradiction between modern cognitive science and the Stoic idea that some things (namely, our judgments) are “up to us.” It is, on the contrary, the case that modern science tells us about the anatomical bases and physiological mechanisms underlying prohairesis. It is this congruence between early Stoic intuitions about human psychology and nature and modern science that make their philosophy still so useful today. We can, therefore, agree with Epictetus when he says:

“You are not flesh or hair but volition; if you keep that beautiful, then you will be beautiful.” (Discourses III.1.40)