Anger: The Alternative Aristotle

Did Aristotle say we should moderate our anger or replace it with empathy?

Everyone agrees that excessive anger can do harm. However, at least in the west, the majority of people today assume that mild or moderate anger can potentially be helpful. Especially if it’s channelled toward what they consider to be constructive ends, such as protesting against perceived injustice. In academic literature, the view that a certain amount of anger can be justified is often traced back to Aristotle, who, it is generally believed, taught that anger can be healthy in moderation, in accord with his famous ethical doctrine of the Golden Mean. In contrast to this view, the Stoics, in the west, and the Buddhists, in the east, traditionally adopt a harder line that rejects all anger as unhealthy.

Indeed, in On Anger, Seneca repeatedly opposes the Stoic theory of anger to what he considers to be the standard Aristotelian position. He argues that whereas Aristotelians believe that anger can sometimes be appropriate, the Stoics reject all anger as intrinsically unhealthy.

Aristotle says: “Anger is necessary, nor can any struggle be carried to victory without it: it must fill the mind and kindle the spirit, but it must be employed as a foot soldier, not the general.” — Seneca, On Anger, 1.9.2

Aristotle does not actually say this, however, or anything like it, in any of his surviving works. One possible interpretation of this is that Seneca may be quoting later authors in the Peripatetic tradition and retroactively attributing their view to Aristotle, the founder of their school. Nevertheless, Seneca insists that what he considers to be the conventional Aristotelian view stands in stark contrast to that of the Stoics.

Aristotle stands as a defender of anger and forbids us to excise it: he says it’s a spur to virtue, and removing it will leave the mind defenseless and too sluggish and supine to undertake great actions. — Seneca, On Anger, 3.6.3

The doctrine being attributed to Aristotle by Seneca came to be known as metriopatheia (the moderation of certain passions) as opposed to Stoic apatheia (the elimination of certain passions), where anger would be a key example of the sort of passion concerned. As we shall see, Aristotle’s surviving remarks on anger are fragmentary, appear inconsistent, and therefore lend themselves to different interpretations.

What Aristotle Seems to Believe

The key passage in which Aristotle describes the relationship of the wise and virtuous person to anger reads as follows:

Concerning anger [περὶ ὀργῆς], the mean is gentleness [πραότης]. [...] The excess is irascibility [or quickness to anger, ὀργιλότης], the deficiency an incapacity for anger [ἀοργησία]. — Ethica Nicomachea, 1125b26–29

First, we should note that Aristotle is not simply referring here to “anger” as an emotional state, which modern psychologists call “state anger”, but rather to our disposition or proneness to anger, which modern psychologists call “trait anger” — he’s talking, in other words, about our character. Second, we should note that ancient thinkers (with the exception perhaps of Plato) viewed our emotions, or “passions”, such as anger, not as feelings occuring in a distinct part or region of the mind but as psychophysiological states encompassing the whole psyche.

Aristotle here defines the virtuous ideal, or “mean”, as lying somewhere between excessive “irascibility”, by which he means proneness to anger, and its opposite, “inirascibility”, the complete inability to experience anger. It’s important to understand that the “mean” (mesotēs) in Aristotle really refers to a disposition somewhere in the middle ground between two extremes. It’s not necessarily dead in the centre but at a point, relative to our circumstances, as determined by practical wisdom (phronimos). To put it another way, it’s what a wise man would choose to do and therefore avoids certain extremes. Hence, Aristotle’s “mean” may be asymmetrically related to the extremes, and closer to one end than to the other. Moreover, he does not even consistently hold that the “mean” must differ only in quantity or degree from the extremes — it is often a qualitatively different attitude or emotion.

For instance, the alternative to too much or too little anger-proneness is described in the passage above as praotes, usually translated as “gentleness”, although, as we will see this word is somewhat ambiguous and difficult to translate into English. This term was used in Greek to describe domesticated animals, as well as humans who are virtuous and even tempered. We might say, by analogy, that the praotes individual has “tamed”, “civilized”, or “domesticated” his passions, or perhaps that he is a civil human being. The established translation, however, is “gentle”. He is gentle rather than excessive and harsh. A “gentleman” as we might once have said, in English, is not prone to bouts of excessive rage, nor is he completely insensitive to wrongdoing.

Another key passage that is usually cited when attributing the doctrine of metriopatheia to Aristotle appears to clarify what he means by praotes:

The gentle person [πρᾶος] is angry at the right things, toward the right people, in the right manner, at the right time, and for the right length of time—as reason prescribes. — Ethica Nicomachea, 1125b31–35

In this passage, Aristotle clearly states that the gentle person does become angry but in accord with reason, which is sometimes taken to provide conclusive support for the claim that he endorses metriopatheia or the moderation of anger. Anger should not be eliminated but rather controlled and moderated by reason, he seems to be saying here.

That should decisively settle the debate. However, there are two problems with this interpretation. First, Aristotle appears, as we shall see, to contradict this passage elsewhere. Of course, it’s quite possible that he changed his mind or unintentionally contradicted himself. There’s a second problem with this passage, though. Aristotle famously employed a philosophical method he called endoxia, which consists in clearly describing conventional wisdom first before analyzing a concept philosophically. He often starts by describing what the majority of people believe. That’s exactly how the passage above is introduced in his writings: not as his own considered opinion but as the position adopted by most non-philosophers. Arguably, therefore, it’s meant to be descriptive not normative.

Today we would refer to this as the assumption about anger made in our common “folk-psychology”. It would therefore be circular to argue that Aristotle’s writings support the view held by the majority of people because, in fact, in this passage he may just have intended to describe what the majority of people already believe: that anger is appropriate in moderation. Moreover, even this popular definition leaves open the question as to precisely where and when anger is appropriate — and the answer, upon reflection, could be that anger is “seldom or never” appropriate.

Photo by Howard Senton on Unsplash

The Problems for Metriopatheia

These two quotes, as we have seen, may appear decisive at first glance but on closer inspection leave some questions unanswered. Aristotle does not necessarily believe that praotes entails having moderate anger because it could be a qualitatively different emotional disposition — it may have some things in common with anger without actually being a form of anger. In addition, Aristotle’s description of praotes as entailing the moderation of anger in accord with reason is presented as a description of popular opinion, a starting point for reflection, rather than his own philosophical conclusion.

First of all, elsewhere in discussing his doctrine of the mean, Aristotle states that praotes, gentleness or civility, has more in common with the total absence of anger-proneness than to its excess.

The excess [irascibility] is more opposed to gentleness than the deficiency [...] gentleness is closer to the deficiency. — Ethica Nicomachea, 1126a29–b4

Even if were to accept the doctrine of metriopatheia or the moderation of anger, Aristotle appears to warn us that his ideal is more akin to the elimination of anger than to some sort of centre ground.

Moreover, elsewhere in Aristotle’s writings we find several passages that appear to directly conflict with this popular interpretation of his views on anger. Most obviously, in the Rhetoric, Aristotle famously defines anger as a desire for revenge.

Let anger be defined as an impulse, accompanied by pain, to a conspicuous revenge for a conspicuous slight directed without justification towards what concerns oneself or towards what concerns one's friends. — Rhetoric, 1378a30–33

There was more or less a consensus among ancient philosophers of different schools that anger should be defined broadly in terms of the desire for revenge. Even Seneca, who opposes the Stoic view of anger to the Aristotelian view, acknowledges that they share a very similar, if not identical, definition of anger.

Aristotle’s definition is not very different from ours: he says that anger is the strong desire to return pain for pain. (The difference between his definition and ours cannot be explained briefly.) — Seneca, On Anger, 1.3.3

Anger is essentially a desire for payback, therefore, or revenge. However, elsewhere in the text we were discussing earlier, Aristotle states that:

The gentle person is not vengeful [τιμωρητικός], but inclined to sympathetic understanding [συγγνωμονικός]. — Ethica Nicomachea, 1126a1–4

That creates a problem for the popular interpretation of Aristotle on anger. If praotes is not vengeful, and anger is essentially a desire for revenge, it appears to follow logically that the gentle person can never be angry.

Moreover, in this passage, Aristotle clearly states that, rather than being vengeful or anger-prone, the gentle person is, by disposition, sympathetic toward others and inclined to pardon their offences. In modern psychology, we typically use the word “empathy” (rather than “sympathy”) to describe what Aristotle appears to mean here, and it would indeed typically be contrasted with the psychology of anger, which is by nature remarkably unempathic.

Conclusion

These passages are problematic for the standard interpretation of Aristotle on anger, which maintains that he considered anger that is moderated by reason to be healthy. Instead, he appears committed to the view that the healthy alternative to anger, praotes or civility, completely replaces the desire for revenge with an empathic disposition, which would exclude it from his own philosophical definition of anger. He may be saying that the wise man, rather than moderating his anger, replaces it with empathic understanding, a fundamentally different attitude.

When read alongside the passages cited earlier, this could be taken to imply that the wise and virtuous alternative to both anger excessive anger and a total incapacity for anger is not a “moderate” or “healthy” version of anger but rather something qualitatively different. Contrary to Seneca, Aristotle doesn’t refer anywhere in his surviving writings to the wise and virtuous man as being moderately angry, or even to him experiencing anger at all. Instead, he makes a point of using the word praotes, meaning civility or gentleness. This appears, perhaps, to be qualitatively different from any form of anger insofar as it is empathic and non-vengeful.

It may seem puzzling why Aristotle would, then, describe it as avoiding the extreme of being totally incapable of anger — although he does say that it has more in common with this than with excessive anger-proneness. Today most people tend, by default, to think of anger primarily in terms of its intensity, something that has been called the “hydraulic” model of emotion. Research in neuropsychology universally rejects this simplistic view. Anger, and other emotions, are complex psychological states composed of different ingredients, such as physiological reactions, certain thoughts and underlying beliefs (cognitions), desires and action tendencies, as well as other feelings. Aristotle, likewise, thought of anger as a complex state of the whole psyche. It would, arguably, make no sense therefore for him to define the “mean” primarily in terms of the raw quantity of anger experienced.

Perhaps what he meant by praotes, civility or gentleness, was a state of the whole psyche that is qualitatively closer to the total absence of anger than to excessive anger. If that includes freedom from excessive anger but also freedom from the extreme of total non-anger, then it may imply that the desire for revenge has been replaced with some other desire. He doesn’t explicitly state what that would be, as far as I’m aware, but based on his claim that gentleness and civility (praotes) entails sympathetic understanding, and an inclination to pardon, he may mean that anger is replaced with a desire primarily for justice and reconciliation, rather than revenge or retribution.