Puzzles of Akrasia
Making sense of how we do what we know is wrong
Introduction
This paper is about the phenomena of akrasia, commonly translated as “incontinence” or “weakness of the will,” understood broadly to refer to all ethical cases where someone deliberately chooses to act in a way considered worse in their own estimation, such as when someone breaks their diet to eat a slice of cake as a paradigmatic example. I will begin my examination by establishing certain premises drawn from Socrates and Aristotle that there is an inherent connection between goodness and desirability, and that all actions aim toward some apparent good. Taking this for granted, we arrive at something close to the Socratic doctrine that no one errs knowingly.
Cases of akrasia appear to contradict this doctrine since people are actively choosing what they know to be worse. I show how Aristotle accounted for at least some cases of akrasia, appealing to ways people with knowledge can also be practically ignorant, and then highlighting how contemporary philosophers like Terrance McConnell, Devin Henry, and Byron Stoyles have argued that there are cases which do not fall neatly into the Aristotelian model, focusing on ‘genuine’ akrasia where knowledge is present at the moment of action. Finally, I evaluate their own attempts at reconciling Aristotelian thought with their counter-examples, representing two general methods we could take of either finding additional ways the genuinely incontinent can be practically ignorant or by finding additional elements necessary for virtuous action which someone may lack even when knowledge is present.
The Problem of Akrasia
Knowledge of the Good as the Desirable
We may begin our investigation, appropriately, by examining the beginning of Aristotle’s Nicomachean Ethics where he defined the good as “what everything seeks” (1094a 1-3). For Aristotle, it was obvious that every action or decision must aim at some end, taken by the actor to be good. By this he did not mean that everything people seek is good in fact, but that whenever we seek something, we do so precisely because it has the appearance of a good. Not everything is choiceworthy, but everything people choose is a result of them holding it to be choiceworthy.
This establishes a strong association being made between ‘goodness’ and ‘desirability.’ To recognize something as good is to see it as desirable, and to desire something means you must recognize it in some way as good. This is just as true for the wicked as it is for the righteous, both seeking what is good in their own eyes. The virtuous pursue what is choiceworthy in fact, while the vicious pursue only things which are only good in appearance, or they pursue a lesser good at the expense of a greater one. Any failure in moral virtue is therefore also a failure in intellectual virtue.
This reasoning led Socrates to conclude that being virtuous was strictly a matter of knowledge, while all vices are just different forms of ignorance. The vicious lack some or all of the relevant knowledge needed for virtuous action. We might formulate a general rule about action: If anyone chooses option X over option Y, it is because they hold X to be more choiceworthy than Y, as more desirable, and therefore hold it to be the greater good. Whether someone is right about something being the greater good is a completely separate question. Given this, the Socratic conclusion about the real nature of virtue and vice seems to inevitably follow.
This can be seen in the virtue of temperance and the vice of intemperance, regarding the right dispositions toward pleasures and pains. Generally speaking, pleasure is a good, although not the chief good, and is only good to pursue when accompanied by other goods, or at least not accompanied by some greater evil. Virtuous behavior therefore requires moderation towards pleasures, pursuing them in the right ways and at the right times. A slice of cake may be delicious, but may be detrimental to our health when enjoyed in excess. A temperate woman will keep her general health in view, and therefore decline to eat cake except in the right times and occasions, such as a celebration, and then only in an appropriate amount. By contrast, an intemperate man will have incorrectly determined pleasure is more important than their health, or perhaps do not seriously understand how eating the cake will be detrimental to their health, sacrificing what is in reality the greater good for the sake of a lesser one. For the sake of simplicity, I will be returning to this example throughout my essay, and assume that any cake-eating is being done in an inappropriate occasion or amount.
There is nothing mysterious about this dynamic. Both the temperate and the intemperate do what is best in their own eyes, but only one side sees clearly and therefore acts rightly. The intemperate man failed to act rightly because of their incorrect belief that pleasure was the greater good. This is why Aristotle wrote that intemperate man acts without regret. They get the excess of pleasure that they were seeking, and they sought it precisely because it was excessive (1150a 19-23).
The Socratic Doctrine vs Akrasia
Where we do encounter a greater mystery, and therefore a greater need for philosophic insight, is in cases of akrasia or “incontinence,” instances where people fail to properly exercise self-control and give in to temptation. This is puzzling because, like the virtuous, the incontinent seem to really know that option X is better than option Y, but in spite of this fact they, like the vicious, choose Y over X.
Continuing on with our previous example, an incontinent man knows that eating cake is bad for them and would go against their diet, but would still pick up a slice regardless. The incontinent are distinct from both the virtuous and the vicious, acting in ways they know they should not and seeing it as a flaw in themselves. As Aristotle put it, “the vicious person does not recognize that he is vicious, whereas the incontinent person recognizes that he is incontinent” (1150b 35). Unlike the intemperate, the incontinent do have regret, knowing they have acted poorly.
How do we make sense of this, given our general rule of action aiming toward the apparent good? If someone really knows that X is better than Y, then why would they pick Y over X? Does not the very act of them choosing Y over X prove that Y appears to them as more choiceworthy? The very existence of akrasia seems to contradict this rule.
Cases of akrasia are also hardly rare, and its possibility is so clearly taken for granted that it is implicit in our common way of discussing ethics, such as when we chide someone by saying “you know better than that.” We must either deny the existence of akrasia, as Socrates held, arguing that any sufficient investigation of akrasia will reveal it to ultimately be caused by ignorance, or we must deny our general rule of action.
Aristotle’s Answer for Akrasia
Attended vs Unattended Knowledge
Aristotle’s investigation of this question is undoubtedly the most well known, especially in Book VII of his Nicomachean Ethics, which is why we are using the Greek term akrasia to describe it in the first place. To properly understand modern discourse about akrasia, we should begin with Aristotle’s own solution to this puzzle and then call out its strengths and weaknesses.
In Aristotle’s account, Socrates’ main objection to akrasia stemmed from how inappropriate it would be to say someone who does wrong has knowledge that is wrong. As he put it, “it would be terrible, Socrates used to think, for knowledge to be in someone, but mastered by something else, and dragged around like a slave” (1145b 22-25). Ethics of slavery aside, this illustrates the issue clearly since someone being “dragged around like a slave” is being overpowered and made to do something against their will. Knowledge, in contrast to a merely true belief or supposition, implies a kind of strength with the appropriate kind of grounding.
A large part of the puzzle of akrasia is that, despite this strength and certainty, we are acting against it, as if we were being forced. If someone merely suspected that something might be the greater good, it is less surprising that this suspicion might be set aside for the sake of a more immediately present lesser good.
Aristotle objects that Socrates’s view of akrasia plainly contradicts our everyday experience. A theory is worthless if it cannot account for the facts, no matter how nice it seems. Yet Aristotle also seems to share the same fundamental premise that led Socrates to this conclusion, agreeing that every action seeks some good. Even as Aristotle wants to distinguish himself from Socrates, he only does so in a qualified way, finding senses where someone might have and not have knowledge simultaneously. Specifically he wants to argue that the incontinent have “the knowledge that his action is wrong, without attending to his knowledge” (1146b 33-35).
Wrong action only seems “extraordinary” if the agent both has the knowledge and attends to it in the relevant moment. If the agent is not attending to this knowledge, then the agent not acting in line with their knowledge is not extraordinary (1146b 35-37). If I ‘know’ I’m supposed to be at a meeting at 4 PM, but this knowledge slips my mind until I am suddenly receiving text messages asking where I am, then I will be embarrassed, but I won’t find it philosophically mysterious that I acted against my knowledge. This is perhaps weakening our meaning of knowledge compared to the strength Socrates gave the term, but it immediately highlights ways we can explain our everyday experiences of akrasia.
The Practical Syllogism
This may also happen in more complex ways. To properly reason about our actions, Aristotle proposes that we use a kind of practical syllogism, using at least two premises to arrive at a conclusion about how we should act (1147a 1-24). We have a universal or major premise which establishes a general rule of action and a particular or minor premise about the particular kind of situation we are in. Finally, from these two premises, we arrive at a conclusion about what we must do. For example, if someone held the major premise “everything sweet should be avoided” and the minor premise “this cake before me is sweet,” then they may conclude “I should avoid this cake.”
This theory complicates the way our knowledge relates to our actions, and consequently introduces more points of failure where this process can be interrupted, especially by our feelings and appetites. Someone might have the major premise of the syllogism in mind, but not attend to the minor premise that would apply it to their own situation. Likewise, we might have particular knowledge about something which corresponds to an irrational appetite we have which, if the good universal premise is not active, leads us to act against reason.
While it seems like this process can be interrupted in a number of ways, such as simply forgetting it, Aristotle is especially concerned with ways it is being interfered with by deeper problems of our character, especially by our passions. Aristotle compares the intemperate man to someone who is sleeping, drunk, or mad. They cannot be said to completely lack the knowledge they would have while sober, but because of their particular physiological conditions they are unable to use it at a particular moment. Our passion or appetites can have similar effects. If someone is too focused on their hunger, lust, greed, anger, and so on, this can cloud their judgment and prevent someone from being able to complete the appropriate practical syllogisms.
To head off a confusion, we can also note that akrasia is frequently presented as the incontinent being overpowered by particularly strong feelings, but this is not strictly necessary. All this theory requires is that, whatever level of passion someone experiences, it is enough to interfere with their reasoning. Someone doesn’t have to be starving for their feelings of hunger to influence their thinking. They simply need to see the temptation and take it without much more thought. Hence the reason Aristotle recognizes that the incontinent can be overcome “even by [pleasure and pains] which most people overcome” (1150a 10-15).
From this description, it actually seems that Aristotle is simply presenting a more nuanced and elaborated version of Socrates’ theory of akrasia, or perhaps is only defining his terms in a slightly different way. Aristotle found ways someone might have knowledge of what is better while still acting against that knowledge, but only by finding senses in which we could also simultaneously deny that someone has knowledge. He is finding ways we might “both have knowledge in a way and do not have it” (1147a 14-15).
Problems with Aristotle’s Account
McConnell on Aristotle’s Inconsistencies
The issue with this theory is that it appears to be rather limited. If someone forgets they are on a diet, then we can easily explain how they might break their diet by showing how they are not attending to their knowledge. But we run into harder cases where people do seem to be attending to their knowledge even as they act against it. Contemporary discourse around akrasia frequently focuses on these sorts of cases as a way to expand on Aristotle, showing areas where his theory is incomplete. Someone might say “this will ruin my diet” even as they pick up a slice of cake. Aristotle himself seems to recognize that these sorts of cases do occur, even as they are not readily explained by his analysis of the practical syllogism. Terrance McConnell has highlighted several places where Aristotle appears to be inconsistent with his own theory of qualified ignorance.
Firstly, Aristotle claims that the incontinent know that they are incontinent. As Aristotle claimed, “the vicious person does not recognize that he is vicious, whereas the incontinent person recognizes that he is incontinent” (1150b 35-36). But if they know they are incontinent, then they also know they are acting against the general principle. It would be ridiculous for someone to say “Yes, I know I am acting incontinently right now, but that’s only because my passions are preventing me from recognizing I’m in the kind of situation where that general rule applies.” This would be equivalent to me saying “Yes, I should be going to that meeting right now, but I’ve currently forgotten about that meeting, which is preventing me from leaving.” The fact that I’m bringing it up proves I have not forgotten about it at all.
Secondly, Aristotle distinguishes two forms of akrasia: impetuosity (propeteia) and weakness (astheneia). He claims that “the weak person deliberates, but then his feeling makes him abandon the result of his deliberation; but the impetuous person is led on by his feelings because he has not deliberated” (1150b 19-22). But the model of akrasia we saw before was precisely one where people could not deliberate because of their inactive knowledge. This appears to be impossible in the above model since the explanation hinges on the incontinent man being unable to complete deliberation because of their inactive knowledge. Any case of akrasia caused by weakness seems exactly like a situation where knowledge is being “dragged around like a slave.”
Thirdly, Aristotle argues that the incontinent are in a state of conflicting desires, that “incontinent people have impulses in contrary directions” paralyzing them like someone drawn both to the left and the right (1102b 14-25). But once again, this kind of conflict can only occur if the incontinent are actively using the knowledge of both the universal premise about what should be done and the particular premise recognizing they are in such a situation. We cannot appeal to ignorance, even of the more limited kind where we are simply not attending to the knowledge we have, to explain all instances of akrasia.
Henry and Stoyles on Genuine Akrasia
The major problem for the three above cases is that they present scenarios where the incontinent seem to be actively aware that what they are doing is wrong at the moment they are acting, and therefore contradicts any appeal to even a qualified form of ignorance. This is a frequent point of attack, or at least a difficulty we see contemporary philosophers trying to resolve in Aristotle.
We see this not only in McConnell’s analysis, but also in that of Devin Henry and Byron Stoyles. They distinguish between “drunken akrasia,” where someone knows something is wrong but is not exercising that knowledge at the moment of action, presumably because they are impaired from doing so by their passions and leading Aristotle to compare it to drunkenness, and “genuine akrasia,” where someone is exercising the knowledge that what they are doing is wrong even in the moment of action.
Henry and Stoyles claim that Aristotle allowed for genuine akrasia, but did not allow for it at all in his Nicomachean Ethics. Instead, they search in his less read Eudemian Ethics, especially the passages of 1224b 15-21 where he states that the incontinent man enjoys what they are doing, but also simultaneously feels a “prospective pain because he thinks he is doing a bad thing.”
However, we’ve seen from McConnell that there was already good reason to suspect that genuine akrasia exists in the Nicomachean Ethics as well, if the above mentioned problematic cases could not be reduced to drunken akrasia. But McConnell also shows how this is possible to at least some extent.
Defending Aristotle’s Account
Time Lags in Knowledge
The first and perhaps most obvious solution to these problematic cases is to say that there is a ‘time lag’ in the knowledge of the incontinent. The first difficulty we examined was Aristotle claiming that the incontinent are able to recognize themselves as incontinent. However, this does not need to imply that they do so in the moment they are acting, and instead only indicate their awareness of a general disposition toward incontinence, or awareness of it in retrospect. This lines up with Aristotle’s comparison to the drunkard, who eventually sobers up and can then recognize what they’ve done as wrong.
This also helps to explain akrasia of weakness. The issue we saw before was the fact that in these instances, the incontinent man seems to be completing their deliberation but then acting against it, meaning they are exercising their knowledge at the moment of action. But there is no need to say that this deliberation is happening at the relevant moment of action.
If someone reasons about what to do well before the moment of action, or reasons about an action which is available to them for a long period of time, then their deliberation might slip from their mind. Someone can tell themselves they will not eat cake before arriving at some party, but forget this commitment once they arrive or over the course of the evening. In both cases, the incontinent person will have deliberated, yet still acted against this deliberation. Akrasia of weakness is therefore not identical with genuine akrasia.
Knowledge Without Desire
We are still left questioning the remaining cases. How do we make sense of cases of genuine akrasia, like when someone says “I shouldn’t eat this” even as they pick up a slice of cake, chastising themselves even as they take a bite? This is a common enough experience that we cannot dismiss its possibility, and we find it not only in Aristotle’s Eudemian Ethics, but also apparently in the Nicomachean Ethics, especially where Aristotle says that the incontinent person feels impulses drawing them in contrary directions. This is the most mysterious and extraordinary form of akrasia given our initial Socratic thesis.
If we are going to account for genuine akrasia within an Aristotelian framework, we need to not only find a way where we can say someone both knows and does not know something at the same time, but also a way in which this knowledge is being exercised at the moment of action, or we need to find a feature necessary for action to take place which might be absent when knowledge is present. Henry and Stoyles take the latter approach, arguing that, while the incontinent have knowledge of the good, they lack the desire for it.
The argument goes that both incontinent and continent individuals are identical in terms of belief, rightly holding that what is in-fact the greater good to be the greater good. They also share a desire for improper pleasures. But they differ in terms of action, with the incontinent indulging in these improper pleasures like the intemperate do, while the continent resists these pleasures to pursue the goods of temperance. The basis for this difference must be something other than their beliefs, and Aristotle has argued that thought is impotent without desire (433a 1-4).
Henry concludes that the actions of the genuinely akratic can be explained because they lack any desire for the proper goods of temperance. It is only the continent or “enkratic” person who “feels an emotional pull from both sides” and has a “genuine conflict of desires.” Stoyles agrees with much of Henry’s analysis, but softens it saying someone may simply lack sufficient desire for the proper goods of temperance. The genuinely incontinent man might really desire the goods of temperance on some level, and therefore feel the “pull from both sides,” but tend to have their irrational desires win out.
Full and Complete Knowledge
McConnell takes an alternative approach, arguing that these cases which Henry would call genuine akrasia instead really do represent a lack of “full and complete knowledge.” He points to Aristotle’s Politics, where he shows that there are three things which make men good: nature, habit, and reason (1332a 39-40), and also that reason and habit are not always in harmony (1334b 10-12). If habit is necessary for full and complete knowledge, then we might explain cases of genuine akrasia by showing there is a lack of habit.
We can certainly think of cases where it might be said to be necessary. If I declare that I know how to play the piano, simply knowing how to read music and which keys correspond to which notes would be insufficient. I’d need to have practiced enough to gain the habit, becoming actually proficient at playing piano. Knowledge is not merely a matter of holding theoretical propositions, but also being able to practically carry out certain actions.
There are also good reasons to think Aristotle held habit to be an important part of full and complete knowledge as well, integrating it into our character. Aristotle argues, within his general examination of akrasia, that someone simply saying something does not prove they know it, since “those who have just learned something do not yet know it, though they string the words together; for it must grow into them, and this takes time.” This means that the incontinent who states the correct answer about what is right and wrong says “the words in the way actors do” (1147a 18-24). They might have some idea of the principles involved, but this does not imply they actually fully grasp the reasoning that leads to it, or all the implications of it. To truly integrate this fully, and therefore to truly know it, it needs to be made part of their habitual character.
The virtuous person has full and complete knowledge in the highest sense, having it entirely integrated into their character. The continent or enkratic would, perhaps, just have sufficiently integrated this knowledge into their character so that they still struggle against their baser desires, not having fully developed the habit, but where their reason ultimately wins out. The incontinent person finally has insufficiently integrated this knowledge, meaning we should not expect their actions to always perfectly line up with what they repeat or even believe on some level.
McConnell vs Henry and Stoyles
Both of these answers provide ways in which cases of genuine akrasia can occur, allowing people to simultaneously feel conflicting desires towards different ends or how they might act against knowledge they are actively using. However, McConnell’s answer appears to be superior, not only in making sense of Aristotle’s own position, but also theoretically, since it also presents a good framework for why someone might lack sufficient desire for the good, and neither Henry nor Stoyles ever explain why someone might really know what is good and yet not desire it.
McConnell did not actively explore the ways in which knowledge of the good might directly entail desire for the good though, and this seems to be the crucial point for determining which explanation is superior. In my own presentation of the puzzle of akrasia, I began by elaborating how Aristotle views, and for good reason, that every action and decision pursues some good. This established a firm link between the good and the desirable. This implies that, if someone “fully and completely” knows that something is good, then desire for it is inevitable. Likewise, if someone did not desire something which is essentially desirable, then this must be counted as a failure in some part of their knowledge.
Henry and Stoyles have simply assumed that knowledge of what is desirable need not be attended by desire itself. If we take a stronger view of knowledge, as held by Socrates and apparently held by Aristotle as well, then even these cases of “genuine” akrasia reveal a kind of ignorance.
If we were to reject this stance, I believe the only defensible position would be to deny that what McConnell is referring to as “full and complete” knowledge should be referred to as knowledge at all. Instead, we might look for other terms like wisdom or prudence. I believe Aristotle can be read in both ways here. He certainly at times does at times imply that fully having knowledge means integrating it entirely into our character, as McConnell has pointed out. This is precisely why we can say that in cases of drunk akrasia that someone might both have and not have knowledge at the same time, since it is missing from parts of their life. This is also why he concludes that “even the result Socrates was looking for would seem to come about” (1147b 13-15) when we consider the ways someone might state the correct position of knowledge without actually having knowledge.
Aristotle vs Socrates: Knowledge, Wisdom, and Prudence
Yet when Aristotle gets closest to restating Socrates’ position, he names prudence or wisdom as the intellectual virtues that are incompatible with incontinence, not knowledge. He writes that the same man cannot “be at once both prudent and incontinent” and that “a prudent person must also at the same time be excellent in character, [and the incontinent person is not]” (1152a 7-10). We also see in his Eudemian Ethics that Aristotle was explicitly aware of his similarity to Socrates here and still distinguished his position, saying that “it is clear that wisdom and excellence go together … and the Socratic saying that nothing is stronger than wisdom is right. But when Socrates said this of knowledge he was wrong. For wisdom is an excellence and not a species of knowledge” (1246b 32-36).
Having evaluated these arguments, exploring the nuances and subtleties of the different forms of akrasia, I have arrived at a position which is at least adjacent to, if not ultimately identical with, the Socratic position. In every form of akrasia we have discovered some form of ignorance or intellectual failure. The idea of goodness inherently implies some aspect of desirability, and it follows that any decision someone makes between options must be because it appears to them as the greater good.
This is, in some sense, true even in cases of akrasia. The human mind and soul are complex things which can come into conflict with one another. If relevant knowledge is not engaged at the relevant moment of action, or has not yet been fully incorporated into our character by means of habit, our actions may conflict with what we “know” to be the better option. This does come, as Henry and Stoyles suggested, from a lack of desire, but such a lack of desire also implies some lack of complete understanding of things which are really desirable. This lines up with our notions of knowledge, which implies a level of strength and conviction which goes beyond a merely true belief, and to some extent must become integrated with our character. Once someone has “full and complete” knowledge, as McConnell has put it, they seem to have risen to a level where we can say they have prudence or wisdom, in which wrong action would be impossible, as Aristotle and Socrates agree.
Works Cited
Aristotle, and Jonathan Barnes. The Complete Works of Aristotle: The Revised Oxford Translation. Vol. 2. Princeton University Press, 1984.
Aristotle, and Terence Irwin. Nicomachean Ethics. Hackett Pub. Co, 1999.
Henry, Devin. “Aristotle on Pleasure and the Worst Form of Akrasia.” Ethical Theory and Moral Practice, vol. 5, no. 3, Sept. 2002, pp. 255–270.
Kraut, Richard. “Alternate Readings of Aristotle on Akrasia.” Stanford Encyclopedia of Philosophy, Stanford University, 2022, plato.stanford.edu/entries/aristotle-ethics/supplement1.html.
Kraut, Richard. “Aristotle’s Ethics.” Stanford Encyclopedia of Philosophy, Stanford University, 2 July 2022, plato.stanford.edu/entries/aristotle-ethics/.
McConnell, Terrance. “Is Aristotle’s account of incontinence inconsistent?” Canadian Journal of Philosophy, vol. 4, no. 4, June 1975, pp. 635–651, https://doi.org/10.1080/00455091.1975.10716082.
Stoyles, Byron J. “Aristotle, Akrasia, and the place of desire in moral reasoning.” Ethical Theory and Moral Practice, vol. 10, no. 2, 24 Jan. 2007, pp. 195–207, https://doi.org/10.1007/s10677-006-9050-6.