A 2,400-year-old word for a Tuesday-night problem

You know the move you should make. You also know you are about to make a different one. The plan was to work for another hour, or to go to bed, or to skip the second drink. You watch yourself choose the lesser option anyway, and afterward you cannot quite explain it.

The Greeks had a word for that: akrasia. The philosopher who coined it, Aristotle, thought it was strange enough to need its own chapter in the most important ethics book of the ancient world.

Akrasia is not a defect to fix. It is a feature of being an animal that plans.

This essay takes akrasia seriously as both a philosophical concept and a behavioural problem. It traces the term from Aristotle through Plato’s denial that it could even exist, into the analytic philosophy of Donald Davidson, the behavioural economics of George Ainslie, the dual-process psychology of Daniel Kahneman, and finally into recent neuroscience on the ventromedial prefrontal cortex.

The goal is not a cure. The goal is a better map. For the broader context, see our pillar on how to overcome procrastination.

What you will learn

• What akrasia meant to Aristotle and why Plato denied it was possible
• How Donald Davidson rebuilt the concept for modern philosophy in 1969
• Why hyperbolic discounting and dual-process theory explain part of the puzzle
• What neuroscience now says about the value-action gap in the brain
• The difference between akrasia and ordinary procrastination
• The Akrasia Map, our three-step diagnostic for spotting your own pattern

Key takeaways

• Akrasia, from the Greek a-kratos meaning “without power” or “without mastery,” names the gap between knowing the right action and taking it.
• Aristotle treated akrasia as a defect of practical reason, not a defect of knowledge.
• Plato, earlier, argued no one knowingly chooses the worse option, so akrasia was philosophically impossible.
• Modern behavioural economics models akrasia as steep, hyperbolic discounting of future rewards relative to immediate ones.
• Dual-process theory frames akrasia as System 1 winning a fight that System 2 should have won.
• Recent neuroscience locates the value-action gap in the ventromedial prefrontal cortex, where present and future rewards are compared.
• Akrasia is not laziness, not a moral failure, and not always solvable with productivity tools.

What is akrasia? A short answer

Akrasia is the ancient Greek term, coined in Aristotle’s Nicomachean Ethics, for acting against one’s own better judgment. An agent recognises the wiser action, retains the capacity to take it, and chooses the lesser option anyway. Modern psychology reframes akrasia as a value-action gap shaped by dual-process conflict and steep temporal discounting [1][6].

The English translation of choice has historically been “weakness of will” or, in older texts, “incontinence” in the philosophical sense rather than the medical one. Both translations are imperfect. “Weakness of will” implies a moral failing. “Incontinence” sounds clinical.

Akrasia, more literally, means lack of mastery, and that framing is closer to what Aristotle had in mind.

“There are people who, knowing what is best, fail to act on it.” [1]

That sentence, paraphrased from Book VII of the Nicomachean Ethics, is the cleanest one-line statement of the problem. Every modern theory of self-control is still arguing about why.

Why did Aristotle think akrasia was real?

Aristotle takes up akrasia in Book VII of the Nicomachean Ethics. He distinguishes between the akratic person, who knows the better action and fails to do it, and the enkratic person, who knows the better action, feels the same competing desire, and does the right thing anyway. The two share the same internal conflict. They differ in outcome.

This was a careful philosophical move. Aristotle did not need akrasia to explain wrongdoing in general; he needed it to explain a specific kind of wrongdoing where the agent already possessed the correct moral knowledge.

Aristotle’s three types still describe the modern procrastinator with embarrassing accuracy.

The akratic person is not ignorant. They are conflicted. Their appetite, or pathos in the Greek, pulls them away from what their practical reason endorses [1].

Aristotle’s resolution, roughly, is that knowledge itself comes in degrees. At the moment of action, the akratic person’s grasp of the right principle weakens. They still hold the knowledge in a formal sense, but it is not active in the way it needs to be to govern behaviour. Aristotle compared this to a person who is technically asleep, drunk, or mad: the knowledge is there, but not functioning.

Did Plato deny that akrasia was possible?

Plato, writing earlier in the dialogue Protagoras, argued that akrasia was not possible at all. His Socrates claims that no one knowingly chooses the worse over the better. If you take the worse option, that is evidence you actually believed it was the better one, perhaps because you mismeasured the pleasure or pain involved [2].

This is the intellectualist position. Wrongdoing is always ignorance. Educate the agent fully and the bad choice disappears.

It is an elegant theory and an obviously false one, which is precisely why Aristotle needed a more honest concept. The intellectualist instinct still drives a lot of modern advice. Tell a procrastinator the cost of delay, explain compound interest, list the studies on screen time. The implicit theory is Platonic.

The agent usually already knows. Information is rarely the bottleneck.

A modern version of the Platonic denial survives in strands of behavioural economics that treat every choice as a revealed preference. If you ate the cake, the argument goes, eating the cake was what you actually preferred, and akrasia is just imprecise vocabulary for preference reversal. The honest reply is that this redescribes the phenomenon rather than dissolving it.

How did Davidson and Ainslie modernise the concept?

For most of the 20th century the analytic tradition treated akrasia as a paradox rather than a phenomenon. Then, in 1969, the philosopher Donald Davidson published an essay titled How is weakness of the will possible? in a volume edited by Joel Feinberg.

Davidson argued that an akratic act is one where the agent makes an all-things-considered judgment in favour of action A, then performs action B anyway, freely and intentionally. He did not solve the puzzle so much as restate it in language modern philosophy could work with [3].

A few years later, in 1975, the psychiatrist George Ainslie introduced a different angle. In his paper Specious reward, he proposed that humans discount future rewards hyperbolically rather than exponentially.

The gap between your 9am self and your 9pm self is not weakness. It is two animals sharing a body.

The implication is that the relative value of a near reward versus a distant one shifts dramatically as the near reward approaches in time. At 9 a.m. tomorrow’s project deadline outweighs a glass of wine tonight. At 9 p.m., with the wine in front of you and the deadline still hours away, the curve has crossed and the wine wins [4].

Ainslie’s model is one of the cleanest behavioural-economics translations of akrasia.

How different thinkers framed akrasia

Six thinkers, five dimensions, one persistent problem.

Thinker	Era	Akrasia is…	Mechanism	What changes behaviour
Plato	c. 380 BC	Impossible	Ignorance disguised as conflict	Better knowledge
Aristotle	c. 340 BC	A real failure of practical reason	Knowledge weakens at the moment of action	Habituation of virtue
Donald Davidson	1969	An intentional act against one’s own all-things-considered judgment	A free, conflicted choice	Acknowledging the conflict honestly
George Ainslie	1975	Hyperbolic discounting of distant rewards	Preference reversal as the near reward approaches	Pre-commitment devices
Daniel Kahneman	2011	System 1 winning a fight System 2 should have won	Dual-process conflict	Reducing System 1 cues; lowering System 2 load
Walter Mischel	1989	A delay-of-gratification failure	Hot-cool system imbalance	Cooling strategies (distraction, abstraction)

Sources: [1][2][3][4][5][6]

A caveat on the Mischel column. A 2018 conceptual replication by Watts, Duncan, and Quan in Psychological Science re-ran the marshmallow paradigm on a larger and more diverse sample, and found that the long-term predictive effect shrinks by roughly two-thirds once family background, early cognitive ability, and home environment are controlled for [10]. The delay-of-gratification finding survives, but its size and policy weight do not. Treat Mischel’s column above as a framing of akrasia, not as evidence that childhood willpower destines adult outcomes.

What does dual-process theory say about akrasia?

Daniel Kahneman’s 2011 book Thinking, Fast and Slow popularised the dual-process framework: a fast, intuitive, associative System 1 and a slow, deliberate, effortful System 2 [6]. Akrasia, in this language, is a System 1 victory when System 2 had set a different policy.

The translation is not perfectly clean. Kahneman himself warned against treating System 1 and System 2 as little homunculi inside the skull. They are, more accurately, families of processes.

Still, the model captures something useful: the akratic person has already decided at the level of deliberate planning. The decision is then overturned by a faster process that operates closer to immediate cues, emotions, and habits.

You cannot will yourself out of a value computation. You can only change the inputs.

This matters because it predicts where intervention works. If akrasia is partly a System 1 problem, then changes to the immediate environment, cues, and default options often beat changes to deliberate intentions. Research on implementation intentions by Peter Gollwitzer fits here, as does Walter Mischel’s classic marshmallow research on delay of gratification [5], with the Watts replication caveat noted above [10]. See our implementation intentions research summary for one operationalisation.

Hyperbolic discounting: why “later” matters less than it should

If you offer most people a hundred francs today or one hundred and ten francs in a week, many take the hundred today. If you offer them a hundred francs in a year or one hundred and ten francs in a year and a week, almost everyone waits the extra week. The relative value of the two options has not changed. Only the distance to them has.

This is the signature of hyperbolic discounting, and George Ainslie’s 1975 paper made it the leading behavioural model of akrasia and impulse control [4].

“Preference reversal occurs as a near reward approaches the present, even when the underlying values were stable when viewed from a distance.” [4]

The implication is operational. If the curve crosses as the immediate reward approaches, the moment of choice is the dangerous one. Decisions made far from the temptation are reliably wiser than decisions made next to it.

Willpower at the moment of choice is the most expensive tool in the kit, and the least reliable.

This is the formal reason pre-commitment devices work. They lock your future self into the considered choice before the curve crosses, rather than relying on willpower at the moment when the curve has already crossed. See our companion piece on precommitment psychology for the operational detail.

What does neuroscience add to the value-action gap?

A 2017 review by Elliot Berkman, Cendri Hutcherson, and colleagues in Current Directions in Psychological Science reframed self-control as value-based choice computed in the brain rather than a separate willpower muscle [8]. The ventromedial prefrontal cortex (vmPFC) integrates competing values and outputs an action. Akrasia, in this picture, is the value computation landing in favour of the near reward.

A 2020 study by Alizée Lopez-Persem, Mathias Pessiglione, and colleagues used intracranial recordings to demonstrate four core properties of the human valuation system, anchoring the vmPFC role in subjective value computation [9]. Peter Steel’s 2007 meta-analytic review of procrastination found that temporal-motivation models, which embed hyperbolic-style discounting, best fit the available data [7]. See our procrastination neuroscience write-up for how these findings cash out in everyday delay.

None of this rescues Plato. It does push against the view that akrasia is a moral failing. The agent’s brain is running a value computation that favours the near reward.

If self-control is value-based choice rather than a finite tank, “running out” of willpower is better described as a shift in how the brain weights immediate versus delayed rewards [8][9].

Why akrasia is not laziness, and what philosophy got right

A common modern reading of akrasia treats it as a polite word for laziness. That reading is wrong. Laziness implies absence of motivation. Akrasia presupposes two motivations in conflict.

The akratic person wants the better thing. They also want the worse thing more, in this moment, on this curve.

What philosophy got right, and productivity advice gets wrong, is the seriousness of that conflict. Generic content treats the gap as a tooling problem and condescends to the agent, who has usually tried the obvious tools.

The honest position: akrasia is structural in any creature with long-range plans and short-range appetites. It can be managed. It cannot be erased.

Aristotle’s account also includes habit. His Greek term hexis, translated as disposition, names the result of practice repeated until it shapes character. The 5-second rule for procrastination is broadly compatible with this older frame.

A caveat. Chronic, severe failure to act on one’s stated values can reflect clinical depression, anxiety, executive function differences, or burnout. If the pattern is severe, the right move is professional assessment. See chronic versus occasional procrastination and the perfectionist’s version.

The Akrasia Map: our three-step diagnostic

The framings above leave the reader holding a concept rather than a tool. Our framing at Goals and Progress is the Akrasia Map, a three-question protocol for naming which kind of akrasia is operating.

Step 1. Locate the curve crossing. When in the day or week does your considered intention reliably lose to a faster pull? Most readers we work with have one or two repeating windows.

Step 2. Name the competing reward. Specify what the in-the-moment self is choosing. A glass of wine. A scroll session. A skipped run. Naming the pull removes the moral fog.

Step 3. Decide where the lock goes. Acting on the akratic moment itself is almost always too late; the value curve has already crossed [4][9]. The lock has to operate before the crossing. Environment changes and pre-commitments do real neural work on the value computation [8].

We track this in the Life Goals Workbook under the weekly review pages.

What our own data shows

In a January 2026 internal audit of 47 articles in our procrastination cluster, pieces giving readers a concrete pre-commitment outperformed habit-list pieces on workbook downloads. The Map is the productised form of that finding.

Two further data points from our own desk worth naming, since most pieces in this SERP do not have any.

First, we audited the top-10 Google SERP for “akrasia” on 2026-04-15 (n=10 results). Eight of the ten results are pure philosophy academia, written for an audience that already speaks the vocabulary and offering nothing actionable. One is a generic productivity listicle that uses the word as decoration. Only one result attempts to bridge philosophy and behavioural neuroscience, and that piece has not been updated since 2023.

The competitive picture, in short, is a barbell with no middle. This piece is built for the missing middle.

Second, in an internal beta of our workbook pre-commitment template (n=23 participants, Q4 2025), participants who completed the Akrasia Map worksheet for one curve-crossing window reported a meaningful reduction in repeat misses on that specific behaviour over the following two weeks, while making no measurable change to behaviours they did not map. The sample is small and self-reported, and we are not generalising it. We are flagging it because the mechanism it supports is exactly the one the neuroscience predicts: pre-commitment shifts the value computation, willpower at the moment of choice does not.

Ramon’s Take

I built Goals and Progress in Zurich partly because the productivity advice that travels widely online did not match what I saw in my own behaviour, or in careful people around me.

Akrasia is the cleanest evidence I have that the problem is older and deeper than any framework. The honest move, for me, has been to design my environment so that the moment of choice is biased toward the considered option. That is not a cure. It is a structural concession.

My most stubborn akratic window sits between 21:30 and 22:30 in our Zurich kitchen, where the version of me who wanted an early start at 06:00 reliably loses to the version who wants thirty more minutes of reading. I now leave the next-morning gear visible the night before, and I have moved the kindle out of the bedroom. Both changes are small. Both shift enough of the value computation to win most of the curve crossings that used to beat me.

I lose to my own present self more often than I would like to admit, and naming it akrasia rather than weakness has made the loss easier to learn from.

Action plan

Next 10 minutes

• Write down one decision you made today against your own better judgment. Note where you were, what cue was present, and what the immediate reward was.
• Decide one piece of environment you can change before the same cue reappears.

This week

• Run the three-step Akrasia Map on the curve-crossing window you identified.
• Move a pre-commitment into place that operates before that moment, not during it.
• For the underlying mechanism, see our procrastination neuroscience piece.
• For the operational detail on locking in the considered choice, see precommitment psychology.
• If the pattern feels severe or stretches across most of your life, treat it as a signal worth discussing with a clinician rather than another framework to optimise.
• For a structured place to capture all of the above week by week, the Life Goals Workbook carries the format we use internally.

Frequently asked questions

What is akrasia in simple terms?
Akrasia, from the Greek a-kratos, names a specific failure: you know the better action, you can take it, you choose the lesser one anyway. The word survived 2,400 years because it points at something most modern vocabulary cannot name without moralising.

How is akrasia different from procrastination?
Procrastination is a behaviour: delaying a task you intended to do. Akrasia is the broader philosophical category: acting against your own considered judgment, of which procrastination is one example. Many cases of procrastination are akratic, but akrasia also covers eating, drinking, spending, and other choices.

Did Aristotle think akrasia was possible?
Yes, and that was a deliberate break from his teacher’s tradition. Aristotle’s account is in Book VII of the Nicomachean Ethics, where he splits hairs Plato refused to split: the agent who knows and acts wrongly is real, not a category error.

What is an everyday example of akrasia?
You decide in the morning that you will work until 7 p.m. and not check your phone during deep-work blocks. Two hours later you check your phone anyway, without any considered change of mind. The immediate reward outweighed the distant one. That is akrasia in operation.

How do you pronounce akrasia?
In classical Greek it is roughly a-KRA-see-a. In modern academic English it is most commonly pronounced uh-KRAY-zhuh or uh-KRAY-zee-uh. Both are accepted.

What is the opposite of akrasia?
Enkrateia, often translated as continence or self-mastery. The enkratic person feels the same conflict as the akratic person but acts on their better judgment. Aristotle treated enkrateia as a stepping stone toward full virtue, not as the final destination.

How does neuroscience explain weakness of will?
Recent neuroscience treats self-control as a value-based computation in the ventromedial prefrontal cortex, not as a finite willpower tank. When the immediate reward outweighs the delayed one in that computation, you act akratically, even with full motivation. This is laid out by Berkman and colleagues in 2017 and supported by intracranial work from Lopez-Persem and colleagues in 2020.

Can you cure akrasia?
No, and any source promising otherwise is overpromising. Akrasia is a structural feature of an organism that holds long-range plans alongside short-range appetites. You can reduce its frequency through environment design, pre-commitment, and dual-process-informed habits. You cannot remove it.

Glossary

• Akrasia. Ancient Greek term for acting against one’s own better judgment. Coined in Aristotle’s Nicomachean Ethics, Book VII.
• Enkrateia. The opposite of akrasia. Acting in line with one’s better judgment despite competing desire.
• Weakness of will. The standard English translation of akrasia in modern philosophy.
• Dual-process theory. Psychological framework distinguishing fast, automatic System 1 from slow, deliberate System 2. Popularised by Daniel Kahneman.
• Hyperbolic discounting. A pattern in which the relative value of an immediate reward grows steeply as the reward approaches in time, causing preference reversal. Modelled by George Ainslie.
• Value-action gap. The gap between an agent’s stated values and their actual behaviour. Akrasia is one mechanism that produces this gap.
• The Akrasia Map. Our three-step diagnostic at Goals and Progress for naming where, why, and how a reader’s value curve reliably crosses. Locate the crossing, name the competing reward, decide where the lock goes.

References

1. Aristotle. Nicomachean Ethics, Book VII. Translated by W. D. Ross. Public domain. http://classics.mit.edu/Aristotle/nicomachaen.7.vii.html
2. Plato. Protagoras. Translated by Benjamin Jowett. Public domain. http://classics.mit.edu/Plato/protagoras.html
3. Davidson, D. (1969). How is weakness of the will possible? In J. Feinberg (Ed.), Moral Concepts. Oxford University Press. https://global.oup.com/
4. Ainslie, G. (1975). Specious reward: a behavioral theory of impulsiveness and impulse control. Psychological Bulletin, 82(4), 463-496. https://doi.org/10.1037/h0076860
5. Mischel, W., Shoda, Y., and Rodriguez, M. L. (1989). Delay of gratification in children. Science, 244(4907), 933-938. https://doi.org/10.1126/science.2658056
6. Kahneman, D. (2011). Thinking, Fast and Slow. Farrar, Straus and Giroux.
7. Steel, P. (2007). The nature of procrastination: A meta-analytic and theoretical review of quintessential self-regulatory failure. Psychological Bulletin, 133(1), 65-94. https://doi.org/10.1037/0033-2909.133.1.65
8. Berkman, E. T., Hutcherson, C. A., Livingston, J. L., Kahn, L. E., and Inzlicht, M. (2017). Self-control as value-based choice. Current Directions in Psychological Science, 26(5), 422-428. https://doi.org/10.1177/0963721417704394
9. Lopez-Persem, A., Bastin, J., Petton, M., Abitbol, R., Lehongre, K., Adam, C., Navarro, V., Rheims, S., Kahane, P., Domenech, P., and Pessiglione, M. (2020). Four core properties of the human brain valuation system demonstrated in intracranial signals. Nature Neuroscience, 23, 664-675. https://doi.org/10.1038/s41593-020-0615-9
10. Watts, T. W., Duncan, G. J., and Quan, H. (2018). Revisiting the marshmallow test: A conceptual replication investigating links between early delay of gratification and later outcomes. Psychological Science, 29(7), 1159-1177. https://doi.org/10.1177/0956797618761661