Skip to main content

Verified by Psychology Today

Temptation and the Will

Are you free to do the right thing?

Key points

  • Effort and ability drive performance.
  • The decision to reduce effort is explicable, though it may appear to be free.
  • The burden of proof lies with those who make extravagant claims.

The will cannot be unchained. The will is the chain. – Hoca Camide

Alberto Salazar, one of the finest long-distance runners of the 1980s, was once asked by a TV reporter why he hadn’t run faster and won the race. “If I had run faster,” Salazar explained [I am quoting from memory], “I would have killed myself.” The reporter’s question struck me as obnoxious at the time [it still does], and I liked Alberto’s no-nonsense answer, which with its clarity and brevity put the reporter in her place, although she did not seem to realize that at the moment.

Alberto understood that the energy available for effort or work is a limited resource (Baumeister et al., 1998). When you run on empty, you risk damaging the system. Not everyone and not all psychologists enjoy the benefits of Alberto’s wisdom. Whenever assertions of the freedom of the will well up, the assumption that we could have acted differently is not far behind.

Alberto made it clear that he could not have acted differently by running faster. Had he been on a training run, however, just jogging along, he arguably could have run faster, and does this not prove that the will is free? Before we answer this question, consider a well-worn finding from empirical psychology.

Mischel and Ebbesen (1970) showed that children differ greatly in their ability to forego the instant gratification of eating a marshmallow when a larger plate of mallows beckons in the future. The experimental set-up created a psychological conflict for the participating kids, a crisis of temptation.

Decades of research have shown that the ability to manage this crisis predicts a suite of desirable life outcomes. Good fortune comes to those who know how to wait.

One interpretation of the marshmallow effect is that each kid resisted the temptation to eat the sweet just as long as they were able. If the task is equally hard for all kids – no guarantee – the time of delay achieved is a clean measure of the ability to resist consumption. If each kid’s effort is at its maximum, effort is aligned with the ability and thus totally confounded with it. Here, ability is the maximum effort the kid is capable of. Of course, some kids might have said to themselves, “Sure, I could wait longer, but I choose not to.” If so, the recorded delay is an impure measure of ability.

What degrades the measure of ability by reducing the exerted effort below maximum levels? Might it be statistical noise, a different psychological force, or the freedom of the will? Noise, or randomness, is boring, and what can really be said about it? A different psychological force making early consumption more likely even when the reservoir of effort energy has not been depleted is a more interesting candidate. But it is easier to postulate than to pin down.

An economist might settle for saying – in typical post-hoc fashion – that the kid’s utility function has changed, perhaps by increasing the value placed on the yummy treat or by a jump in the temporal discounting rate.

A psychologist would wish to see non-circular evidence for a preference shift toward consumption and away from effort exertion. Such evidence is hard to come by. It is difficult to tell, in other words, whether a kid has exerted all the effort it was capable of exerting at that moment.

When Salazar was jogging, we said – and he would have agreed – that he could run faster but chose to jog. Salazar rationally attained his goal of jogging. Is this then the point where free will shows itself? Salazar could run faster or slower if he had the corresponding goals. This is obvious enough, but the question is whether he freely chose which goal to have, and what would that mean?

On what grounds would he choose this goal over another? If there are grounds, we can search for a lawful explanation (e.g., he has formed a habit of jogging each Friday morning and the corresponding goal comes to mind as he awakens); if there are no grounds and if goal-emergence is not random, then we are back to the perennial question of how free will can come out of nowhere when out of nowhere the will must come to be free.

Burden of Proof

A naturalistic explanation of behavior and other events appeals to necessary and sufficient causes and to chance when causes run out. Free will is claimed to be a third way, but to my knowledge, no one has offered a theory that explains how something can come out of nothing (and not by chance) and be by virtue of its obscure origin the most morally significant moment.

Ask an advocate of free will (perhaps yourself) to provide proof or evidence for its existence, and you might find that the advocate shoots back by challenging you to prove its non-existence. There is no need to play burden-of-proof ping pong (Pinker, 2021; reviewed in Krueger, 2022a). The burden lies with the more extravagant claim. Compared to the necessity-plus-chance model of naturalism, the necessity-plus-chance-plus-free-will model is the more extravagant. It must show that its third parameter cannot be reduced to the first or the second.

Beyond Moralism

The doctrine of free will has done much damage to human happiness because it opens up all of our actions to condemnation (Reginster, 2021; reviewed in Krueger, 2022b). See, for example, Genschow and Vehlow (2021) on the poignant case of victim-blaming. Where Mischel and Ebbesen found evidence for the causally active personality trait of delay of gratification, a moralist insists that the eager consumer of marshmallows should have inhibited his desire because he could have.

To paraphrase David Hume (1711-1776), we can’t infer a should from a could. Well, we should not, although we could. Spinoza (1632-1677), anticipating the better angels of psychological science, was content trying to understand people; he had no desire to judge them.


I asked Hoca Camide if he had passed the marshmallow test. “I never went to the lab,” Hoca said, “I had too much on my plate.”

Moving on with the topic of temptation, I asked him what he thought of Augustine’s Da mihi castitatem et continentam, sed noli modo. “Father Augustine,” Hoca mused, “must surely have read Ecclesiastes. But if he understood it, he would have known that his plea was unnecessary.”


Baumeister, R. F., Bratslavsky, E., Muraven, M., & Tice, D. M. (1998). Ego depletion: Is the active self a limited resource? Journal of Personality and Social Psychology, 74(5), 1252–1265.

Genschow, O., & Vehlow, B. (2021). Free to blame? Belief in free will is related to victim blaming. Consciousness and Cognition, 88.…

Krueger, J. I. (2022a). Rationality now! Review of ‘Rationality: What it is, why it seems scarce, why it matters’ by S. Pinker. American Journal of Psychology.

Krueger, J. I. (2022b). Nietzsche’s last will. Review of ‘The will to nothingness. An essay on Nietzsche’s On the genealogy of morality’ by B. Reginster. American Journal of Psychology.

Mischel, W., & Ebbesen, E. B. (1970). Attention in delay of gratification. Journal of Personality and Social Psychology, 16(2), 329–337.

Pinker, S. (2021). Rationality: What it is, why it seems scarce, why it matters. Viking.

Reginster, B. (2021). The will to nothingness: An essay on Nietzsche's On the Genealogy of Morality. Oxford University Press.