canine guilt

It would have been so pointless to kill himself that, even if he had wanted to, the pointlessness would have made him unable.

~ Kafka; The trial

Those who argue that the will is free must explain from what. At the limit, the will is free from everything. There are no constraints, and no prior causes. This view of free will is interesting, radical, and illogical. If it were true, the will would be an uncaused cause, entirely unpredictable, inexplicable, and yet, not random. When people say a person acted out of free will because he could have acted otherwise they seem to accept this radical and false notion of freedom unless they explain which constraints they allow. Likewise, when people say they need the concept of free will in order to be able to punish others for their misdeeds (in good conscience), they must explain how the constraints they do allow mitigate responsibility and punishment.

This is tough, and my impression is that most people who have responded to this issue have not considered the implications of what they are saying. My colleagues Andrew Monroe and Bertram Malle (2010) did an interesting set of studies to see what ordinary people mean when they say an act was freely willed. In the first study, they asked participants to “explain in a few lines what you think it means to have free will.” They coded the answers into three categories. 65% of the respondents referred to the ability to make a decision or choice; 33% referred to doing what one wants; and 29% referred to acting without internal or external constraints. Of these, the first category is irrelevant because it begs the question of free will. Of course, a decision or choice needs to be made, but was it free? The second category refers to desire, which one might also call motive, drive, or impulse. In other words, the second category refers to the will and again begs the question of freedom. This leaves the third category, which refers to constraints. And here it gets murky. Monroe & Malle note that what matters is “overcoming (internal or external) constraints” (p. 214) but refer to “acting without internal or external constraints” in Table 1 (p. 215; emphasis added).

If there are no external constraints (coercion, social pressure), a person can do what he wills (but not will what he wills, says Schopenhauer). In that sense, the will is socially unconstrained and thus free. To say that there are no internal constraints makes no sense because the will itself is a constraint. It compels the person to act. Hunger compels eating and prolonged wakefulness compels sleep. Only when there is an internal conflict between two or more wills (to sleep and to keep driving) is there a question of whether one will can overcome the other; or rather, which will will win, and we know that one will—because one must. Why is it then that we credit the person who manages to stay awake with having exercised free will, whereas we blame the person who falls asleep at the wheel with having failed to exercise his free will, because, we say, he could have acted differently?

Judgments of free will are moral judgments. They reflect moral values. For this reason, they are unsuitable as a basis for the scientific study of will. Since, save for a handful of hard-shell libertarians, most sane people now agree that metaphysically free will is fiction, not fact, we could get on with the business of studying “the will”; that is, all the forces that make us act. This is the rich field of motivational psychology. It is scientifically tractable because it operates without assuming unmeasurable or illogical constructs. It is a good way to do science.

Once we get rid of the idea that we must support some notion of moral responsibility, we can return to the saner question of how to shape human (or animal) behavior. If we can train dogs to be housebroken by using a regular regimen of walks and biscuits, why would we think that we can educate our children only by giving them a guilty conscience?

In this context, I find the notion of “moral outrage” and “altruistic punishment” to be false advertising. In behavioral economics, we see that many cooperators will punish defectors to bring them back in line. This tends to work, and a behaviorist would not be surprised. The moralistic terminology is an unnecessary and misleading superstructure. Are we to believe that cooperators punish defectors only if they think that they freely chose to defect? If so, cooperators appear to believe that they can change the ways of freely acting agents with mechanistic means. "He was free when he defected against me, but I will compel him now to cooperate". No. Punishment works precisely because the punished deed was not free either.

Monroe & Malle write that people “believe that the actor could have acted differently” (p. 216) even if they were told that the universe’s causal clockwork is deterministic. To maintain this assertion, people need to either deny desire (the will itself) or claim that the balance of competing wills was different from what it was. In other words, they have to deny the premises they were given. In Monroe & Malle’s second study, participants were presented with the neuropsychological claim that action is preceded and explained by specific patterns of brain activity and asked how they would respond. The majority responded by simply reaffirming choice. E.g., “even though you have neural impulses, your free will allows you to look over those impulses and decide for yourself” (p. 218). This is compatibilism by assertion. But outside of Wonderland, just saying so does not make it so. The question researchers need to ponder is whether they really want to look to folk psychologists for guidance on the trickiest questions. Why allow Joey from Springfield to help define the constructs of the field, and ignore Schopenhauer?

Back to Genesis

Where’s all this hand-wringing over free will and moral responsibility coming from? I think it’s Genesis 3:1 to 3:6. Eve, the snake, the apple, you know the story. Interestingly, temptation is psychologically outsourced and represented by the snake. Even more interestingly, Eve recognizes temptation and she knows that she’s not allowed to nosh from the tree because god said so. She knows that she’s breaking the rule before eating the fruit that gives her knowledge of right and wrong. Yet more interestingly, the knowledge that she and her helpmate Adam gain from consuming that apple strudel gives birth to the emotion of shame (over nakedness), not guilt. Today, we still love to blame and punish miscreants inasmuch as they had foreknowledge; if they knew what they were doing. And yet still more interestingly, foreknowledge has absolutely nothing to do with any real or imagined ability to act differently. God obscured things – as usual – but Eve gave us food for thought.

Monroe, A. F., & Malle, B. F. (2010). From uncaused will to conscious choice: The need to study, not speculate about people’s folk concept of free will. The Review of Philosophy and Psychology, 1, 211-224.


Moralism is not to morality what rationalism is to rationality.

You are reading

One Among Many


Gratitude has a dark side.

Psychology for Real

Are research and reality different worlds?

Small Samples, Big Hopes

Mind your N’s and p’s.