Skip to main content

Verified by Psychology Today


Should We Blame Criminals?

Behaviorism, punishment, and blame.

Behaviorism is no longer the cutting edge of psychology, possibly because psychology’s edges have grown rather fuzzy. Behaviorism is out of fashion partly because the behavioristic conditions for doing science are tough[1]. Data must be “third-party-accessible”, no more ‘personal truth’ or ‘lived experience’; experiments must be repeatable; single-subject studies are favored, and statistics are distrusted.

Second, “behaviorism” now means, for many, B. F. Skinner’s ”radical” variety, which has both positive and negative features. On the positive side, Skinner was always concerned with empirical results, with controlling behavior rather than just observing it. On the negative side, the emphasis on control and Skinner’s aversion to theory soon led to a rather simplistic philosophy that, at times, downplayed empirical data. Skinnerians came to ignore, or even deny, the idea of internal states (other than the motivational state), to focus on positive reinforcement and to oppose punishment in any form despite evidence that it works[2].

I hope to show that an updated version of behaviorism can help sort out a legal squabble over blame: when and whether it is legitimate to punish someone guilty of wrongdoing.

Source: eamesBot/Shutterstock

To blame someone for a crime is to justify some form of punishment. Radical behaviorism claimed that punishment is an ineffective way to control behavior; many others oppose punishment on moral grounds; a third group offers metaphysical distractions involving free will, or political ones involving egalitarianism and the social safety net. Since this is a scientific account, I will set aside for the moment the moral and metaphysical claims and concentrate on the effectiveness of punishment.

In fact, punishment can be as effective as positive reinforcement. Neither is without its limitations. Both suffer from similar defects: the change in behavior may fade if the contingency is withdrawn—no more reward or punishment. In both cases, subjects will counterattack: try to get the food without doing the work, evade punishment in clever ways, like the “breakfast in bed” rat who learned to avoid foot-shock by lying on his furry back. But under suitable conditions, both reward and punishment can affect desired behavioral change. There is no scientific reason to exclude punishment as a means to control behavior.

The real objection to punishment as a way to control behavior is moral: punishment is itself “bad.” But experimental psychologists and law professors are equally unwilling to say so directly. Skinner delegitimized punishment with a faulty empirical argument. As we will see, some law professors achieve the same end by invoking “theory of mind” introspection and behavioral determinism to attack the notion of “personal responsibility” on which most punishment arguments are made to depend.

The debate in a legal context is an epistemological one, about whether free-will and/or determinism exists and, if they do not, is blame and punishment justified? It is not difficult to discern behind this debate a moral objection to retributive punishment, not to mention a favorable attitude towards political positions such as egalitarianism. For example, legal philosopher Ronald Dworkin is criticized for accepting the legitimacy of blame because it is (apparently) inegalitarian, he “performed for egalitarianism the considerable service of incorporating within it the most powerful idea in the arsenal of the anti-egalitarian right: the idea of choice and responsibility[3].” So, personal responsibility, which is a philosophical position, is seen by some legal scholars as just a cover for inegalitarianism. What is clear, however, is that “anti-blame” often reflects political ideology as much as an unsullied epistemology.

The anti-blamers are both vocal and prestigious. For example, Stanford legal philosopher Barbara Fried introduced her contribution Beyond Blame to an interesting 2013 debate as follows:

The philosophy of personal responsibility has ruined criminal justice and economic policy. It’s time to move past blame.

This may strike some as a bizarre idea. Why not blame a miscreant who clearly knew what he was doing (stealing, say), knew it was wrong, but did it anyway?

Determinism is the chief counter-blame argument. If an individual’s constitution and personal history determines his every action, then the criminal “couldn’t help himself”, his criminal behavior was inevitable. A Stanford biologist says: “Our growing knowledge about the brain makes the notions of volition, culpability, and, ultimately, the very premise of the criminal justice system, deeply suspect.” Evidently, “science” casts doubt on the naïve claim that the criminal is responsible for his crime. If he is not “responsible,” is it right to blame/punish him?

The legal arguments involve free will, determinism, and the never-ending conflict between compatibilists and incompatibilists. Free will is easily disposed of: it is not a scientific notion. It cannot be proved or disproved empirically. If you watch a bird in the wild, hopping about looking for food, it seems free, making up its own mind about where to hop next. How would you know it was not? "Well, that’s just a bird," you might say. We can just ask a person: do you feel free? Most times, if he is not under duress, coerced, or restrained he will say “yes.” But how conclusive is that? People can be indoctrinated, hypnotized, possess “false consciousness,” not be in full control of either their consciousness or behavior. You may fail to recall a name that you know perfectly well; yet in an hour or two, it pops up and you don’t know why. You are not “free to recall” even familiar words. Free will is a feeling, not a scientifically provable fact.

Free will may be unverifiable, but how convincing is the determinism argument? Suppose that there are laws—either physics or fate— which totally determine your behavior. Are you free or not? Some say yes (compatibilists) some not (incompatibilists). Compatibilists say blame and punishment are legitimate. But there are plenty of incompatibilists, people who argue that if behavior is determined, praise and blame are illegitimate.

Fried began her piece by quoting well-known political philosopher James Q Wilson:

Does the fact that biology determines more of our thinking and conduct than we had previously imagined undermine the notion of free will? And does this possibility in turn undermine, if not entirely destroy, our ability to hold people accountable for their actions?

She could have enlisted the aid of an older ally, Skinner, who in 1971[4] transferred responsibility away from the man to his past and present environment:

A person is responsible for his behavior, not only in the sense that he may be justly blamed or punished when he behaves badly, but also in the sense that he is to be given credit and admired for his achievements. A scientific analysis shifts the credit as well as the blame to the environment, and traditional practices can then no longer be justified.

Wilson, a compatibilist, answered his question “no.” But the weight on the other side is substantial: according to Fried, Skinner, and many others, from Richard Dawkins to famous Leopold and Loeb attorney Clarence Darrow, if behavior is causally determined, praise and blame—and punishment— are inappropriate.

But this argument is wrong, for the following reason: the effectiveness of punishment, as a remedy and deterrent, depends upon the predictability of behavior. There is no point in punishing a crazy person like Daniel M’Naghten who killed a British official in 1843 evidently under a delusion of persecution. But for most people, the prospect of punishment will deter and being punished may reform. Most people will respond to reward and punishment in the ways that behaviorists have mapped out. So determinism, far from ruling out punishment, is necessary for punishment to have any effect. Skinner’s argument is really a defence of legal punishment not a dismissal of it.

What about blame, which is a moral issue? We may punish a dog for stealing food from the table, but we don’t blame him, don’t treat him as morally defective, the way we would blame (wrongly, according to Fried) a human thief. Punishment, the treatment for the crime, may be similar for both, but for the dog, the moral angle is lacking. Moral issues lie outside science. Science can tell how to gain a moral end; it cannot tell us the end. This is an issue for another time, but this quote from a neuroscientist and a philosopher hints at the problem:

New neuroscience will affect the way we view the law, not by furnishing us with new ideas or arguments about the nature of human action, but by breathing new life into old ones… “[It] can help us see that all behavior is mechanical, that all behavior is produced by chains of physical events that ultimately reach back to forces beyond the agent’s control... Other neuroscientists hope to see a general attitude ‘shift from blame to biology.”

The sentiment is familiar: determinism means no personal responsibility. But the epistemology assumes an agent. For science, there is no agent, just a series of causes and effects. Agency has a bearing on the retributive function of legal punishment, as opposed to its deterrent effect. The idea of an agent comes not just from introspection but from religious sources. It requires a moral or religious argument, not a scientific one.


[1] Staddon, J.(2021) The New Behaviorism: Foundations of behavioral science. (3rd edition) Philadelphia, PA: Psychology Press.

[2] Staddon, J. (1995) On responsibility and punishment. The Atlantic Monthly, Feb., 88-94.

[3] G. A. Cohen (1989)

[4] Skinner, B. F. (1971) Beyond freedom and dignity

More from John Staddon, Ph.D.
More from Psychology Today