Criminology, Hypothesis, Science

Is There a Biological Case for Criminal Justice Reform?

Tell a woman the story of Christ’s crucifixion, and if she doesn’t cry, she’s a witch. In the 1500s, women who failed this test were burned alive.

The test—clearly and painfully faulty—was not publicly questioned until Dutch physician Johann Weyer wrote “De Praestigiis Daemonum” (or “On the Tricks of Demons”) in 1564. Weyer correctly argued that many older women couldn’t cry due to atrophy of their lachrymal glands. Prosecutors were presented with a dilemma: reform the witch trial system or potentially kill innocent women. A paltry sum of sanity was brought to the system, and hundreds of potential deaths were prevented.

Robert Sapolsky

Stanford University neurobiology professor Robert Sapolsky believes that today’s U.S. criminal justice system has similar biological blind spots. Just as witch prosecutors didn’t know about the lachrymal glands’ connection to tears, we don’t fully understand an untold number of connections between DNA, the brain, hormones, and other dynamic aspects of the human body. In Sapolsky’s latest book, Behave: The Biology of Humans at Our Best and Worst, he argues that our justice system falls into the same traps as the witch prosecutors of the fifteenth century—we send people to prison, even death, based on incomplete evidence.

For example, 351 people have been exonerated via DNA evidence by the Innocence Project—including 20 who sat on death row. They’re lucky, but how many innocent lives have met unfortunate ends? Or consider the cruel fact that 2 million people with mental illness are booked into jails each year. Once inside, 83% of mentally ill inmates have no access to treatment, per the National Alliance on Mental Illness, likely keeping them in a cycle of recidivism.

“The current criminal justice system needs to be abolished and replaced with something that, while having some broad features in common with the current system, would have utterly different underpinnings,” Sapolsky writes in Behave.

After centuries of judicial burnings and killings based on faulty evidence, Sapolsky believes that we’ve set ourselves apart from our ancestors with a simple phrase: “It’s not her; it’s her disease.” A systematic focus on disease and treatment, rather than evil and punishment, would save lives and prevent needless suffering.

However, the brain and biology remain largely misunderstood in U.S. law. This misunderstanding stems from a common belief in mitigated free will, Sapolsky writes, a belief that biology influences our actions, but doesn’t make our final decision. Our final decisions are made somewhere beyond biology, according to mitigated free will believers, by something that allows us to control our actions. We may not be able to completely control our actions during seizures, while we’re drunk, or when a brain tumor is pressing on our frontal cortex, they say, but we can give a vehement “no” to the voices in our head whispering at us to kill our family.

Mitigated free will is how most people make sense of this confluence of biology and free will, Sapolsky writes. If someone believes that biology determines 99.9 percent of their actions and also believes that free will can determine 0.1 percent of their decisions, Sapolsky says they are “tacitly invoking a homunculus operating outside of the rules of science.”

If free will does exist, new scientific discoveries are whittling it into smaller pieces, Sapolsky writes. Two hundred years ago, we learned that the frontal cortex affects our control of appropriate social behavior, changing how we consider brain damage. Less than 70 years ago, we discovered that the root of schizophrenia lies in biochemistry, not hell or heaven. With more biological discoveries to come, can we confidently use mitigated free will arguments to decide who lives, who dies, and who goes free?

“[If] you believe that there will be the accrual of any more knowledge, you’ve just committed to either the view that any evidence for free will ultimately will be eliminated or the view that, at the very least, the homunculus will be jammed into even tinier places,” Sapolsky writes;

And with either of those views, you’ve agreed that something else is virtually guaranteed: that people in the future will look back at us as we do at purveyors of leeches and bloodletting and trepanation, as we look back at the fifteenth-century experts who spent their days condemning witches, that those people in the future will consider us and think, ‘My God, the things they didn’t know then. The harm that they did.’

This homunculus—mitigated free will—has no biological proof, Sapolsky argues, but we do have biological proof that our decisions and actions are influenced by a combination of our DNA, brain, hormones, nervous system, how we grew up, and what happened while we were still growing in our mother’s uterus. A single split-second decision can be years in the making; it is multifactorial, containing no simple answers, as Sapolsky argues throughout Behave.

Most people believe that this argument allows criminals to escape culpability. Typically, a question from the dissenter follows: “If we can’t hold criminals responsible for their actions, are we expected to let them run loose in the streets?” Clearly, no rational person wants violent criminals to wreak havoc in the streets, Sapolsky writes. Someone who commits a heinous crime must be kept away from citizens, even if their prefrontal cortex is damaged; neuroscience and biology cannot erode the safety of society. Violent individuals, Sapolsky writes, “can no more be allowed to walk the streets than you can allow a car whose brakes are faulty to be driven. Rehabilitate such people if you can, send them to the Island of Misfit Toys forever if you can’t and they are destined to remain dangerous.”

This follows even without the available medical diagnoses to find the exact problem with a violent individual. Sapolsky writes that the mechanics from the broken car analogy wouldn’t call the vehicle “evil” if they couldn’t pinpoint a mechanical issue. Perhaps there’s a problem with how the car was built, perhaps the car is affected by an unknown pollutant, or perhaps there is a diagnostic issue that simply has not yet been solved. Many people find this comparison dehumanizing, Sapolsky writes, but it “is a hell of a lot more humane than demonizing and sermonizing them as sinners.”

Instead of allowing criminals to run free, Sapolsky offers a suggestion that he believes would “change everything”: remove the punisher mindset from the criminal justice system. Punishment feels good for all people, Sapolsky says, not just for blood lusters and witch hunters. It’s a deep, atavistic pleasure: “Put people in brain scanners, give them scenarios of norm violations. Decision making about culpability for the violation correlates with activity in the [dorsolateral prefrontal cortex],” Sapolsky says. “But decision making about appropriate punishment activates the [ventromedial prefrontal cortex], along with the amygdala and insula; the more activation, the more punishment.”

Even if punishment is necessary to improve behavior in violent individuals, there’s no place for punishment as a virtue, Sapolsky writes. The reason is simple: We used to punish people for having seizures; now we give them medication, take their car keys away, and tell them not to operate heavy machinery.

“We’ve successfully banished the notion of punishment in that realm,” Sapolsky writes. “It may take centuries, but we can do the same in all our current arenas of punishment.”

Sapolsky admits that removing the punisher mentality may create a practical challenge as there are studies that both support and contradict the hypothesis that criminal punishment discourages law breaking. Perhaps some amount of conventional punishment will still be necessary to keep peace in society, he writes, but that punishment must be stripped of the beliefs that people deserve to be punished and that punishing them is virtuous. Bloodlust, science, and criminal justice cannot mix.

Filed under: Criminology, Hypothesis, Science


Hal Conick is a writer and journalist based in Chicago. You can find his writings at and send him tips and notes on Twitter at @HalConick.


  1. Robert Heinlein, in Starship Troopers, has a passage for one of the Morals and History courses, or whatever it’s called, where the Professor questions the student about training a puppy. This is an anecdotal explanation of why the term “juvenile delinquent” is meaningless because it implies the juvenile is responsible which is like saying the puppy is responsible for having an accident on the kitchen floor — no, it is the fault of the master/adult who is responsible for training the puppy/juvenile on what the acceptable norms are.

    I bring this up because i’ve had this discussion with peers when we’d see articles in the paper about students assaulting teachers. We universally would agree that we would have never EVER considered doing that because of our parents. Just the threat of “wait until your father gets home” was enough to train behavior in the teenage years. But now, any sort of training is equated with child abuse. Is it any wonder that our juveniles grow up to be out of control (criminal) adults in ever increasing #s? It isn’t that our criminal justice system has morphed to reward (financially) incarceration. Yes, it may have, but a big part of the reason we see police confrontations is the lack of training in the juvenile years — the training now deemed “abuse” and banned or at least shunned.

    Heinlein’s passage explains it this way. When a puppy does something you follow 3 steps: 1.) make sure they know you are angry/upset with the behavior (you yell “NO!”). 2.) You make certain the puppy knows what you are upset about (rubbing their nose in it…and it must be done right away, not later on if you happen to find it). 3.) Physical discipline (current mode is to kennel them up, keep them away from you but for children it’s a smack on the rump). This 3rd step is because humans for thousands of years have evolved (biologically) to view physical discomfort as something to be avoided. If we remove that then we remove the reinforcement mechanism of training. Positive versus negative reinforcement can be debated. I use positive with my kids…do something bad, no computer for a month which makes her react like i’m yanking her fingernails — but even that is subverted by her teachers who then send me notes home insisting I allow her to use the computer to improve her math skills. I can only imagine their reaction if she said “oh, Daddy spanked me for being bad.” I’m sure DFACs would be at my door in a heartbeat.

    • Carl Sageman says

      It’s worth reading the diary of Charles Panzram. He was told no and given a smack on the bum. It didn’t work. So they escalated it (his diary is quite fascinating about how far people went). Charles responded by sexually assaulting many boys and men and murdering many people. The more they punished him, the more he sought revenge.

      While I don’t know anyone quite as bad as Panzram, I know plenty of people who don’t respond to loud voice and smack.

      So, let’s revisit the past and ask what the article asked. What happened when we tried the blanket “yell and hit”? We created serial killers. If we analysed these people and determined they were physically damaged, maybe, just maybe, the “yell and hit” wasn’t the right answer.

      I agree with the article 100%. Seek to understand if there’s a physical issue that makes it impossible to discipline using regular methods. If that’s the case, then come up with an alternate strategy (which could be isolation to protect society).

      I have a family member who falls into the category of problematic. Society has punished him repeatedly as if it will “solve” he problem. Society has paid dearly (financially) for the one show fits all. I would suggest a broader range of analyses may provide a better outcome for society. This doesn’t mean holiday camps, it means exactly what the article suggests, seek to understand if there are physical causes. These people will not be deterred, no matter what you do to them.

      • “What happened when we tried the blanket “yell and hit”? We created serial killers.”

        Provide evidence that serial killers are good examples of violence and not just the ones that capture our attention. (Alternative example: We would save far more lives by preventing automobile deaths than aviation deaths, yet plane crashes capture more attention.)

        Provide evidence that serial killers were created by discipline. (The discipline itself is why they killed, not just the excuse they latched on to for self-justification.)
        Provide evidence that they would not have become serial killers without discipline. (Discipline is directly causal for the result.)
        Provide evidence that the absence of discipline would mean the absence of serial killers.

        Provide evidence that the reason that evil people face discipline when growing up is not correlated to their propensity to commit evil acts when they are adults.

        I’ve not seen a study that shows any controlled experiments, nor do I know how one would effectively test this specifically. I doubt one could identify two serial killers at birth and raise them in a discipline/non-discipline environments.

        However, I can point to violence against teachers by students and other anti-productive behavior. I’m willing to believe, until evidence proves otherwise, that in general, lack of discipline is not helpful.

        I suspect that students who observing and/or participating in violence are more likely to commit violence than students who do not. I suspect that what causes serial killers to act in the manner they do, is not applicable to the wider population. I suspect that establishing violence against teachers as a unpunished norm, is not a beneficial lesson.

    • I see what you’re saying, but it’s not as black and white as that. People (and animals) are not well understood enough to be treated as machines – each will have their own predispositions to different behaviours so treating any misdemeanour as a fault of the parents is too broad a generalisation. A set of incompetent parents may produce a well-behaved, conscientious child and very good parents may give birth to a wrong ‘un whom no amount of good parenting can fix.

      Taking your dog analogy: consider a spaniel and a labrador. If a pup of each is subjected to the same amount of training, when they’re adults the labrador will (generally) be much better behaved than the spaniel. They’re just (generally) born with different temperaments.

  2. See also:

    No, You Don’t Have Free Will, and This is Why

    Though it’s important to note that semi-irrational blind justice does serve a function: it serves as a deterrent to wrong doers. If you know that the aggrevied is likely to go bonkers in the event of a transgression against them, it makes you think twice about it.

  3. Carl Grover says

    I agree on many points here, yet don’t see any financially feasible remedy. Perhaps one step has been coining prisons as “corrections”, but that, to me, is a misnomer. Many people in the system are truly, and diagnosed, mentally ill. They far too often are likely held to the same behavior standards in prosecution, incarceration, and prisoner management. Where it gets tricky, is the poor. There are studies that show impoverished people have lower dopamine levels, and consequently, lowered inhibition to crime. If we are to label that as a disorder, we need to be able to identify that, and adjudicate based on it. You could likely say that about far too large of the prison and jail populations. I highly respect the seizure narrative, as that speaks volumes about our weakness for categorizing the unknown as criminal. We should constantly remain vigilant in avoiding that. However, like vaccines as an example, a certain percent will negatively affected by them, but the value to society, outweighs them. Grim, of course, but most reading this participate in them, despite the cost that small portion of the society has paid.

    If we switch our mindset away from punishment, that could potentially be semantical, unless there is broad change to adjudication and incarceration. Let’s say we call it “time out” and there are programs to identify illness, use mental or pysical care more appropriately, and seek to better rehabilitate based on those deficits – will this not now be the new “punishment”? Does anyone suppose this type of approach will yield better results for offenders, and society at large? We can hope, but it sounds like a costly endeavor.

    Overall, I feel like, even though we may focus on illness in this regard, we would still be treating symptoms. We need to get to the root cause – education, culture, and early illness identification, treatment, -and management.

    • I think there is a simple remedy. Stop attempting a 1-size-fits-all model to criminal justice. Legislate which crimes are misdemeanor, which are felony, and which are capital and then leave it at that. Stop adding in “x to y years and $$$ fine” as sentencing guidelines. Those tiers simply govern the trial rules like for a misdemeanor you can plea out to a judge imposed fine and have no attys, for felony you get a public defender, for a capital crime the State pays for a private atty for the indigent.

      Just like in civil trial, if found guilty then determine what makes society “whole” based upon the legislated severity of the crime. Juries do it all the time and can do it in the sentencing phase. Will some scream it isn’t fair? Absolutely, because two criminals, same crime, can get vastly disparate punishment — no different than the wide ranging awards in slip-falls or in wrongful death. Drug dealer kills another drug dealer?

      Part of the beauty of this system is that the “unknown sentence” is a deterrent. It’s the “would you like to plead guilty and make suggestions of your sentence? or leave it up to Dad to decide?” Of course, those suggestions could be “nah, not good enough….jury?” This allows a jury to also say “20 years in a mental health facility” for people like the S.C. church shooter too.

      Perfect? No. But it solves some of the issues in the current system where sentencing legislation was based upon worst-case strawman and there is a requirement for an actual mental diagnosis in order to sentence to a mental healthcare facility.

  4. Yandoodan says

    Everything old is new again. “Crime as disease” dominated criminal justice theory in the 60s and 70s. Curiously, it remained unchangeable dogma within the Carter Administration just experts within the field were rejecting it.

    And it wasn’t new then. C.S. Lewis gave a full-throated attack on it in his 1945 satire, That Hideous Strength. Prison punishment is limited by law, as to both length and conditions of incarceration. Imprisonment as cure is open-ended in length and its conditions determined by the doctors. Paraphrasing from memory, ‘Experiment on prisoners and all the Mrs. Grundys will attack you, but put ’em in an experimental prison and watch everyone approve.’

    One last thing. How does lack of free will fit in? Doesn’t it mean that doctors without free will are acting on people without free will? That neither has any claim to acting in the proper manner? That there is no such thing as a proper manner — only brute power? Shouldn’t it be clear that the doctors are merely the more powerful automatons?

  5. Victoria says

    Despite the insistence that the notion of “free will” has been disproven, the evidence in no way supports such a sweeping statement. Experiments like Libet et al., Soon et al., etc. use actions with no consequential significance, sometimes on neuroatypical brains. When the determinists can show that planning in advance of consequential outcomes and other higher level functions procede from unconscious, materially-determined processes, then they would have have even a basis for even beginning to suggest we lack free will.

    Even one of Lisbet’s own experiements showed an ability to consciously override the precognitive impulses. If that doesn’t qualify as “free will” then it just proves that the denialists use semantics to adavance their agenda.

    Sapolsky is just a testament to blind arrogance. I mean given the low quality of the evidence in question, the notion we should be discussing major societal-level structual reforms is obscene.

  6. Fabio says

    Didn’t Sam Harris make already *exactly* the same argument in his book Free Will? This author may have expanded the proposals on criminal justice reforms, but the main point is identical.

  7. augustine says

    Within this article is an example of what Scott Adams calls “thinking past the sale”. We are invited to think about the problems around behavior and punishment after a quote that handily dismisses any notion of sin or sinners. Will science and evolutionary theory move us beyond all our moral dilemmas?

    “… remove the punisher mindset from the criminal justice system.” I did not see anything in the article describing what punishment is. My understanding is that the incarceration itself is the “punishment” since it removes the offender from normal social connections. In theory there is no bloodlust or “punisher mindset” here on the part of victims or the legal/prison system. If one wishes, the whole process can be viewed in cold, pragmatic terms.

    Removing vengeful feelings from the criminal justice system would be a noble goal I think but now we are back to morality. Asking for such a transformation in the minds of victims and their families is really asking for them to adopt the mind of Christ. Again, noble, but this has little to do with science.

Comments are closed.