The Washington PostDemocracy Dies in Darkness

Denmark doesn’t treat its prisoners like prisoners — and it’s good for everyone

Prisoners prepare their own meals, wear their own clothes and leave each day. It's led to lower recidivism.


The most interesting thing about Scandinavian prisons? Many are barely prisons at all.

Our research team spent six weeks conducting intensive research in Danish prisons. We were struck by the sight of prisoners wearing their own clothes, cooking their own meals and having private family visits as often as once a week. At these “open” prisons, there are no barbed wire fences, solid walls with gun towers or secure perimeters. Instead, prison officials allow for the possibility that a prisoner might escape. Better that than a prisoner take a staff member hostage trying to breach a secure perimeter, they explain.

As we shared meals with prisoners and with staff and observed their days, we noticed something else more surprising: Danish prison officials assume some prisoners will get in fights, smuggle in drugs, commit suicide or escape. They expect and accept imperfections in their system. Their response to these unpleasant, unsanctioned events is rarely a crackdown, a drastic change in policy or even a public apology. For instance, even though prison officials know that drugs are smuggled into prisons through various body cavities, they do not conduct searches of prisoners or prison visitors. Prison officials prioritize dignity over implementing more stringent drug enforcement policies.

Likewise, when one prisoner stabbed another with a knife in a communal kitchen where prisoners prepare their own meals, prison officials responded by anchoring each knife to the wall with a foot-long steel cord. Then a prisoner used a vegetable peeler to stab another prisoner. So officials anchored the vegetable peeler to the wall, too. Prison officials prioritize prisoner independence over implementing stricter violence reduction policies, like removing the knives entirely.

One senior prison official in Denmark described a recent escape to us: A prisoner in the country’s highest security prison got access to a burqa from a visitor, put it on and walked out of the prison. He was never found. The prison official speculated that the escaped prisoner returned to his home country of Morocco. The media covered the escape, but the stories were funny. As the head of the prison guards union in Denmark said: “If you want to, you can escape … if you take that away … you will get a much harder prison.”

And, in truth, mistakes are rare. In 2014, only one prisoner escaped from a “closed” Danish prison with a secure perimeter. Of the 60,000 prisoner leaves granted each year in Denmark, only 3 percent either violate the terms of their leave or fail to return to prison.

Serious violence is also rare in Danish prisons. There were only three suicides and five other deaths throughout the Danish prison system in 2013, as compared to 4,446 deaths in U.S. jail and state prison facilities that same year. (These numbers also speak to the disparate scales of the prison systems; in both countries fewer than 1 percent of the prison population dies each year).

Our research suggests that prison staff do indeed feel safe in these environments, as imperfect as they are.  Recidivism is also relatively low among released Danish prisoners, hovering around 27 percent, half of the average recidivism rates reported across various U.S. jurisdictions.

In spite of low violence and low recidivism rates, the Danish prison system grapples with both ethnic inequities and human rights abuses. About 40 percent of prisoners in Denmark are not ethnically Danish; this is almost four times the percentage of non-Danes in the general population. And Danish prisons, much like U.S. prisons, have faced criticism for being too quick to put prisoners in solitary confinement for extended periods of time.

Still, the value of Denmark’s example to a reform-minded public lies not in replicating its particular strategies or techniques but in adopting its broader ethos — one that grants prisoners dignity and allows room for error.

This is a lesson that the United States needs to learn. The United States has the highest incarceration rate in the world, yet we have little to show for either the money invested or the lives lost to this system. U.S. prisoners wear anonymous facility garb, eat mass produced food in assembly cafeteria lines, and spend hours on end in tiny, bleak cement cells. As President Obama noted this past week, as many as 100,000 prisoners across the United States are housed in solitary confinement. Hundreds of these prisoners are released directly to the streets every year, often with dangerous consequences: two went on shooting rampages upon release in 2013.

Officials say a zero tolerance policy is the only way to ensure safety in a facility full of felons. But in reality, such policies do little. Prisoners use drugs, escape and recidivate. In spite of invasive search routines for prisoners and visitors alike, prisons across the United States report problems with contraband from drugs to cellphones to prison-made knives. Even though U.S. prisoners are not permitted to have knives or prepare their own food for safety reasons, in 2011 the Supreme Court found that one California prisoner died unnecessarily every week — lives lost not to violence, but to medical negligence. And when a prisoner escaped from Clinton Correctional Facility in New York in June 2015, more than 60 prisoners complained of a backlash of abusive beatings. Danish prison officials say that their prisoners act out less because they are treated humanely; they, too, are allowed to make mistakes.

Remedying contraband problems, health-care failures, violence and abusive practices will require more than a value shift toward rehabilitation, normalization or dignity. To roll back mass incarceration and humanize our prisons, as a nation we’re also going to have to start getting comfortable with failure.

Our research was supported by the National Science Foundation under Grant Nos 1455971, 1455413, 1455091. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.