by Gwen van Eijk

As COVID-19 spreads through prisons, governments are focusing on public safety rather than the health and human rights of people in prison – and it is resulting in irresponsible and unfair policies.


Countries have responded to (expected) outbreaks of the coronavirus in prisons in various ways, for example by releasing individuals, enforcing isolation and suspending leaves and visits. Early April, the British Ministry of Justice announced to release 4,000 ‘low-risk prisoners’ from prisons in England and Wales – a total of five percent of the entire prison population – in order to combat the spread of COVID-19. However, just two weeks later, the plan had been put on hold, because six men, all described as ‘low risk offenders’, had been released by mistake. By that time, at least 15 prisoners and 5 prison staff in England and Wales had died of COVID-19 and another 321 prisoners and 293 staff had tested positive.

The above illustrates how governments are struggling to respond swiftly to the coronacrisis in prisons, as they are prioritizing public safety over the health of prisoners. In this blog I discuss how the pandemic is highlighting and exacerbating the problems of ‘risk management’.

Social distancing

In countries where prisons are overcrowded – among which the UK – reducing the number of incarcerated people is crucial: ‘social distancing’ behind bars is difficult, if not impossible. In many countries, people share their cell with others. Even in the Netherlands, where the incarceration rate is low, 30 percent of prison cells are usually occupied by two to six people, though cell-sharing has been reduced to 12 percent in the last two months. English and Welsh prisons would have to release 16,000 of 83,000 people – four times as many as was planned – to eliminate cell-sharing. 20,000 prisoners are especially vulnerable through old age or pre-existing health conditions.

Reducing the prison population is not just about protecting the health of prisoners. Prison staff are at a high risk of contracting the virus; many prisons do not provide masks or preventative testing. This puts families and communities at risk as staff move in and out of prisons daily. Freeing up space and resources in prisons could also make life a little easier for those still inside, which could help maintain peace (which is important, as protests and riots among prisoners have erupted worldwide).

Public safety

However, as authorities consider releasing people, they want to release only ‘low risk offenders’. In the US, this means excluding all ‘violent offenders’ from early release – ignoring the fact that in state prisons two out of three in this category are over age 55 and many have pre-existing health conditions. For example, Governor Tom Wolf of Pennsylvania announced to release 1,500 to 1,800 individuals from state prisons, but emphasized: ‘I have no interest – and I want to make this crystal clear – in releasing violent criminals from our system’. In New York City, high-level District Attorneys objected to plans of Governor Andrew Cuomo to release prisoners because it would include people sentenced for violent crimes.

Several US-based scholars have called for releasing ‘violent offenders’ because only by including them it will be possible to release a sufficient number of people. Moreover, they point out, recidivism is much lower than the socio-political discourse around ’violent offenders’ suggests. The problem is that ‘people convicted of violent crimes are in a special category that deserves less compassion and harsher treatment’, writes legal scholar John Pfaff. Excluding them from release not only ignores the facts, it ‘reflects a poor moral choice’ as it views those who have committed violence as ‘not fully human’.

Risk assessment

In addition to focusing on ‘non-violent offenders’, authorities promise to release only individuals that are considered ‘low-risk’. Sometimes ‘high-risk’ is just a synonym for ‘violent’. However, in some jurisdictions, authorities rely on risk assessment tools to evaluate an individual’s risk of recidivism. For example, the British Justice Secretary announced that no ‘high-risk criminals’ would be considered for release and that ‘all prisoners will face a tough risk assessment’. In the US, the release of elderly and sick prisoners from federal prisons will be informed by their risk score as assessed by the actuarial tool PATTERN. Texas (pdf) is also using risk assessment for decisions about early release and it is likely that other jurisdictions have similar guidelines.

Actuarial tools have informed decisions about release for decades, and ever since the rise of ‘actuarial justice’ in Europe, North America and elsewhere, concerns have been raised about class and racial/ethnic bias (I have written about socioeconomic bias in risk-based justice here and here). As incarceration affects the well-being and resources of individuals, the impact of bias extends beyond the decision about one’s liberty. These should be important reasons for authorities to refrain from relying on actuarial tools for decisions about release.

Concerns about biased risk assessment are even more urgent now that release decisions may be a matter of life and death. A coalition of 220 American organisations wrote to Attorney General William Barr, who is responsible for federal prisons, to express concern about the use of the tool PATTERN which is ‘scientifically unverified’ and biased against ‘Black people, Latino people, poor people, unhoused people, and people with mental illness’. For example: using PATTERN, ‘only 7 percent of black men in federal prisons would be considered low-risk enough to get out—compared with 30 percent of white men’.

Life and death

Meanwhile, as authorities are struggling to identify people to release ‘safely’, the prison population in most US states has remained stable, raising the question whether prisons will be releasing significant numbers at all – and whether by then it will be too late. As I am writing this, at least 373 American prisoners have died from COVID-19, while at least 25,239 prisoners have tested positive. Many of the ‘corona clusters’ in the US are in prisons.

In the UK, only 57 people have been released and the plan to release 4,000 has been ‘shelved’. By late April there had been 21 deaths in English and Welsh prisons; fortunately, much less than the number that was anticipated. However, since the outbreak of COVID-19, many prisoners have been locked up for 23 hours a day and put in isolation if they showed symptoms. While this has likely prevented the spread of the virus, enforcing ‘social distancing’ in prisons in this way leads to de facto solitary confinement. This raises concerns about humane treatment, especially if the restrictions would stay in place for the next 12 months, as seems to be plan.

The focus on public safety results in a slow, irresponsible and unjust response by authorities to the coronacrisis in prisons. It has been said often in the last months: the pandemic brings to light and exacerbates many social problems that already existed in society. Likewise in prisons, risk management is problematic in ‘normal’ times, as it encourages inhumane and unjust treatment of people in prisons, especially those who are vulnerable. The pandemic is making such problems a matter of life and death.

Note: Thanks to Professor John Monahan of the University of Virginia School of Law for sharing information about the use of risk assessment in corona-related release from US prisons. Those who read in Dutch may be interested in following my blog series on detention and corona on the blog Sargasso.