← Return to search results
Back to Prindle Institute

Implications of Exonerations

black-and-white photograph of empty jail cell

I expect that in the near future we will know, for certain, that at least one innocent person has been executed in the United States. This should not come as a surprise. There have been many cases where those on death row are found to be innocent; indeed by some estimates more than 4% of death row inmates may be innocent. One major reason we have not previously proven someone’s innocence is that there is rarely the political will to continue investigating post execution.

What I want to investigate in this piece is what this should mean for the use of the death penalty. The BBC, in its Ethics Guides, notes that the “most common and most cogent argument against capital punishment is that sooner or later, innocent people will get killed, because of mistakes or flaws in the justice system.” But just how cogent an argument is this?

Let me lay my own cards on the table; I am emphatically anti-death penalty. I think the unnecessary killing of any human being is monstrous, and that the state may only use lethal force when combating an active threat, not as punishment for a past threat. However, I would hold this view even if there was no chance of an innocent person ever being executed; my objection is grounded in an invariant pro-life commitment. Thus, I think I am particularly well placed to assess the ethics of this question. I’m not in favor of capital punishment, and so am not looking for any rationalization to dismiss the argument from innocent execution; but nor do I feel a rationalizing compulsion to think the argument works — my own commitments would persist regardless. Of course, this all may just be a second-order rationalization about my own objectivity, but I’ll put aside worries about meta-level rationalizations for another post.

So here is a simple argument from the possibility of executing the innocent to the wrongness of the death penalty:

  1. If we use the death penalty, eventually an innocent person will be killed. (innocence premise)
  2. Executing the innocent is always morally wrong, no matter how good the consequences. (absolutism premise)
  3. Therefore, if we use the death penalty, we will eventually do something wrong that cannot be justified by the goods of executing the guilty.

Now, there is a certain intuitive plausibility to this argument. Suppose I thought that the death penalty has a strong deterrent effect, and so executing the guilty would save many lives (see this article for a defense of this argument; note there have been, as far as I can tell, several persuasive replies); I still would not think it is permissible to frame an innocent person and execute them to get that same deterrent effect. Thus, it is wrong to kill the innocent, even to bring about the valuable execution of the guilty.

The problem with this argument, is that, by parity, it seems to suggest we should never punish anyone:

  1. If we imprison people, eventually an innocent person will be imprisoned. (innocence premise)
  2. Imprisoning the innocent is always morally wrong, no matter how good the consequences. (absolutism premise)
  3. Therefore, if we imprison people, we will eventually do something wrong that cannot be justified by the goods of imprisoning the guilty.

This argument, as far as I can tell, has the same surface level plausibility as the previous one. If I knew the only way to capture a murderer was to, for some reason, imprison an innocent person for life, it would not be permissible to imprison that person.

So what has gone wrong with these two arguments? I think it is that there is a difference between killing or imprisoning the innocent as an intended means to punishing others, and foreseeing that the innocent may be killed or imprisoned as a consequence of a broader policy of punishment. While there may be an absolute prohibition on killing or imprisoning the innocent, that does not mean there is a prohibition on anything that could lead to that as a result.

To articulate the innocence argument against the death penalty, one needs to show that a) the fact an innocent person will be killed means we cannot use the death penalty, and b) the fact an innocent person will be punished does not mean we cannot use any schema of punishment (since I assume few people would accept that the inevitable punishment of the innocent means all punishment is unjust).

In fact, the problem for the innocent argument is even more profound. Remember that study I cited at the beginning, saying that as many as 4% of death row inmates are innocent? The way that study worked is that they compared the exoneration rate of those who stayed on death row (about 4%) to those who were shifted to life in prison. Because more scrutiny is given to death penalty cases (especially as the guilty approach execution), those who are actually to be executed (and not just on death row) are more likely to be exonerated. So, the study most commonly cited to support the claim that some innocent will be executed, actually shows that likely more innocent people would be punished if we switched from the death penalty to life in prison (since we are less likely to identify the innocent without the scrutiny provided to death penalty cases).

So, what can we point to in order to say that executing the innocent is uniquely bad — bad in a way that imprisoning the innocent is not?

Barbarity. Perhaps the thought is that the death penalty is, in some way, so much worse than life in person, that we cannot take any risk with the innocent being killed, even if we can take risks on an innocent person going to jail for life. But I’m not sure this is quite right. In other parts of life, we don’t treat risks of death as categorically worse than other risks. Anytime you drive a car there is a disturbingly high chance an innocent person might die (far more innocent people die in car accidents than are executed); but I don’t think we treat such risks as categorically distinct from other types of risks. You might, for instance, choose a small risk of death to avoid a much larger risk of having to spend the rest of your life locked in prison.

Irreversibility. The first thing we might suggest is that the death penalty is irreversible. If you kill someone you cannot bring them back to life, but if you imprison someone you can always let them go later. But it is not quite that simple. For one thing, you really cannot reverse a prison sentence. Even if you are eventually released, you do not get back those twenty years spent in prison. The punishment cannot be reversed, all we can really do is shorten it if we discover you are innocent.

Permanency. Is that the solution then? Is the reason the death penalty is so bad because it is permanent? Perhaps the thought is that if we cannot be certain someone is guilty, no permanent punishment is justified. But here, again, this does not seem quite right. After all, I still permanently lose my twenties and thirties to prison, even if I get released on my fortieth birthday.

And note too, the risk of permanence is not the same thing as permanence. Just because someone could be released does not mean they will. And we know that, since executions receive greater scrutiny, an innocent person is more likely to be sentenced to life in prison than sentenced to death.

Reparability. Perhaps it is not that the death penalty is permanent, but that it cannot be repaired. Sure, someone imprisoned till they are forty permanently, loses out on their thirties, but at least the state can do something to make it up to the person falsely imprisoned. For example, often those falsely imprisoned are given financial compensation from the state.

I actually find this argument somewhat convincing, but we have already seen that as a society we don’t accept the broader implication. After all, it is death, not execution, that is irreparable. But we have already seen that we don’t treat risks of death as categorically different from other types of serious risks. If car accidents just resulted in serious bodily injury, we could imagine making some reparation for such injury. Since they sometimes result in death, sometimes no such reparation is possible. But, again, it seems we don’t see that fact as particularly dispositive when assessing the ethics of driving.

It is a terrible and tragic thing to execute the innocent. But, I think, that is just a subset of the terrible and tragic thing that it is to punish the innocent. Perhaps we should strengthen our criminal standards for conviction (I’m quite sympathetic to that line of thought) so that fewer innocent people are punished. But I’m not sure. At the very least, if we accept as inevitable that the innocent will be punished it gives us a categorical reason to select some punishments over others.

Wildfires and Prison Labor: Crisis Continues to Expose Systemic Inequity

photograph of lone firefighter before a wildfire

As around a dozen wildfires continue to grow in California, the smoke has reached Nebraska. The two major wildfires that are occurring in Northern California are the second- and fourth-largest fires in state history. The status of Big Basin Redwoods State Park, California’s oldest, is changing daily. The oldest trees have seen many fires, but the current threat has been particularly devastating. California Governor Newsom has asked for help from as far away as the east coast of Australia in order to gather more firefighters.

Reaching out so broadly is a result of how central the issue of sufficient firefighters has become. State prison officials shared at a press briefing that due to COVID-19 quarantining and early release measures, California was unable to use its usual contingent of incarcerated firefighters during this year’s wildfire season.

The failure to address land management issues and the increasingly dire effects of climate change have led to the disastrous fire seasons both this year and in the recent past. However, the reliance on dangerous work being done by underpaid and under-protected incarcerated people in order to ensure the safety of others and conserve precious resources is a part of a systemic trend.

The US has more people in prison — in absolute numbers as well as by percentage of the population — than any other country on earth. This statistic alone should give us pause and encourage us to reflect on the purpose of isolating such a large amount of our population. But this year, our handling of the pandemic and, now, these wildfires highlights further issues with mass incarceration, beginning with the justification for imprisoning so many members of your population in the first place.

Society could be aiming at a few different goals when it takes people from society and places them in prison for violating the law. One goal might be sanctioning citizens that have “harmed” society based on a somewhat loose notion of “just deserts”: the individual has done wrong so they deserve punishment, and isolation is seen as the appropriate form of that punishment. Other justifications of incarceration as punishment are based on deterrence: by isolating someone who violates the law, we hope to make it less likely that this person or others — who are aware of the incarcerating policy — will do so again in the future. Incarceration could also be construed as means to rehabilitate someone who has not performed to the standards that the law suggests society deems necessary. In this case, the isolation is supposedly meant to be a constructive time to become able and willing to conform to societal standards more adequately in the future. Finally, incarceration could be a way of isolating someone from society to prevent further harm. (A version of justice that doesn’t fit this model is “restoration,” which focuses on the effects of violating a statute and allows those harmed by the violation to initiate a process where there is opportunity to share concerns, make amends and future plans, and potentially forgiveness, in hopes of healing the part of society that was in fact impacted by the violation).

Most regard the standards for our penal system to be a mix of these goals. A prison sentence might make sense as a mix of deterrence and rehabilitation in a particular case of sentencing, or in the mind of a particular legislator. When considering the labor that those who are serving sentences in prison perform, however, the justification for their punishment plays a crucial role in determining what conditions are appropriate.

Those in favor of the starkly different working conditions for non-incarcerated employees and prison labor use a variety of explanations. Appeals to the need to maintain facilities and rationale of providing job training for people who eventually will “reenter society” are the strongest justifications for employing incarcerated people. However, from these considerations, the working conditions and pay structure that exist today in US prisons do not follow. From the justifications for incarceration as a form of punishment, it is unclear why the human rights protections that guarantee safe working conditions and fair wages would be forfeited along with the freedom of movement that the punishment itself constitutes. For rehabilitation purposes, working while in prison can aid the transition once released, but differentiated pay scales and lax safety protocols appear punitive and demand human rights attention. For the retributive (“desert”) model, the presumption that someone deserves worse or inadequate working conditions on the basis of being incarcerated would need to be considered along with their original sentencing.

Further, the working conditions and wages that make up the structure of prison labor create a market designed to exploit the incarcerated. People in government-run correctional facilities perform jobs that are necessary for the prison to function, but do so without labor protection or the possibility of unionizing. And they do so for a fraction of what those performing the same tasks outside of prisons would earn. As a prison laborer, often the guarantee of safe working conditions simply does not apply.

The disparities in pay and working conditions for incarcerated employees creates an exploitative market, where prisons are incentivized to keep costs low and private businesses can reap great benefit. This economic structure does not simply exploit a marginalized incarcerated population, but, given the structural racism in the US justice system, further imbalances racial inequalities in a society already saturated by racist institutions.

Further, there are a variety of incarcerated employees who perform manufacturing jobs outside the prison, manufacturing products that are sold to government agencies and corporations. This work can include answering calls in a phone bank, constructing furniture, warehouse work, farm work, and, in California, front-line firefighting. The fire season and the pandemic have laid bare the living and working conditions in our prisons.

Throughout the pandemic, the cramped living conditions and poor quality of healthcare have put prisons at particular risk of experiencing a COVID-19 outbreak. Calls to attend this heightened danger have been neglected throughout the national emergency. In California, the nature of the pandemic has led to shutdowns across many of the prisons, making the incarcerated firefighters unable to respond to fires.

What this has meant for this record-breaking year in California is that they cannot rely on the state’s “primary firefighting ‘hand crews.’” According to the Sacramento Bee, “Inmate crews are among the first on the scene at fires large and small across the state… Identified by their orange fire uniforms, inmates typically do the critically important and dangerous job of using chainsaws and hand tools to cut firelines around properties and neighborhoods during wildfires.”

The owners of the prisons actively market their workers to private businesses, emphasizing the low wages their employers would be able to pay, how many have “Spanish language skills,” and how this labor pool is one of the “best kept secrets.”

In 2018, there was a three-week nationwide strike over the work conditions in prisons. Prisons are paying incarcerated people less today than they were in 2001. Prison jobs are unpaid in Alabama, Arkansas, Florida, Georgia, and Texas, and maximum wages have been lowered in at least as many states. Further, in many states the wages that incarcerated employees earn do not accurately reflect “take-home” pay; prisons deduct fees (like garnished wages) for what prison laborers have cost them during their stay. Because it costs money to incarcerate people, the claim goes, their wages should contribute to the running of the prison. As Vox reports, “Most prisons also deduct a percentage of earnings to help cover a prisoner’s child support payments, alimony, and restitution to victims. But at 40 cents an hour, that seems impractical.”

There is no getting around the fact that hiring incarcerated employees is a cost-saving measure. It’s estimated that the “Conservation Camp Program, which includes the inmate firefighters, saves California taxpayers tens of millions of dollars a year. Hiring firefighters to replace them, especially given the difficult work involved, would challenge a state already strapped for cash.” As the Managing Editor at Prison Legal News told Newsweek, “Prisons cannot operate without prison labor. They would simply be unaffordable.” So, incarceration in its current state requires exploitation and unethical labor practices that we wouldn’t accept outside of prisons, and are inflicting disproportionately based on systemically racist institutions.

There is also the option that doesn’t seem to get enough attention: If incarceration costs so much, wouldn’t it be cheaper to have less incarceration?

The Complexities of Reforming Indiana’s Bail System

This article has a set of discussion questions tailored for classroom use. Click here to download them. To see a full list of articles with discussion questions and other resources, visit our “Educational Resources” page.


Every year, thousands of bills are written and proposed during Indiana’s legislative session. The Indiana General Assembly takes place during the first few months of the year, and is a chance for state representatives to advance their agenda. Many Americans pay more attention to what happens at the federal level, but state and local government also has a large influence on the lives of citizens. The 2017 session, Jan 3 through April 29, is taking place during a budget year, and in the wake of an extremely contentious and important state and national election. Legislation authored this session ranges from bills that deregulate environmental protection to resolutions aimed at honoring professional athletes. One bill that has not gained much attention, however, raises numerous ethical concerns in regards to criminal justice and the prison system.

Senate Bill 228, authored by Senator Michael Young, aims to reform Indiana’s approach to bail and release after arrest. This bill involves the rules regarding pretrial risk assessment system which assist courts in assessing an arrestee’s likelihood of: (1) committing a new criminal offense; or (2) failing to appear.

At first glance this bill doesn’t seem particularly unusual. After all, using risk assessment in the criminal justice system doesn’t sound particularly radical or unethical. However, a closer examination of the bill reveals a large ethical dilemma. According to S.B. 228, the Indiana Supreme Court will revolutionize the standards for bail and release. Instead of basing bail and release off of a standard punishment for crime, it will rely upon risk assessment “based on empirical data derived through validated criminal justice scientific research” regarding individuals and the groups to which they might belong.

Proponents of the bill would argue that this not only humanizes those arrested on bail but also saves taxpayers money. Ideally, fewer people would be stuck waiting for their bail to be posted or for their trial, and therefore less taxpayer money would be dedicated to detaining those arrested. It is estimated that at any given time, there are nearly half a million Americans detained in local jails awaiting their trial, which costs approximately $17 billion every year. Many also argue that the notion of bail is outdated, and inherently favors wealthy individuals over poorer ones, further reinforcing societal inequalities surrounding income. Evidence-based risk assessment has been implemented in Kentucky, and supporters point to the fact that the average arrest rate for released defendants has declined. Additionally, many legislators are aiming to improve the reputation of the United States, which has one of the highest prison populations in the world, based on the fact that this bill would most likely result in less pretrial prisoners.

But will abolishing bail and relying upon risk assessment truly improve the stark inequalities present in the criminal justice system? Risk assessment aims to allow those who are detained for non-violent crimes and are not repeat offenders out of jail before their trial. But what are the complications of individualizing criminal offenses? Though fewer people would ideally be sitting in jail awaiting pretrial, those who are detained may be treated differently than those who committed the same crimes, or even more serious ones. If two people are arrested for the same crime, shouldn’t they be treated the same regardless of differences in criminal record and history?  

Factors such as race, criminal arrest record, or even gender could influence how risk assessment is measured and change how two people who commit the same crime are treated. A report on bail and pretrial risk assessment admits that “researchers have documented that racial bias can influence how juvenile offenses are described in post-arrest narrative reports, which could influence pretrial release decisions.” Though one could argue that educating law enforcement officials about implicit bias could eliminate this problem, S.B. 228 does not encourage or mandate doing so. Additionally, there has been increased debate about whether or not crime statistics can stand alone if they do not take into account racial and socioeconomic inequalities.

Though controversial, S.B. 228 passed its Senate Committee vote 8-1, passing an amendment on February 19. Though it is too early to tell if S.B. 228 will make it through the legislative process and become law, it should be considered seriously. The ethical implications of transforming pretrial requirements to individual considerations as opposed to a standard should not be taken lightly.