What Is Prison Abolition and How Would It Work?

Prison abolition is a political movement that aims to gradually eliminate prisons and jails by building community-based ways of handling harm and by shrinking the conditions that funnel people into cages in the first place. It is not a plan to open every cell tomorrow. It is a long-term project that pairs decarceration — fewer people locked up, fewer facilities, shorter sentences — with investment in the services that make incarceration unnecessary: housing, mental health care, addiction treatment, crisis response, and community accountability processes that address harm without a prosecutor.

Understanding the movement means understanding both halves. One half is a critique of why reform alone cannot work. The other half is a set of concrete practices, some already operating at scale, that show what a smaller carceral footprint can look like.

What Abolitionists Actually Mean

No serious abolitionist argues for releasing every incarcerated person overnight. The framework describes a long-term process of building alternatives that make incarceration unnecessary for more and more categories of conflict, while simultaneously shrinking the physical and legal capacity of the carceral state. The guiding question is not “what do we do with people in prison right now” but “what would we need to build so that prisons become obsolete over time.”

That gradualism is often confused with ordinary criminal justice reform. The distinction turns on a concept abolitionists call non-reformist reforms, a term that originated with French philosopher André Gorz in the 1960s and that Ruth Wilson Gilmore brought into abolitionist thought. A reformist reform eases suffering inside the system without challenging the system’s size or existence. A non-reformist reform reduces the system’s power and reach. Adding air conditioning to a jail is reformist: the cages get more comfortable, but the cages remain. Ending cash bail is non-reformist: it closes a mechanism that feeds people into pretrial detention.

The distinction matters because abolitionists argue that well-intentioned improvements often become permission slips for expansion. A newer facility with better lighting and mental health pods is still more bed space, and more bed space tends to get filled. The U.S. prison population roughly quintupled between 1980 and its peak around 2009, even as individual facilities were marketed as upgrades over the ones they replaced. For abolitionists, that pattern is the point: a system that grows in response to its own reforms cannot be reformed into something small.

How It Would Work in Practice

Abolition operates on two tracks at once. The first is decarceration, the practical work of releasing people and reducing the inflow. The second is building alternatives, so that fewer disputes, crises, and harms are routed into the legal system to begin with.

Transformative Justice

Transformative justice is a community-driven approach to addressing harm that deliberately avoids police, prosecutors, and courts. Where the legal system asks what law was broken and how to punish the person who broke it, transformative justice asks what conditions allowed the harm to happen and how to change those conditions so it doesn’t happen again. The framework treats harmful behavior as a symptom of structural failures — housing instability, untreated trauma, addiction, poverty — rather than as evidence of individual moral deficiency.

In practice, this takes the form of community accountability processes. Organizations like Creative Interventions and the Building Communities, Ending Violence project at DePaul University have developed models in which people connected to both the person who caused harm and the person who was harmed work through a conflict together. The group identifies concrete steps: stopping the behavior, taking responsibility for its impact, respecting the wishes of the person who was hurt, and getting support for underlying issues. Participants sometimes draft written agreements. These carry no legal weight, but they rely on something that often proves more durable than a court order: the pressure of relationships with people who know you.

The model has real limits that honest advocates name. One is the collision with mandatory reporting laws. Social workers, therapists, teachers, and medical professionals in every state are legally required to report certain types of harm to authorities. A transformative justice process that involves a counselor hearing about child abuse or domestic violence puts that professional in an impossible position: participate in the community process and risk losing their license, or report to the state and collapse the framework. There is no clean resolution. The second limit is sustainability. Accountability processes can stretch over months and depend on volunteers. Participants burn out. People move. A person who causes harm may disengage, and a neighbor has no enforcement mechanism beyond social pressure. These are not reasons to dismiss the model, but they explain why it currently operates at a small scale and works best in tight-knit communities with existing trust networks.

Restorative Justice

Restorative justice overlaps with transformative justice but differs in one critical way: it often operates within or alongside the existing legal system rather than outside it. The core mechanism is victim-offender mediation, where a trained facilitator brings together the person who was harmed and the person who caused the harm for a structured conversation. The person who was harmed can ask questions, describe the impact, and state what they need. The person who caused the harm has to sit with that impact directly, instead of watching a prosecutor handle it from across a courtroom.

The outcomes data is strong. Research published by the Administrative Office of the U.S. Courts found that roughly 80 to 90 percent of mediation sessions that reach a meeting produce an agreement, and roughly 80 to 90 percent of those agreements are completed. In a multi-state study, 79 percent of crime victims who went through mediation reported satisfaction with how the system handled their case, compared to 57 percent of similar victims who went through conventional court proceedings.1United States Courts. The Impact of Victim-Offender Mediation: Two Decades of Research Those numbers push against a common objection — that victims prefer punishment. The research suggests many victims prefer answers and accountability more.

Restorative justice also shows measurable effects on reoffending. Controlled evaluations across multiple jurisdictions found reoffense rates roughly one-third lower among participants than among similar non-participants, and subsequent offenses tended to be less severe. Adult program studies suggest recidivism reductions in the range of 10 to 25 percent compared to conventional processing. Completion of a restorative agreement frequently results in dismissed charges or reduced sentences, giving participants a direct path to repairing the harm without a prison term or a permanent record.

Legal Levers Available Now

Abolitionists distinguish between the long-term vision and the immediate work of getting people out of cages. Decarceration is the practical side of that work, and it runs through legislatures, courts, and sentencing policy. Four levers do most of the work.

Bail Reform

Cash bail is one of the clearest examples of wealth determining freedom. Someone who can post bail goes home to await trial; someone who cannot stays locked in a jail cell for weeks or months before any conviction. Federal law already prohibits judges from imposing financial conditions that result in pretrial detention.2Office of the Law Revision Counsel. 18 USC 3142 – Release or Detention of a Defendant Pending Trial State systems have historically relied heavily on money bail, and the daily cost of holding someone pretrial runs roughly $85 per person for basic needs, with costs rising sharply for people with mental health conditions. Illinois became the first state to eliminate money bail entirely in 2023, and several other states have significantly restricted its use.

Mandatory Minimums and Sentencing Reform

The Comprehensive Crime Control Act of 1984 created the United States Sentencing Commission and introduced mandatory minimum sentences for drug offenses and firearm crimes, sharply limiting judges’ ability to tailor sentences to individual circumstances.3Congress.gov. S.1762 – Comprehensive Crime Control Act of 1984 The result was a regime in which a low-level drug courier could receive the same prison term as a cartel organizer.

The First Step Act of 2018 rolled some of this back at the federal level. A 20-year mandatory sentence dropped to 15 years for offenders with one prior qualifying conviction, and a life sentence dropped to 25 years for those with two or more. The law expanded the “safety valve” so that more low-level, nonviolent drug offenders could receive sentences below the mandatory floor, and it eliminated the practice known as “stacking,” where prosecutors could impose a 25-year mandatory sentence for a second firearm offense even when both offenses were charged in the same case. The law also made the Fair Sentencing Act of 2010 retroactive, allowing courts to resentence people convicted under the old crack cocaine disparity.4Congress.gov. The First Step Act of 2018: An Overview Separately, it created a system of earned time credits for incarcerated people who complete evidence-based recidivism reduction programs, with credits applying toward earlier transfer to supervised release or prerelease custody.5Office of the Law Revision Counsel. 18 USC 3632 – Development of Risk and Needs Assessment System Abolitionists view the First Step Act as useful but modest: it affects only the federal system, which holds a small fraction of the total incarcerated population, and it excludes people convicted of dozens of enumerated offenses from earning credits.

Compassionate Release

Federal law allows courts to reduce a sentence when “extraordinary and compelling reasons” justify it, or when the incarcerated person is at least 70 years old, has served at least 30 years, and has been determined by the Bureau of Prisons not to be a danger to the community. Before the First Step Act, only the Bureau of Prisons could initiate a compassionate release motion. The 2018 law changed that, allowing incarcerated people to petition courts directly after exhausting administrative remedies or waiting 30 days from the date they submitted a request to their warden.6Office of the Law Revision Counsel. 18 USC 3582 – Imposition of a Sentence of Imprisonment That procedural change matters because the Bureau of Prisons had historically been a bottleneck, denying or ignoring requests from aging and terminally ill prisoners.

Closing Facilities

Decommissioning prisons is the most concrete form of decarceration because it physically removes the capacity to incarcerate. When a facility closes, the state transfers or releases the remaining population, reassigns staff, and repurposes or demolishes the building. Some legislators have attached sunset provisions to prison funding to prevent facilities from being quietly reopened after a political shift. The logic is simple: as long as empty beds exist, the system tends to find bodies to fill them.

Divesting From Prisons, Investing in Alternatives

Abolitionists frame divestment from prisons and reinvestment in social infrastructure as two halves of the same strategy. The median cost of incarcerating one person for a year across all states is roughly $61,000, with individual states spending anywhere from under $20,000 to well over $100,000. That money buys a locked cell, meals, guards, and minimal programming. The same amount directed toward housing, mental health care, or job training addresses the conditions that lead to incarceration in the first place.

Crisis intervention teams offer one of the clearest working examples. The CAHOOTS program in Eugene, Oregon, has operated for more than 30 years, dispatching medics and crisis counselors to calls involving mental health emergencies, substance use, and homelessness instead of armed police. In 2021, the program handled over 16,000 calls that would otherwise have gone to police, saving the city an estimated $2.2 million per year. The teams carry no weapons and make no arrests. Programs modeled on CAHOOTS have since launched in cities across the country, reflecting a growing recognition that a person in psychiatric crisis does not need someone with a gun and the power to arrest. They need someone trained to de-escalate and connect them to care.

Housing programs follow a similar cost logic. Permanent supportive housing for people experiencing chronic homelessness typically costs far less per person per year than incarceration, while also reducing emergency room visits, psychiatric hospitalizations, and repeat arrests. Addiction treatment and community health clinics serve the same preventive function. None of these alternatives require the abolitionist label to justify themselves on the merits. The abolitionist contribution is connecting them into a single framework: each dollar moved from cages to care is a step toward making the cages unnecessary.

The Hard Question: Serious Violence

Every conversation about abolition eventually arrives at the same place: what about people who commit murder, rape, or other serious violence? Abolitionists take this question more seriously than critics assume, though the answers are less tidy than a sentencing table.

The first response is statistical. The majority of incarcerated people are not serving time for the most serious violent offenses. Drug crimes, property offenses, technical probation violations, and low-level assaults account for a large share of the prison population. Decarceration could proceed enormously far before ever reaching the cases that dominate public fear. Legal scholar Allegra McLeod has argued that the question of the “dangerous few” can largely be deferred because meaningful decarceration would, by political necessity, only proceed gradually.

The second response targets root causes. Research consistently links concentrated violence to concentrated poverty, lack of economic opportunity, untreated mental illness, and participation in criminalized markets like drug trafficking. A society serious about reducing violence has to be serious about reducing inequality, funding mental health care, and creating legal livelihoods for people currently surviving through illegal ones. Decriminalizing certain drug markets, for instance, would eliminate much of the violence associated with the drug trade without putting anyone in a cell.

The third response is the most uncomfortable for advocates and critics alike. Even within abolitionist scholarship, there is an acknowledgment that some form of restrictive intervention may remain necessary for people who pose imminent, ongoing danger to others. Angela Davis framed abolition not as “immediately eliminating the ability to invoke the criminal process in certain instances of grave interpersonal harm” but as a project that strives to make punitive responses unnecessary while approaching their use “with deep conflict and ambivalence, even shame.” That is not a dodge. It is an admission that abolition is an aspirational framework operating within real constraints, not a blueprint that resolves every edge case on day one.

What Stands in the Way

Even where political will exists to reduce incarceration, structural features of the current system resist change. Roughly three-quarters of private prison contracts include minimum occupancy clauses requiring the state to keep between 80 and 100 percent of beds filled. If the incarcerated population drops below the threshold, the government pays a penalty to the corporation for each empty bed. These clauses create a perverse financial incentive: the state loses money when fewer people are locked up. Companies like GEO Group and CoreCivic have built their business models around guaranteed demand, and their lobbying operations work to maintain it.

Prison labor is another entrenched interest. Incarcerated workers across the country often earn below a quarter per hour for regular institutional assignments, with averages around $0.14 to $0.63 per hour for non-industry jobs and $0.33 to $1.41 per hour for prison industry work. Some states pay nothing at all. The Thirteenth Amendment’s punishment clause, which permits involuntary servitude “as a punishment for crime,” provides the constitutional basis for this arrangement.7Congress.gov. U.S. Constitution – Thirteenth Amendment This labor functions as a subsidy to the state and to private industry, creating economic stakeholders with a direct interest in maintaining the incarcerated population.

Rural communities that depend on prisons for employment present a different kind of barrier. When a correctional facility is the largest employer in a county, closing it threatens the local tax base, school funding, and the livelihoods of corrections officers, administrative staff, and the businesses that serve them. Any serious decarceration strategy has to include an economic transition plan for these communities, and the absence of such plans has historically been one of the most effective arguments against facility closures.

Common Criticisms

Prison abolition faces criticism from both law-and-order conservatives and progressive reformers who share part of the diagnosis but reject the conclusions. Three objections come up most often.

The deterrence argument holds that fear of imprisonment keeps people from committing crimes, and that removing the threat would embolden harmful behavior. Abolitionists counter that deterrence research shows diminishing returns: severity of punishment matters far less than certainty of consequence, and the extremely high incarceration rates in the United States have not produced correspondingly low crime rates compared to countries that imprison far fewer people.

The public safety objection is more visceral. People who have been victimized by violence often want the person who harmed them physically separated from the community, and telling a survivor of assault that their attacker should participate in a community accountability circle rather than go to prison can feel like a second violation. Abolitionists respond that the current system fails most victims too: the majority of violent crimes never result in an arrest, let alone a conviction, and incarceration without treatment or accountability does not prevent reoffending after release. That response, however accurate, does not fully resolve the tension. The movement’s relationship with victim advocacy remains one of its most contested internal debates.

The feasibility critique is perhaps the most practical. Transformative and restorative justice processes require community infrastructure, trained facilitators, and sustained participation that does not exist in most neighborhoods. Scaling a model that works in a tight-knit activist community to a city of millions is an unsolved problem. Critics argue that abolitionists are better at describing the world they want than at building a credible path from here to there. The strongest abolitionist response is not theoretical but demonstrative: programs like CAHOOTS, restorative justice diversion courts, and community violence intervention initiatives are producing real results at growing scale, even if they remain far from universal.