The Supreme Court case that could turn homelessness into a crime, explained

Grants Pass v. Johnson could make the entire criminal justice system far crueler. It also tests the limits of judicial power.

The Supreme Court will hear a case later this month that could make life drastically worse for homeless Americans. It also challenges one of the most foundational principles of American criminal law — the rule that someone may not be charged with a crime simply because of who they are.

Six years ago, a federal appeals court held that the Constitution “bars a city from prosecuting people criminally for sleeping outside on public property when those people have no home or other shelter to go to.” Under the United States Court of Appeals for the Ninth Circuit’s decision in Martin v. Boise, people without permanent shelter could no longer be arrested simply because they are homeless, at least in the nine western states presided over by the Ninth Circuit.

As my colleague Rachel Cohen wrote about a year ago, “much of the fight about how to address homelessness today is, at this point, a fight about Martin.” Dozens of court cases have cited this decision, including federal courts in Virginia, Ohio, Missouri, Florida, Texas, and New York — none of which are in the Ninth Circuit.

Some of the decisions applying Martin have led very prominent Democrats, and institutions led by Democrats, to call upon the Supreme Court to intervene. Both the city of San Francisco and California Gov. Gavin Newsom, for example, filed briefs in that Court complaining about a fairly recent decision that, the city’s brief claims, prevents it from clearing out encampments that “present often-intractable health, safety, and welfare challenges for both the City and the public at large.”

On April 22, the justices will hear oral arguments in City of Grants Pass v. Johnson, one of the many decisions applying Martin — and, at least according to many of its critics, expanding that decision.

Martin arose out of the Supreme Court’s decision in Robinson v. California (1962), which struck down a California law making it a crime to “be addicted to the use of narcotics.” Likening this law to one making “it a criminal offense for a person to be mentally ill, or a leper, or to be afflicted with a venereal disease,” the Court held that the law may not criminalize someone’s “status” as a person with addiction and must instead target some kind of criminal “act.”

Thus, a state may punish “a person for the use of narcotics, for their purchase, sale or possession, or for antisocial or disorderly behavior resulting from their administration.” But, absent any evidence that a suspect actually used illegal drugs within the state of California, the state could not punish someone simply for existing while addicted to a drug.

The Grants Pass case does not involve an explicit ban on existing while homeless, but the Ninth Circuit determined that the city of Grants Pass, Oregon, imposed such tight restrictions on anyone attempting to sleep outdoors that it amounted to an effective ban on being homeless within city limits.

There are very strong arguments that the Ninth Circuit’s Grants Pass decision went too far. As the Biden administration says in its brief to the justices, the Ninth Circuit’s opinion did not adequately distinguish between people facing “involuntary” homelessness and individuals who may have viable housing options. This error likely violates a federal civil procedure rule, which governs when multiple parties with similar legal claims can join together in the same lawsuit.

But the city, somewhat bizarrely, does not raise this error with the Supreme Court. Instead, the city spends the bulk of its brief challenging one of Robinson’s fundamental assumptions: that the Constitution’s ban on “cruel and unusual punishments” limits the government’s ability to “determine what conduct should be a crime.” So the Supreme Court could use this case as a vehicle to overrule Robinson.

That outcome is unlikely, but it would be catastrophic for civil liberties. If the law can criminalize status, rather than only acts, that would mean someone could be arrested for having a disease. A rich community might ban people who do not have a high enough income or net worth from entering it. A state could prohibit anyone with a felony conviction from entering its borders, even if that individual has already served their sentence. It could even potentially target thought crimes.

Imagine, for example, that an individual is suspected of being sexually attracted to children but has never acted on such urges. A state could potentially subject this individual to an intrusive police investigation of their own thoughts, based on the mere suspicion that they are a pedophile.

A more likely outcome, however, is that the Court will drastically roll back Martin or even repudiate it altogether. The Court has long warned that the judiciary is ill suited to solve many problems arising out of poverty. And the current slate of justices is more conservative than any Court since the 1930s.

Grants Pass’s litigation strategy is bizarre

One reason why this already difficult case is being needlessly complicated is that Grants Pass made some odd strategic decisions when it brought this case to the Supreme Court. While the city’s primary argument seems to attack one of the fundamental principles of American criminal law, there is probably much less to this argument than an initial read of their brief would suggest.

Robinson was an Eighth Amendment decision. It held that this amendment, which prohibits “cruel and unusual punishments,” does not permit the government to punish mere “status.” Instead, as mentioned, criminal laws must target some “act” committed by a defendant.

The city’s primary argument is that Robinson erred in this decision. The Eighth Amendment, it claims, “focuses not on the nature of a criminal offense, but the sentence imposed for it.” So, under this approach, California did not violate the Eighth Amendment in 1962 when it made merely existing while experiencing addiction a crime, so long as it was not imposing an excessive sentence on that addiction. Similarly, the amendment would forbid Grants Pass from imposing the death penalty on homeless people — because such a harsh punishment would be excessive — but it wouldn’t forbid a city from making existing while homeless a crime.

On the surface, this is an extremely consequential argument. If the Supreme Court should agree that mere status can be criminalized, that would open the door to thought crimes and allow states and localities to effectively banish entire classes of people they deem undesirable.

But there is probably less to this argument than it initially seems. As the city notes in its brief, some scholars argue that even if being arrested for a status crime does not violate the Eighth Amendment, it does violate two other provisions of the Constitution, which forbid the government from denying “life, liberty, or property, without due process of law.” So even if a majority of the current justices agreed that Robinson misread the Eighth Amendment, that doesn’t necessarily mean that the government can criminalize status.

Moreover, the idea that government may only punish voluntary actions, and not status, is hardly some newfangled idea invented by liberal justices in the 1960s. It has deep roots in the common law, the body of judge-made law that developed in English courts over many hundreds of years and that still shapes much of US law. In their brief, the unhoused plaintiffs quote a 1754 lecture by an English legal scholar who said that “no action can be criminal, if it is not possible for a man to do otherwise. An unavoidable crime is a contradiction.”

There’s even a Latin term, “actus reus,” that refers to the criminal act that someone typically must commit before they are charged with a crime. This is one of the most basic concepts in American criminal law. Virtually any law student who has completed the first week of their introductory course in criminal law will be familiar with this term.

So, while it is theoretically possible that the current Supreme Court could eliminate the requirement that someone commit an actus reus before they can be criminally punished, that seems unlikely. This is such a foundational principle in US criminal law that even this Court is unlikely to disturb it.

The line between “status” and “action” is often blurry

Yet while the Court is unlikely to say that people can be declared criminals simply because of who they are, the line between what constitutes a law criminalizing “status” and a law criminalizing action can be quite blurry at the margins.

Consider Powell v. Texas (1968), which asked whether an alcoholic who claimed to have an irresistible urge to drink could be charged with a crime for being drunk in public. Leroy Powell, the defendant in this case, claimed that arresting him for being drunk was no different than arresting someone addicted to drugs simply for being addicted, because his drunkenness was an unavoidable consequence of his status as someone with alcoholism.

The Court, however, rejected this argument — albeit in a close 5–4 decision.

Writing for himself and only three other justices, Justice Thurgood Marshall wrote the Court’s lead opinion in Powell. That opinion leaned heavily into Marshall’s doubts that Powell’s alcoholism was a truly an “irresistible compulsion to drink and to get drunk in public” that was so strong he was “utterly unable to control” his drinking.

Justice Byron White, meanwhile, cast the fifth vote against Powell but did not join Marshall’s opinion. Citing Robinson, White argued that “if it cannot be a crime to have an irresistible compulsion to use narcotics,” then “I do not see how it can constitutionally be a crime to yield to such a compulsion.” He also wrote that “the chronic alcoholic with an irresistible urge to consume alcohol should not be punishable for drinking or for being drunk.”

Ultimately, White voted against Powell because Powell was convicted of public drunkenness — the justice reasoned that, even if Powell could not avoid drinking, he could have remained at home. But White’s approach has fairly obvious implications for the Grants Pass case.

That case involves a web of local ordinances that, the Ninth Circuit determined, punish homelessness in much the same way that a ban on drinking punishes an alcoholic who genuinely is incapable of not drinking. Among other things, these ordinances include strict limits on where people can sleep and prohibit anyone from using “material used for bedding purposes” on public property — a provision that, the city claims, permits it to cite anyone who so much as wraps themselves in a blanket while sitting on a park bench.

Violators face a fine of at least $180, an enormous amount for someone who cannot afford housing, and the penalties escalate quite quickly for repeat offenders.

Because everyone has to sleep eventually, and because Grants Pass is too cold in the winter for anyone to sleep outside without a blanket or similar protection, the Ninth Circuit reasoned that Grants Pass’s web of ordinances effectively makes it impossible to live while homeless in Grants Pass — thus criminalizing the status of being homeless.

One way that the Supreme Court could resolve this case is to reject White’s conclusion in Powell that there is no difference between a law that criminalizes status directly and one that does so indirectly by criminalizing an involuntary act that arises out of their status. That would be a huge blow to unhoused people, as it would fundamentally undermine the Martin decision.

Even under White’s framework, moreover, Robinson only protects individuals who have an “irresistible compulsion” to drink alcohol. It follows that Robinson should only protect people who cannot voluntarily sleep anywhere except for places where Grants Pass’s ordinances effectively forbid them from sleeping.

And this distinction between voluntary and involuntary action presents the biggest problem for the unhoused plaintiffs in Grants Pass.

The biggest problem with the Ninth Circuit’s decision, briefly explained

The Ninth Circuit determined that people are protected by Robinson only if they are “involuntarily homeless,” a term it defined to describe people who “do not ‘have access to adequate temporary shelter, whether because they have the means to pay for it or because it is realistically available to them for free.’” But, how, exactly, are Grants Pass police supposed to determine whether an individual they find wrapping themselves in a blanket on a park bench is “involuntarily homeless”?

For that matter, what exactly does the word “involuntarily” mean in this context? If a gay teenager runs away from home because his conservative religious parents abuse him and force him to attend conversion therapy sessions, is this teenager’s homelessness voluntary or involuntary? What about a woman who flees her violent husband? Or a person who is unable to keep a job after they become addicted to opioids that were originally prescribed to treat their medical condition?

Suppose that a homeless person could stay at a nearby shelter, but they refuse because another shelter resident violently assaulted them when they stayed there in the past? Or because a laptop that they need to find and keep work was stolen there? What if a mother is allowed to stay at a nearby shelter, but she must abandon her children to do so? What if she must abandon a beloved pet?

The point is that there is no clear line between voluntary and involuntary actions, and each of these questions would have to be litigated to determine whether Robinson applied to an individual’s very specific case. But that’s not what the Ninth Circuit did. Instead, it ruled that Grants Pass cannot enforce its ordinances against “involuntarily homeless” people as a class without doing the difficult work of determining who belongs to this class.

That’s not allowed. While the Federal Rules of Civil Procedure sometimes allow a court to provide relief to a class of individuals, courts may only do so when “there are questions of law or fact common to the class,” and when resolving the claims of a few members of the class would also resolve the entire group’s claims.

But that’s not true in Grants Pass. A case involving a queer teen who fled his parents’ home is materially distinct from a case involving a woman who sleeps outside because she cannot find a shelter that will allow her to bring her dog. That does not mean that both of these individuals should not prevail in court. But the Federal Rules of Civil Procedure require them to bring separate legal proceedings that can address the unique facts of their unique cases.

The courts probably aren’t going to provide much help to homeless people in the long run

Grants Pass is hardly the first time the courts have been asked to intervene in a complicated question of anti-poverty policy. The best-known example is probably San Antonio Independent School District v. Rodriguez (1973), which challenged a public school funding scheme in Texas that tended to provide much more money to wealthy school districts than to poorer ones. The Court turned away this suit in a 5–4 decision.

In the decades after Rodriguez, however, many state supreme courts broke with their federal counterparts and ordered their states to spend more on education, to provide more resources to poor districts, or to otherwise implement a more equitable finance system. As of 2019, plaintiffs bringing Rodriguez-like suits in state courts had prevailed in 23 states.

But these cases are difficult to litigate and often require multiple trips to the state supreme court over the course of many years. Frequently, after a state supreme court issues a decision calling for some change in the state’s funding scheme, the legislature makes some small changes and then drops the issue until a court orders them to act again.

In Arkansas, for example, school finance reformers won a state supreme court victory in 1983 declaring that the state’s school finance system bore “no rational relationship to the educational needs of the individual districts” and then had to return to court nearly two decades later. Seventeen years after its initial decision, the Arkansas Supreme Court found that the wealthiest school districts were still spending nearly twice as much per pupil as the poorest districts.

Even if Martin survives contact with the Supreme Court, anti-poverty advocates are likely to face even more difficulties trying to wield it to mitigate the problem of homelessness than those same advocates have faced in school finance cases. Because the law restricts when courts can provide class-wide relief to anyone experiencing homelessness (or even to “involuntarily homeless” people), enforcing Martin is likely to become a long, slow slog of individual cases attempting to rescue individual criminal defendants from an individual arrest for sleeping outside.

Of course, the courts could relax the rules governing when judges can provide class-wide relief. But such a relaxation would have implications far beyond homelessness policy and would likely do far more to empower the judiciary’s far right than it would to help anti-poverty advocates.

Imagine, for example, what Matthew Kacsmaryk, the Trump-appointed judge who tried to ban the abortion drug mifepristone and who routinely hands down court orders implementing right-wing policy preferences, would do if he were handed a new power to issue class-wide relief to any group of people he wants to help out.

So, with so many ways that Grants Pass could end very badly for homeless people — and for criminal defendants generally — the case is unlikely to end well for them.