On July 24, 2025, President Trump signed an executive order titled “Ending Crime and Disorder on America’s Streets.” The order reversed several decades of federal policy toward the homeless, which had been geared toward creating more permanent housing and more voluntary services. Going forward, the focus is to be on mental illness and substance abuse, which afflict a large share of the homeless population. This new “Behavioral Health Treatment First” approach replaces the “Housing First” approach emphasized by the Biden administration, and puts less emphasis on voluntary services and more on involuntary civil commitment.
The federal government’s discovery of mental illness as an important societal issue is more of a re-discovery than a new development. Pushing mental illness into the forefront is the fact that so many American cities today are plagued by homelessness. The homeless population in 2024 was approximately 275,000, the highest ever recorded, bringing with it crime spikes and public health emergencies. The mentally ill form a large share of this population. Yet mental illness has been a longstanding problem. Our mistake over the last half-century has been to ignore it, to cover it up with well-intentioned slogans, or to try to deal with it on the cheap. Unfortunately, there are no shortcuts to a more effective policy that will preserve public safety while also helping mentally ill patients.
An Inescapable Problem
Mental illness was sufficiently important to form the subject of the first major sociological study. In 1897, Emile Durkheim published Suicide. The great French sociologist said suicide was not a problem to be fixed so much as a fact of life. In an individualistic, market-oriented world, Durkheim argued, some people will feel a sense of not belonging or fitting in. Therefore, they will inevitably drift into despair and take their own lives.
Surprisingly little has changed. In 2023, approximately 50,000 Americans committed suicide, more often men than women, a ratio similar to what Durkheim observed in Europe during his time. This represents an increase of about 30 percent over the last twenty years. Yet the suicide rate has fluctuated between 10 to 15 per 100,000 Americans for almost a century. The debate over how much blame to apportion between social factors and mental illness also continues along the same lines. Durkheim emphasized social factors, such as a failure to socially integrate or the feeling of anomie arising from community breakdown, but he did give psychopathology its due—especially alcoholism, which he admitted was a predisposing factor.
The debate’s broad outlines now cover homicide as well. Researchers question whether mental illness or social factors play the crucial role in causing extreme violence. One study says people from the same poor neighborhood, whether mentally ill or not, have similar rates of violent behavior. In contrast, another study says mental illnesses such as schizophrenia, bipolar disorder, and depressive disorders make violent behavior three times more likely, independent of socioeconomic background. Like Durkheim, researchers on both sides give substance abuse its due, noting that it combines with serious mental illness to make violent behavior three times more likely still.
The American public cares about this debate. They see homeless people living on the streets, many of them mentally ill. They read about mentally ill people, homeless or not, assaulting or murdering innocents. They may have been assaulted themselves. I was, when a mentally ill person slammed into me fifteen years ago while I was peaceably walking down a New York City street. They hear that many of the dangerously mentally ill were once housed in asylums, but that asylums have mostly disappeared. They wonder why we can’t just bring them back and stick all the crazy people in them.
The real truth about mental illness and violence differs somewhat from the perception, yet three facts endure. First, nothing has really changed on the mental illness front. Mental illness existed in Durkheim’s time; it exists today. Second, the technology for managing severe mental illness has changed over time—somewhat. Third, everything has changed on the cultural front, especially with today’s emphasis on “autonomy.”
The word autonomy has existed for centuries, but has meant different things at different times. In ancient Greece it referred to city-states that governed themselves. The philosopher Kant expanded the definition to include individual self-governance through reason. The European romantics who followed stressed feeling over thinking. For them, autonomy meant being one’s own person. In this version, autonomy overlaps with the idea of “authenticity”.
The romantic conception of autonomy lay dormant in the U.S. for over a century. Then, in the 1960s, it began to penetrate most institutions in American life. In the university, for example, appeals to authenticity challenged the very notion of reality, as some academics began to argue that every individual had their “own truth,” based on lived experience. In health care, the old Kantian idea of autonomy led to the doctrine of informed consent. The new romantic version fueled the patient-centered care movement, in which patients went from being informed about medical decision-making, to participating in medical decision-making, to sharing in medical decision-making, to making medical decisions unilaterally, with the doctor dutifully approving whatever the patient decided. The romantic version of autonomy created a kind of imperial self, in which a person, at least in theory, lives, thinks, and feels however he or she wants, unfettered by authority or social influence, as long as he or she doesn’t interfere with anyone else trying to live, think, and feel the way they want.
These three enduring facts—that the mentally ill have always been with us, that the technology for treating the most severe cases has evolved only somewhat, and that the romantic notion of autonomy governs much of American life today—frame the debate over how to manage the severely mentally ill. They put limits on what we might accomplish in terms of keeping the severely mentally ill from attacking others or hurting themselves.
From Responsibility to Authenticity
In early nineteenth century America, the severely mentally ill were typically placed in prisons or shelters for the poor. The goal was not to treat them so much as to safeguard the community. To the degree that American culture prized autonomy, it was of the Kantian kind, meaning people capable of thinking for themselves and governing themselves. Thought to lack these capacities, the mentally ill were called “mad” and targeted for “taming.”
The technology available at that time perfectly matched the cultural prejudice. Prison was a logical place for housing “mad” people. Mechanical restraints for “taming” were also plentiful. They included such gruesome devices as straitjackets, manacles, chains, and hot irons for branding.
Change in these attitudes came in the second quarter of the nineteenth century. Both a new culture and a new technology supported it. Evangelical Protestantism during the Second Great Awakening, from 1780 to 1830, preached against the old harsh predestination. People could improve, the movement declared. Applied to the mentally ill, the idea ushered in what has been called the Moral Treatment Era. The mentally ill, it was thought, retained their capacity for reason after all; they simply needed compassionate care in an improved social and physical environment to get their wits back. Just as the penitentiaries and reformatories replaced the old jails, the new building for offering such care for the mad would be the asylum.
Although kinder than prisons, asylums still ran on involuntary admissions. While the mentally ill were thought to have a latent capacity for reason and self-control, at the moment of admission they were presumed unable to make decisions themselves. Family members and psychiatrists pushed them into asylums whether they agreed to go or not.
From the perspective of technology, the asylum perfectly matched the new culture’s aspirations. Often located in a quiet area with attractive grounds, the asylum offered inmates a retreat where they could enjoy rest and refuge. Other technologies reinforced the asylum’s goals. New sedatives—medicines to induce peace and relaxation—entered the psychiatrist’s toolbox. These pharmaceutical replacements for physical restraint included opium, camphor, bromides, and, later, choral hydrate.
The mid-nineteenth century immigration surge, along with growing urbanization, thrust more people into close proximity with one another, including the severely mentally ill. Moral treatment became an expensive luxury rather than something that could be practiced on a mass scale. A two-tiered system emerged. Private asylums remained quiet retreats with treatment as the goal of admission. Close to where I live is the Sheppard Pratt hospital, originally the Sheppard asylum, established in 1893—to this day a beautiful place to sit under a tree and eat lunch. These asylums had plentiful staff and other resources. They existed for acute cases who were expected to improve rather than for the chronically mentally ill.
But for the large numbers of poor and immigrant patients, especially those suffering from permanent psychotic disorders or chronic alcoholism, a different building, with a different purpose, came into being. Public asylums took these people in, with their goal being containment rather than treatment. Expanding enormously over the rest of the century, these public asylums shared much in spirit with the earlier prisons that housed the chronically mentally ill. Indeed, prison inmates and the chronically mentally ill often dwelled together in the same asylum.
At this juncture, keep in mind the three facts described above.
First, the severely mentally ill endure. The names of their illnesses may change—the term “psychosis” arose at mid-nineteenth century, replacing the term “mad”—but the disorders remain.
Second, the primary technology for managing the severely mentally ill was the building, whether the prison, the poor house, the private asylum, or the state institution. Sometimes the building existed to treat the mentally ill; indeed, private asylums were presumed to be beneficial. Other buildings existed to “tame” the mentally ill, or just to contain them. But in all cases the building’s purpose was to separate the mentally ill from the main body of society. The building harmonized the government’s two roles that might have otherwise conflicted—its obligation to protect the incapacitated from themselves and its obligation to protect citizens from the incapacitated.
Third, the Kantian notion of autonomy that prevailed during this period laid the philosophical basis for mental illness as a treatable condition. Through rest and relaxation, it was thought, patients might regain their rational decision-making powers and their moral agency. Nevertheless, even well-off private asylum patients were presumed to lack reason and autonomy at the time of admission. Some were given the flimsy diagnosis of “moral insanity,” thereby confirming the link between autonomy and moral agency. True, compared to “taming the mad,” practically any treatment represented progress of a kind. Nevertheless, the Kantian notion of autonomy still provided a philosophical basis for involuntary commitment.
These three basic facts governed the care of the severely mentally ill in the U.S. well into the twentieth century. Court decisions and legislation corrected some of the abuses. A person’s right to liberty was increasingly considered before commitment to an institution, allowing the potential in-patient a right to a trial and a lawyer. Commitment standards were also tightened. Much of the decision-making power in commitment was taken away from psychiatrists and put into the hands of judges and magistrates.
Yet the reforms remained faithful to the Kantian principle of autonomy. Although the state rather than private practitioners would decide if a potential inmate’s reasoning powers were compromised, the decision to commit turned on whether that person had sufficient powers of reason to make decisions as a moral agent. As for technology, little changed. Potential inmates may have had to wait in jail when a lawyer was unavailable to represent them. And whether it was a jail cell or a bed in a state institution, separation—in the form of confinement to a building—remained the operative concept.
In 1951, the National Institute of Mental Health published the “Draft Act Governing Hospitalization of the Mentally Ill” to correct some of the abuses spawned by these earlier reforms. Much of the decision-making power surrounding commitment was put back into the hands of psychiatrists. Still, the basic structure of caring for the severely mentally ill remained intact, with the building as the primary technology for either treatment or containment, and the old Kantian principle of autonomy supplying the system its theoretical basis. By 1953, the U.S. psychiatric inpatient population reached 559,000, an all-time high.
The War on Normal
Important change occurred during the later 1950s and ‘60s, although consistent with the three basic facts outlined above.
First, severe mental illness remained, even though the medical nomenclature had evolved. The term “schizophrenia” was coined in 1908. Freud used the term, while associating the disease with another phrase that emerged around that time, “dementia praecox,” which he linked to a process he called “decathexis,” and that somehow involved the libido.
More revealing, the neutral-sounding phrase “mental illness,” which had been around since the eighteenth century, began to displace the older terms once used to describe the severely mentally ill, such as “insane” or “lunatic.” In the process, schizophrenia and depression ceased to be matters of morality. Under the old idea of autonomy, self-governance achieved through reason made moral agency possible. This idea connected mental health with morality. With the more objective phrase “mental illness,” morality ceased to be associated with psychiatric conditions altogether. A new, less judgmental understanding of autonomy stood ready to take the old one’s place.
The shift in understanding was not limited to medical circles. Whether it was sociologist Erving Goffman in The Presentation of Self in Everyday Life railing against conformism and people’s obsession with being “normal,” feminist Betty Friedan railing in The Feminine Mystique against suburban women forced to stay at home rather than pursue careers, singer Pete Seeger railing in his song “Little Boxes” against a society that compelled people to all be alike, or beatniks in urban areas railing against the prevailing dress codes, hair styles, and other norms of the day, the goal was the same: let people be themselves, let people fully express their individuality.
The sentiment penetrated the mental health industry. Books like One Flew Over the Cuckoo’s Nest (1962) and movies like The Snake Pit (1948) portrayed the asylum as a bureaucratized, nightmarish institution that brutalized psychiatric inpatients, who were often guilty of nothing more than trying to be themselves, and who were sometimes less crazy than the medical staff who cared for them. Containing the mentally ill, separating them from civil society, seemed increasingly inhumane and unfair to intellectuals and institutional administrators. The state’s two roles—caring for the severely mentally ill and protecting citizens—began to conflict. While protecting citizens required the severely mentally ill to be separated from civil society, allowing the severely mentally ill to be themselves demanded the opposite.
This was not simply an anarchic impulse to let it all hang out. Technological advances made a new, less restrictive approach to the severely mentally ill seem realizable. Until the 1950s, the primary technology for managing these people was the building, meaning some venue for containment, with the goal being separation. In 1950, new medications called antipsychotics came on the market, such as chlorpromazine, sold under the trade name Thorazine. Rather than sedate and relax people, as medications during the moral treatment era had, the new antipsychotics treated hallucinations, delusions, and other incoherent thoughts directly. They made irrational people seem more rational.
Those with decision-making power leveraged the success of these medications to take the principle of autonomy to another level. If patients took antipsychotics, then their separation from civil society seemed unnecessary because they would no longer pose a threat, either to themselves or others. They could be their quirky selves and live freely but safely, without confinement to a building. This new policy, known as community-based outpatient treatment, could return the state’s two roles to harmony.
By 1960, asylums were widely criticized for offering “little effective treatment.” Their critics said they merely housed the mentally ill, which was true. But that had once been good enough. The asylum had not changed. The culture had.
Consistent with the change, President John F. Kennedy signed the Community Mental Health Centers Act in 1963 to help move the severely mentally ill from inpatient psychiatric hospitals out into the communities. Deinstitutionalization began and continued for decades. By the 1990s, the inpatient population in the U.S. had declined to 30,000. In 2017, the total number of psychiatric inpatient beds in the U.S. was just 37,209, despite a more than doubling of the population since the 1950s.
But nature quickly trumped culture and technology. True, the new antipsychotics worked. But not perfectly—even diminished symptoms such as hallucinations could still be severe. Besides, some patients forgot to take their medicine or simply refused to do so. The new community mental health centers were insufficiently funded, which helps to explain their failure. Yet the strategy that led to their creation had a basic flaw: without separation, the danger posed by the severely mentally ill could not realistically be contained.
Legislators and the courts seemed to grasp the problem. In 1964, the concept of “dangerousness” entered mental health law. But the concept actually made the public less safe, not more. In the old “treatment” model of mental health, severely mentally ill patients could be forced into institutions and made to take their medication. In the new “dangerousness” model, a mentally ill person had to be shown to be an “imminent threat,” to either himself or others, before commitment could occur. Even then, the person could only be committed for a short time.
But the dangerously mentally ill do not always telegraph their tendency toward violence. Sometimes their dangerousness can only be judged in hindsight, after they have committed a violent act. Unlike the treatment model, the dangerousness criterion risked delaying forced institutionalization until the moment mentally ill people grew demonstrably suicidal or physically violent, by which time it was too late. Even then, psychiatrists were required by the new model to recommend the least restrictive level of care possible.
Some legislators pushed a hybrid model called “outpatient civil commitment.” Instead of locking up a mentally ill patient inside a building, the patient would be allowed to roam free, but only within his or her community. A patient would have to follow a regular schedule of mental health checks and have access to medication—but not necessarily be forced to take it.
The New York State version of outpatient civil commitment was passed in 1999 as Kendra’s Law, named after a young woman who was killed when a man with schizophrenia pushed her onto the subway tracks. In this system, the eligible patient needed to have a history of violence, thereby exposing the community to at least one dangerous episode, before he could be committed. In addition, the eligible patient had to be judged noncompliant with treatment. In other words, he had to be the kind of patient who needed to be forced to take his medication. Finally, outpatient civil commitment could only last a year, after which an arduous renewal process had to be conducted.
According to the new ideal of autonomy, the self should be allowed to express itself in all its uniqueness and peculiarity, to feel and live as it wants. This ideal seemed always to hover around the authorities, threatening to reprimand them if they dared push people too far, and causing them to err on the side of timidity when dealing with the mentally ill. The official language used in Kendra’s Law demonstrates this point. Rather than call the commitment procedure “involuntary outpatient commitment,” the law euphemistically renamed it “assisted outpatient treatment,” suggesting that no one was forcing anyone, and that the prospective patient was not being contained so much as assisted on his or her journey to self-actualization. The word “treatment” provided additional cover for an action that some might construe as intrusive or compulsory.
Professionals today debate the benefits of outpatient civil commitment. Supporters say it results in fewer arrests; opponents say it offers few therapeutic benefits. This arcane professional debate matters little to the public, which cares less about whether some small statistical improvement may have been achieved here or there, and more about the gross dangers posed by the severely mentally ill who they walk past every day. The homeless are in every city. Anywhere from one-third to two-thirds of the homeless have a mental illness that makes a person more prone to violent behavior. The incidence of antisocial personality disorder, for example, is thirty times higher among the homeless than in the general population. The incidence of psychotic disorders is also thirty times higher. The incidence of substance abuse disorder, which combines synergistically with other mental illnesses to make a person more prone to violence, is three times higher.
True, people with mental illness are responsible for only 5 percent of societal violence and criminal behavior. And most people with severe mental illness are not violent. Yet the public can take precautions against most other forms of violence by avoiding certain neighborhoods or cities, or by remaining indoors after dark. In contrast, an attack by a mentally ill homeless person can occur anywhere, at any time, and at random.
Just during the time I spent writing this essay, a severely mentally ill person, found muttering and screaming, pushed an ER doctor onto the subway tracks in New York. In Colorado, a sex offender previously found mentally incompetent tried to kidnap a child. In Austin, Texas, a severely mentally ill man who called himself “Jesus” randomly shot and killed four people in a Target parking lot. None of the usual precautions against crime or violence would have prevented these cases, which explains why they are particularly terrifying to the public even though they are statistically rare.
Getting Real about Mental Illness
What is to be done?
The three basic facts outlined above frame what is possible.
First, the severely mentally ill will always be with us. Lessening their numbers is not an option.
Second, fighting against the romantic conception of autonomy will be extremely difficult, although not impossible. For seven decades, a belief in the imperial self has burrowed deep inside our institutions, becoming almost canonical. Indeed, its understanding of the good life only strengthens with each passing decade. Once leading the assault on dress codes at the airport, or on the prejudice against tattoos and piercings in white-collar work, it now underlies the drive to legalize cannabis, or to legitimize the quasi-recreational use of psychedelics.
From the autonomy perspective, the hallucinations and visions experienced during psychosis are less discrete manifestations of mental illness and more psychological experiences sitting on the same spectrum as normality. If stupefaction through cannabis and psychedelics represents a kind of personal declaration of independence, then why not other altered states of mind, including hallucinations and visions? All of us live in a state of illusion, argue proponents of the romantic conception of autonomy. All of us are crazy in a crazy world; all of us have the right to feel secure in our illusions, free of judgement, whether it be the stupefied or the mentally ill. The Insane Liberation Front, a group formed in Portland, Oregon, in 1969, the “Psychosis Validation Coalition,” and the “Festival of Creative Psychosis,” are extreme examples of this attitude taken political form over the years. Yet the same attitude resonates with many people, to one degree or another.
The American people want two things that are in conflict. They do not want to see the mentally ill deprived of their freedom; at the same time, they want to feel safe. The latter has led to some pushback against mental health policy steeped in the romantic notion of autonomy. New York City mayor Eric Adams has sponsored the Compassionate Interventions Act, a proposed state law that would expand involuntary commitment for those suffering from substance abuse disorder. The act would allow judges, acting on physician advice, to forcibly move some substance abusers into hospitals and undergo treatment. Over thirty states have similar laws. The act is part of Mayor Adams’s “End the Culture of Anything Goes” campaign to strengthen the involuntary commitment of the severely mentally ill.
Such legislation suggests the tide may be turning. But the devil is in the details. Some of the involuntary commitment laws passed by states focus on substance abuse rather than on mental illness. Others do just the opposite. Even when mental illness is covered, involuntary commitment is often for only a short period of time. In Connecticut, for example, it can last only 15 days. In some states the period of time has recently been extended—in Alaska, for up to two years. Nevertheless, states continue to err on the side of keeping the severely mentally ill living within the community.
The third option for managing the severely mentally ill involves technology. I am not referring to the newer antipsychotic medications, which, for all practical purposes, hardly differ from the old ones. Such medicines target different receptors and have fewer side effects, but psychotic patients still must swallow them or accept them as injections. If they refuse to take them, the medicines won’t work. Non-drug therapies for psychosis have also come on the market, including non-invasive brain stimulation, brain implants, and virtual reality therapy for psychosis treatment. But, again, patients must accept them. And that is exactly what the severely mentally ill may be unwilling to do.
Throwing another monkey wrench into the mix is the popular distrust of the medical and public health establishments more generally, which arose during the COVID pandemic. With forced vaccinations and other mandates still uppermost in people’s minds, it is highly unlikely that the public will support forcing new medications and invasive treatments on people in the near future.
Instead, it is an older technology—the building—that offers some hope. Containment and separation have always been the optimal method for keeping the severely mentally ill from hurting themselves or others. To this day, the approach continues, although in secret. Studies show that the mentally ill are arrested more often than people without mental illness when they encounter the police under similar circumstances. Why? The police admit that taking the mentally ill into criminal custody rather than to a hospital emergency room makes it easier to get these people into long-term care—that is, into a building, even if just a prison. Given the difficulty in securing civil commitment these days, as well as the short involuntary hospitalizations achievable at best, prison has become the unofficial method of choice for containment and separation.
The future of managing the severely mentally ill may involve a fusion of this old technology—the building—and the new romantic conception of autonomy. Rather than forcing a severely mentally ill person into a prison, an asylum, or a hospital, another option might be voluntary placement in a kind of building yet to be conceived. Health and Human Services Secretary Robert Kennedy, Jr., recently spoke of “wellness farms” as a place where substance abusers might go and live for a time and detox. To the degree that admission would be involuntary, defenders of the romantic conception of autonomy accused Kennedy of wanting to force people into institutions, especially minorities and the disabled, to “control” them, perhaps to “experiment” on them, and to deprive them of their freedom to be themselves. Yet if a building could be made more attractive, such that the severely mentally ill would want to live there—that is, if they volunteered to go—it might harmonize the need to separate the severely mentally ill with the romantic conception of autonomy. If the mentally ill refuse to go because they prefer to live on the street or in a park, then a street or a park can be built next to the building. A whole living complex can be envisioned—so long as there is separation.
This approach has the disadvantage of being expensive and, with community opposition to establishing new mentally health facilities likely, politically difficult. To deal with the latter problem, the facilities can be located far away from populated areas; there is no rule that says they must exist in urban or suburban areas. Yet there are no cheap or easy solutions. At the same time, it is probably not by chance that the title of President Trump’s executive order includes the word crime. The word captures the climate of fear in many of America’s urban areas. Today, the public feels insecure, and it will likely support almost any level of expenditure to get back its sense of security. This is true in domestic policy as much as it has traditionally been in defense policy. The administration has shrewdly invoked people’s fear of crime to get the public behind the new direction in mental health policy.
In some ways, Durkheim’s advice for reform at the end of Suicide comports with the idea of separating the severely mentally ill from the main body of society. He writes: “We have shown that it is not necessary… to restore, artificially, social forms which are outworn and which could be endowed with only an appearance of life [like the asylum], or to create out of whole cloth entirely new forms without historical analogies [like the wellness farm]. We must seek in the past the germs of new life which it contained, and hasten their development.” For the severely mentally ill, we need a new building, drawn from the old but not the old, a building that blends the old idea of containment with the new conception of freedom.
About The Author
