Mental Health Policy: What Each Position Is Protecting

In 1955, the United States had approximately 559,000 psychiatric hospital beds — about one for every 300 Americans. Over the following five decades, in a process called deinstitutionalization, those beds were systematically eliminated. By 2016, the number had fallen to roughly 37,000, a decline of over 93% against a population that had nearly doubled. The official story of deinstitutionalization is a liberation narrative: the state hospitals were brutal places, Thorazine and the community mental health center could replace them, and returning the mentally ill to their communities was a humane reform. The actual record is more complicated. The community mental health centers were never fully funded. Thorazine suppressed symptoms but did not replace the structure that hospitals, however imperfect, had provided. The patients did not return to communities — many had no communities to return to. They returned to the street.

Today, the largest mental health institution in the United States is the Los Angeles County Jail, which houses an estimated 3,000 people with serious mental illness at any given time. Cook County Jail in Chicago, Rikers Island in New York, and the San Francisco County Jail follow. The criminal justice system — built to punish, not treat — has become the default response to psychiatric crisis, and its inadequacy for that function is visible in every data point: people with serious mental illness are incarcerated at roughly four times the rate of the general population; they spend longer in pretrial detention than those without mental illness charged with similar offenses; and they cycle back through the system at high rates because incarceration does not address the conditions that brought them into contact with it.

The debate about what to do is not a dispute between people who care about the mentally ill and people who don't. It is a dispute between people who have drawn different conclusions from the same failure — and who are, in many cases, protecting different things that the failure endangered. What the biomedical advocates, the civil liberties tradition, the criminalization critics, and the social determinants position are each protecting is not obvious from their surface arguments. This map tries to get underneath them.

It is also one of the places where the healthcare cluster's larger contradiction becomes hardest to ignore. On paper, mental health care is supposed to be covered like other medical care; in practice, parity rules are weakly enforced, insurer networks are thin, and many counties still do not have enough psychiatrists, therapists, crisis beds, or mobile teams to make "access" meaningful. When voluntary care is this fragile, coercion and jail become fallback institutions. So the argument is not only about liberty and safety. It is about whether serious mental illness will be treated as a basic public obligation that the system must organize itself to meet, or as a nominally covered service that people are expected to find in an underbuilt market.

What the involuntary care and institutional capacity position is protecting

The people the system has abandoned by calling abandonment liberation. The psychiatric capacity position — associated with E. Fuller Torrey, the Treatment Advocacy Center, and, more recently, California's CARE Court legislation — begins from an observation that the civil liberties framework tends to elide: the most severely mentally ill are not declining treatment because they have thought carefully about their options. They are declining treatment because their illness has impaired the very capacity for judgment that the right to refuse requires. Anosognosia — the neurological condition in which a person with psychosis lacks insight into their own illness, not as a psychological defense but as a symptom of the illness itself — affects an estimated 40–50% of people with schizophrenia and 20–40% of those with bipolar disorder. A person who does not believe they are ill cannot meaningfully consent to or refuse treatment for an illness they do not believe they have. The involuntary care tradition is protecting the claim that leaving these individuals alone on the street in the name of autonomy is not respecting their freedom — it is abandoning them to a disease that has taken their freedom already.

The families and communities that bear what policy externalizes. The capacity argument has a distributional dimension that is frequently lost in the debate. When a person with untreated schizophrenia is left in the community without services, the costs do not disappear — they are transferred to family members, to neighbors, to emergency departments, to police officers who are not trained to respond to psychiatric crisis, and ultimately to jails. Pete Earley's memoir Crazy: A Father's Search Through America's Mental Health Madness (2006) documents this transfer through one family's experience: his son's untreated bipolar psychosis led to a felony breaking-and-entering charge — not to steal anything, but because he was psychotic — and to years of cycling through jails and homeless shelters rather than the psychiatric care that would have addressed the underlying condition. The involuntary care tradition is protecting the claim that the burden of untreated serious mental illness is not borne by the people who have chosen to enforce the right to refuse, but by the people closest to those who are suffering — and by the people suffering themselves.

The capacity floor below which autonomy becomes a fiction. The philosophical core of the involuntary care position is a challenge to how liberalism deploys the concept of autonomy. John Stuart Mill's harm principle — the state may not interfere with an individual unless they are harming others — was formulated with a specific precondition: "Those who are still in a state to require being taken care of by others must be protected against their own actions." The psychiatric capacity tradition is protecting the application of this precondition to severe mental illness: that a person floridly psychotic, unable to recognize their own condition, unable to meet their basic needs, and deteriorating on the street has not made an autonomous choice to live that way. Kendra's Law in New York — named for Kendra Webdale, killed when Andrew Goldstein, a man with untreated schizophrenia who had sought hospitalization thirty-two times without success, pushed her in front of a subway train — provides court-ordered outpatient treatment for people who meet specific criteria of treatment non-compliance and psychiatric history. The tradition is protecting the claim that intervention in these cases is not tyranny but care.

What the civil liberties and recovery-oriented care position is protecting

The history of psychiatry as a coercive instrument. The civil liberties tradition — rooted in the patient rights movement of the 1970s, in Thomas Szasz's The Myth of Mental Illness (1961), and in the reform litigation that produced the constitutional right to refuse treatment — begins not from abstraction but from a specific historical record. State psychiatric hospitals were not primarily treatment institutions; they were custodial institutions where people were indefinitely confined, often for conduct that had nothing to do with dangerousness: being gay, being poor, being a difficult woman, being a political dissident. The Soviet Union institutionalized dissidents as "sluggishly progressive schizophrenics." In the United States, the diagnosis of "drapetomania" — the purported mental illness that caused enslaved people to flee captivity — was published in the New Orleans Medical and Surgical Journal in 1851 and treated as legitimate medical science. The civil liberties tradition is protecting the memory that psychiatric coercion has a documented history of serving interests other than the patient's — and that expanding the legal authority for forced treatment requires extreme caution about who will be subjected to it and under what circumstances.

The evidence that coercion undermines the therapeutic relationship and future care-seeking. The recovery-oriented care position does not simply oppose involuntary treatment on principle — it disputes its effectiveness. Research on coercive psychiatric interventions consistently shows that hospitalization against a patient's will is associated with higher rates of treatment avoidance afterward, greater difficulty forming therapeutic alliances, and in many cases a worsening of the person's relationship with the care system that they will need for the rest of their life. The Hearing Voices Network and the Intentional Peer Support model — both drawing on the experience of people who have received psychiatric care — emphasize that what people with serious mental illness most often report needing is not more coercion but more continuity: a stable relationship with a care provider, a place to live, and treatment that they have chosen. The recovery tradition is protecting the claim that the choice between "involuntary treatment" and "no treatment" is a false binary produced by chronic underinvestment in voluntary services — and that the solution to people not accessing care is not to remove their right to refuse but to make care accessible enough that refusal is no longer the default.

The difference between formal parity and actual care. This argument has become sharper, not weaker, as parity law has matured. The Mental Health Parity and Addiction Equity Act and the Affordable Care Act established the principle that behavioral health benefits should not be treated as second-class coverage. But the April 2024 federal parity rule was needed precisely because plans have continued to use nonquantitative treatment limits, inadequate networks, and utilization-management practices that leave mental health care technically covered yet functionally scarce. The recovery position is protecting the claim that "treatment refusal" is often misdescribed. People do not only refuse care because they reject it. They also fail to find a clinician, fail to find one who takes their insurance, fail to get an appointment before the crisis deepens, or fail to stay in care after a coercive encounter confirms their distrust. In that system, expanding compelled treatment without repairing voluntary access risks confusing institutional scarcity for patient noncompliance.

The demographic reality of who will be involuntarily treated. The civil liberties tradition draws on the same distributional lens that strengthens the structural argument in many policy debates: who will bear the costs of expanded coercive authority? The data are consistent across decades of research. Black Americans are far more likely to be diagnosed with schizophrenia than white Americans presenting with identical symptoms, far more likely to be involuntarily hospitalized, and far more likely to be subjected to more restrictive interventions. The lived experience of many Black Americans with serious mental illness is that the coercive mental health system and the criminal justice system are continuous — that both deploy state force against them, that both are heavily racialized in their application, and that neither has demonstrated, for this population, that force produces better outcomes than its absence. The civil liberties tradition is protecting the claim that "who gets to decide" is not a neutral technical question but a political one — and that the populations most subject to forced psychiatric intervention are among the populations whose distrust of the systems exercising that power is most historically warranted.

What the criminalization critique is protecting

The exposure of a category error that costs lives. The criminalization critique — advanced most systematically in work by researchers like Henry Steadman, in reform advocacy by organizations like the Vera Institute and the Treatment Advocacy Center (despite their significant other disagreements), and in journalism by writers like Ken Stern and David Brooks — begins from the observation that jails and prisons are the worst possible environments for the treatment of serious mental illness, and that their emergence as the largest psychiatric institutions in the country is not a policy that anyone chose. No policymaker designed a system in which people with untreated schizophrenia cycle between emergency rooms, homeless encampments, and county jails. It emerged from a series of smaller choices — closing psychiatric beds, underfunding community mental health, criminalizing the behaviors that untreated psychosis produces — whose aggregate effect was to route people with psychiatric crises into the only system that could not refuse them. The criminalization critique is protecting the observation that this is a category error: that jail is not treatment, that the conditions of incarceration actively worsen psychiatric illness, and that the system has been allowed to operate this way because the people most harmed by it have the least political power to demand something different.

The investment in diversion and alternatives that the current system forecloses. Crisis Intervention Team (CIT) training — developed in Memphis following the 1987 police shooting of Joseph DeWayne Robinson, a young Black man with schizophrenia who was shot while wielding a knife he had demanded police use to kill him — has demonstrated that training officers in de-escalation, mental health recognition, and community resources can substantially reduce both arrest rates and use of force in mental health encounters. Mobile crisis teams, like the CAHOOTS program in Eugene, Oregon, dispatch mental health workers rather than police to calls that do not involve active violence. Mental health courts, which operate in over 400 jurisdictions, divert people with serious mental illness from incarceration into treatment and supervision — with recidivism rates consistently lower than standard prosecution. The criminalization critique is protecting the claim that these alternatives are not expensive experiments but proven models that have been chronically underfunded because the people who benefit from them cannot advocate effectively for their own resources.

The racial accounting that the policy debate typically suppresses. The criminalization of mental illness is not racially neutral in its application. Black Americans experiencing psychiatric crisis are more likely to be arrested than white Americans experiencing identical presentations; they are more likely to be killed in those encounters; and they are less likely, once incarcerated, to receive psychiatric treatment. The Vera Institute's 2019 analysis found that Black Americans with mental illness were 4.5 times more likely to be incarcerated than white Americans with mental illness. The criminalization critique is protecting the claim that the choice to route psychiatric crisis through the criminal justice system rather than the health system is not a politically neutral technical failure but a decision with racialized consequences — and that any honest accounting of the costs of the current system must include who is bearing them.

What the social determinants and upstream investment position is protecting

The evidence that poverty, trauma, and housing instability cause mental illness — not only correlate with it. The social determinants position — drawing on epidemiological research in the tradition of Mervyn Susser and Ezra Susser's work on schizophrenia, on adverse childhood experiences (ACE) research, and on the World Health Organization's Commission on Social Determinants of Health — challenges the framing that mental illness is primarily a neurological condition requiring primarily medical intervention. The ACE study, conducted by Kaiser Permanente and the CDC in the 1990s, found dose-response relationships between adverse childhood experiences (abuse, neglect, household dysfunction) and adult mental health outcomes that could not be explained by genetic transmission. Children who experienced four or more categories of adverse childhood experiences were seven times more likely to be diagnosed with depression and twelve times more likely to report suicidality. The social determinants position is protecting the claim that the upstream causes of mental illness — poverty, violence, food insecurity, housing instability, childhood trauma — are amenable to policy intervention in ways that downstream psychiatric treatment is not, and that a system focused exclusively on treatment after illness onset is addressing the last stage of a long causal chain.

The housing-first evidence as a proof of concept. Pathways to Housing, founded by psychiatrist Sam Tsemberis in New York in 1992, developed the Housing First model on a simple premise that violated conventional wisdom: give people with serious mental illness and substance use disorders permanent housing first, without requiring sobriety or treatment compliance as preconditions, and provide voluntary support services afterward. The conventional model — a "treatment first" continuum that required demonstrated psychiatric stability before housing access — operated on the theory that housing was a reward for treatment engagement. Housing First treated housing as a platform for treatment engagement: stable housing made everything else more possible. The evidence accumulated across multiple cities and a decade of randomized trials was striking. Tsemberis's initial study found 80% housing retention after two years, compared to 30% in the control condition. Mental health outcomes improved. Substance use was no worse and in many cases better. The cost per person housed was lower than the cost per person cycling through emergency rooms, shelters, and jails. The social determinants position is protecting the Housing First evidence not just as a program but as a demonstration that stable conditions produce better psychiatric outcomes — that the causal arrow runs both ways, and that addressing social conditions is not a soft alternative to psychiatric treatment but part of what treatment actually requires.

The workforce and capacity deficits hidden inside "choice." The upstream argument is not only that people need housing, income stability, and safety. It is also that the care system itself has been allowed to operate with a scarcity that would be intolerable in a domain treated as essential infrastructure. Federal workforce projections continue to show substantial shortages of psychiatrists, psychologists, counselors, and social workers over the next decade, especially in rural and low-income communities. SAMHSA's crisis-care guidance and the 988 buildout both assume a continuum that includes call centers, mobile crisis teams, and short-term stabilization options. But in many places the call is reachable before the care is. The social determinants position is protecting the claim that this shortage is not a neutral fact of nature. It is a policy choice to underbuild a care workforce, then describe the resulting absence as individual failure, family burden, or public disorder.

The critique of the treatment-without-context model. The social determinants tradition challenges not just the insufficiency of the psychiatric model but its implicit politics. When the policy response to a population experiencing high rates of psychiatric illness is to increase psychiatric services, the framework naturalizes the conditions producing the illness: poverty becomes a "stressor" rather than a policy choice; childhood trauma becomes an "adverse childhood experience" rather than the product of specific failures of protection; housing instability becomes a "precipitating factor" rather than the direct consequence of decades of disinvestment in affordable housing and mental health infrastructure. The social determinants position is protecting the claim that calling a consequence a cause — treating the mental illness as the problem to solve rather than the signal that the conditions producing it are unaddressed — is not politically neutral. It is a frame that locates the problem in the individual rather than the conditions, and that immunizes those conditions from the accountability that the reframing would require.

What the argument is actually about

Whether anosognosia or the treatment system has failed more completely. The sharpest disagreement between the involuntary care position and the civil liberties position is empirical as much as philosophical: does the population of people with untreated serious mental illness on the street consist primarily of people who lack capacity to recognize their need for treatment (the involuntary care claim) or people who have tried the treatment system and found it inadequate, traumatizing, or actively harmful (the civil liberties and recovery claim)? Both things are true of some people. The contested question is the proportion — and that proportion matters enormously for what the right policy response is. A system dominated by anosognosia calls for more capacity to intervene over refusal; a system dominated by rational treatment avoidance calls for better voluntary services. The research does not resolve this cleanly, partly because the definitions are contested, and partly because the population is heterogeneous in ways that aggregate statistics obscure.

Whether the right level of analysis is the individual or the system. The involuntary care tradition and the social determinants tradition are not simply in conflict — they are analyzing the same phenomenon at different levels. The psychiatric capacity position looks at the individual: this specific person is deteriorating, lacks insight, and needs intervention. The social determinants position looks at the population: these patterns of deterioration are not randomly distributed but track poverty, race, trauma history, and housing instability in ways that cannot be explained by individual neurology alone. Both observations are correct. The policy disagreement is about which level of analysis should organize the response — and about whether individual-level interventions can address a population-level pattern, or whether population-level interventions are necessary to change the individual-level outcomes.

Whether the mental health system can be reformed or needs to be replaced. Underneath the specific policy debates — involuntary treatment thresholds, psychiatric bed capacity, diversion programs, housing investment — is a deeper disagreement about the nature of the mental health system itself. The involuntary care tradition wants more of the existing system: more beds, more legal authority to compel treatment, more resources for a recognizable psychiatry. The recovery tradition and the social determinants tradition, in different registers, want something different: peer support instead of or alongside clinical hierarchy, social investment instead of or alongside medical treatment, housing as health care rather than a separate domain. These are not disputes about resource levels but about what kind of institution should receive the resources — and that question exposes deeper disagreements about what mental health is, what causes it, and what healing requires.

Whether mental health care is an obligation the system must guarantee or a benefit it may only partially deliver. This is where the map plugs most directly into the broader healthcare cluster. If behavioral health care is treated as a public obligation, then parity enforcement, crisis-response capacity, bed supply, and community follow-up are not peripheral implementation details. They are the substance of the promise. If it is treated as a service delivered through ordinary market channels, then shortages, narrow networks, and uneven geography become unfortunate but acceptable features of the landscape. The mental health debate becomes especially combustible because the people least able to navigate scarcity are also the people most likely to encounter the state coercively when scarcity turns into crisis. That makes market failure show up not only as untreated illness, but as forced treatment, family collapse, homelessness, and jail.

Beneath the surface: not a dispute about whether people with serious mental illness deserve care — almost everyone agrees they do — but about what care means, who gets to define it, and who bears the cost when the answer is wrong. The involuntary care position protects the people most completely disabled by their illness, the ones for whom the right to refuse has become another name for abandonment. The civil liberties position protects the history of psychiatric coercion and the populations most at risk of being subjected to it again. The criminalization critique protects the observation that the current system is not a policy but an accident — and a racialized one. The social determinants position protects the upstream truth that mental illness is not randomly distributed and that its clustering is not fate. The depth of the challenge is that all four are protecting something real — and that the conditions producing the crisis (chronic underfunding of voluntary care, the de facto criminalization of poverty and psychiatric disability, the political invisibility of the most severely affected) make every individual reform insufficient without the others.