An idea can disappear from the textbook and remain inside the institution.

The history of neurodivergence is not a clean sequence in which superstition was replaced by medicine, bad medicine was replaced by good medicine, and discrimination was replaced by inclusion. It is better understood as a history of changing authorities that repeatedly claimed the power to define which minds were normal, educable, credible, independent, useful, dangerous, defective, curable, or worthy of protection.

Eugenics is now formally rejected as pseudoscience and as a profound ethical failure. The vocabulary of “idiots,” “imbeciles,” “morons,” “feeblemindedness,” “mental defect,” “racial hygiene,” and hereditary unfitness has been removed from legitimate contemporary clinical practice. Autism is no longer classified as childhood schizophrenia. The “refrigerator mother” theory has been discredited. Federal law now recognizes disability rights, educational access, and the right to participate in society. These changes are real and consequential.^{[1][2][18][19][22][36]}

But removing a word does not automatically remove the social theory that gave the word power. A bureaucracy can stop saying unfit while still organizing people according to assumptions about productivity. A clinician can stop saying mental age while still speaking to a disabled adult as though the adult were a child. A school can use the language of inclusion while removing a student from ordinary participation because their communication, movement, or distress is interpreted primarily as “behavior.” A service system can reject institutionalization while still requiring disabled adults to surrender privacy, choice, movement, money, or decision-making in order to obtain support.

This is the article’s central claim: the oppression experienced by autistic, neurodivergent, intellectually disabled, psychiatrically disabled, and otherwise differently minded people today cannot be understood only by looking for openly hateful beliefs. It also operates through inherited classifications, professional habits, educational models, legal presumptions, diagnostic prototypes, and ordinary language. Those structures can preserve the logic of an older hierarchy even when the people using them sincerely believe they have rejected that hierarchy.

That distinction matters because it makes the analysis more, not less, serious. If oppression required an explicit commitment to eugenics, it would be easy to identify and reject. The harder problem is the survival of assumptions that appear neutral because they have been naturalized: that there is one correct way to communicate; that eye contact signals honesty; that speech signals intelligence; that independence is the measure of adulthood; that distress is noncompliance; that atypical social behavior reflects deficient empathy; that disability reduces credibility; or that professional interpretation automatically outranks lived experience.

The result is what this article calls lexical modernization without conceptual reform: the terminology becomes more respectful while the underlying hierarchy remains partially intact.

Three different kinds of continuity have to be separated

This framework is also why the article refuses two easy stories. The first is the triumphalist story in which old cruelty was simply replaced by modern science. The second is the totalizing story in which every diagnosis is merely oppression. Both erase evidence. Diagnostic language can give people access to services, explanation, community, and legal protection. The problem is not description itself. The problem is the historical tendency to let a description expand until it becomes a theory of the person's worth.

Village, household, hospital, refuge: there was never only one historical way to live with difference.

Before the modern psychiatric hospital consolidated diagnosis, treatment, custody, labor, and recordkeeping behind one administrative wall, European communities developed many overlapping ways of responding to poverty, illness, cognitive difference, epilepsy, madness, dependency, age, pilgrimage, homelessness, and social exclusion. These arrangements were not uniformly humane, and they should not be romanticized. But they matter because they demonstrate a point often forgotten in modern debates: segregated institutional custody was a historical choice, not the inevitable social response to neurological difference.

Some historical arrangements placed people inside households and ordinary neighborhoods. Others created charitable settlements. Others built self-contained villages whose residents worked, worshiped, ate, and lived together. The same village form could therefore produce radically different relationships to power. A village might increase belonging, or it might remove a population from public life. A workshop might support meaningful occupation, or it might become compulsory labor. A charitable foundation might secure housing, or it might make support conditional on obedience, religion, class status, or institutional rules.

PLATE 01 / GEEL

PLATE 02A / FUGGEREI

PLATE 02 / YORK

The disability village: refuge, segregation, or both?

In the nineteenth century, village-like institutions became a recurring answer to disability. Their planners often objected to overcrowded asylums and poorhouses. They imagined cottages, farms, workshops, churches, schools, and open space. This could look more humane than a locked ward. But the fundamental question was not architectural. It was political: could the resident leave, choose, refuse, marry, control wages, make relationships, and remain a citizen of the surrounding community?

PLATE 03

PLATE 04

Neuerkerode and Bethel are particularly important because they destroy a simple good-institution/bad-institution story. Religious and charitable communities could create shelter, work, social worlds, and long-term belonging. They could also participate in classificatory systems that judged reproduction, heredity, and social value. Under a coercive state, the records generated for care could become records used for selection. The same file that arranged support could be repurposed to decide sterilization or transfer.

Craig Colony and the seductive idea of the therapeutic colony

New York acquired roughly 1,900 acres in 1894 to establish Craig Colony for Epileptics, one of the earliest American public colony systems devoted to epilepsy.^[57] The colony form offered farms, cottages, work and distance from overcrowded urban institutions. Yet American “colonies” for people labeled epileptic or “feebleminded” became inseparable from segregation and, in some jurisdictions, sterilization policy. Their existence illustrates a recurring pattern: institutions are often created in the language of protection and rehabilitation, then gradually become mechanisms for managing a population considered difficult to accommodate in ordinary society.

To compare historical care models, separate the building from the relationship.

A picturesque village can still be coercive; an ordinary apartment can still be institution-like if someone else controls visitors, bedtime, money, food, communication, and movement. Conversely, a specialized setting can sometimes provide chosen community, expertise, and safety. The useful historical comparison is therefore not “institution versus community” as architecture alone. It is a set of questions about membership, exit, privacy, decision-making, labor, family contact, and who sets the rules.

Geel matters because support was distributed through civic life rather than concentrated behind one wall.

Geel's family-care tradition developed from a religious pilgrimage culture associated with Saint Dymphna into a long-running system in which boarders lived with local families while psychiatric services operated around the town. Historians and clinicians have repeatedly emphasized that the model changed across centuries and should not be romanticized. Boarders remained subject to supervision and social hierarchy. Yet Geel offers a conceptual contrast to the nineteenth-century total institution: the supported person could be a recognizable participant in streets, shops, households, and routines shared with other residents rather than primarily a patient contained within a hospital.^[50]

The lesson is not “copy medieval Geel.” The lesson is that ordinary membership can itself be therapeutic and protective. Contemporary systems often treat community participation as an outcome that happens after treatment. Geel reverses the sequence: community is part of the environment in which support occurs.

Fuggerei offers a different comparison: sometimes the relevant intervention is not psychiatric at all.

Because the Fuggerei was founded as social housing rather than a disability service, it helps expose a category error that continues today. A person may be distressed because they lack safe housing, money, food, transportation, predictable routines, or social protection. If every problem encountered by a neurodivergent person is routed through clinical services, social deprivation can be misread as individual pathology. Historical charitable housing reminds us that material conditions are not secondary to mental health; they are often part of the causal environment.^[54]

This power audit also clarifies why Neuerkerode, Bethel, the York Retreat, Geel, and colony systems cannot be placed on a simple humane-to-inhumane line. Each belongs to a different moment and institutional tradition. What matters for the argument of this article is that care models repeatedly oscillate between two impulses: make a place where difference can belong, and make a place where difference can be removed from everyone else. Sometimes the same institution contains both impulses.

When “difference” became a problem to be prevented from reproducing.

The movement’s importance to disability history is direct. American eugenicists categorized people as “unfit,” “feebleminded,” “defective,” or “antisocial.” Disability, poverty, sexuality, ethnicity, race, and perceived intelligence became entangled. Indiana enacted the first compulsory sterilization law in 1907. By the 1930s, more than thirty U.S. states had sterilization laws, and at least 60,000 people were involuntarily sterilized under state eugenic programs over the twentieth century. People with disabilities were among the populations disproportionately targeted.^[1][2][3][5]

Psychometrics changed the scale of sorting

Alfred Binet’s early intelligence test was designed for an educational purpose: identifying children who needed additional help. In the United States, figures such as Henry H. Goddard imported and transformed intelligence testing into a much broader system of classification. Goddard promoted hereditarian theories of “feeblemindedness,” popularized family-pedigree narratives, and introduced moron as a technical category. Categories such as idiot, imbecile, and moron became associated with purported levels of intellectual capacity. The tests appeared to provide a number where previously there had been judgment.^[3]

Numbers can increase precision, but measurement does not eliminate ideology. If the test is culturally narrow, if the construct being measured is treated as more comprehensive than it is, if language proficiency is confused with reasoning, if disability is confused with moral worth, or if a score is used to determine citizenship rights, reproductive rights, education, or freedom, the instrument can become a mechanism of oppression.

This history is especially important to contemporary neurodivergence because modern systems continue to rely heavily on standardized measurement. Standardized tools can be clinically useful. The lesson from eugenics is not “measurement is evil.” The lesson is that measurement becomes dangerous when a limited measure is allowed to stand in for the whole person and when the category determines how much humanity, autonomy, credibility, or opportunity the person is allowed.

Buck v. Bell: when diagnosis became state power

In 1927, the U.S. Supreme Court upheld Virginia’s compulsory sterilization statute in Buck v. Bell. Carrie Buck had been committed to the Virginia State Colony for Epileptics and Feeble-Minded and was designated “feebleminded.” The Court accepted the state’s eugenic rationale. Justice Oliver Wendell Holmes Jr.’s infamous five-word conclusion—“Three generations of imbeciles are enough”—compressed an entire hierarchy into a sentence: the state could decide that some family lines should end.^[4]

Later historical research demonstrated the profound injustice surrounding Buck’s case, including the manipulation of evidence and the social circumstances that were medicalized as hereditary defect. The broader lesson is larger than one court opinion. A diagnostic label became evidence of reproductive unworthiness because the society using the label already believed that dependence, poverty, sexual vulnerability, cognitive difference, and institutionalization revealed inferior heredity.

After Nazi atrocities exposed the extreme consequences of racial hygiene and state eugenics, explicit eugenic ideology became discredited. But the end of a movement does not erase the institutions, professional disciplines, records, testing practices, and cultural habits that grew while it was influential. That is why the next part of the history—Nazi psychiatry and early autism research—requires both connection and distinction.

PLATE 05 / GALTON

Eugenics did not start with the Nazis—and it did not start with genetics as we know it

To understand why eugenics spread so effectively, it helps to move backward before Galton. Nineteenth-century European psychiatry was already preoccupied with heredity. French psychiatrist Bénédict Augustin Morel's mid-century theory of degeneration proposed that hereditary “taint” could worsen across generations, blending psychiatric observation with moral, social, and biological judgments. The theory was scientifically unstable, but it supplied a powerful story: poverty, alcoholism, criminality, sexual behavior, intellectual disability, and mental illness could be imagined as visible symptoms of a declining family line.^[60]

That story mattered because it changed the scale of concern. If a person's distress was an individual medical problem, the state might ask how to treat that person. If it was evidence of hereditary degeneration, the state could begin asking how to protect the population from that person's descendants. This is the conceptual move that makes eugenics distinct: the target of intervention becomes the imagined future population.

Positive and negative eugenics: reward one future, prevent another

Historical eugenicists often distinguished between positive eugenics—encouraging people judged desirable to reproduce—and negative eugenics—discouraging or preventing reproduction among people judged undesirable. That vocabulary makes the movement easier to see as a complete social program rather than only a sterilization campaign. Better Baby competitions, “Fitter Families” contests, marriage advice, heredity exhibits, and pronatalist messaging helped teach the public what a supposedly desirable family looked like. Sterilization laws, segregation, institutional commitment, and marriage restrictions targeted the other side of the hierarchy. The American Eugenics Society used fairs, publications, public exhibits, and economic arguments about the cost of caring for “mentally ill” children to popularize the framework.^[1]

The public pedagogy is crucial. Eugenics did not spread only because legislators secretly imposed it. It spread because experts, universities, museums, fairs, textbooks, churches, philanthropists, social reform organizations, newspapers, and professional societies helped make its premises feel modern and reasonable. A population learned a lexicon in which “fitness” sounded empirical and dependency sounded hereditary.

Why eugenics could look “progressive,” modern, and humanitarian to its supporters

Another reason eugenics is easy to misunderstand is that it cannot be confined to one modern political camp. In the United States and Britain, eugenic arguments circulated among physicians, scientists, social reformers, philanthropists, administrators, clergy, educators, and politicians who disagreed about many other questions. Progressive-era faith in expertise and administrative efficiency could combine with hereditarian assumptions: if crime, poverty, disability, institutional dependency, or psychiatric illness were primarily inherited, then preventing the birth of people predicted to have those traits could be framed as a more “efficient” solution than changing working conditions, housing, education, poverty, healthcare, or social exclusion.^[76][78]

This is the bridge between eugenics and administrative systems. The movement offered officials a way to convert difficult social questions into population-management problems. Why is an institution overcrowded? could become why are “defective” families reproducing? Why does this child struggle in school? could become what hereditary grade of intelligence does this child possess? Why are people poor? could become what kind of stock produces dependency? Structural problems were repeatedly relocated into bodies and family lines.

PLATE 05B / PUBLIC PEDAGOGY

“Feeblemindedness” was powerful precisely because it was elastic

The historical category usually translated as feeblemindedness did not correspond neatly to a modern diagnosis. It could capture intellectual disability, learning differences, developmental delay, poverty, perceived sexual impropriety, delinquency, institutional dependency, language barriers, low test performance, and behavior authorities found difficult. Women and girls could be targeted through sexual and moral judgments; immigrants could be judged through tests they did not understand; poor families could have structural deprivation interpreted as inferior heredity. The term operated as a bridge between a medical description and a social accusation.

This is one reason historical terminology must be read as governance, not merely vocabulary. A label's effect came from the decisions attached to it. If “feebleminded” meant you could be segregated, prevented from marrying, sterilized, removed from ordinary schooling, or denied legal credibility, then the diagnosis functioned as a change in civic status.

Psychology supplied the score; psychiatry supplied categories and captive archives

The relationship between eugenics and psychiatry was institutional rather than metaphorical. British psychiatric history documents psychiatrists publicly supporting eugenics, mental-deficiency legislation, and debates over sterilization; psychiatric hospitals simultaneously generated the clinical populations and family histories on which hereditarian claims could be built.^[44] In psychology, intelligence testing could be redirected from identifying educational support needs toward large-scale classification, with scores treated as stable markers of hereditary capacity.^[47] At London's Maudsley Hospital in the interwar period, historical research shows that intelligence testing intersected directly with legal certification under the Mental Deficiency Act: children scoring below specified levels could be referred for certification and institutional care.^[48]

This does not mean that psychological testing itself is eugenics, or that psychiatric diagnosis is inherently eugenic. It means that those tools were historically embedded in systems in which classification could change a person's freedom. The ethical lesson is about what happens after the score.

PLATE 06 / THE NETWORK

The Eugenics Record Office: turning family life into a database

The Eugenics Record Office at Cold Spring Harbor, established in 1910 under Charles Davenport and directed operationally by Harry Laughlin, trained fieldworkers, distributed questionnaires, constructed pedigrees, and collected information about traits it treated as hereditary. Its records included categories such as “feeblemindedness,” criminality, alcoholism, epilepsy, and psychiatric conditions. The ambition was not simply to study families. It was to create a body of evidence that could guide population policy.^[1]

The pedigree chart gave social judgment a visual grammar. Circles and squares, shaded or unshaded, made complex lives look like a transmissible binary property. Poverty could disappear from view. Trauma could disappear. School access could disappear. Institutional bias could disappear. The diagram encouraged the reader to see a family line rather than a social history. The methodological error was not merely insufficient data; it was the conversion of multidimensional circumstances into hereditary essence.

Immigration: disability language became border technology

Eugenics also connected disability classification to national belonging. Harry Laughlin's congressional testimony and eugenic analyses supported arguments for restricting immigration, particularly from southern and eastern Europe. The Immigration Act of 1924 formalized a quota system built around national-origin hierarchies and exclusion.^[1] At ports of entry, medical inspection and intelligence testing could turn unfamiliar language, education, poverty, bodily difference, or test performance into evidence that someone was undesirable as a future citizen. Disability history and immigration history therefore overlap: the category of the “unfit” was never only about medicine. It was also about who the nation imagined should compose its future population.

Buck v. Bell was a pipeline, not an isolated bad opinion

Carrie Buck's sterilization is often remembered because of Holmes's notorious sentence. But the decision is more informative when treated as the visible endpoint of an institutional pipeline. A young woman entered a custodial institution. Experts assigned a hereditary category. Family members were represented through the same classificatory lens. The institution sought sterilization. A statute authorized it. Courts deferred to professional claims. The Supreme Court constitutionalized the state's power. At every stage, an earlier classification acquired more authority until a social judgment became irreversible surgery.

This is why eugenic history matters to contemporary systems even when the outcome is no longer sterilization. A modern administrative pipeline can still magnify an early assumption: a school description becomes a medical note; the note becomes a service eligibility code; the code shapes a risk assessment; the assessment affects guardianship, housing, crisis response, or hospital treatment. The ethical question is whether each stage independently tests the assumption or merely inherits it.

Nazi racial hygiene was a radicalization of an international movement, not its point of origin

When Nazi Germany enacted the 1933 Law for the Prevention of Offspring with Hereditary Diseases, it mandated forced sterilization for categories including schizophrenia and “hereditary feeblemindedness.” The regime ultimately sterilized hundreds of thousands of people. Eugenic “racial hygiene” then escalated into the systematic murder of institutionalized disabled patients under the so-called “euthanasia” program.^[45][46] This was a genocidal dictatorship using eugenics with a scale and murderous purpose that should never be flattened into an analogy for every coercive medical practice.

At the same time, treating Nazism as though it invented eugenics creates another historical error: it allows Britain, the United States, Scandinavia, Canada, and other countries to imagine the ideology as foreign. American sterilization statutes, eugenic research, immigration restrictions, and public campaigns were already well established before Hitler came to power. Nazi racial hygienists operated inside an international field and radicalized it through totalitarian law and mass killing.

1945 did not switch eugenics off

The Holocaust discredited explicit racial hygiene, and scientific criticism had already exposed the weak hereditarian methods of many American eugenicists. Yet sterilization laws and practices persisted. California's statute remained on the books until 1979; Alexandra Minna Stern's historical work connects approximately 20,000 sterilizations in California state institutions with later sterilization abuses affecting working-class Mexican-origin women at Los Angeles County hospital in the 1970s.^[58] Across the United States, the twentieth-century history of reproductive coercion disproportionately affected disabled people, poor people, women, racialized communities, Indigenous people, immigrants, and institutionalized populations.

The postwar lesson is therefore not that every later sterilization was “the same” as early eugenics. It is that an established administrative justification can migrate. Words such as racial hygiene may disappear while arguments about dependency, cost, irresponsible reproduction, presumed incapacity, or the quality of a future child's life can continue to pressure disabled people.

The most durable inheritance is not a word. It is a decision architecture.

Historical eugenics becomes relevant to the present when we can identify a repeated sequence of institutional reasoning. First, a category is treated as a stable essence. Second, the essence is used to predict a person's future. Third, the prediction is generalized from one domain to the whole person. Fourth, experts are granted greater authority than the person being classified. Fifth, restriction is justified as protection, efficiency, prevention, or social necessity. The vocabulary may now be clinically respectable, but the architecture of the decision can still reproduce the older hierarchy.

For autistic, neurodivergent, intellectually disabled, psychiatrically disabled, nonspeaking, multiply disabled, or otherwise differently minded people, this can appear today as a presumption that support needs cancel adulthood; that a communication difference cancels credibility; that a diagnosis predicts parenting ability; that low measured performance predicts global incapacity; that distress proves dangerousness; that “challenging behavior” explains away environmental harm; or that inclusion must be earned by approaching a neurotypical norm. These are not identical to compulsory sterilization. They are connected historically because they reuse the same conversion error: a description of difference is turned into a ranking of personhood and then into permission for someone else to decide.

CASE FILE / BUCK v. BELL · 1927

How a label became legal authority over Carrie Buck’s body.

What happened: Virginia selected Carrie Buck as a test case for its sterilization law. She was institutionalized and characterized as “feeble-minded.” Her mother had also been institutionalized, allowing eugenic advocates to narrate disability, poverty, sexual vulnerability, and family circumstances as hereditary degeneration. The U.S. Supreme Court upheld the law in Buck v. Bell.

Why this case belongs in an autism/neurodivergence history: not because Buck can or should be retrospectively diagnosed, but because the case demonstrates the administrative pathway this article is tracing. A social and clinical classification was treated as evidence of hereditary prediction; hereditary prediction was treated as justification for state control of reproduction.

What gets lost in shorthand: the familiar phrase “three generations” can make the case sound like a scientific finding. It was a legal and institutional construction produced through the categories and power of its time.

01OBSERVEdifference becomes a case →02CLASSIFYa label compresses the person →03PREDICTcategory becomes presumed future →04RANKfuture becomes social value →05CONTROLauthority moves to the institution/state

Select a step to see what changes when description becomes authority.

Related source: [149]

Nazi psychiatry made “social usefulness” a matter of life and death.

German and Austrian eugenics did not emerge from nowhere. Nazi racial hygiene drew upon broader international eugenic ideas while radicalizing them through a totalitarian state committed to racial purification, compulsory sterilization, and mass murder. Disabled adults and children were targeted under so-called “euthanasia” programs. Psychiatry, pediatrics, genetics, and public administration participated in deciding whose lives were considered burdensome or “unworthy.”

This history is directly relevant to autism because Hans Asperger conducted his work in Nazi-era Vienna. Archival scholarship by historian Herwig Czech has shown that Asperger adapted to the Nazi medical system, used language aligned with racial-hygiene priorities, and referred some children to institutions connected with the killing program. The older heroic story in which Asperger straightforwardly protected autistic children from Nazism is not supported by the full archival record.^[16]

Yet even here the important point is not to turn autism history into a morality play with one villain. The more revealing issue is the sorting principle. Children judged capable of becoming socially useful could be described in ways that preserved educational possibility. Children judged to have more profound disabilities could be categorized as burdens with devastating consequences. Usefulness became a condition of mercy.

That principle has a recognizable modern echo whenever disabled people are defended primarily by listing exceptional talents: the autistic programmer, mathematical savant, artist, engineer, musician, or worker with extraordinary concentration. Positive representation can correct stereotypes, but it can also unintentionally reproduce a hierarchy in which neurodivergent people must demonstrate unusual economic or intellectual value to justify inclusion.

This distinction is essential for a neurodiversity framework. Neurodiversity does not require denying disability, suffering, communication difficulty, intellectual disability, epilepsy, severe sensory distress, self-injury, or the need for lifelong support. It rejects the inference that greater support needs imply lesser personhood. A society is not neurologically inclusive only when it celebrates the neurodivergent people it finds impressive.

The bureaucratic horror was built from ordinary-looking forms

The T4 killing program is often remembered through gas chambers and transport buses, but its administrative beginning was paperwork. Institutions received registration forms asking about diagnosis, duration of hospitalization, ability to work, criminal history, and related information. Reviewers could make life-and-death decisions from files. The United States Holocaust Memorial Museum describes the program as the systematic murder of institutionalized people with mental and physical disabilities and places it within the radicalization of eugenic “racial hygiene.”^[46]

The detail about work capacity matters to the argument of this article. “Usefulness” did not remain an abstract moral judgment; it became an administrative data field. Once a state has accepted that a person's right to live depends on economic utility or genetic value, a diagnosis can become evidence in a lethal calculation.

Bethel shows why institutional history must resist heroic myths

Bethel's own contemporary historical project documents a complicated record. More than 1,665 people were forcibly sterilized in its hospitals under Nazi law.^[52] At the same time, its leadership opposed the systematic T4 killing program on Christian grounds; when a state commission examined residents, 446 people were considered potentially endangered, but the anticipated mass deportations did not occur before the centralized program was halted.^[53] This is historically uncomfortable because it resists a clean binary between resistance and complicity. An institution could oppose murder while participating in coercive sterilization. That is exactly why ethical analysis must examine specific powers and practices, not the moral reputation of an institution as a whole.

PLATE / HADAMAR, 1945

Aktion T4 was not merely “Nazi ideology applied to hospitals.” It was a medical-administrative killing system.

The Nazi “euthanasia” program targeted disabled children and institutionalized adults whose lives were classified as burdensome, hereditary threats, or “life unworthy of life.” The United States Holocaust Memorial Museum documents how physicians and administrators used questionnaires sent to hospitals and care institutions. The forms gathered diagnoses, duration of institutionalization, ancestry, and—revealingly—capacity for work. Medical “experts” reviewed paperwork and selected people for transfer. The process converted murder into administrative workflow.^[46][114]

This is one of the most important historical connections in the entire article. The killing decision did not require a doctor to know the person. It could be made from a file. Once a human life was compressed into diagnosis, productivity, institutional duration, and hereditary status, the document became a proxy for the person. Bureaucratic distance made moral catastrophe easier to administer.

The child “euthanasia” program shows how developmental medicine could be captured by a hierarchy of future value.

Disabled infants, children, and adolescents were reported to health authorities and transferred to special wards where thousands were murdered through medication overdose, starvation, or neglect. The logic was prospective: the child's predicted future dependency was treated as evidence against the value of their future life. The historical warning for modern neurodevelopmental practice is not that contemporary early intervention resembles T4. It is that predictions about future burden are ethically different from descriptions of present support needs.^[114]

A child who needs intensive lifelong support may indeed require substantial public resources. A rights-based system treats that fact as a planning obligation. A eugenic system treats the same fact as evidence that the life itself has negative value. The difference between those positions must remain explicit whenever public discourse turns disability into cost-benefit arithmetic.

The six T4 killing centers connected psychiatric institutions to technologies later used in the Holocaust.

Beginning in 1940, centralized killing centers at Brandenburg, Grafeneck, Hartheim, Sonnenstein, Bernburg, and Hadamar used carbon monoxide gas to murder disabled people transferred from institutions. The program's formal centralized phase was halted in 1941 after public protest, but killings continued by medication, starvation, neglect, and decentralized methods. USHMM estimates that at least 250,000 people with disabilities were murdered in the broader Nazi “euthanasia” program between 1939 and 1945. Personnel and techniques developed in T4 later contributed to the machinery of mass murder used against Jews and other targeted populations.^[113][114]

Hadamar demonstrates why “just following the medical record” is not ethically neutral.

At Hadamar, almost 15,000 people were killed between 1941 and 1945. During the initial gassing phase, approximately 10,000 patients were murdered with carbon monoxide. In the later phase, thousands more died through lethal medication and deliberate neglect. Victims included disabled Germans, elderly people, forced laborers, children, and others classified as expendable. Physicians and nursing staff did not merely fail to protect patients; members of the healing professions became direct agents of killing.^[115]

That history does not justify treating modern clinicians as suspect by association. It does justify a permanent ethical rule: professional expertise cannot be allowed to become a substitute for the equal moral status of the patient. Medicine can determine that a condition is severe. It cannot determine that a human being's claim to life, communication, intimacy, or community is proportionally reduced by severity.

The postwar break with eugenics was incomplete because institutions, personnel, and categories did not vanish in 1945.

After the war, Nazi racial hygiene was discredited internationally, but coercive sterilization continued in parts of the United States for decades, and disability institutions remained powerful across Western countries. The language of explicit racial purification became politically toxic; the administrative questions of dependency, “mental deficiency,” productivity, institutional placement, and reproductive control did not disappear at the same speed.^[58][59][71]

“Never again” requires more than rejecting murder.

The narrowest lesson of Nazi disability history is obvious: killing people because they are disabled is an atrocity. The more demanding lesson concerns the sequence that made the atrocity administratively thinkable: define a population as biologically defective; portray support as social waste; treat professionals as authorized judges of future value; separate people from ordinary community; convert individuals into files; then make rights conditional on the classification. Modern democracy rejects the final outcome. A serious historical education asks us to recognize the earlier steps as well.

The diagnosis changed because the category changed—not because autistic people suddenly appeared.

The Diagnostic and Statistical Manual of Mental Disorders is often treated in popular culture as a catalog of fixed diseases. Historically, it is better understood as a professional classification system that changes as evidence, theory, consensus, and clinical priorities change. Autism is one of the clearest examples.

Historical reviews of autism diagnosis show that DSM-III was a decisive break: research had increasingly demonstrated that autism and schizophrenia had different developmental courses, family patterns, and treatment implications. Later manuals broadened and reorganized the category as clinicians encountered presentations that did not fit Kanner’s narrow prototype.^[14][17]

This matters politically as well as clinically. A diagnostic threshold is a border. On one side, a person may qualify for explanation, accommodations, insurance coverage, school services, legal recognition, workplace adjustments, specialist care, or a coherent account of lifelong difference. On the other side, the same person may be described as difficult, anxious, oppositional, obsessive, socially immature, rude, lazy, unstable, overly sensitive, or simply “not trying.”

Every major diagnostic revision therefore changes more than terminology. It changes who becomes legible to the system.

The prototype problem

Even when formal criteria broaden, clinicians and educators may continue using older mental prototypes. Autism research and diagnostic practice historically centered boys and children with more externally visible presentations. Contemporary research documents the ways girls and women may camouflage, have interests that appear socially typical in content, or present in ways that clinicians fail to recognize. Camouflaging is also reported across genders and can contribute to exhaustion, poor mental health, and delayed recognition.^[31][32][39]

The same issue affects adults, racial and ethnic minorities, people with intellectual disability, nonspeaking people, people with psychiatric comorbidities, and people whose support needs fluctuate by environment. A manual can be updated while the clinician’s mental image remains decades behind.

From categorical error to social consequence

A mistaken category is not merely a semantic mistake when institutions use it to allocate power. If autistic shutdown is interpreted as defiance, the response may be discipline rather than reduced sensory load. If atypical pain communication is interpreted as anxiety, physical illness may be missed. If direct speech is interpreted as hostility, a worker may be penalized. If an adult’s reliance on AAC is interpreted as low comprehension, consent may be routed through someone else.

That is why critique of historical DSM criteria is not an argument for abolishing diagnosis. Diagnosis can provide access, legal protection, self-knowledge, and treatment of co-occurring conditions. The critique concerns diagnostic imperialism: the tendency to let a clinical description become a total theory of the person.

The DSM is also an administrative document

The DSM is usually discussed as a clinical taxonomy, but in the United States a diagnosis can also function as a passport into or out of insurance coverage, disability documentation, accommodations, research cohorts, school evaluations, service systems, and legal arguments. That does not make the manual eugenic. It does mean that diagnostic wording has consequences beyond the consulting room. The more resources depend on a category, the more the category becomes a gate.

This creates a paradox for neurodivergent communities. Diagnostic categories can pathologize difference, yet they can also provide a name for experience, protection from discrimination, access to services, community, and a basis for demanding accommodation. Abolishing categories without replacing the rights attached to them can therefore harm the very people a depathologizing reform intends to help.

From “disease entity” to spectrum does not automatically change the power relation

DSM revisions can broaden criteria, replace stigmatizing names, recognize dimensionality, and improve reliability. But terminology reform does not guarantee relational reform. A clinician can use modern DSM-5 language while holding a nineteenth-century assumption that the professional knows the person's inner life better than the person does. A school can use “autism spectrum disorder” while treating autistic communication as noncompliance. A service agency can use “support needs” while making support conditional on proving incapacity. The relevant historical question is therefore not only what did the manual say? but what kinds of authority did institutions attach to the category?

The walls came down. Some of the institutional logic moved with us.

During the twentieth century, exposés of institutional abuse, family activism, disability-rights organizing, new medications, fiscal pressures, and changing legal standards contributed to deinstitutionalization. Large state hospitals and developmental institutions declined. In education, the 1975 Education for All Handicapped Children Act—later IDEA—opened public-school access to millions of disabled students who had previously been excluded or underserved.^[18][36]

The Americans with Disabilities Act of 1990 explicitly recognized a national history of isolation and segregation and identified continuing discrimination in employment, housing, education, transportation, communication, institutionalization, health services, and public life.^[19] In 1999, the Supreme Court’s Olmstead v. L.C. decision recognized unjustified institutional isolation as a form of disability discrimination under Title II of the ADA under specified conditions.^[20]

These are profound civil-rights achievements. But deinstitutionalization is not complete merely because a person sleeps in a house instead of a state hospital.

An institutional relationship is defined not only by architecture but by power: Who chooses when you wake? Who decides what you eat? Who controls transportation? Who can enter your room? Who manages your money? Who can approve relationships? Who determines whether you may leave? Who interprets your distress? Who records your behavior? Who can punish noncompliance? Who speaks for you in medical appointments? Who decides which risks you are permitted to take?

When those powers remain concentrated in other people or systems, institutional logic can survive in community settings. The same is true in schools, group homes, day programs, hospitals, residential treatment, guardianship arrangements, and even well-intentioned family systems.

Guardianship and the presumption of incapacity

The National Council on Disability has documented major concerns about guardianship, including erroneous assumptions about disabled people’s capacity, insufficient evidentiary foundations in some capacity determinations, inadequate monitoring, and underuse of less restrictive alternatives.^[21][38] Guardianship may be necessary in some circumstances, but the disability-rights critique asks whether support is being used to enable decision-making or to replace the person as decision-maker.

This is one of the clearest places where an old hierarchy survives in contemporary legal clothing. The historical category of the “non-consenting adult” often merged childhood, cognitive disability, and mental incapacity. Modern law is more sophisticated, yet disabled adults can still encounter a presumption that dependence justifies substituted authority.

Deinstitutionalization changed geography faster than it changed authority

Moving from a large state hospital to a group home, family home, school program, residential campus, or community service can be a major civil-rights gain. But institutionalization is not defined only by brick walls. Sociologist Erving Goffman's concept of the “total institution” emphasized environments in which sleeping, working, eating, recreation, and authority are organized under one administrative regime. A small setting can reproduce elements of that structure if residents cannot choose schedules, visitors, food, communication, relationships, or exit.

This is why the phrase community-based should describe a distribution of power, not merely a street address. A person can live in a suburban house and still be functionally institutionalized if every meaningful choice requires staff permission. Conversely, a person with profound support needs can exercise substantial autonomy when support is built around reliable communication, supported decision-making, personal relationships, control of routine, and meaningful access to public life.

Deinstitutionalization was not one event, one policy, or one moral victory.

In the United States, the decline of large state hospitals accelerated after World War II through several overlapping forces: public exposure of abusive conditions, changing psychiatric treatments, new federal involvement in mental health, litigation, disability-rights organizing, fiscal incentives, and a growing conviction that people should not be warehoused simply because they were disabled. The 1963 Community Mental Health Centers Act represented enormous optimism that community-based services could replace long-term institutional care. But the actual transition was uneven. Hospital populations fell faster than comprehensive housing, outpatient care, income support, transportation, crisis services, and long-term community supports were built.^{[116][117][136]}

This distinction matters because “deinstitutionalization failed” is often used as an argument for rebuilding institutional solutions. The historical record is more specific: in many places, the institution was reduced without fully funding the community. A policy can therefore be correct about the right to community life and still fail in implementation.

Willowbrook exposed what “care” could mean when people had almost no exit power.

Willowbrook State School on Staten Island became nationally notorious for overcrowding, neglect, and dehumanizing conditions. It was also the site of hepatitis research in which institutionalized children were deliberately infected; historical ethics reviews continue to discuss the compromised consent and exploitation of a captive, highly vulnerable population. The significance is not only that conditions were terrible. Willowbrook shows how institutional dependency can weaken ordinary safeguards: parents may be desperate for admission, residents may have no practical ability to leave, and researchers or administrators may control access to resources.¹³⁴

The modern ethical lesson is broader than research consent. When one organization controls housing, care, medication, transportation, communication, employment, and access to family, “agreement” inside that system may not mean the same thing as freely revocable consent.

Pennhurst and related litigation transformed institutionalization into a civil-rights question.

During the 1970s and 1980s, litigation over institutions for people with intellectual and developmental disabilities helped shift the public framework away from charity and custodial benevolence toward enforceable rights, education, habilitation, and community living. This was part of a larger disability-rights transformation in which disabled people increasingly rejected the role of passive recipients and demanded political authority over the services governing their lives.^[118]

The 1977 Section 504 sit-in changed the social image of disability itself.

Section 504 of the Rehabilitation Act prohibited disability discrimination in federally funded programs, but implementation regulations were delayed. In April 1977, disabled activists occupied federal offices; the San Francisco occupation lasted 25 days. The Smithsonian's history of the protest emphasizes how the sit-in inverted public stereotypes of disabled people as dependent, apolitical, or incapable of collective action. Cross-disability organizing—and material support from groups including the Black Panther Party—helped sustain the protest until the regulations were signed.^[119]

This moment is conceptually central to the article because it changes the subject of disability policy. Under a paternalistic model, the question is What should society do for these people? Under a civil-rights model, the question becomes What barriers and discriminatory rules prevent these citizens from exercising rights they already possess?

These frameworks overlap rather than cleanly replacing one another. A person may need excellent medical treatment and civil-rights protection at the same time. The danger appears when treatment becomes the condition for citizenship—when the person must become less visibly disabled before ordinary rights are recognized.

Community living can reproduce institutional power at a smaller scale.

Moving from a 1,000-bed campus to a six-person home is a major change in scale, but scale alone does not answer the rights question. If staff still control food, bedtime, visitors, relationships, internet access, money, transportation, and the front door, a small residence can preserve what disability scholars call institutional characteristics. Conversely, a person with 24-hour staffing can live an ordinary community life if the support is organized around their choices rather than around provider convenience.

Olmstead made unnecessary segregation a discrimination issue.

In 1999 the U.S. Supreme Court's Olmstead v. L.C. decision held, under Title II of the ADA as interpreted by the Court, that unjustified segregation of people with disabilities can constitute discrimination. ADA.gov's current community-integration page summarizes the basic principle: state and local services must provide community-based services when appropriate, not opposed by the person, and reasonably accommodated within the public system.^[20][120]

The deeper historical connection is striking. Nineteenth-century institutions often justified separation by asserting that disabled people needed a different social world. Disability-rights law asks the reverse question: what modifications are required so that the ordinary social world is accessible? That is not merely a change of location. It is a reversal of who must adapt.

CASE STUDY / DEINSTITUTIONALIZATION

Closing a building is not the same as redistributing power.

AN INSTITUTION IS MORE THAN A BUILDINGWhere do you live?Who sets your schedule?Who can enter your room?Who controls your money?Who decides who you see?Can you refuse?Can you leave?Who believes your complaint?

The ADA integration mandate and Olmstead matter because unnecessary segregation can be discrimination. The contemporary question is not “large institution or small group home?” It is whether services enable life in the community with meaningful choice, access, relationships and participation. See Community Integration source →

Related source: [152]

The old hierarchy survives most efficiently when it no longer sounds old.

Language is not merely cosmetic. Words shape attention. They tell the listener which features of a person matter and what kind of explanation should follow. Yet language reform can fail when a new term is inserted into an unchanged sentence.

Consider the historical sequence from idiot, imbecile, and moron, through mental deficiency and mental retardation, to intellectual disability. The shift toward intellectual-disability terminology was important enough that the United States enacted Rosa’s Law in 2010 to replace “mental retardation” in federal law.^[22][33] But respectful terminology does not automatically prevent a person with intellectual disability from being treated as childlike, denied choice, or excluded from decision-making.

The same phenomenon appears in autism. “High functioning” can become a reason to deny support. “Low functioning” can become a reason to deny autonomy. “Support level” can be useful when describing current needs, but can become another global rank if people treat it as a fixed measure of intelligence, humanity, or future potential. “Behavior” is a legitimate descriptive term, but in institutional settings it can erase communication: pain behavior, escape behavior, refusal behavior, attention-seeking behavior. The noun centers what observers can see while removing the person’s reason for doing it.

The goal is not to create a prohibited-word list. Any term can become oppressive if it is used to collapse a person into a rank. The more durable reform is conceptual: describe specific needs, specific contexts, specific risks, specific preferences, and specific accommodations rather than converting a multidimensional person into a single status.

Euphemism can modernize the hierarchy without removing it

Disability terminology changes partly because communities reject words that have accumulated contempt. That is healthy. But a recurrent linguistic process—sometimes called the “euphemism treadmill”—means the contempt can migrate into the replacement term if the social status of the people remains unchanged. A clinically respectable word can become an insult; a new word replaces it; the underlying hierarchy then begins attaching to the new word.

The solution is therefore not to stop updating language. It is to update the conceptual grammar beneath the language. “Support level” should answer a practical question about a context, not become a caste. “Behavior” should prompt a functional and communicative inquiry, not erase intention. “Safety” should describe a specific risk and evidence-based response, not become a universal warrant for control. “Capacity” should be decision-specific and support-sensitive, not a global judgment of personhood.

Language does not merely offend. It can compress a decision into a noun.

Clinical and educational shorthand is useful because professionals need concise records. The danger is semantic compression: a phrase that originally describes one situation becomes a portable explanation for everything the person does. “Behavioral” can swallow pain. “Low functioning” can swallow uneven skills. “Poor insight” can swallow disagreement. “Noncompliant” can swallow inaccessible communication, fear, side effects, trauma, or a competent refusal.

Rosa's Law demonstrates that terminology can and should change. In 2010, federal law replaced references to “mental retardation” with “intellectual disability.”^[22] But history also shows why vocabulary reform cannot be the entire project. A respectful noun can still carry an old decision rule if the person described by it is presumed globally incompetent.

The “euphemism treadmill” explains only part of the problem.

Words for disability often become stigmatized because society uses them as insults, prompting communities and professions to adopt new terms. But changing terms is not merely cosmetic; names can remove demeaning assumptions and improve precision. The deeper problem arises when stigma migrates into the replacement word because the hierarchy attached to the underlying category remains. If “intellectual disability” is eventually used to mean “permanent child,” the problem is not that respectful terminology was pointless. The problem is that the social theory beneath the term was not changed.

Grammar itself can reveal who has agency.

Compare two chart sentences:

The first may be factually defensible, but it deletes environment, sequence, and interaction. The second does not excuse harm; it makes the event analyzable. This difference is central to double empathy. Behavior never occurs in a vacuum, yet records often preserve only the disabled person's contribution to an interaction.

“Special needs” can soften language while obscuring ordinary rights.

The phrase can be affectionate and is still widely used. But many needs described as “special” are ordinary human needs delivered through a different access route: communication, education, privacy, healthcare, housing, safety, friendship, sexuality, meaningful work, rest, and control over one's body. The adjective can unintentionally make an accommodation sound like an optional favor rather than the means by which an ordinary right becomes usable.

Terminology matters most when it becomes executable.

The word in a magazine may wound. The word in a chart can also trigger a protocol. “Dangerous,” “lacks capacity,” “not educable,” “behavioral,” “noncompliant,” “requires supervision,” or “unable to benefit” can become instructions to other institutions. This is why language reform belongs inside policy and documentation standards, not only etiquette training.

Modern hospitals are not Victorian asylums. The power question, however, did not vanish.

It would be historically careless to describe every modern psychiatric hospital as an asylum under a new name. Contemporary inpatient psychiatry includes evidence-based treatment, legal protections, multidisciplinary care, crisis stabilization, and clinicians who work under ethical standards radically different from nineteenth-century custodial institutions. Hospitalization can save lives.

But the history of institutionalization remains relevant whenever a person enters a setting where ordinary liberties are curtailed. Acute psychiatric care necessarily raises difficult questions about consent, risk, involuntary treatment, seclusion, restraint, communication, medication, visitors, discharge, and the weight given to professional judgment. Those questions become especially complex when the patient is autistic, intellectually disabled, nonspeaking, psychotic, multiply disabled, or communicating under extreme distress.

Deinstitutionalization also produced an uneven landscape. The decline of large state hospitals did not automatically create sufficient community housing, accessible healthcare, supported employment, crisis services, or long-term assistance. Historians of U.S. psychiatric services have described cycles in which people moved among hospitals, community systems, homelessness, emergency departments, jails, and prisons. The location of confinement can change while the problem of inadequate support remains.^[6][7]

For autistic people, emergency and inpatient environments can intensify distress through fluorescent lighting, unpredictable routines, loss of familiar communication supports, touch, alarms, crowded waiting areas, rapid questioning, and staff unfamiliar with autistic presentation. A shutdown may be interpreted as refusal. A meltdown may be interpreted only as aggression. Stimming may be treated as behavior to suppress. A support person may be excluded without an accessible alternative. The clinical problem and the accessibility problem then amplify one another.

The historical lesson is not “never hospitalize.” It is: the greater the power imbalance, the more deliberate the system must be about communication access, supported consent, least-restrictive care, sensory conditions, review, and the patient’s own account.

That principle applies beyond autism. People living with schizophrenia, bipolar disorder, intellectual disability, traumatic brain injury, dementia, epilepsy, developmental disabilities, and other forms of cognitive or psychiatric difference have all been affected by histories in which diagnostic status could reduce civil standing. A neuroinclusive future cannot correct one diagnosis by creating another class of people whose testimony matters less.

The hospital chart can reproduce an old asymmetry of knowledge

In a modern emergency department, a diagnosis in the chart may shape interpretation before the clinician has heard the patient. Autistic distress can be attributed to autism; pain can be redescribed as behavior; atypical affect can be read as indifference; a shutdown can be mistaken for refusal; literal language can be taken as evasiveness. None of these errors requires malice. They emerge when the category becomes a stronger source of knowledge than the person.

This is where the history of the case file becomes immediately contemporary. The record should preserve information, not become a substitute witness. Good disability care treats prior diagnoses as context, asks what is different from baseline, makes communication accessible, reduces sensory load, allows processing time, and distinguishes psychiatric symptoms from pain, medication effects, seizure activity, sleep loss, trauma responses, and other medical causes.

A hospital can become inaccessible before any clinical decision is made.

Autistic adults report barriers from fluorescent lighting, alarms, television, crowded waiting rooms, touch, uncertainty, rapidly delivered questions, interruptions, and pressure to answer in real time. Research comparing autistic and non-autistic adults has found greater sensory discomfort and communication barriers, while participants frequently requested clearer information, slower pacing, and written material.^{[24][125][126]}

These are not hospitality preferences. Sensory overload and processing pressure can change the very behavior clinicians use as evidence: speech may reduce, pain communication may become less organized, eye contact may disappear, movement may increase, or the person may shut down. An inaccessible assessment can therefore contaminate its own data.

Diagnostic overshadowing can occur when a known label becomes the explanation before the examination.

In intellectual-disability research, diagnostic overshadowing is usually defined as incorrectly attributing psychiatric or behavioral symptoms to the known intellectual disability. A recent systematic review found inconsistent evidence and emphasized limitations in the literature, so the concept should not be invoked as proof whenever clinicians make an error.^[128] The practical safeguard is still valuable: every new symptom deserves its own differential diagnosis.

For autistic patients, the parallel risk is that pain, gastrointestinal illness, neurological change, medication effects, trauma, or other medical problems are interpreted through an autism/anxiety lens. The reverse can also happen: an autistic person's escalating overload or communication crisis may be treated only as a medical mystery without recognizing environmental causes. Good care requires both possibilities to remain open.

“Capacity” is decision-specific; communication difficulty is not a global incapacity test.

Hospital staff sometimes face urgent questions about informed consent. The ethically and legally relevant question is not “Is this patient autistic?” or “Do they have an intellectual disability?” but whether the person can understand, appreciate, reason about, and communicate a choice regarding the particular decision under the applicable law—with communication supports and reasonable modifications where required. A person may need help understanding a complex surgery while being fully able to decide who may enter their room or whether they want a particular support person present.

ADA effective-communication obligations reinforce a complementary principle: communication access must be made effective for the person rather than judged by whether they can use the provider's preferred channel.^[94][95]

Behavioral escalation in healthcare should trigger a clinical and access differential, not only a security response.

Psychiatric hospitalization raises an additional power problem because the treatment environment can control liberty.

Modern involuntary-treatment law contains procedural protections that did not exist in Victorian asylums, and psychiatric hospitals provide life-saving care. Historical comparison should not erase that. The relevant continuity concerns the extraordinary authority activated when a professional conclusion can limit movement, privacy, communication, or refusal. That makes accurate interpretation, accessible communication, independent review, and the least restrictive effective intervention especially important.

The hospital chart deserves the same historical scrutiny as the asylum casebook.

Records are necessary for continuity and safety. They can also preserve interpretive error. Once “combative,” “poor historian,” “drug seeking,” “behavioral,” or “lacks insight” appears in a chart, later clinicians may read future encounters through that frame. Epistemic-injustice research with autistic people documents how professional interpretations can displace the person's account and shape access to care.^[26][127]

A rights-compatible chart therefore distinguishes observation from inference. “Patient yelled” is observation. “Patient was manipulative” is a theory. “Patient repeatedly asked the same question after receiving rapidly delivered verbal instructions” is a richer observation that suggests a testable intervention: slow down and provide the information in writing.

Do not replace an old label. Replace the decision rule.

If the problem is lexical modernization without conceptual reform, then a successful response cannot consist only of preferred terminology. It requires changing the questions institutions ask.

The aim is not to reverse the hierarchy and declare neurotypical people deficient. That would preserve the same structure with different winners. The aim is to make neurological difference ordinary enough that variation does not automatically become a rank.

Acceptance also does not mean romanticizing disability. Some autistic people need 24-hour support. Some people have intellectual disabilities that substantially limit independent functioning. Some people experience dangerous self-injury, debilitating anxiety, epilepsy, psychosis, severe sleep disturbance, gastrointestinal disease, catatonia, or other conditions requiring serious medical attention. A rights-based neurodiversity framework can acknowledge all of that while refusing the historic equation more impairment = less personhood.

That equation is the part that must be deprogrammed.

A de-eugenicized system is not a diagnosis-free system

The opposite of eugenic classification is not pretending differences do not exist. People need accurate diagnosis, specialized expertise, medical treatment, disability income, communication support, accessible education, and protection from discrimination. The reform target is the conversion of description into rank.

A rights-compatible diagnostic system can say: this person has epilepsy without treating them as socially dangerous; this person has an intellectual disability without deciding that all decisions belong to someone else; this autistic person has high support needs without assuming they lack preferences; this person has psychosis without treating every physical symptom as psychiatric; this student needs intensive support without making segregation the default.

Replace global judgments with domain-specific questions

“How functional are they?” should become: functional for which task, in which environment, with what communication, after how much sleep, under what sensory load, with which supports? “Do they have capacity?” should become: which decision, what information format, what time, what communication partner, and what support would make the decision accessible? “Can they be included?” should become: what environmental changes, staffing, technology, peer support, curriculum adaptation, or communication access would make participation possible?

Reform has to operate at the level where power is allocated.

Replacing a stigmatizing term is worthwhile. Updating diagnostic criteria is worthwhile. Closing an abusive institution is worthwhile. None is sufficient if the downstream decision rule remains unchanged. A serious reform agenda therefore asks each system what it currently treats as evidence of incapacity, danger, educability, credibility, or readiness for inclusion—and whether that evidence is actually valid.

A de-eugenicized service system should be able to describe need without constructing a hierarchy of lives.

This is harder than it sounds because resource allocation requires comparison. Public agencies have to determine who qualifies, what level of service is funded, and how scarce resources are prioritized. The answer cannot be to stop measuring need. It is to make the purpose of measurement explicit: measure support intensity to allocate support, not to infer personhood.

A person needing two-to-one staffing may have very high support needs. That fact does not tell us whether they enjoy music, understand political conversation, want a romantic relationship, can choose clothes, recognize familiar people, make some healthcare choices, prefer one staff member, or experience a rich internal life. Global labels erase exactly the information individualized support requires.

Professional education should include the history of the concepts it still uses.

Training in psychiatry, psychology, education, social work, medicine, law enforcement, and disability services often teaches current terminology without teaching why older terminology became dangerous. Historical literacy changes how professionals hear apparently ordinary phrases. A clinician who knows the history of “mental age” is less likely to use it casually with a 35-year-old. A school administrator who knows the history of custodial education may be more cautious about equating quiet compliance with learning. A genetic counselor who knows eugenic history can distinguish reproductive information from judgments about which lives ought to exist.^[62][63]

Participatory design changes who is allowed to define the problem.

The disability-rights principle “Nothing about us without us” is not simply a request for representation. It is an epistemic safeguard. People who experience a system can identify failure modes invisible to its administrators: the form that cannot be completed with AAC, the sensory environment that makes an assessment invalid, the behavior plan that rewards masking, the policy that forces a person to perform incapacity to qualify for support. Autistic self-advocacy organizations have specifically argued for meaningful self-advocate participation in autism research priorities and governance.^[118][132]

Reform also requires enough material support to make rights real.

It is easy to celebrate community inclusion while underfunding accessible housing, direct-support labor, transportation, AAC, respite, healthcare, income support, and crisis alternatives. Rights without infrastructure can become another form of abandonment. The history of deinstitutionalization demonstrates that closing a harmful setting and building a viable alternative are separate policy tasks.^{[116][136]135}

The endpoint is not independence understood as “needs nobody.” Human beings are interdependent. The goal is self-determination with whatever support is genuinely required: a social order in which needing more assistance does not reduce the person's claim to adulthood, privacy, communication, community, or dignity.