What Then Shall We Do: The Work Horwitz Left

Allan V. Horwitz has done something rare in the sociology of medicine. He identified a structural problem with precision, named its institutional causes without exaggeration, and refused the convenient resolutions available on both sides of the debate. His central claim stands: the DSM model, by detaching symptom checklists from social context, transformed contextually intelligible responses to loss, failure, and dislocation into treatable disorders. The expansion served pharmaceutical companies, insurance administrators, clinical psychologists seeking parity, and patient advocacy organizations simultaneously. The cost was the erosion of a cultural vocabulary for enduring normal suffering without medical supervision.
But his framework stops short in six places where truth and social need press beyond where Horwitz went.
The first is genetics. Horwitz’s distinction between normal suffering and genuine disorder requires a boundary. He gestures toward severe psychoses with plausible biological substrates as the paradigm cases of real disorder while treating most depressive and anxiety presentations as contextually expectable responses. That distinction is morally powerful and intuitively clear. It becomes empirically unstable once behavioral genetics is taken seriously rather than bracketed. Twin and adoption studies show heritability estimates for major depressive disorder clustering around thirty to forty percent, higher for recurrent and severe forms. Gene-environment interaction research, particularly findings on differential stress reactivity across genetic variants, shows that two people experiencing identical losses can diverge sharply in trajectory not because one’s grief has been misclassified but because their regulatory systems have different thresholds. A person with high polygenic loading for depression may develop something chronic and disabling from the same loss that another person processes adaptively. Whether that constitutes harmful dysfunction in Wakefield’s sense, a system responding to real inputs but with miscalibrated intensity, or normal variation in a continuous trait is precisely the question Horwitz’s framework cannot answer because it lacks the conceptual apparatus to use heritability evidence. Ignoring this romanticizes distress as normal in cases where it is not functioning as it should. Overstating it collapses back into the biological reductionism Horwitz resisted. The synthesis required is neither compromise nor capitulation. It treats genetic variation as modulating the probability and intensity of dysfunctional responses rather than as a binary disease marker, and it requires specifying which social contexts interact with which genetic backgrounds to produce genuine dysfunction versus adaptive mourning. Without that specification, the sorrow-versus-sickness line stays rhetorically sharp and empirically blurry.
The second is reflexivity. Horwitz applied coalition analysis to biological psychiatry with precision and without mercy. He showed how the beliefs of DSM committee members were convenient for the institutional arrangements that formed and sustained them. That analysis has to be turned inward. Medical sociology has its own convenient beliefs. It is trained to see expansion, labeling, and power. It is rewarded for discovering medicalization. That formation does not make its conclusions wrong. It makes them situated in exactly the way Horwitz showed psychiatric conclusions to be situated. If both sides are operating from formation-shaped perceptions rather than unmediated empirical sight, then the argument for reform cannot rest on claiming that sociology occupies a position of clarity that psychiatry lacks. It has to rest on identifying the specific distortions each formation produces and correcting for both. Horwitz never applied the four diagnostic questions to his own coalition. What beliefs did his institutional base at Rutgers, the American Sociological Association, and the sociology of mental health subfield require him to hold? The answer is that underweighting biological constraint and overinterpreting institutional motive were both structurally convenient for a medical sociology making its jurisdictional claim against psychiatry.
The third is operationalization. Horwitz proved that context matters diagnostically and that erasing contextual criteria produced systematic inflation. What he never supplied was anything a clinician could use under real constraints. Fifteen minutes. Limited history. Insurance requirements. Liability risk. Saying context matters is correct and not operational. The DSM filled that vacuum with checklists because checklists work administratively, not because anyone believed they were conceptually adequate. The critique remained at the level of truth and never reached practice because it produced no tools. The extension requires building structured contextual probes that are fast, teachable, and reliable enough to survive institutional use. It requires training curricula that rebuild contextual judgment as a clinical capacity acquired through formation rather than applied from a rulebook. Turner’s point about tacit knowledge is exactly relevant here. Wakefield tried to capture explicit criteria that would substitute for clinical judgment. Turner would say the judgment is not fully capturable in propositions because it is a trained perceptual skill. Rebuilding it requires rebuilding the formation conditions that produce it, which is a generational project in clinical education, not a revision of diagnostic manuals.
The fourth gap is cultural. The deepest harm of diagnostic inflation is not overprescription. It is the attrition of non-clinical frameworks for making sense of suffering. When grief becomes major depressive disorder and ordinary fear becomes anxiety disorder, the language for endurance, mourning, adaptation, and moral struggle thins. People lose access to the interpretive resources that once made suffering bearable without being pathological. They are left with treatment but no framework for living through difficulty with meaning intact. Fixing that is not a DSM project. It is a cultural project requiring intervention in schools, media, peer formation, and the informal social environments where people first learn what suffering means. If that work does not happen, even corrected diagnostic criteria will not stop the demand for diagnosis, because diagnosis supplies identity, validation, community, and recognition that people will seek regardless of its accuracy when no other framework is available for sharing pain.
The fifth is political. Once diagnoses become identities, you cannot tighten categories without producing harm to people who have organized their lives around them. Horwitz’s framework implies that many of those identities rest on inflated categories. The people who would lose recognition, community, insurance coverage, and legal protection from a narrowing of categories are not abstractions. They are people who found in the diagnosis real social goods that the non-clinical alternatives Horwitz defends were not available to supply. Any reform that does not account for that political reality and design transitions that do not strand those people will fail not because it is wrong but because it is incomplete. This requires a kind of political sociology that is largely absent from Horwitz’s work, an account of how diagnostic categories become coalition membership cards and what you offer people when you take the card away.
The sixth is the hardest to state because it threatens the entire reform project. The boundary between normal suffering and pathological dysfunction may not admit a stable, fully explicit rule. Horwitz identified the problem with the DSM’s erasure of contextual criteria. Wakefield proposed a philosophical account of harmful dysfunction as the replacement. The criteria that should govern the distinction are partly constituted by trained clinical judgment that resists full propositional capture. The DSM tried to replace that judgment with checklists and lost something real. Wakefield tried to recover it through philosophical analysis. What may be required is admitting that some parts of this boundary are irreducibly judgment-dependent, not arbitrary but not fully codifiable, and that reform therefore points toward training and formation rather than better definitions. A clinician trained to perceive context first, to read suffering against the life that produced it, to distinguish grief that is doing its work from grief that has stopped functioning, is not applying a rule. He is exercising a capacity. That capacity is what needs to be rebuilt.