The field turns its lens on itself
Mary Doherty, Sebastian Shaw, Eddie Chaplin and Simon Baron-Cohen published a short essay in The Lancet in July 2026, in the journal’s “Art of Medicine” strand — a commissioned perspective rather than original research. It opens against a familiar political backdrop: the resurgent invocation of an “autism epidemic” and the recycling of discredited theories of causation. But the notable thing about the piece is not its politics. It is the manoeuvre underneath it — psychiatry taking its newest diagnostic lens, the neurodiversity paradigm, and turning it not on another cohort of patients but on itself.
That move has a precise shape. First-order observation is a discipline classifying the world in front of it: psychiatry sorting its patients into categories. Second-order observation is a discipline observing its own observing — turning the instrument back on its own history, its own founders, its own authority. It is rare, because the apparatus of diagnosis is built to look outward, and looking inward requires it to concede that it was ever a particular kind of mind doing the looking in the first place. This essay is second-order. It is the field examining the hand that drew its categories.
The authors say almost as much in their own vocabulary. They write that “psychiatry’s discomfort with neurological difference deserves renewed scrutiny”; they invite the profession to “re-examine long-standing assumptions about competence, empathy, and authority”; they describe themselves as “witnessing a cultural shift in how autism is conceptualised”. This is a field narrating its own turn while making it — an unusual thing to catch in print.
Spitzer as the mechanism, not the diagnosis
The device the essay uses to stage that turn is a single figure: Robert Spitzer, the architect of DSM-III, the classification that became the working “bible” of modern psychiatry. The authors assemble the contemporaneous accounts of him — a monotropic, decades-long focus on the manual, an apparent difficulty registering how others were responding to him, a habit of treating everyone identically regardless of rank. Steve Silberman’s assessment, which the essay leans on, was that DSM-III emerged from a mind showing many of the hallmarks of autistic intelligence.
The irony writes itself. The person who built the system that pathologised cognitive difference may have embodied precisely that difference. Constructing a vast classification system is itself an act of hyper-systemising — the pattern-seeking cognition the manual went on to catalogue as disorder. The apparatus, at its origin, may have been authored by the very cognitive style it excludes.
The essay’s restraint is where the real argument sits. It explicitly refuses to diagnose Spitzer retrospectively — the dead cannot speak for themselves, the biographical evidence is fragmentary, and exhibiting autistic traits is not the same as needing a diagnosis. That refusal is not caution getting in the way of the point. It is the point. The claim is not that Spitzer was autistic; it is that the categories were drawn by particular minds at a particular moment — historically contingent, not natural law. To pin a posthumous label on him would simply reproduce the classificatory move the essay is holding up for scrutiny.
From there the authors ask their sharpest question: if people on either side of an arbitrary diagnostic cutoff have more in common than apart, what does the cutoff actually mark? I’ve made a version of this argument before through the dress of archetypes — that what varies across the line is manifestation and accommodation, not the actual, underlying substrate. Here it does a different job. It is offered as evidence of the field interrogating the validity of its own boundary-drawing.
And there is a final turn of the screw. Because the criteria require clinically significant impairment, they cannot register autism that is well accommodated — the thriving autistic person stays invisible to the instrument until the supportive scaffold is removed. Spitzer, by that logic, is the ideal specimen: a mind that, thriving in a role built around its strengths, the criteria would never have caught at all.
Why a discipline seeing itself is rare
Second-order observation matters because it is the precondition for self-correction. A discipline that can only look outward (i.e. psychiatry) cannot revise itself; it can only accumulate onto itself. The turn inward is the moment a field becomes able to ask whether its own frame and foundations were ever the neutral instrument it took itself to be.
It is worth distinguishing two modes of the apparatus describing itself. I’ve written before about the unwitting kind — a review reproducing its own logic while believing it reports neutral fact, blind to the fact that it is specifying itself rather than the world. This essay is the opposite mode: deliberate self-observation, a field knowingly naming its own discomfort as the thing that requires examination. It is the same apparatus in both cases — blind in one mode, self-aware in the other — and the difference between them is the whole distance a discipline has to travel to change.
Psychiatry has been made to look at itself before, but usually forced from outside, or in hindsight once the harm was done. What is newer here is self-generated reflection from within the affirming wing of the field, offered voluntarily and from the most prestigious platform in medicine. The essay reaches for a precedent: homosexuality was removed from the DSM in 1973 when clinicians inside psychiatry challenged its inclusion. That is the lineage the authors mean to extend — the field correcting itself from within. Fair play, and fair enough.
The part of itself it still can't see
A mirror, though, shows only what it is angled to show. The self this essay sees is the eminent one — the autistic psychiatrist, the DSM architect, the pattern-seeking contributor to human progress. To argue against deficit framing, it reaches for the register of autistic accomplishment, even autistic genius.
That register, as we know, is double-edged. It flatters the exceptional and does comparatively little for everyone else — least of all autistic people with intellectual disability, non-speakers, and the high-support-needs population that the ageing literature was recently found to have excluded from its entire evidence base. A field that recognises itself only in its most accomplished members has not, in any full sense, yet seen itself.
So the honest reading holds both halves at once. The turn is real and genuinely rare — a discipline observing its own observing is the precondition for changing it, and it deserves to be marked as an event rather than waved through as a nice gesture. But the reflection is partial, and it flatters the part that was always going to be easiest to celebrate. The field has begun to see itself. What it sees, for now, is the flattering portion of the mirror — and what it still cannot see is most of the people it was built to classify. That needs to come next.
Citations
Doherty, M., Shaw, S. C. K., Chaplin, E. & Baron-Cohen, S. (2026) — Embracing neurodiversity in medicine: insights from the history of psychiatry — The Lancet
Murray, D., Lesser, M. & Lawson, W. (2005) — Attention, monotropism and the diagnostic criteria for autism — Autism
Shaw, S. C. K., Fossi, A., Carravallah, L. A. et al. (2023) — The experiences of autistic doctors: a cross-sectional study — Frontiers in Psychiatry
Silberman, S. (2015) — NeuroTribes: The Legacy of Autism and the Future of Neurodiversity — Avery
Spiegel, A. (2004) — The Dictionary of Disorder — The New Yorker
