Chapter null
Introduction
Introduction
Theory of Mind began with one task in one room.
In 1985, Simon Baron-Cohen, Alan Leslie and Uta Frith sat children in front of two dolls. A doll called Sally puts a marble in her basket and leaves. A doll called Anne moves the marble into a box. Sally comes back. Where will Sally look for her marble?
Most of the typically developing children, and most of the children with Down syndrome, said the basket. Sixteen of the twenty autistic children pointed to the box. They knew where the marble was, and they pointed to it. Four of the twenty autistic children pointed to the basket.
From that afternoon came a claim that travelled much further than the study ever could. Autistic people cannot represent what another person knows. Autistic people cannot take another perspective. Autistic people lack empathy. Autistic people lack insight.
That claim entered schools, therapy rooms, assessment frameworks, inspection materials, courtrooms and training programmes across Europe. It is applied before professionals have met the person. It is applied in place of asking. It is handed to families as an explanation. It is handed to institutions as permission to stop.
One of its own authors has since said that in those early years they took a very narrow view of theory of mind, and that the mind blindness theory had something going for it but missed a lot, particularly the role of emotions. That is Simon Baron-Cohen, speaking about his own theory.
In 2012, the autistic researcher Damian Milton proposed the Double Empathy Problem. Communication difficulty runs in both directions. When two differently disposed people meet, each struggles to read the other. Non-autistic people are no better at reading autistic people than autistic people are at reading them. The research since has supported him.
In 2019, Morton Ann Gernsbacher and Melanie Yergeau published a paper whose title says everything: Empirical Failures of the Claim That Autistic People Lack a Theory of Mind.
The theory has been questioned by its own authors, challenged by autistic researchers and contradicted by peer-reviewed evidence. It is still in use. Every day. In rooms where the person it describes has no idea it is being applied.
I am autistic. I never changed. That was never the fight.
For fifteen years I have been investigating what is allowed to become action. The same pattern keeps appearing. Experts who stopped learning. Frameworks people are not allowed to question. And each new professional in front of us who needs to be taught, again, what autism is, what has happened, and what is happening right now.
Het Passer College is a school in the Netherlands for young people with autism and anxiety. Its anti-bullying protocol, the pestprotocol, is published on the school's public website. It sets out the framework the school uses to explain why autistic students become involved when bullying happens.
The framework rests on three cognitive theories of autism. Each is applied to every student.
On Theory of Mind:
"Het inlevingsvermogen van de leerlingen is beperkt waardoor inschatten wat de ander bedoelt en probeert lastig is (TOM)."
The capacity for empathy of the students is limited, which makes it difficult to estimate what the other means and tries. (Theory of Mind.)
On executive function:
"Daarnaast verloopt het plannen en organiseren van eigen gedrag niet bewust (executieve functie)."
In addition, the planning and organising of their own behaviour does not happen consciously. (Executive function.)
On central coherence:
"Verder is door de CC niet het geheel te overzien, maar worden details uit de situatie gepikt en gezien die niet de overkoepelende bedoeling van de ander weergeeft."
Furthermore, because of central coherence, the whole cannot be taken in. Details are picked out of the situation and seen, and they do not reflect the overall intention of the other person.
When bullying happens, the protocol directs staff to look for its cause, which "kan liggen in bovengenoemde theorieën": may lie in the theories above.
As of September 2026, the statements are still there.
All three apply to all students. None of them is a note about an individual. Together they form the school's published account of what autistic young people are able to understand, plan and perceive. They sit in a document about bullying, and the protocol points to them when the cause is sought. So when an autistic student is hurt by another person, part of the explanation has already been placed inside the student.
The framework describes every student before any particular student is known. The theory has already answered the question.
A school for autistic students sets out, in its published anti-bullying protocol, that its students have limited empathy. The sentences name no exceptions.
A framework like this has a particular shape. When a child pushes back against it, the pushback can be read as proof: they cannot understand our perspective. When a parent questions it, the question can be read as proof: they cannot see what we, the professionals, can see. The challenge becomes evidence. The resistance is absorbed and returned as confirmation.
The child who objects is described by Theory of Mind. The parent who questions is described by Theory of Mind. The researcher who challenges it is described by Theory of Mind.
See. Theory of Mind.
A theory that reads every challenge as proof of its own claims is a closed system. It cannot be falsified from the inside. Anyone who questions it has already been explained by it.
That is what a framework like this says on the page. What happens when one is used in a room can be seen in a different place.
A special education centre, a different place from the school above. Our son was there for help with anxiety. At the end of term, the question was whether he should do another term. I said we could ask our son.
The therapist said: there is no point. He is autistic.
I asked her to say it again.
There is no point. He is autistic. Autistic people do not have insight.
I asked her to say it again.
There is no point. He is autistic. Autistic people do not have insight.
The third time, she smiled. As if this was a known thing. A settled thing. Something the room already understood.
Three times. Same room. His coaches and teachers present. No one on her side of the table questioned it.
The framework had already decided what our son could and could not know about himself. She applied it before she asked him. She applied it in place of asking him.
22 May 2026. London. A march to 10 Downing Street.
Autistic people, families, campaigners. Chris Packham among them, naturalist and broadcaster, openly autistic for years. The march called for the removal of text from Ofsted's national training materials for school inspectors. The text said that children with autism are at increased risk of susceptibility to extremism, because they are more likely to develop special interests.
The text was removed in July 2026. Thirteen months after it was first reported. One month after the march.
I spent five days with autistic people in London. I brought a camera. What I watched was insight. Precise, layered, specific. People who could read a government statement and point to the exact word where the meaning had been shifted. People who could read other minds with great accuracy, and who could not always get the people across the table to read theirs.
Then the news about Uta Frith arrived. In an interview with The Times, she had been asked about Chris Packham.
"He cannot be autistic in my opinion," she said. "It is just absolutely absurd to me that somebody who can communicate as brilliantly as he does." And then: "That is what I call a blatant contra-indication."
Later that month she told Holden Thorp, the editor-in-chief of Science, that she regretted the remark. She also told him that in the 1990s she would have diagnosed Packham herself.
The reason she gave, as Thorp reports it, was that it had been "an off-hand remark" and she "didn't recognize how newsworthy it would be." Her regret concerns how far the words travelled. Many professionals say sentences like hers in closed rooms, to parents and to children, expecting them to go no further. The therapist's sentence earlier in this Introduction was one of them. Hers reached a newspaper.
Uta Frith co-authored the 1985 paper that gave the field Theory of Mind.
The Essay
In the summer of 2026, Uta Frith published an editorial in Psychological Medicine. Its title is a question: Autism spectrum disorder: has it lost its meaning and is it leading to misdiagnosis?
It was received on 3 July, revised on 3 July and accepted on 6 July. It appeared online in early August.
The essay argues that the autism spectrum has widened too far. It names the drivers as cultural: inclusion, harm avoidance, the acceptance of masking, self-diagnosis, a search for identity, and reliance on self-report. It says that when autistic people's accounts of their own experience are used as diagnostic evidence, reliance on them "can be treacherous." It proposes that people diagnosed in adolescence and adulthood may belong to a different category altogether, and that the spectrum should be split.
It is being read right now by clinicians, researchers, teachers and commissioners. It will shape assessment waiting lists, school policies and the conversations families have across tables like the ones described above.
Forty-one years after the marble moved from the basket to the box, one of the theory's founders has returned to the question of who autistic people are. The method is the same. Watch from outside. Decide from outside. Set aside what the person says about themselves.
This book begins there.
How This Book Works
The essay has nine sections. This book reads them one by one, in her order. Her sixth section, on the drivers of change, is the longest, and it takes two chapters.
1. The Abstract 2. Changes in the diagnostic criteria 3. Is autism real? 4. The uneven rise in prevalence 5. A split in the spectrum 6. Drivers of changes in the autism concept: the cultural drivers 7. Drivers of changes in the autism concept: lived experience, "hidden" girls and masking 8. The tension between social and medical models of disability 9. Can comorbidities rescue the spectrum? 10. Toward diagnostic precision
A separate chapter, Her Words, Their Words, reads the studies the essay cites in full and sets what she says beside what their authors wrote. A closing chapter gathers what the reading found.
Each chapter sets out what she says, in her own words. Then it asks what the section never asked. Then it reads the section through Permission Architecture. Then it sets out what the year of sessions will show, in rooms with real people. Every chapter closes on the same two headings:
What Is Expert Still Stopping, Forty Years Later
What This Still Ignores
The Map I Bring
Theory of Mind asks one question. Can this person represent what another mind is thinking?
This book brings two frameworks to that question. Permission Architecture asks what happens in the room and in the person when understanding is being judged. Epistemic injustice asks what happens to a person's standing as someone who knows.
Permission Architecture
Permission Architecture asks a wider set of questions than Theory of Mind. What route was open? What route was closed? What did the room cost? What did the title cost? What closed before the person could explain it? Who else was in the field, and what did their presence do to what became reachable?
The framework rests on one statement.
Human action is not only chosen. It may also be permitted, shaped, blocked, or released through whole-system biological architecture.
Four ideas from it run through this book.
Capability, access and permission. These are three different things. Capability is what a person is able to do in principle. Access is whether that capability can be reached in this room, in this state, in front of this person, right now. Permission is whether the reached capability is allowed to become visible action.
A person can be fully capable and still blocked. A person can have the words and lose them the moment a title enters the room. A person can speak fluently in a studio and lose speech entirely in a clinic. Theory of Mind has spent forty years scoring visible action and calling the score capability.
The Striatic Veto. The striatum is a group of structures deep in the brain involved in selecting which actions go ahead and which are held back. The Striatic Veto proposes that action, speech and engagement can be inhibited here before the person has any conscious explanation ready.
The explanation comes afterwards. A person says they froze. They say the words were there and then were gone. These accounts may be accurate reports of a veto that had already happened lower down, before the part of the system that builds sentences had anything to work with. A test that scores only the final answer will never see the veto that shaped it.
D1 and D2. Within the striatum, two families of dopamine pathways are often described. In this work, D1 pathways are explored in relation to action, exploration and moving towards. D2 pathways are explored in relation to evaluation, holding back and deciding whether action should proceed. D2 pathways learn from experience. When a route has been costly again and again, the system learns to hold it back earlier.
Permission is relational. Permission is built between people, in the field they share. A second person standing at a clinical distance cannot simply read it off the first.
The tester is part of the field being measured. The therapist who says "there is no point" is closing the very route she then judges. The expert who decides a man is too fluent to be autistic is deciding what the field between them is allowed to mean. A title in the room changes the architecture of the room.
Epistemic Injustice
The philosopher Miranda Fricker named this in 2007, in Epistemic Injustice: Power and the Ethics of Knowing. Epistemic injustice is a wrong done to someone in their role as a knower. The person is unfairly blocked from being believed, from making sense of their own experience, or from contributing knowledge.
It has two forms.
Testimonial injustice. The short version is: I don't believe you.
A person's word is given less credibility than it deserves because of who they are. Fricker calls this a credibility deficit, and traces it to prejudice about the speaker's identity. Gender, age, accent, mental health history and neurotype can all lower the credibility a person is granted before they have said a word.
The knowledge exists. It is not trusted.
A child explaining something complex and being written down as confused. A parent describing their own child and being told the professionals can see what the parent cannot. An autistic adult describing their own inner life and being told that fluency disproves it. There is no point. He is autistic. Autistic people do not have insight.
Hermeneutical injustice. The short version is: you don't even have the words for this yet.
This form is quieter and deeper. A person cannot fully understand or explain their experience, because the shared concepts, language and frameworks that would make it understandable have not been built, or have been built by other people, for other purposes.
The experience is real. The interpretive tools are missing.
For decades there was no shared word for masking. No shared word for the Double Empathy Problem. No shared word for sensory overload. Autistic people had the experiences long before they had the words. Many of those words were built by autistic people themselves, in autobiographies, in forums and in research of their own, and only later entered the clinic.
Theory of Mind sits on both forms at once. In the room, it produces testimonial injustice: the person's account is outweighed by the framework's prediction. At the level of the field, it produced hermeneutical injustice: it arrived before there was any shared language to question it, and it supplied the only words available. Families were handed "lacks theory of mind" as the explanation of their child. Autistic adults were handed it as the explanation of themselves.
Where the two meet
This is where the two frameworks meet.
Testimonial injustice happens between people. Permission Architecture shows what it leaves behind inside a person. Each time an account is doubted, corrected or dismissed, the route that carries that account into words becomes more costly to use. D2 pathways learn the cost. The Striatic Veto begins to act earlier. The next account comes out more hesitant, more partial, more shaped to what the room expects, or does not come out at all.
The professional then records a person who cannot clearly describe their own experience. The framework reads that as proof.
The injustice creates the closure. The closure is then used as evidence for the injustice.
What This Book Is For
This book holds two things at once. Some autistic people, in some rooms, do find another person's perspective hard to reach. And access changes with state, room, person, history and cost, for everyone. So the book asks the question good science asks of any claim: does this hold up outside the one narrow condition where it was first observed?
It also turns to the people the theory never asked.
From September 2026, a year of open research lets them show. The sessions are living observation, with portraits alongside them, across the Netherlands, the United Kingdom and Europe. People arrive with their own stories, in their own order. They show; everyone else looks. Children, adolescents and adults. People diagnosed in childhood and people diagnosed in adulthood. People still waiting. People who have never been near a formal process. People who speak and people who do not. Parents. Professionals who quietly know that what they see every day does not match what they were trained to believe.
One person showing their life is a story. A thousand people showing the same patterns, the same costs, the same gaps between what the professional recorded and what the person knew, are evidence.
This is Volume One. It rests on fifteen years of research and education in this area, and it reads the essay against the published evidence. The stories themselves are only beginning, and they are not in these pages yet. This volume is the introduction to them. It sets out why they are needed, and what the theory and the essay left out that only the people themselves can show. The stories will follow in the volumes after this one.
The essay says that reliance on autistic people's accounts of themselves can be treacherous. This book treats those accounts as primary evidence, and tests them the way every other form of evidence deserves to be tested: openly, carefully, and in the company of the people who gave them.
The theory never asked the people it claimed to understand.
This research does.
People are my rockpools.
Your stories become education.
Curiosity before conclusion.