Most of my work is with families whose children have been described as "fine in school", while what they see at home looks very different, often for years. These are children whose nervous systems have been working flat out to get through each day, and who arrive home with nothing left. It's the work I've become known for, and it shapes what I notice. I'm also AuDHD myself (autistic and ADHD), diagnosed late, and the parent of an AuDHD child, so I come to this as part of the community I'm writing about.
The divide I want to write about has been building for a long time. Autistic children's needs vary enormously, and they change over time and from one environment to another. Some children communicate through speech, some through signing, a device, sound or behaviour, and many through a mix that shifts with how regulated they feel. Some need an adult close by to stay safe, some need help with eating, sleeping, washing or getting dressed, some have a learning disability, epilepsy or chronic pain alongside being autistic, and some hold everything together at school and come apart at home. Many children's lives involve several of these at once.
For years, in support groups, on social media and in conversations with families, I've heard parents from across all of this describe feeling unseen: their child missing from the public conversation about autism, their child's struggles doubted because no one else witnesses them, or simply worn down by having to fight for every piece of support. Instead of bringing families together, that shared experience too often turns into comparison.
It surfaced again this weekend, when the journalist John Harris published a long piece in the Guardian asking whether the idea of a single autism spectrum has been stretched too far. It's built largely around an interview with the autism researcher Professor Uta Frith, but it also brings in many of the other voices in this debate: researchers who want autism split into subtypes and researchers who say that has been tried before, autistic adults diagnosed late with very different views on what should happen next, a mother whose daughter needs support all day at school, and John himself, writing as the parent of an autistic son who also has learning disabilities. Reading it, and the reaction to it, brought all of this into sharp focus for me. I don't think I've always held it as carefully as I'd like to, so this piece is my attempt to do just that.
The article, in its own words
In March, after Frith's interview with Tes and her conversation on Naomi Fisher's Let's Talk Neurosense podcast, I wrote a long response to her claims about masking, late diagnosis and non-speaking autistic people. Reading John's article, what struck me most is how much of the answer to her is already in it, often in the words of the people he interviewed, and sometimes in her own.
Frith calls masking "a non-scientific concept, if ever there was one", and dismisses the research on it because "subjective experience is not reliable". The same article quotes her verdict on Chris Packham: "He cannot be autistic in my opinion." When she explains her worries about where autism is heading, she cites "emails from parents". Opinion and anecdote are apparently enough to question someone else's diagnosis, while autistic people's own accounts of their lives aren't enough to count as evidence. It's a standard nobody applies elsewhere, either: pain scales, depression assessments and much of psychiatry rely on what people report about themselves. Masking also has a substantial body of peer-reviewed research behind it, including validated ways of measuring it and studies linking sustained masking to burnout and poorer mental health, and the Royal College of Psychiatrists names it in its guidance on autistic adults.
Her own former student, Francesca Happé, describes meeting people who seem to show nothing at first, then seeing "the mask almost literally slip" when they're tired. Masking that one of the field's leading researchers can watch happen is hard to call unfalsifiable.
Abigail Stow, diagnosed this year, was once told by a GP that she "can't possibly be autistic" because she made eye contact, had a boyfriend and held down a job. That's the same reasoning Frith applies to Packham, and Abigail's experience shows what it does once it reaches a GP's surgery: years without understanding, for someone who now describes work as "very masky" and comes home needing nobody to talk to her. The skills that years of masking produce are treated as proof that there's nothing to mask. Many of the women now being diagnosed in adulthood passed through assessments built largely around how autism shows up in boys, and as the autistic research consultant Ann Memmott has put it, rising late diagnosis is what finding the people who were always there looks like.
Frith's own history points somewhere else too. She describes being fascinated, early in her career, that the autistic children she met "could do things I would not expect children who were then called mentally retarded to do", sometimes completing a jigsaw with the picture face down. Her work began with the discovery that autistic children had been underestimated. In March, she questioned whether non-speaking autistic people's written accounts are really their own. A line drawn around "real" autism risks underestimating those same children all over again.
When the idea of a spectrum first arrived, she says she was "more than happy", partly because "we had more people to work with". Now she says there are "too many cases that were not anticipated to appear". Something not being anticipated doesn't make it wrong. It may simply mean nobody was looking.
None of this stays in an interview. What someone with Frith's standing says arrives, with its caveats stripped away, in assessment rooms, school meetings and council decisions, as one more reason to trust what professionals see in a moment over what autistic people and their families describe over years.
What the article reveals without meaning to
The mother John interviews, whose bright, curious daughter needs support all day at school, says something that I think sits at the centre of all of this. People, she says, are "seeking a diagnosis so their child is treated as an individual". She doesn't blame them. She worries that it leaves children like hers lost among so many others.
She's describing what happens when support is rationed. When support is scarce, it gets rationed, and it tends to be rationed by deciding whose needs to believe, and how to read them. Even when a child's needs are obvious to anyone who spends time with them, families still have to fight: their child's needs are put down to behaviour, they're told needs can be met with whatever is already there, they're left waiting for a specialist place, or they find their parenting questioned instead of their child supported. Families whose children mask are told their child is fine. Being seen has never guaranteed being understood, and being understood has never guaranteed support. In every one of these situations, families are being asked to prove something the system should already be listening for.
This is where so much of the comparison comes from. When support depends on proving need, families are pushed to show how hard things are, again and again, in meetings, reports and assessments, often against an assumption that they're exaggerating. A parent describing their child's hardest days to a panel, and a parent insisting that their child's struggles are real even though school doesn't see them, are both doing exactly what the system demands of them. Neither is taking anything from the other. The scarcity belongs to the system, and the comparison is what it leaves families holding.
Running through the reaction to John's article, and through years of conversations before it, is exhaustion with that comparison: who is "more disabled", whose life is harder, whose child counts. Again and again, people say, in different words, that families are far stronger fighting broken systems together than fighting each other.
Language becomes part of the rationing. Words like "mild" and "high-functioning" were never accurate, and they're still used to decide whose needs sound serious enough, sorting children into those who seem to need less and those who seem to need more, which is exactly the comparison families then end up having with each other. Talk of autism as a superpower can feel painfully far removed from caring for a child around the clock with no respite, while being told a child is "fine in school" erases the child coming apart at home. Each can hear the other's experience as a dismissal of their own, when both are describing what happens when a child isn't understood, and when the system responds by asking for more proof.
Many autistic people share the frustration with that lighter picture. Abigail Stow cringes at people calling themselves "neuro-spicy", and John worries that talking about autism as a difference rather than a disability can overlook his son's life. There's a name for the watered-down version of the neurodiversity paradigm that so many people are reacting to, "neurodiversity lite": the quirky, superpower framing that keeps the friendly language while leaving out disability, distress and the support people need. Humour about our own lives can be a real lifeline for autistic people, and nobody should have to perform their struggles to be taken seriously. The problem comes when the light version becomes the whole public picture, because it leaves out a great many children's lives, and it was never what the neurodiversity movement was asking for. The neurodiversity paradigm has always recognised that being autistic in a world not built for you can be disabling, and that people need real support, so the families pushing back against the light version and the autistic people pushing back against it are, far more often than not, on the same side.
A solution that argues against itself
The way forward the article explores comes from Simon Baron-Cohen, who wants autism split into subtypes: "type one is autism with an intellectual disability and a language delay. Type two is autism without an intellectual disability." Some families and some autistic adults would welcome that, including Hina Shaikh, one of the autistic adults John spoke to, who was diagnosed in her forties and is still without any support. A name can make a need harder to overlook, and I understand the pull.
The strongest objections come from inside the article. John says that if his son were labelled "type one", understood as lower functioning, "that's going to other him". Baron-Cohen replies that "nobody wants to be othered" and that "we must never prejudge what someone is capable of", yet a type defined by intellectual disability and language delay does exactly that, because many autistic children who don't speak, or whose development is uneven, don't have a learning disability at all. In the discussion around the article, one parent described professionals underestimating her children for years, and how often children like hers get past locks and safety measures nobody expected them to manage, because no one had recognised what they were capable of. Happé, meanwhile, points out that the field has already tried splitting autism with Asperger's, and that it didn't work, "because everything blurs, and because people change."
The timing makes this more pressing. The government says EHCPs will stay, but what an EHCP is would change. At the moment, an EHCP is built around one child: it sets out that child's needs and the specific provision to meet them, the council has a legal duty to secure that provision, and families have routes to challenge it if it's wrong or isn't delivered.
Under the proposals, a child would only have an EHCP if they need one of a small number of nationally defined "Specialist Provision Packages", which, in the government's own words, would "collectively define 'complex needs'". Much of the detail of each child's day-to-day support would move into an Individual Support Plan written by their school or setting, with far fewer ways for families to challenge it, and every other child with SEND would have only that. If support is increasingly decided by which package a child fits, sorting autistic children into types would give families one more reason to compete with each other, and give children one more line to fall on the wrong side of.
We've been here before. For much of the last century, children were sorted into fixed categories, the category decided what they were offered, and some children were judged "unsuitable for education at school" altogether. It took the Warnock Report in 1978, and the Education Act 1981, for support to start from what each individual child needed, and EHCPs were built on that shift. Sorting children back into groups, whether by type of autism or by package, takes us a step back towards a system that failed children and families on every side of today's divide.
What our children share
Baron-Cohen says something else in the article that I think is right: that different groups of autistic people are "equally needy; it's just that the needs are different". Underneath those differences, there's a great deal our children share.
Every one of them has a nervous system trying to make sense of a world that wasn't built with them in mind, and needs that come with that: around sensory input, movement, predictability and rest, and around communication, whether that's through speech, signing, a device, sound or behaviour. Distress is what happens when those needs go unmet for too long. For some children, that distress is out in the open, and it's too often treated as something to manage instead of something to understand. For others, it's held in all day and shows up at home, or in burnout months later. Either way, distress is communication, and the most useful question any adult can ask is what the child is telling them about what they need, and about the environment around them.
One autistic parent in the discussion around the article, raising children whose needs look very different from one another, described her worries for her children who mask, about their mental health and about them navigating a world with no support, alongside her worries for her son who needs someone with him all the time, about what happens if his support isn't right, because he may never be able to put it in place himself. Those are different fears, and both are real. Mary Doherty, one of the autistic doctors John interviews, puts the same truth from the other side: asked whether she feels an instinctive affinity with autistic people whose lives look very different from hers, she says yes.
Our children are also misjudged in both directions. Some are underestimated: assumed not to understand, not to have a view, not to be capable of more, and then kept from the very experiences, choices and ways of communicating that would let them show otherwise. Others are assumed to be coping because they look settled, and expected to keep going until they can't. Both come from the same habit of judging a child by what adults happen to see in a particular moment, instead of getting to know the child.
What helps is the same too: adults who presume competence, who treat every form of communication as real, who give the child a say in their own life, and who, before asking a child to change, look honestly at what in their surroundings, routines and expectations could reasonably be adjusted. That starts with curiosity about what's making things hard, with changes made where they're reasonable and possible. None of that depends on which "kind" of autistic a child is.
It's telling that the voices most sceptical of autistic experience tend to doubt our children all at once. Frith has questioned both whether masking is real and whether non-speaking autistic people's written accounts are their own. Those doubts land on very different children, and they come from the same place: a reluctance to believe what autistic people, and the families who know them best, say about their own lives.
What bridging could look like
If the divide grows out of rationing, then one of the most important things families can do is refuse to let the rationing set us against each other. That's far easier to write than to live, especially when you're exhausted and fighting for your own child, and it asks something of all of us, me included.
For me, it means listening more closely to families whose children's lives look different from the ones I spend most of my time supporting, and noticing when my own work shapes what I see. For all of us, it means believing one another's accounts, whether they describe what happens behind closed doors or what happens in plain sight and still goes unsupported. It means not measuring each other's lives against our own, and giving one another the benefit of the doubt when we're all running on empty. It also means directing our frustration where it belongs: at a system that makes families prove their children's needs, instead of at the families standing next to us in the same queue.
Underneath it all, we're asking for the same things: for every child to be believed and understood as they are, for support that meets them where they are and is decided one child at a time, and for that support to be backed by rights families can enforce. Those are exactly the things the SEND reforms put at risk, and they have always been easiest to take away from families who are busy arguing about who deserves them most.
John's article ends with Francesca Happé hoping that all sides can start from "the assumption of goodwill" and talk without being afraid. I share that hope. Goodwill comes far more easily when families aren't being made to compete to be believed, which is why the argument about who counts as autistic and the fight for every child's legal rights are, in the end, the same fight.
The hurt families describe comes from different directions, and it points to the same place. If we can recognise that in each other, we become much harder to divide, and much harder to ignore.
References
• Harris, J. (2026, 1 November). 'The term is becoming unrecognisable': have the experts who believed in a single autism spectrum changed their minds? The Guardian.
• Canary SEND (2026, March). "Masking Is a Myth, Late Diagnoses Are Wrong, and Non-Speaking Autistic People May Be Lying": What Uta Frith Got Wrong.
• Cage, E. and Troxell-Whitman, Z. (2019). Understanding the reasons, contexts and costs of camouflaging for autistic adults. Journal of Autism and Developmental Disorders, 49(5), 1899–1911.
• Hull, L. et al. (2019). Development and validation of the Camouflaging Autistic Traits Questionnaire (CAT-Q). Journal of Autism and Developmental Disorders, 49(3), 819–833.
• Raymaker, D. M. et al. (2020). Defining autistic burnout. Autism in Adulthood, 2(2), 132–143.
• Loomes, R., Hull, L. and Mandy, W. P. L. (2017). What is the male-to-female ratio in autism spectrum disorder? A systematic review and meta-analysis. Journal of the American Academy of Child and Adolescent Psychiatry, 56(6), 466–474.
• Royal College of Psychiatrists (2020). Autism in adults: guidance for psychiatrists.
• Department for Education (2026). SEND reform: putting children and young people first, Part three: Specialist Provision Packages.
• Warnock, H. M. (1978). Special Educational Needs: Report of the Committee of Enquiry into the Education of Handicapped Children and Young People. HMSO.
• Education Act 1981; Children and Families Act 2014.
