Is Autism a Natural Kind? *

I first saw an autistic child during my psychiatry term as a medical student, many years ago. This was a slightly-built 10yo boy who stood in the corner of the room, smiling faintly, occasionally flapping his hands or tapping himself. He did not have the stiffness (waxy flexibility) of catatonia but he had no speech at all, hardly responded to people around him and was never aggressive. He occasionally asked for food or a drink by mimicking but he was not toilet-trained and required total care. I was introduced to him by the staff and had to lead him around by his hand. He seemed happy enough and after a few minutes, he climbed on my hip like a baby, wet himself, and hung on tightly, murmuring repetitively. He had no relevant medical history and no evidence of brain damage. The incidence of this dire condition was said to be one in 3,000 but nobody had any idea what caused it. I saw him again, perhaps 6 years later, in the adult mental hospital but apart from being much bigger, nothing had changed.

We are now told that “autism” is a spectrum of disorder (ASD), ranging from barely perceptible to the extremes above. According to 2018 figures from the US Centers for Disease Control [1], the incidence of ASD in 8yo children is 2.3% (one in 44), with boys outnumbering girls 4:1. Obviously, the overwhelming majority of them are only marginally affected; extreme cases fortunately remain very rare. However, there are two major problems with these types of surveys. Firstly, the signs and symptoms of ASD are highly subjective, heavily influenced by the attitudes of parents and teachers, and by what the psychiatrist believes. Second, they overlap with so many other conditions that the individuals often end up with several diagnoses, so saying what came first is little better than flipping a coin. That’s the polite view: the realist asks: “Does this thing even exist?”

There is a very clear example of this in a case commentary in the online news service The Primary Care Companion (i.e. general practice), part of the bigger site Psychiatrist.com. An intelligent 13yo girl was admitted after a serious overdose which followed being prescribed three antidepressants in 32 weeks. She was given three diagnoses: Major depression with psychotic features (MD-P), generalised anxiety disorder (GAD) and autism spectrum disorder (ASD), quite a load to be carrying at thirteen. The ASD was diagnosed on “core deficits,” including:

… poor eye contact, impairments in the social pragmatic use of language, difficulties in establishing friendships, and poor prospective-taking skills (ability to consider a situation from a different point of view)… rigid routines, (narrow) interests including preoccupation with technology, and unusual sensory responses to loud noises and tight clothes.

Now hang on: apart from the clothes, aren’t these also the symptoms of anxiety? They are, so the same batch of symptoms justifies two diagnoses. And doesn’t severe, untreated anxiety, which she clearly has, lead to depression? It does, just because being anxious is such a burden that eventually life becomes unbearable, especially when the drugs from the clever psychiatrists aren’t helping.

The central problem is this: Psychiatry doesn’t get anxiety at all. Psychiatrists have no idea what anxiety is, how bad it is or what to do with it. Even with this case, they just sent her for 6 months of “supportive psychotherapy,” whatever that is. Indeed, publicising this failed case is about the same as shouting from the rooftops “We don’t know what we’re doing.” The case points directly to the theoretical (and intellectual) black hole at the centre of psychiatry: What’s mental disorder? What’s personality disorder? How do these concepts relate to each other and to normality? Don’t ask a psychiatrist. Even though these are absolutely fundamental issues, you won’t get an answer, just because they haven’t got one. To understand it, we have to go back a few hundred years to Sweden where the biologist Carl Linnaeus took on the scientific establishment over whether whales are fish.

Linnaeus (1707-1778) is considered one of the greatest scientists of all time. Before him, biology (known then as natural philosophy) was a hopeless mess, with everybody having a private system of classification. Should bats be classed with birds, or were birds (warm-blooded) closer to mammals or to reptiles (cold-blooded)? Nobody could say although everybody had an opinion, and the arguments raged without end. This was also the age of European exploration, of the Americas, Africa, Asia and the Pacific, so the known variety of plant and animal life was expanding explosively, except nobody knew what to do with them.

Linnaeus had a prodigious intellectual appetite and, after a few years of study and still aged only 31, he published his System of Nature, which introduced the binomial classification we use today. Every living thing could be slotted neatly into a cascading formula with four levels (now six), the last two (genus and species) becoming the scientific name, as it were, for each species (like Homo sapiens). Probably his most radical move was to put monkeys and humans on the same branch, which caused huge offence, especially to monkeys, but Linnaeus persisted and scientists quickly adopted his concept. Whales, of course, moved to the mammalian line; living in water was no longer enough to be classed as fish.

What he was doing was sorting everything into groups known as natural kinds, which means just what you expect. That is, he grouped things according to their inherent nature and not by what humans may prefer. Linnaeus based his descriptive system mainly on observable anatomy – apes and humans are practically identical inside - although now we rely on immunology and genetics to split nature at its joints, as they say (e.g. recent news that there are actually three species of funnel web spiders). In his era, and still today, people wanted to believe humans were entirely distinct from all other living creatures, special in God’s eyes, halfway between angels and animals. It was profoundly shocking to be told we are just bald monkeys. Fortunately, and unlike Charles Darwin who took the idea further, Linnaeus was greatly rewarded by his nation and died a hero.

The relevance for psychiatry is this: do depression, anxiety and autism and all the rest represent “natural kinds,” or are they artefacts of an entirely false order humans are trying to impose on nature? These conditions may be related, in the sense that tigers and pussycats are related, but that’s not enough. We need to understand what if anything they have in common, why they’re separate, and how they stand in relation to the rest of the universe.

Fast forward to about the 1880s, when three sorts of mental disorder were recognised: mad, bad and sad, or psychotic, personality-disordered, and depressed, as we would now say. Over a long period from the 1890s, German psychiatrist Emil Kräpelin split the “mad” group in two: the slowly deteriorating dementia praecox, or dementia of youth (now schizophrenia), and the cyclical manic-depressive (bipolar) group, who had repeated bouts but got better in between. For Kräpelin, mental disorder was a physical disease of the brain resulting from hereditary degenerations. He wasn’t interested in what his patients had to say, only in the course of their condition. He rejected all philosophical or metaphysical matters of the mind, such as Freud’s system of psychoanalysis, and was convinced male homosexuality was a vice caused by masturbation so it should be punished. After his death in 1926, this “organic” principle was embraced by the Nazis as the foundation of their “race science” including the idea that homosexuals should be punished, if not murdered.

The history is significant because if recognisable mental disorders have distinct biological causes, then they genuinely are “natural kinds,” and any similarities are coincidental, not causative. For example, the fact that whales and fish have fins is coincidental, a product of their environment (called convergent evolution), but has no causative role. The fact that whales have lungs and not gills means they must breathe air regularly, as we do, so our relationship as mammals is causative, not coincidental.

With the girl described above, the standard view, which the authors clearly adopted, was that her separate diagnoses were coincidental to each other and were not causally-related. Each diagnosis had a separate and distinct underlying cause in her brain, and therefore needed to be treated separately. This theme, the search for “natural kinds” of mental disorder, dominates psychiatry to the present and is responsible for the relentless expansion of the numbers of psychiatric diagnoses. For academic psychiatrists, the grand prix d’honneur is to have your pet obsession recognised as a new condition by the DSM committee. Just as with funnel web spiders, splitting mental disorders into ever-smaller splinter groups is the name of the game. The problem with the tradition Kräpelin left us is that, for him and his generations of successors, the belief came before the evidence. That is, they firmly believe all mental disorders are natural kinds of things; all they need now is the evidence.

Look at rocks: granite and marble are both very hard and heavy and can be carved and polished to a mirror-like surface, as distinct from sandstone, which is light and can’t. However, they are not a unique kind of their own because granite is an igneous rock, formed from molten lava, while marble is metamorphic, formed from layers of sandstone compressed under intense heat and pressure for aeons. Their relationship as heavy, hard, sharp rocks is coincidental, not causative.

Modern psychiatry starts with the idea that all mental disorders are distinct just because they have distinct genetic causes. Any similarities are coincidental and can be dismissed because the doctrine says “It’s all genetic.” As a result, psychiatry today is intensely biological in nature. The glittering prize awaiting any researcher, the Holy Grail, is to be the first to come up with a specific biological cause for a recognised disorder. The amounts of money involved, and the time and the intellectual effort, are enormous, a sort of Manhattan Project of the brain (not the mind, mind you). Thomas Insel, longest serving director of the NIHM, estimated that in his 13 years at the helm, the agency had disbursed something like $20billion for basic research in biological psychiatry. In the 12 years since, they would easily have matched that, if not more, plus all the money that went before 2001, plus all the money spent elsewhere … pretty soon, you’re talking big money, as they say. Very big money (see Substack for Jan 21^st).

Trouble is, as Insel was forced to admit, it was money down the drain. All that monumental effort has achieved nothing. We are no closer today to proving a genetic or biological cause for even a single mental disorder than we were fifty years ago, and all the outcome statistics for mental disorder hadn’t budged. Worse still, the number of diagnoses is heading for the moon. If researchers can find a slight difference between two groups of patients, that’s enough to demand a new diagnostic category, just because the difference is biological, and biological differences are causative. That’s exactly what happens when the belief comes before the evidence. The fact that this only happens in psychiatry should be enough to alert people to the possibility that the whole process of diagnosis is an artefact but no, the diagnostic horse has the bit between its teeth and was last seen going over the hill.

Should we therefore stop to reconsider the whole project? Should we be asking: “Is it true that mental disorders are natural kinds, or is that impression caused by an evidence-free human opinion?” Take the case above: the unhappy girl’s “poor eye contact” was allocated to the category of ASD. Why? Who says? Why wasn’t it seen as avoiding direct gaze as a typical part of severe anxiety? She’s only got one pair of eyes. Do we allocate them to her ASD or her GAD? This is where the whole notion breaks down because the decision is made on opinion, not objective fact. This could have been avoided if the architects of the diagnostic systems had listened to the late Bishop William of Ockham, whose name is attached to a renowned razor, also known as the principle of parsimony. This says that the number of explanatory entities must not expand beyond the minimum required to do the job. In practice, it means where there are two competing theories, the simpler explanations is to be preferred. Therefore, instead of saying the poor girl had rotten genes causing three distinct but coincidental conditions, why don’t we say they are all causally related? For example, we could say her problems started as quite a severe anxiety state, which caused her to…:

…avoid eye contact, interfered with her schooling and social life, made her stutter and stumble over her speech, led to rigidity in her routines due to fear of criticism, avoiding anything new in case she failed, was constantly preoccupied with her troubles but could forget everything when playing with technology (which she did very well because she was bright and put the effort into it), startled easily with sudden noises or bright flashing lights, and so on.

These are all absolutely stock-standard symptoms of anxiety. But with a bit of shuffling and rewording, the same symptoms suddenly become a diagnosis of ASD - if that’s what the psychiatrist wants. The symptoms always support what the psychiatrist believes. Finally, anybody with this level of trouble will eventually become depressed. That is, a single entity accounts for the lot. But, you object, isn’t that still a biological disease of the brain? No, there is no evidence for this. Insel couldn’t point to any positive results from his massive spend, yet the reason for failure is staring him in the face: anxiety is not a “mental/brain disease,” it is a personality problem. Unfortunately, once again, the belief comes before the evidence. The American DSM system doesn’t recognise an “anxious personality disorder” because they want anxiety to be a mental disease, then they can throw drugs at it.

Can we give a parsimonious account of an anxious personality? Sure can: anxiety is the (psycho-somatic) response to the perception of a threat. This is universal: every creature on earth has a threat response, even earthworms. The threat response is hard-wired into us. An anxious personality exists when the person habitually reacts to neutral events in the environment as though they were a threat, i.e. for whatever reason, the unfortunate person misclassifies events as dangerous and then reacts accordingly. There is no biological defect in the brain or body [2]. Anxious people are simply switching on their alarm systems too often and too intensely. The mind and body are reacting correctly to inappropriate alarm signals. Treatment may involve drugs but is primarily directed at correcting the false perception of the world, because that’s the cause. Trouble is, if it’s a personality problem, it means psychiatry has failed and it has to go to the psychologists.

When the inventor Thomas Edison was looking for a material to form the filament of the electric light globe, the project didn’t go well and people mocked him for failing. He replied: “I haven’t failed. I’ve just found 10,000 ways that won’t work.” After 75 years and $75billion, even the most obdurate academic must be secretly wondering whether they’re on the right path, whether the 10,001st trial will be right. Psychiatrists, of course, would rather pull their tongues out with pliers than admit their “biomedical model” (the one they’ve never written) has failed so we can probably look forward to another few years of the same old same old. If only they’d read Linnaeus more closely. Even though practically the whole world wanted humans standing alone at the top of the tree, he stuck us next to our cousins, the apes. Everybody tried to prove him wrong but we’re still there. When the flow is going nowhere, as Insel admitted, it’s time to start thinking about going against the flow.

My contribution is this: “autism spectrum disorder” is not a “natural kind.” It is an artefact which can be given full account within a psychological model of anxiety as a primary personality disorder. Now let’s see if anybody can prove that wrong.

*Betteridge’s law says "Any headline that ends in a question mark can be answered by the word no."

References:

1. Maenner MJ, et al. Prevalence and characteristics of ASD among children aged 8 years in 11 sites, 2018. MMWR Surveill Summ. 2021;70(11):1–16. PubMed CrossRef

2. McLaren N (2018). Anxiety: The Inside Story. Ann Arbor, MI: Future Psychiatry Press.

The whole of this material is copyright but can be quoted or copied if the author is acknowledged.