The Science and Fiction of Autism 
by Laura Schreibman.
Harvard, 293 pp., £17.95, December 2005, 0 674 01931 8
Show More
Send in the Idiots, or How We Grew to Understand the World 
by Kamran Nazeer.
Bloomsbury, 230 pp., £12.99, March 2006, 0 7475 7910 5
Show More
Show More

Autism is devastating – to the family. Children can be born with all manner of problems. Some begin life in great pain that can never be relieved, but at least there is a child there. An autistic child – and here I am talking about what’s known as core autism – is somehow not there. ‘Nobody Nowhere’, as the title of Donna Williams’s autobiography (1992) has it. Very often physically healthy (though there is a high incidence of other problems) he – and it is usually he – just does not respond. It is not merely that he does not learn to speak until years after his peers, and then inadequately. He has no affect; he never snuggles. He is obsessed with things and order, but does not play with toys in any recognisable way, and certainly does not play with other children. He mercilessly repeats a few things you say. With no comprehension. He has violent tantrums, not the usual sort of thing, but screaming, hitting, biting, smashing. This alternates with a placid gentleness, maybe even a smile – but not really for you. Serious Down’s syndrome is pretty bad too, but despite all the difficulties, physical and mental, there is a loving little child there. That is what is so dreadful about core autism: your child is an alien. Parents who guide their autistic infant through to adulthood, who create a human being who can be loving, who can to some extent compensate for his deficits, who can find some dignity and maybe a modest type of respected work – they are, in my opinion, heroes.

Many parents will be angry with what I have just said. ‘It is not like that at all. Peter is the most lovable little boy. We understand his need to have everything just so and we know he has trouble playing with other children. It’s a shame we cannot take him out much, because he gets disturbed and people don’t understand. But his grandparents adore him. He will learn to talk quite soon. He is making a lot of progress – after all, he is autistic.’ It helps to have a name for something inexplicable, to be able to meet other families, to get help from social services and special schools. Things are easier than they once were, but bringing up a child with core autism continues to demand a lot of courage and persistence.

Laura Schreibman has spent much of her career caring for and working with autistic children and their families. She is the director of the Autism Research Program at the University of California, San Diego. But her real qualification is a lifetime of working with the children, and she acutely knows the difficulties of the families. For anyone close to an autistic person who wants to learn about the problem, this is almost certainly the best available manual. Schreibman’s target readership is the front line.

What do we know? What should we do? What can we hope for? These are the questions that beset anyone hit by autism – not just the autistic person, but their family and friends. Schreibman does her best to answer them. What do we know about autism? Not much. Worse, there are a lot of people who claim to know a lot about autism, but their proposals about what to do are radically incompatible with each other. Many hold out bountiful hope. They quarrel, these experts, they hate each other. Who to trust? Schreibman describes pretty much every current approach to autism, and examines how well supported by evidence each proposed theory is, and how well each therapy works. There is no better straightforward source of answers than this book. Schreibman is blunt about what she thinks is fiction. She is also blunt about our present state of ignorance.

To show how hard things are, let me run through a few simple assertions. (1) Autism is a developmental disorder. True, but true by definition. Autistic children are among those who do not develop mentally and socially in the way that most children develop. (2) Autism is a neurobiological disorder with which a child is born. There is ample reason to believe this is true, but at present its value is chiefly corrective. It implies that autism is not caused by bad parenting, or bad diet, or measles vaccine. There are lots of research programmes aimed at finding out what is wrong with the brain and its chemicals, but despite all the hype for this or that project, we have only just begun. (3) Autism is a genetic disorder. Yes, there is some evidence about homozygous twins, but once again, for all the talk about this or that genetic site, we are only just beginning. One paper says the disorder is genetic, but the defective genetic material would be different in every single case. There is certainly no simple genetic solution like that for Down’s syndrome, which is caused by an extra chromosome 21. (4) More males than females are autistic. Yes, four out of five cases, and that holds true for core autism and for the most generous definition of the disorder. (5) The symptoms show up very early, and are usually clear by the age of 30 months. Yes. Leo Kanner, who in 1943 introduced the diagnosis of childhood autism, went further. He thought you could tell in the first few months of life. Ordinary infants move their heads and position themselves so they can be picked up by mum, as if expecting to be held and loved. Autistic infants do not do so, or so Kanner found.

Over the past fifteen years everyone has got to know about autism. Autism will figure this year in dozens, maybe hundreds of cheap novels, thrillers and maybe a good book or two, just as multiple personality did fifteen years ago. (Thank goodness that’s gone!) As well as core autism we now have the autistic spectrum. We have Asperger’s. We have ‘high-functioning’ autists. The success of the high-functioning, their foibles and their triumphs, tends to make the general reader think, ah, so that is what autism is like. Mark Haddon’s The Curious Incident of the Dog in the Night-Time acts, among other things, as a wonderful means of raising awareness. But life is not always like that. Downers don’t sell unless there is something sensational to tell. There are any number of more or less factual books about any number of subjects to make you feel wretched, but I have never read a book more wrenching than Jeanne-Marie Préfaut’s Maman, pas l’hôpital! (1997), written by a woman who murdered her 23-year-old autistic daughter.

The word autism was introduced at the beginning of the 20th century by the Swiss psychiatrist Eugen Bleuler, the man who made up the word schizophrenia, disseminated by his book of 1911, The Group of Schizophrenias (note the word ‘group’). The book also spread the word autism he had made up a little earlier, to indicate the total self-absorption of some of his patients. His examples were pretty bizarre, and make you think that the bourgeoisie of those days were even nuttier than we can imagine. The noun was used regularly, if not frequently, in German-language (including what we would now call Eastern European) psychiatry in the interwar years, but only, I think, in connection with adults and usually associated with schizophrenia. German-language child psychiatry was evolving rapidly, however, often in the form of Heilkunde, a combination of psychiatry and pedagogy almost unknown in the West.

Leo Kanner was a Galician Jew (hence Austrian) who trained in Berlin. During the early and turbulent days of the Weimar Republic he went out to practise in rural South Dakota. He later went east and found his home at Johns Hopkins University, where he established the first child psychiatry clinic in the US. He wrote the first English-language textbook of child psychiatry, which, in successive thick editions, became the standard. It drew heavily on existing German texts. Kanner was not only a brilliant clinician, but also a caring and gentle one. In 1943 he published the diagnosis of childhood autism, with 11 case studies. In 1944, Hans Asperger in Vienna published a series of similar cases. In the beginning autism was paired with childhood schizophrenia. The ideas were divorced in 1979. The Journal of Autism and Childhood Schizophrenia, founded in 1971, became the Journal of Autism and Developmental Disorders, with editorials saying what a stupid idea childhood schizophrenia was. Some people are seriously rethinking that issue today, partly because we now wonder if, after all, schizophrenia itself is precisely one thing. Remember that for Bleuler it was a group.

How come, at exactly the same time, but separated by an ocean and a war, we get two very similar, absolutely new childhood disorders? Not so surprising. Asperger, a generation younger than Kanner, had trained under August Homburger, the author of one of Kanner’s main German textbooks, who wrote about childhood schizophrenia and other developmental disorders. They came from the same medical culture (each had served in the Austrian army, although in different wars).

How come no one had made this diagnosis before? There are precedents in the German literature, even if they are not so clear-cut. The important point is that in the 1930s there was an American guru named Arnold Gesell, who had a role similar to that of Dr Spock in the 1960s. In such books as The Mental Growth of the Pre-School Child: A Psychological Outline of Normal Development from Birth to the Sixth Year, including a System of Developmental Diagnosis (1925), he told parents exactly how their children should develop, the age at which they would be able to tie their shoelaces etc. Middle-class people knew their Gesell. Thanks to him, you knew when there was something awry with your child’s development. To start with, you would notice the absence of a welcoming posture when you took your infant into your arms. Parents of such children in Baltimore were able to go to the new child psychiatry clinic.

Kanner noted that the parents who came to see him were rather uptight professional people. Hence an inadvertent tragedy. He said the mothers were like refrigerators, cold and distant. Psychoanalysis was dominant in American psychiatry. The shrinks moved in: it was the mothers who needed treatment, not the children. Lord knows how much pointless pain this caused, how many families were destroyed, how many children damaged further. There is a standard paragraph in almost all books on autism, including those under review, trashing Kanner. Yes, he was responsible for the beginning, but if you want someone to hate, I suggest you switch your antipathy to Bruno Bettelheim, whose book The Empty Fortress: Infantile Autism and the Birth of the Self (1967) cemented the idea that autism should be treated by psychoanalysis.

Kanner was not wholly wrong, by the way. Simon Baron-Cohen, the best-known British autism researcher today, has a research programme to investigate the extent to which the family trees of autists are made up of engineers, scientists and other abstract thinkers whose lives revolve around cold structures rather than human empathy. Baron-Cohen finds that what he calls ‘male’ as opposed to ‘female’ attributes, abstraction and distance from human relationships, tend to run in families with an autistic child. This statistical claim rides well with Kanner’s initial observations, which is not to say it is a universal fact. The monstrous error lay less in Kanner’s clinical description than in people abusing it by supposing that you should work on parents, e.g. by psychoanalysis, and that success with the parents would alleviate the child’s autism.

What about Asperger? He was very well established in the German-speaking world after the war, and his work was well known in the Soviet Union. But he was unknown in English until, in 1981, the British psychiatric worker Lorna Wing introduced his ideas. Asperger’s now often means something like high-functioning autism, with the usual social deficits and obsession with order, an obsessive literalness, and an inability to pretend or imagine, but no or fewer difficulties in learning to speak. There are debates about whether it has the same basic causes as autism, or whether it is something different. It has become a widely used term of common speech. A New York psychiatrist told me a joke, not quite a joke: ‘Twenty years ago, women of a certain age would come to us, and say: “Doctor, my husband is a jerk, what should I do?” Now they say: “Doctor, my husband has Asperger’s, what should I do?”’

We are getting to part of the reason there is suddenly so much autism all around us. The criteria for being autistic have radically expanded since Kanner’s day. This is not just a conceptual change, but also a discovery that took time. When there was no diagnosis of autism, you could not be said to have grown out of it, or recovered, or been cured. Kanner followed up his first patients. None recovered; two survived thanks to the dedication of their families; most were in institutions, against which Kanner raged. But once there was a diagnosis, and more attention to care, some children did grow up to have viable social lives. The most famous is Temple Grandin, who has a strong empathy with animals and has helped to make radical improvements in US slaughterhouses. Kamran Nazeer’s is the latest addition to the list of fascinating autobiographies by socially competent former autists. Thanks to all these stories, many misfit adults now recognise themselves as autistic, or so they say. It really helps to be able to put a label to your oddities. It brings a kind of peace: so that is what I am. Therapists are glad to encourage this. There is even a new movement afoot: we are all right, we are just different from you, we do some things better than you, you do some things better than we do. But let’s not forget the vast majority of core autists who at best barely compensate for their difficulties.

I am told by a British activist and philanthropist that in the 1960s, when her child was born, parents pressed for the autism diagnosis because it was the only form of learning disability not classed as ‘ineducable’ in Britain. In fact autism has become an advocacy disorder, which contributes to the increased rate of diagnosis. Militant middle-class parents fought to have their problem taken seriously. In the United States the provisions for ‘special education’ are very generous, partly because John Kennedy had a severely retarded sister. With special education in place, parents of autistic children fought long and hard for public awareness, and they got it. Today a child with learning and social problems will get more attention if he is labelled autistic. North Carolina has a remarkable statewide programme for children with serious difficulties. The acronym is TEACCH, Treatment and Education of Autistic and Related Communication-Handicapped Children. The ‘related’ part of that gets filed as autistic when details are not being attended to.

The support system in Britain is pretty good. More children are diagnosed early, simply because ordinary teachers are more likely to refer a difficult child who may be autistic, and parents will encourage that because there is no longer any stigma and there is more institutional support. The raising of consciousness by autism advocates and others, including novelists, has immeasurably improved the lives of autists not only by aiding the provision of better services but also by making it possible for others to accept them as they are.

I believe that such factors can explain the greatly increased prevalence of autism diagnoses. If you really believe there has been an increase in the incidence of autism itself, then that must be a very important clue to its causes – which is exactly what is said by the fringe groups. What changed in the environment to produce more autists? The introduction of MMR vaccine, they say. Or more mercury in the food chain. And so on. Schreibman rightly says there is little sound evidence for these hypotheses, but if you think autism is becoming more common you had better look for a reason. Since we know nothing about the causes of autism, every superficial clue should be attended to. This is one of the merits of Baron-Cohen’s research. Although he goes out on some limbs that a robin would not dare to perch on, he asks more of the right kind of simple question than most researchers. The fact that most autists are male must be a clue. There must be something important about the inability of autistic children to make eye contact; it is even reported that they are much better than most of us at noticing changes in the lower part of a face, and much worse in the upper parts. One kind of autism research has to start with these raw puzzles.

Almost by definition, autistic children do not well understand other people’s feelings. A vigorous school of cognitive science holds that many human capacities are innate and modular, that a distinct neural organisation will correspond to each group of abilities. According to this view, autistic children lack the mental module that enables us easily to understand others. They lack ‘a theory of mind’. Many clues point in that direction. There are batteries of tests that indicate autists do not attribute beliefs to other people in the ordinary way. They have defective ‘mirror neurons’, it’s said, which produce dispositions parallel to those in people we see and hear. There are the observed idiosyncrasies about eye contact. The theory of mind approach, well established in Britain by Uta Frith, Baron-Cohen, Alan Leslie and others, is not much welcomed by either Schreibman or Nazeer, however.

Nevertheless, here is yet another pointer to the increased awareness of autism: there are a lot of psychologists and cognitive scientists who think that autism is a key to understanding the human mind. Moreover, an unusual number of major scientific contributors, such as Lorna Wing, are also parents and passionate advocates. We read a lot about autism not only because of its increased profile but also because it is scientifically in vogue. Nazeer, having been on the receiving end, notes that autistic children are subject to endless tests, and sardonically warns of experimenter-effects on the responses of the children.

So what do we know about autism? Not much. What should we do? Schreibman describes a lot of treatment systems. The most radical was introduced by Ivar Lovaas in Los Angeles forty years ago. He tried psychoanalysis, which did no good, so he switched to intensive operant conditioning. It is the purest behaviourism. He calls it Applied Behavioural Analysis. At the earliest possible age, even 30 months, the child begins to work with a trainer, seven hours a day, six days a week. Anything good, a single sound at the right time, a look in the right direction, is reinforced with smiles and even sweets. Bad behaviour is rigorously discouraged. The Lovaas method costs about £40,000 a year. The claim is that after this gruelling preparation, the child with an assistant can start in the state school system, and by the age of 12 will be fully integrated. A lot of experts do not accept this.

What behaviour do you reinforce? The operant conditioning emphasises language. Schreibman notes a problem. Many children with core autism seem to learn only rote sentences that they have been trained to repeat. Modern grammatical theory began with Chomsky’s demolition of the behavioural approach to language acquisition proposed by B.F. Skinner. It is as if such children are a horrid parody of Skinner. It is as if they really do lack the language faculty, which, according to Chomsky, is innate in most human beings. Nevertheless, before Lovaas, no system seems systematically to have helped autistic children. Applied Behavioral Analysis opened doors.

Schreibman reports that numerous variations on the theme of behavioural modification do help children. She describes the current evidence for the benefits of each. TEACCH accepts that autism is neurobiological and irreversible, and tries to reinforce whatever strengths the child may have. Schreibman files it among ‘treatments likely to be effective’ but notes some serious reservations. Nevertheless, she writes, ‘today the treatment of choice is that based on the behavioural model. In fact, behavioural treatment is the only treatment that has been empirically demonstrated to be effective for children with autism.’ What one wants is an eclectic mix of basically behavioural methods and a lot of loving human kindness.

Perhaps some cases of autism are, after all, ‘reversible’. Nazeer is the son of what appear to be prosperous, itinerant and very supportive Pakistani parents. He did begin to speak at the age of four, but had the usual inability to comprehend other people and was obsessed with lining his toy cars in the correct order against the wall. He had the good fortune to be sent to a small school for autistic children in New York, run by dedicated and sensitive teachers. They too worked on behaviour, but, it appears, more on social deficits than anything else. Hour after hour, day after day, getting the children to play with each other. Repeatedly running tapes of a conversation: what is Tom saying to Maureen? Why? How does she react? Why? Today they use a computer programme: Ultimate Learning – Fun with Feelings. Nazeer does not mention that there is now a first generation of cheery looking robots intended to interact with autistic children, and save on the inevitable burn-out of many teachers.

Nazeer retains a few oddities, but he has a law degree and a PhD in philosophy and now works in London as a policy analyst for the government. He decided to see what had happened to some of his schoolmates. He tracked down three who were willing to talk at length. A fourth had committed suicide, but the family met with him. One is almost as high-functioning as Nazeer, and works as a speechwriter for Democratic politicians. He has written speeches for quite a few senators, though he has a flat, affectless voice. He was the valedictorian (graduation speaker) for his high school – but another student had to read his speech. A second is a bicycle courier who seems to me to have a low life-expectancy: one of his eccentricities is to head out before dawn towards downtown Chicago; before he hits the city, he rides for a few miles with his eyes shut. He gets on well with his boss; there is a creepy anecdote about a customer mocking him by telling him to deliver a pistol across town, which he does mechanically, because that is what he does. His gay lover exploits him. Finally they break up, which is in its way a socialising experience too. The third has discovered how to use puppets to say things he cannot say. Nazeer argues that what autistic people most need is what he calls ‘local coherence’, and he analyses his friends in terms of their desire to achieve that end. He also has a lot of conversational information and opinions about autism and its theorists. Schreibman is usefully didactic; Nazeer is a charming raconteur with serious things to say.

He returns to his teachers. One is rather hard on him. ‘You’re a mean person,’ he says to her. ‘You’re not autistic,’ she replies. (A fully autistic person would not understand that she is being mean.) What made it possible for him to come out of it? Totally supportive and presumably prosperous parents, a good school at the right time. No really serious problems with acquiring language – age four is not so terrible. Even given these very favourable starting points, his evolution is exceptional. We have no idea why. Did his little school accept only the most promising pupils?

Are all these stories, from Maman, pas l’hôpital! to Kamran Nazeer’s, stories of people with the same neurobiological disorder, but in varying degrees? Or are we really talking about different things? As Nazeer observes, most autistic people do not live independently. For all the three successes he has discovered, plus himself, he says that most autistic adults live with their parents – as did his schoolmate who killed herself – if they have not been put in an institution. And aside from fully socialised people like himself, most of those who manage away from their families live lonely lives. An ageing woman I know has a son who looks after himself in a small cottage: ‘Mum, when you’re gone, I will be all alone.’

We say ‘autistic spectrum’, which is not quite right, because a spectrum is a one-dimensional range. Autism ranges in at least three dimensions: language deficit, social deficit and obsession with order. We should talk of an autistic space. So, are all the individuals that we now place in this space of symptoms in the same neurobiological space? Or should we more cautiously speak of the group of autisms, without implying that in terms of causes they are variations of the same thing? Whatever made Kamran Nazeer as he was at the age of four may have been fundamentally different in kind from what lies behind the children with core autism with whom I began.

It is an enormously important question, for neuro-bio-genetic research, whether we are dealing with one fundamental entity or not. In terms of helping people at present, it may not matter at all. Lorna Wing and her colleagues have a rich questionnaire for identifying people with difficulties in the ‘spectrum’. ‘Our purpose was to develop a clinical instrument, which didn’t ask: “Does this person have autism?”’ But instead: ‘“What problems, what advantages and skills does this person have?” Now that is a very different question from: “Has this person got autism?” It is ever so hard to convey this to people.’ It is a mark of how little we know, that many good answers to the question ‘what should we do?’ simply bypass issues of identifying autism. And what about that last question: what can we hope for? In the short run, significantly better lives for children and adults with autism, and hence for their families. I suspect it will be a very long time before we find out what the neuro-bio-genetic space of autism is.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.


Vol. 28 No. 11 · 8 June 2006

Ian Hacking predicts that parents of autistic children, whom he describes as ‘heroes’, will be angered by his list of all the things that are ‘so dreadful’ about their autistic children (LRB, 11 May). He is right to suggest that autism is a hazy description of a little known condition that may in fact not be a condition at all, but this does not stop him from giving a detailed description of ‘an autistic child’, which is entirely a list of negative characteristics, and concludes: ‘Your child is an alien.’ Does every autistic person share every one of these negative characteristics and no positive ones? As with any diagnosis concerning the mind, everyone can identify ‘autistic’ characteristics in themselves and every ‘autistic’ person will sometimes confound expectations.

Hacking also describes John Kennedy’s sister as ‘severely retarded’. In terms of its level of offensiveness, this is on a par with a white writer casually using the term ‘negro’, but this description crops up in the London Review on a disturbingly regular basis. It is offensive to sum up an individual with a one-word pejorative description of their perceived disability, because it suggests that there can be nothing else to know about them. Compare Stephen Hawking with me: if you take only physical ability into account, he becomes a ‘cripple’; if you consider only intellectual ability, I become a ‘retard’. All the currently widely accepted labels begin with the word ‘person’, which is where your contributors’ thinking on the subject should be encouraged to begin.

Alex Fox

Vol. 28 No. 12 · 22 June 2006

Alex Fox takes Ian Hacking to task for describing John Kennedy’s sister Rosemary as ‘severely retarded’ and suggests the LRB adopt ‘currently widely accepted labels’ (Letters, 8 June). Hasn’t he noticed how fast euphemisms date? And do they help much anyway? Rosemary Kennedy seems to have had mild brain damage from birth, but was capable of travel and a social life, until her parents had her lobotomised and institutionalised. Earlier she had been a pupil at the Coletta School for Exceptional Children. The coyness did not, I suppose, make her life easier to bear.

Hilary Mantel

Vol. 28 No. 13 · 6 July 2006

Ian Hacking’s description of the autistic child as an ‘alien’ is obviously meant to shock (LRB, 11 May). But to what end? There is nothing more ‘alien’ about children with autism than those with other severe developmental disabilities. However, because these children have difficulties with social communication and engagement, a profound state of alienation between such children and their parents and others can develop. This can be reversed once parents learn to understand the nature of the difficulty and adapt the way they communicate with the child. Although this does not magic away the disability, it is brought back into the realm of human meaning.

In fact, far from being applied only to ‘alien’ children, the notion of autism is currently overgeneralised. ‘Autism’ is now invoked to explain everything from intellectual distinction to artistic innovation to a football fan’s obsession with statistics. The seductive (but unproven) theory that autism is an extreme form of ‘male brain-ness’ doesn’t help in this respect. Maleness has, historically, always been associated with increased neurodevelopmental vulnerability of just about every kind. Why this should be so is undoubtedly of interest, but even though some aspects of personality (including those we associate vaguely with gender) appear to have something in common with traits associated with autism, it would be a mistake to equate them. ‘Sociability’, like any distributed trait, has a high and a low end: in a culture that privileges ‘sociability’ so highly it isn’t surprising that the ‘low’ end is scrutinised and even denigrated. States of pathology and disability, not least the notion of ‘madness’ itself, have often been used for spurious cultural ends, and it would be nice to avoid doing that in this instance.

Jonathan Green
University of Manchester

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN

Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences