Autistic autobiography

Abstract

Autism narratives are not just stories or histories, describing a given reality. They are creating the language in which to describe the experience of autism, and hence helping to forge the concepts in which to think autism. This paper focuses on a series of autobiographies that began with Grandin's Emergence. These are often said to show us autism from the ‘inside’. The paper proposes that instead they are developing ways to describe experience for which there is little pre-existing language. Wittgenstein has many well-known aphorisms about how we understand other people directly, without inference. They condense what he had found in Wolfgang Köhler's Gestalt Psychology. These phenomena of direct understanding what other people are doing are, Köhler wrote, ‘the common property and practice of mankind'. They are not the common property and practice of people with autism. Ordinary language is rich in age-old ways to describe what others are thinking, feeling and so forth. Köhler's phenomena are the bedrock on which such language rests. There is no such discourse for autism, because Köhler's phenomena are absent. But a new discourse is being made up right now, i.e. ways of talking for which the autobiographies serve as working prototypes.

1. Introduction

We now watch, hear and read a great deal about what it is like to be autistic—in autobiography, biography and fiction. Autism narrative is a new genre: not expert reports by clinicians or reflections by theorists, but stories about people with autism, told by the people themselves, or their families, or by novelists, or by writers of stories for children.

Since this is a new genre, it is multimedia. Its richest habitat is the blogosphere. Chat rooms are awash with autists chatting. I shall unfashionably stick to print, and will mainly refer to four well-known autobiographies, those by Grandin (1986, 1995, 2005), Williams (1992, 1994), Mukhopadhyay (2000, 2008) and Tammet (2006). The authors are more than ‘talented’, as are for example Nazeer (2006) and Mór (2007). As might be expected, they are not all uniformly ‘high-functioning’; Mukhopadhyay is brilliant in the use of words, but needs help in many other aspects of life. I use these four to explore how talented autistic individuals are affecting our understanding of autism.

I shall not use this occasion to quote much of what they say. Let them speak for themselves. I offer instead a suggestion about their role in the ongoing social and cultural evolution of the autistic spectrum. I take for granted that underneath the spectrum is a family of definite biological conditions, be they neurological or genetic or whatever, but that in an important sense the spectrum itself is a ‘moving target’ that has evolved dramatically (Hacking 2007).

2. Some questions

Our instinct is to treat the words of the autobiographies as literal descriptions, albeit tinged with metaphor and coloured by art. That is to that they are true (or false) according to some pre-existing criteria for describing experiences and sensibilities. I wonder whether they are not instead helping to create a language for talking about what was hitherto unknown. Yet even that is not right, for it implies that there was a continent there to know, but not yet explored.

1. Are the autobiographies and other stories less telling what it is like to be autistic than constituting it, both for those who inhabit the autistic spectrum, and for those who do not?

Despite my few examples, we are concerned with the entire multimedia genre. Different kinds of item influence each other in complex ways. Novelists study autobiographies, whose authors learn from theorists. Parents pick up ideas from novels when they are thinking about their children. We all watch movies and documentaries. A ‘thick’ kind of human being is coming into being, where once there was only a ‘thin’ one. The autistic thin man of yore, or rather the thin child, when not having a tantrum, was a silent self-absorbed creature, alone with bizarre habits.

Our four autobiographers clearly have personalities that are thick, dense or rich, whatever adjectives you please. Can they teach anything about other people with autism, and, in particular, more severely impaired people?

1. Do the autobiographies provide prototypes for describing and thinking of all autistic people? Or is that exactly the wrong way to go, because it suits only the ‘high functioning’, and creates too many false expectations about others?

Oliver Sacks wrote his own engaging account of Grandin, under the title, ‘An anthropologist on Mars’. Grandin had said to him after dinner, ‘Much of the time I feel like an anthropologist on Mars’. (Sacks 1993, p. 259; 1995.) That phrase invites the all too suggestive trope of the alien, which some in the autistic communities favour and some resent. Wittgenstein (2001, p. 190e) thought that: ‘If a lion could talk, we could not understand him’. If a Martian spoke, would we understand it? Only if we shared or came to share some ‘forms of life’, some ways of living together with lions. That is precisely a problem for a person with severe autism. But the evocative phrase, form of life, is never more than a pointer; we shall need to be more specific about what is missing. Some observations by Wittgenstein and by Wolfgang Köhler, the best-remembered pioneer of Gestalt psychology, are unexpectedly helpful in this respect.

3. Texts not people

The people who write autobiographies will not concern us here, only their texts. This has several corollaries. First, it does not matter that in some cases the autist is not the sole author of the final words of the text. Not only have editors had their roles, but also for example Grandin's first volume of autobiography, Emergence, was written ‘with’ a professional co-writer. Mukhopadhyay's recent title, How Can I Talk if My Lips Don't Move? Inside my Autistic Mind (2008), was made up by someone in Arcade Publishing Inc. That is irrelevant. It does not matter for present purposes who contributed what to the final print form of the text.

Second, in my lowbrow sense of the word ‘text’, the outside of a book is part of the book. The cover with its puffs and blurbs will be more impressed on most people who read the book than the middle 100 pages.

Third, we are not concerned with diagnosis. I use the label ‘autism’ in a deliberately ill-informed way, to refer to individuals with some sort of autistic spectrum disorder. Mukhopadhyay is the only one out of our four authors diagnosed as autistic at an early age, a diagnosis confirmed by a thorough examination in childhood by British experts. Although sometimes these four individuals are often taken to be aspergic as opposed to autistic, they themselves regularly use the word autism and are heard by the non-discriminating public as speaking about autism. The UK title of Tammet (2006) uses ‘Asperger's’ where the USA one has ‘autistic’.

Suppose, however, that we were to answer question (ii) in the negative, and hold that the autobiographers are speaking only for the high-functioning end of the spectrum. Then it would be essential to attend to the distinctions along the spectrum that I am deliberately ignoring.

Fourth, our concern is not that of neurologists who may hope to decipher from the words of the autobiographers' clues to what is going on their brains, or the biological causes of their condition. My preoccupation is quite different and far less fundamental. I am concerned with what these words are doing to the public understanding of autism.

4. The tension between high functioning and severe

Alison Singer, vice-president of Autism Speaks, expresses one side of a tension that has bedevilled relationships between parents of autistic children. She contrasts her daughter Jodie, who ‘has classic autism, which falls on the other end of the autism spectrum from Asperger syndrome’, and another girl and friend of the family, Haley.

Many days it is hard to believe that the challenges Haley faces with regard to her Asperger syndrome and those Jodie struggles with are related under the same DSM-IV diagnosis.

At one end of the autism spectrum, we often find lower functioning persons like my daughter who cannot speak, have violent tantrums and can be self injurious, while at the other end we have persons who struggle with very significant, but very different, predominantly social issues.

(Singer, no date)

Singer is one of the two major participants in the video Autism Every Day.1 She went so far as to say, on screen, that sometimes she wanted to end it all by driving her car, with herself and her daughter in it, over the Brooklyn Bridge. For this, she has been much vilified, even to the extent of some critics saying that her child should be taken away from her.

Singer is, then, an activist who speaks for the family-shattering impact of severe autism, and wants to insist that this is a problem quite distinct from that at the other end of the spectrum. I presume she would say that our four autobiographers speak for high-functioning autists, but not for her daughter. I fully respect her point of view, which is shared by myriad mothers on the front lines. I am exploring the alternative possibility, that the spectrum may evolve in such a way that there is thought to be more of a continuum in the descriptions of experience, than there has been. It is not a question, in my opinion, of finding out that there is a continuum, so much as trying to smooth over the contrasts urged on us, and experienced by Alison Singer. Note also that a few children begin as extremely impaired, and somehow grow out of it, to function well, so that a continuum can appear in a single life.

Singer's assertions are of grave practical importance. There is a growing wing of the autistic community that rejects the idea of looking for a cure. That makes sense, Singer implies, for Asperger's people who have a thick if unusual life. It is not fine for her daughter. Likewise Moore, author of George and Sam (2004), the remarkable book about her two severely autistic sons, is very much against integrating her children in regular schools, with the assistance of aides qualified in special education. That may be excellent for children further along the spectrum, but not for her older sons.

5. Autism narrative

Autism stories thrive on stage, radio, TV, cinema and above all on the Internet. Texting is an ideal mode of communication for many autists. Even when we stick to print, there is a vast publishing world of autism narrative beyond autobiography. Moreover, there are many more than a handful of autobiographies. Williams herself has published two more such volumes as well as other books. There is a host of parental biographies.

The most copious domain of printed autism narrative is fiction. Haddon's (2003) The curious incident of the dog in the night-time is only the tip of a book mountain. Reviewing the book, Charlotte Moore wrote that ‘Autistic people are not easy subjects for novelists. Their interests are prescribed, their experiences static, their interaction with others limited.’ (Moore 2003.) In my loaded terminology, these words imply that George and Sam were thin boys, destined to grow up into thin men. This is in no way intended as a criticism of her parenting; instead, it should serve as a reality check.

In an interview a year after Moore's review, Haddon (2004) implied complete agreement, but ironically reversed its connotations. The lives of autistic people are so boring on the face of it, he said, that he modelled himself on Jane Austen, describing the lives of equally boring people, and making them fascinating. He thereby turns some preconceptions upside down. Little could he have known about the amazing subgenre of retroactive autism narrative that was waiting in the wings. No fewer than eight characters of Pride and Prejudice have now been diagnosed, with proud Darcy the prime person with Asperger's or high-functioning autism (Bottomer 2007).

Jane Austen's personages are truly thick, yet, as Haddon playfully insists, most of us would be bored stiff in their socially assigned roles. There is a tension here, for his character Chris is much more high-functioning than George and Sam were said to be. This leads back to question (ii): Should we conceive of George and Sam as living far more complex and indeed more ‘interesting’ lives than we get on first impression after reading Moore's book? I am deliberately using that book here, because it is the parental biography that I, personally, most admire and am most affected by. Thus I am posing here a question that I find both hard and uncomfortable.

Haddon's tale has become a staple textbook in teacher-training courses, in the unit dedicated to working with children with special needs. Thus the well-informed model furnished by an established writer of young adult fiction is transmitted to a generation of teachers and in turn to their charges. The children may then learn that that is how it is to be autistic. Draaisma (2008) has an important analysis of related phenomena, using Haddon's The curious incident of the dog in the night-time among others as a key example.

My own list of autistic fictions has approximately 200 books, and counting. Especially important are the stories for children, some for autistic children, some for non-autistic children and some for all children. They take great pains to describe behaviour and thereby serve as role models. The young adult category is also influential in determining what autism is like and in some sense should be like. They illustrate ‘the norm’ and thereby help stabilize what the norm should be. Detective stories, spy thrillers, science fiction, gothic horror and Harlequin romances all help to reinforce a way of talking about autism.

The story-tellers learn from autobiographies how to tell their tales. But that is a two-way street. Temple Grandin's Emergence was written before the genre got underway, so her self-descriptions are unaffected. Today's autistic child, brought up on children's stories about autistic children, and who in later years goes on to write an autobiography, will give accounts that are textured by the early exposure to role models.

6. ‘Inside’ autism

In his foreword to Grandin (2005), Oliver Sacks wrote that her previous book, Emergence, was

Unprecedented because there had never before been an ‘inside narrative’ of autism; unthinkable because it had been medical dogma for forty years or more that there was no ‘inside’, no inner life, in the autistic. (Sacks, foreword in Grandin 2005, p. 11)

Even before we dip into these books, we find that word ‘inside’ over and over again—on their covers. On the back cover of a current paperback of Grandin's (2005) Emergence:

A remarkable story … uniquely valuable in helping us to see autism from the ‘inside’.

A quotation from People magazine on a paperback of Williams' Nobody Nowhere:

By turns fascinating and harrowing…a riveting autobiography that describes how autism feels like from the inside.

The subtitle of Mukhopadhyay's (2008) is:

Inside my Autistic Mind.

The subtitle of the American edition of Tammet (2006) is:

Inside the Extraordinary Mind of an Autistic Savant.

These examples can be repeated many times without even opening books.

7. Unexceptionable uses of the inside metaphor

Sacks put quotation marks—which I take to be ‘scare quotes’—around ‘inside narrative’, and, a line later on, around ‘inside’. His point was that Grandin's first book was astonishing, because no one familiar with autistic people expected such a text. It is an ‘inside story’ in the simple sense that it is an autistic person telling her own story, and not a clinician or parent writing about her. Sacks also meant that before Grandin, someone who had come across autism in only a casual way might have thought that autistic people had nothing like we would countenance as a thick mental life.

In a puff for Mukhopadhyay's (2003) Sacks said that the book,

is indeed amazing, shocking too, for it has usually been assumed that deeply autistic people are scarcely capable of introspection or deep thought, let alone of poetic or metaphoric leaps of the imagination […].

To use a phrase of Sacks in the same blurb, Mukhopadhyay ‘gives the lie’ to the doctrine that autistic thinking is necessarily literal, and that it resists metaphor. Moreover, at an early age, he seems to have had a great aptitude for storytelling and imaginative play. A primary item in Lorna Wing's ‘triad’ is impairment of imagination. At the age of 5, he is making up stories, starting with one about a lost goat (2003, p. 36). By the age of 8, he well conceptualizes imagination and fantasy. Referring to a time after his grandfather died when he was 4 years old, he wrote: ‘Imagination took shape to lead his mind to a world of fantasies’ (2003, p. 19).

8. ‘Inside the mind’

When ‘inside’ connotes written or spoken by a person with autism, and ‘outside’ connotes written or spoken by an observer, parent, clinician or friend, then the metaphor is benign. But it is also, once the point has been made, rather banal, and hardly worth the constant repetition we have encountered.

Aside from such benign uses, I am cautious about ‘inside the mind’, for reasons presented in Wittgenstein's Philosophical Investigations. This is not the occasion to argue the case or even sketch what is at issue, but it does motivate my approach. It is certainly not a hankering after behaviourism.

A first danger of the ‘inside’ metaphor needs only to be stated to be scotched. It is the idea of ‘a unique insight into the autistic mind’: as if ‘the autistic mind’ were a species of mind. Our four autists have very different minds! Grandin describes herself as thinking in pictures. Mukhopadhyay is dominated by sounds. Tammet sees abstract objects in colour. Hence Williams’ (2005) metaphor of autistic spectrum ‘fruit salads’. To quote a common adage: ‘If you know one autistic person, you know one autistic person’.

Talk of getting inside the mind of another is apt in its place, but unreflective use out of context often suggests a misleading picture of mental life. It suggests that looking inside is much like looking outside, in the way that looking inside a cardboard box after opening it is no different from looking at it from the outside, except for a change in point of view. This goes along with the sense that our own minds are transparent to us (subject to Freudian reservations). We look inside them all the time. That generates the question of how we ever know what is going on in the mind of another person.

This ‘Problem of Other Minds’ has been a topic in English-language philosophy for well over a century. It is proposed that we know what other people are feeling and thinking (etc.) by analogy with our own case. Or we postulate a mind similar to ours as the best possible explanation of the actions of another person. In any event, we must always infer what someone else is thinking (etc.) from their behaviour. That is nonsense. We often know what someone thinks via, or thanks to, their behaviour, but without any need to go through some process of inference. In fact, the emphasis on thinking is misleading; we should think first of knowing what another person is doing, i.e. of immediate recognition of intentions.

Of course sometimes we do have to infer what another person must feel, hope, want, detest, or whatever, from their actions. There are innumerable occasions when we are completely baffled as to what someone else is doing. We have to figure out their intentions and their hidden worries. There are also situations where inference is required, for example the famous false belief tests of Wimmer & Perner (1983).

There can always arise particular difficulties in understanding another person. But the puff writers who talk of getting into the mind of an autistic person do not, for a moment, think that there is the same general question for most people, as there is for autists. I cannot recall an ‘inside the mind’ (or variant thereof) being written on the cover of any non-autistic autobiography that I have examined lately. It could certainly occur, of course, in connection with a biography. ‘Finally we have got inside the inscrutable mind of Vladimir Putin.’ But that is not the norm.

The ‘Theory of Mind’ approach to the ability to understand what other people believe, hope for and are doing has the great advantage, over philosophers’ theories of ‘Other Minds', that it does not imagine that we infer, for example, by analogy, that other people have minds. But it is still preoccupied by inference, thanks to ties with the false belief tests. I cannot ‘see’ that Sally will think the box has Smarties in it, not pencils; I have to infer it from knowledge about what she knows and does not know. That is not a core case of interpersonal relationships. Theory of Mind approaches to autism, driven by false belief tests, focus on complex situations that are parasitic on the bedrock cases of simply knowing what someone else is doing (which may include thinking, feeling, plotting and so forth).

The very label ‘Theory’ has the unfortunate connotation of always getting to the destination by reasoning. More importantly, it does not easily distinguish between, on the one hand, seeing what someone is doing right off and, on the other, inferring or working it out from clues. To say that there is a distinction is not to quarrel with someone who insists that there is always a ‘computation’ leading up to knowledge; it is only to conjecture that, within that paradigm, different types of computation must be involved. Two earlier writers thought that such a distinction was profoundly important.

9. Wittgenstein and Gestalt psychology2

Wittgenstein's aphorism (2001, p. 152e) captures an important fact about other minds:

The human body is the best picture of the human soul.

This and related remarks were foreshadowed in the first edition of Köhler's Gestalt Psychology (1929). Wittgenstein devoted some of his classes to matters arising from that book. They bore fruit in Part II of the Philosophical Investigations. For example, one of Köhler's observations (p. 250) anticipates the aphorism just quoted. It explicitly takes us back to the inner:

[…] not only the so-called expressive movements but also the practical behavior of human beings is a good picture of their inner life, in a great many cases.

Both the philosopher and the psychologist use the metaphor of a picture. We usually see what a picture is of, and do not infer it. I look (for example) at Goya's etching, titled The pitcher's broken. I see a child howling; he is close to a broken jug, and he is being ruthlessly spanked with a slipper by a woman who had been hanging out her laundry.3 I infer from the ironic caption that the child broke the pitcher and that this is Goya's protest against routine cruelty to children, under the guise of fair punishment. But I do not infer what the etching is a picture of.

Likewise I usually just see that a man is in bad humour (Wittgenstein 2001, p. 153e). I note it, and do not infer it. (Of course there are cases when I do have to infer it! But those are parasitic upon cases where I do not.) A kindly boss is upset because he has to reprimand an employee: ‘Viewed from without the official's activity is a picture of his inner perturbation’ (Köhler, p. 254). We see that he is upset. We do not infer it, in the common case, from the way he looks and acts. (Which is not to deny that sometimes we have to infer, guess or divine that he is upset.) Köhler has many more illustrations. They include that of following a person's gaze (p. 250f):

If my attention is attracted by a strange object, a snake for instance, I feel directed toward it and at the same time a feeling of tension is experienced. A friend, even if he has not recognized the snake, will see me and especially my face and eyes directed toward it; in the tension of my face he will have a visual picture of my inner tension, as in its direction he has a direct picture of the direction which I experience.

Wittgenstein was also much interested in pointing and less explicit ways of directing attention.

Another example from both Köhler (p. 252) and Wittgenstein (1980, § 1066) is that of a child, reaching out to touch an animal, but not daring to do so. We see what the child wants to do, and also see that that's a bit too scary. The example originates with Watson's (1926, p. 52) experiment with a child and a white rat, to show something about the transfer of fear reactions. The child becomes scared of anything woolly. Köhler cites but refers only to an ‘animal’, not saying what kind. That made Wittgenstein think of a child and a dog.

Köhler admitted at once, in 1929, that his account ‘gives us neither an altogether new nor an altogether perfect key to another person's inner life; it tries only to describe so far as it can that kind of understanding which is the common property and practice of mankind’ (p. 266, italics added). He hoped for future work ‘when the simpler facts described in this chapter will have found more general acknowledgement’ (p. 267). Enthusiasts for Theory of Mind may be inclined to say that Köhler was describing, in naive terms, just what it is to have a Theory of Mind. In fact, ‘naive’ is a word that Köhler often uses himself with cautious approval, as something more true to experience than what psychologists of his day were saying. Whether or not one should repeat that comparison with today's psychologists, I find it a virtue that his descriptions are not at the level of propositional knowledge that so often comes to the fore in cognitive science. Curiously, both Köhler's approach to these phenomena, and Theory of Mind, originated in studies of the great apes.4

10. Köhler's phenomena, and their absence

Köhler pointed to a wide range of phenomena in which we see and do not infer what a person is doing. Among these are the examples cited: seeing that the child wants to touch the dog, but does not dare; seeing that the friend is startled by something; seeing where that something must be; and seeing that a man is upset by a disagreeable task. Let us call these Köhler's phenomena. You can think of innumerable examples.

Perhaps the future work, for which Köhler hoped, explaining his phenomena, will prove to be research on mirror neurons. They certainly do provide an explanation sketch of how his phenomena might be possible. And yet it is somehow too tidy. A sceptic could suggest that causation runs the opposite way. Electrical impulses reach the mirror neurons precisely because the person has seen that a man is upset, rather than the person seeing that the man is upset because a mirror neuron has been activated.

I shall stick to the phenomena. They are familiar to most people, but are precisely what are not familiar, ‘automatic’, ‘immediate’ or ‘instinctive’ for most autistic people. They are not ‘the common property and practice’ of that part of mankind that is autistic. Expert observers report that autistic children do not see that someone is in a bad humour; they do not follow the direction of a startled person's gaze; they do not readily understand what another person is doing, i.e. they do not easily recognize intentions.

Conversely, most people cannot see, via the behaviour of severely autistic people, what they feel, want or are thinking. Even more disturbing is an inability to see what they are doing: their intentions make no sense. With the severely autistic, it may seem as if they do not even have many intentions. They are taken to be, to repeat my metaphor, thin children who grow up to be thin men and women, lacking a thick emotional life. Or so it has seemed to most people, including many parents and many clinicians. Since Köhler's phenomena do not take place, when most people are in the company of a severely autistic person, we have to compensate for their absence. This is what autism narrative helps to do.

There is a partial symmetry between the autistic and the non-autistic. Neither can see what the other is doing. The symmetry is only partial because we have an age-old language for describing what the non-autistic are feeling, thinking and so on, but are only creating one for the autistic.

Precisely because autistic children do not share in Köhler's phenomena, it is now common practice to try to teach them how to infer the feelings and intentions of other children and adults from their behaviour, from their gestures, from their tone of voice. There are indeed a number of books, posters and videos intended to teach what many people look like when they are happy or sad.

Conversely, ordinary people cannot see what an autistic boy is doing when, to take a banal example, he is furiously flapping his hands. What on Earth is hand flapping? The parent or other outsider knows vaguely that there must be some kind of agitation, yet the child seems so tranquil when hand flapping. Autobiographies tell us how calming it is. So we are now able to infer a bit of what's going on.

Autistic narrative thus comes to our aid. It is striking that although we are told that it takes us into the mind of the autist, in fact autobiographers usually begin with their behaviour, reaching back to childhood. Grandin begins with herself as a 3-year old, having a tantrum and thereby causing her mother to total the car in which they were driving. Then we are told that Temple hated the horrid hat she was required to wear. Thus actions and behaviour are put together not only with common words for emotions, often of fear, shock, assault by the senses, but also with a feeling of peace, of getting to a quiet place. All this works by using ordinary language. Yet at the same time, the autobiographers have to retool linguistic materials made in an age-long community. Köhler's phenomena are the bedrock upon which rests the common understanding of those words. But the phenomena are lacking when we try to empathize with the autistic author. So Grandin and her successors tell stories to help us infer from autistic behaviour the words she would now use to say what is going on.

11. Hypersensitivity

Our autobiographers insist that certain sensations cause huge distress to the author. They give examples to demonstrate they are sensitive to the point of anguish by too bright a light, too loud a sound, too scratchy a surface or other touch; when sensations become overwhelming, this leads to virtual collapse, screams, etc. The too much is ‘painful’. These accounts are strikingly consistent and backed by parental observation. Possibly they are connected to remarkable sensory acuity, such as ‘eagle-eyed’ vision and the like now being studied in the laboratory as reported by Baron-Cohen et al. (2009).

‘Acuity’ sounds neutral, but too much sensation is, for many autists, unbearable, and seems to fit into the category of pain. But the fit is loose; we do not know quite what the words should be. We should listen carefully to the ordinary language of pain, and then note how people with autism try to adapt it to their own experience. In this respect, it may be useful to take seriously Wittgenstein's remarks on the public language of pain in order to cast away the stereotypical—I want to say mechanical—conceptions of pain that so often blind theory to experience. We want something subtler than thresholds, acuity and so forth.

12. A role for autism narrative

Autism autobiography, and autism narrative more generally, is thus playing a remarkable role in the evolution of the autistic spectrum. It is enabling us to try to compensate for the lack of Köhler's phenomena in our interaction with autistic people. The various regimes that help autistic people learn to understand most other people compensate in one direction. They enable the autist to infer from neurotypical behaviour. The narratives teach many of us how to compensate in the other. That is, they suggest what to infer from autistic behaviour which on the face of it means nothing to us.

I suggested at the outset, in connection with question (i), that the autobiographies do not so much describe the mental life of their autistic authors, as constitute it by choosing words from ordinary language to be applied in connection with their behaviour. This is important for question (ii). If the autobiographies are straight descriptions, true or false according to the existing criteria, then it is a plain matter of fact whether those descriptions apply to less high-functioning people. But if we think of the descriptions as constituting autistic experience, it is less a question of fact than of the ways in which we will come to understand the less able.

13. Concluding query on language

There is no space here to do the real job, to look in detail at the ways in which our authors recall and describe their past and present. I invite you to read these texts not as describing well-defined experience, but as creating ways in which to express experiences. I shall end with a puzzle taken from one set of texts. It bears on another question, namely how human speech can be acquired by individuals who have little grasp of the ordinary community of speakers, and then be later used by them to describe their condition before they entered the community.

Our autobiographers imply and sometimes state that they understood what was said around them, long before they could speak. Grandin writes of herself at the age of 3,

Although I could understand everything people said, my responses were limited.(Grandin 2005, p. 22)

Mukhopadhyay communicates in the way developed by his mother and himself. He does not speak. Yet he too reports detailed understanding of what was said around him when he was only 3 years old. He was being interviewed by a psychologist in the presence of his mother. He was diagnosed as autistic. The expert explained that the child was so withdrawn he could not understand what is going on around him. The boy was listening, and a few years later he reported:

‘I understand very well’, said the spirit in the boy.(Mukhopadhyay 2003, p. 23)

This does not sound at all like the reading of Frith and Happé in their paper, ‘What is it like to be autistic?’ Their central thesis is that articulate autists ‘appear to arrive at an explicit theory of mind by a slow and painstaking learning process, just as they appear to arrive at self-consciousness by a long and tortuous route’. (1999, p. 2).

If we take the words of Mukhopadhyay and Grandin as straight descriptions of matters of fact, we have to ask whether the memories were reliable. We might then suppose something like them is true, but only at a later age. We would still have to ask how it is possible for these children to acquire a remarkable understanding of language without yet having participated in dialogue, in babble, and in trial and error.

But suppose we are less concerned with whether as a matter of fact the child did understand, than with a form of words that represents how the autist felt, or seems to remember feeling. If we took this point of view, we might come to judge that less gifted autistic children and adults, who communicate very little, also understand, in a quite specific way, far more than is evident to the outsider. If we were to take this route, it would be a shift, perhaps a radical one, in our conceptions of and relationships to individuals on the spectrum.