Note: ‘Encounters with Autism‘ is an essay that’s been sitting in my drawer for a couple of years. It’s formatted and styled slightly different to my normal practice here, but it seemed relevant to – you know – recent events.

In NBC’s Hannibal (2013-2015), Hugh Dancy plays the role of FBI special investigator Will Graham, an autistic bisexual with an overactive imagination. I’ve fixated on Will Graham for almost a decade. He hated eye contact. He worked to reconstruct the thinking of serial killers: his experience of empathy and social connection was nestled in the context of violence and trauma. The colours of his world were always overexposed, somehow sharper than in real life. In moments of direct trauma, the framerate was lowered, or individual frames were cut out of the edit. His movements became choppy, staccato beats, skipping from one place to the next without registering the in-between. He consumed my attention. In retrospect, I’m not sure why I wasn’t diagnosed earlier.

When I was diagnosed with autism in 2021, I decided to take the label for a spin. I wanted to see what I had missed, what was known and what support would have been available to me. I sat down to read the back-catalogue of Autism, a peer-reviewed scientific journal with a very particular focus. Autism began in 1997, publishing four issues a year. They increased to an annual six issues around the mid-noughties, and eight by 2015. Autism is a flourishing field of study. And yet, as I read, I found myself glad to have been diagnosed later in life. Growing up undiagnosed means I am accustomed to a certain level of harassment and isolation. People were cruel, routinely. But the early pages of this journal have been an education in a new form of stigma. In these pages I found my experience observed, packaged, and categorised by an extensive medical infrastructure. Autism was discussed as a burden, as an inconvenience, as a topic of study leveraged in service of career progression. It felt like my ribs were being pulled apart and put on display as researchers elbowed in with their special interests: the co-occurrence of autism and deafness, or toxic trace elements in the hair of autistic children in Kuwait. At times in these articles, the autistic person exists only by implication:

“Autism is a severely handicapping condition with associated social, communication and behavioural impairments that place extensive demands on parents, teachers and society.”[1]

This sentence recognises no person at the heart of the condition. There is mention of parents but not children, teachers but not students. The autistic person is veiled; at most they exist as an implied vehicle for autism, which is characterised as some sort of malevolent parasite, as if it hollowed out the person and wore their skin like a suit. Having a diagnosis helps me understand myself, but it also exposes me to the prejudices and stigma of the medical establishment. It makes me more of a whole to myself, and maybe less of a whole to them.

In the first season of Hannibal, Eddie Izzard plays Abel Gideon, a surgeon arrested for killing his family. Subjected to psychic driving under institutional care, Gideon is led to believe that he is the infamous Chesapeake Ripper. He escapes, and, in an attempt to reconstruct his shattered mind, begins using his surgical skills to eviscerate the psychologists who tormented him. As they used their skills upon him, so he uses his skills in return. Reading through Autism has been something of an Abel Gideon experience. I observe myself through the eyes of others, through eyes which are not always neutral and which do not always represent best practice. Of course, the rapid change in understanding means that researchers and clinicians are often playing catch-up. “Who would have ever thought,” asked Gary Mesibov in 2006, “that today’s accepted rate of Autism Spectrum Disorder could be anything approaching 1 in 166?”[2] Today’s estimated prevalence is closer to 1 in 36.[3] And although the landscape is changing fast, there remain underlying issues of mistreatment and neglect. In 2022, Michelle Dawson and Sue Fletcher-Watson published a commentary critiquing the lack of proper risk assessment in the literature on early autism interventions. In one review of 150 early autism intervention group designs, they write, a staggering 93% “did not even mention or allude to” the possibility of adverse outcomes for the autistic people involved.[4] Dawson and Fletcher-Watson document a range of lapses, low standards, and negligent reporting in the study and implementation of autism interventions, including instances that have been flagged as violating research ethics. They conclude that “autism researchers should be deeply troubled by this comprehensive failure to apply fundamental standards.”[5] In an attempt to counter these harms, researchers increasingly seek to incorporate autistic communities through “inclusive research design and participatory research,”[6] as collaborators rather than subjects. The most recent research is, increasingly, accountable for past mistakes (and the ways in which those mistakes continue to inform current harms) and searching for ways to work alongside rather than over the top of autistic communities.

At the same time, even these more progressive protocols do not blunt the experience of discovering myself as an object of study. The articles of Autism are overwhelmingly written by researchers to other researchers about autistic people as a third party. I discover myself there as a they and never a you – in Buber’s terms, as it and not thou. “Anxiety may not play a fundamental role in their difficulties, but may instead represent a secondary phenomenon,” muses one paper.[7] “They also respond differently to people than to objects,” observes another.[8] Attention is given to the way in which some autistic people struggle with pronouns, confusing ‘I’ and ‘you’,[9] but not to the operation of power as researchers deploy pronouns to ringfence their objects of study, isolating them from the general population. “These children tend to be highly task focused,” notes one article, “though their tasks may differ from ours.”[10] Overwhelmingly, the language of this research is structured under the logic of us and them. The medical framing justifies and disables, elevates and dismisses. Autistic people are objects to be studied, while agency is reserved for researchers; autistic people have characteristics while researchers are assigned tasks. Thus we read that autistic people have “difficulty … in understanding other people’s minds,”[11] while scientists carry out “research on the inner experience of people with [ASD].”[12] When I struggle to understand them it’s evidence of disability, but when they struggle to understand me it’s a research paper.

Like Abel Gideon, then, I set about reclaiming my sense of self by applying my skills to the people applying their skills to me – by no means a threat; I am not a surgeon. I am a student of literature. My training is in language. I observe the drawing up of boundaries, the linguistic separation of us and them. I study definitions. In one 2002 study, for instance, researchers Ashleigh Hillier and Lesley Allinson defined and characterised embarrassment in a way that led them to conclude that autistic people don’t understand the concept. Their study took four groups of ten children – one set with autism, one set with learning difficulties, and two sets of “typically developing children” matched respectively on “verbal mental age” (around 10) and “chronological age” (around 17).[13] The children in each set were shown six scenarios – two which the researchers defined as “embarrassing,” and four “non-embarrassing” – and asked to decide in each instance whether the protagonist felt embarrassed.[14] This a priori act of definition sits at the heart of the paper’s problem. Hillier and Allinson defined the parameters of embarrassing and non-embarrassing according to their understanding of the experience and interpreted difference as a breakdown in comprehension rather than as a marker of a different point of view.

In their introduction, the researchers acknowledge that embarrassment is a social phenomenon: “The crucial factor leading to someone experiencing embarrassment is their recognition of possible negative evaluation from others.”[15] But they seem to overlook that, as a social phenomenon, embarrassment exists in competing and often mutually exclusive forms. Parents are embarrassed by their sullen children. Teenagers are embarrassed by their uncool parents. In one of my religion papers at university, a heated debate broke out over women’s rights: a mature student in his mid-40s pointed to the first-year girl sitting next to him and declared “Some women want to be submissive. Look at her. She has decided not to say anything for our group, to submit as I speak.” Perspectives on embarrassment change across age and gender, across class and culture. To define embarrassing and non-embarrassing scenarios in advance is to beg the question. It is to begin with a conclusion and pathologise difference. Hillier and Allinson therefore struggle to understand embarrassment outside of the metric of right and wrong. The results of the study showed that the autistic participants matched “typically developing” groups on identifying embarrassing situations,[16] and that autistic participants rated the non-embarrassing scenarios as “significantly more embarrassing” than the “typically developing comparison groups.”[17] Hillier and Allison, leaning on their pre-existing definition of embarrassment, take this deviance as proof that autistic people have “difficulties in understanding embarrassment.”[18] But as an autistic person, I know that certain situations can be experienced as stressful or embarrassing even if they appear innocuous to non-autistic viewers. I personally struggle with haircuts. I do not understand the language or the social conventions, and I am bewildered by the way they expect me to know what I want. Getting my hair cut is distressing, not only as a physical experience, but also as a source of potential embarrassment. It is very difficult to explain the intricacies of neurodiverse experience to a bleach-blond surfer while you’re wearing one of those circus tent things they put you in. We cannot assume that these autistic participants saw something in the non-embarrassing scenarios that caused them to register stress or embarrassment; it might be that they made genuine mistakes, that they did have some level of difficulty with the actual concept. Because of the way the experiment was structured, we simply do not know. The pre-existing delineation of embarrassing and non-embarrassing scenarios has inhibited curiosity about why different people might identify embarrassment in specific situations.

Similar problems of definition emerge in a 2001 article on autistic humour. Authors Abigail Werth, Michael Perkins, and Jill Boucher studied a woman named Grace White, who they suggest may be “a humorist savant.”[19] Throughout the article, the researchers implicitly wrestle with the question of what constitutes humour. Is Grace funny if she intends to be funny, or is she only funny if her audience laughs? The researchers do not address this question directly, leaving their work to oscillate between the two positions. Terms and categories consequently begin to double over. Jokes are recorded indiscriminately upon meeting either threshold: “either because they struck us, or members of Grace’s family, as funny, or because Grace herself signalled that she was intending to be funny.”[20] The article alternately describes Grace as making both jokes and ‘jokes’, in quote marks, as if uncertain whether the term should be allowed its full meaning.[21] In one fascinating instance, the authors dismiss one of Grace’s jokes as unsuccessful due to the fact that only autistic people were laughing. The recorded joke apparently relates to Grace’s special interest: “A broken seatbelt that got mended!” We are told that “Grace’s friend, who also has autism, was present when this latter ‘joke’ was told and both laughed hysterically at it.”[22] In theory, this joke meets the threshold on both sides of our philosophical question. Grace clearly intended to be funny, and her autistic audience found the joke hilarious. And yet the researchers describe the joke as unsuccessful. In the moment, they demonstrate some hesitance around this slippage, writing that Grace’s jokey references to her obsessions are “less successful, at least to non-autistic people.”[23] However, a few pages later, they comfortably dismiss the “failed ‘seatbelt’ joke” in its entirety.[24] The example seems to indicate that these researchers only really consider jokes to be funny when they are funny to non-autistic listeners (or in their words, “normal listeners”).[25] On seeing two autistic people laughing, Werth et al do not exhibit curiosity as to whether their own framework of understanding might be insufficient. Instead, they assess the joke against a foreign set of standards and write it off as a failure. It is often observed that autistic people engage in narrow, rigid thinking; the irony of this article is that it judges an autistic person as unfunny specifically because her humour fails to adhere to a narrow, rigid framework.

In keeping with standard practice for scientific journals, both of these articles routinely deploy the passive voice. Autistic participants are variously recruited, tested, read to, instructed, and assessed, all by some unarticulated ghostly presence. The passive voice is a stylistic decision meant to indicate the repeatability of each experiment, the importance of method over the specific person or people carrying out the study. But what we observe here, I think, is that the question of who studies and who is studied, the question of identity and power, sits at the heart of autism research. Often discussions of autism and identity are reduced to debates about terminology – for instance, whether we should use person-first language (eg ‘a person with autism’) or identity-first language (eg ‘an autistic person’).[26] These debates are not without value, but, like the passive voice, they risk functioning as a veil, obscuring the far more crucial matter of power and control. In a letter to the editor in 2004, Australian researchers Mark Stokes and Naomi Newton suggest that autistic people might be prone to stalking, as (in their characterisation) autistic people are socially inept loners who seek relationships without the skills to maintain or initiate them. “Notably,” they intone, “over the last decade dramatic increases have been recorded in both the incidence of stalking … and the frequency of ASD diagnosis.”[27] Would this bizarre, hateful correlation really be improved by person-first language? The problem is not as simple as adjectives and nouns: it is about our access to the halls of power. It is about research that happens in dialogue – not a one-way practice of observer and observed, but a reciprocal process of exchange. Researchers must speak in anticipation of a reply. They must observe and be observed.

In the first years of Autism, a section was reserved for first-person accounts of autistic experience, “as a response to the criticism that research on autism needs to hear ‘the autistic voice’.”[28] The autistic voice is imagined here as a view from the inside, where an autistic person is permitted to stand alongside researchers and gaze reflexively on the phenomenon of autism. The basic dynamic of observer and observed is retained, but autistic people are invited to join researchers in their position of observational privilege. This framing is insufficient. The autistic voice must exist as challenge, as confrontation, as a second source of agency and activity with overlapping but not identical goals. That is the only position from which we can transcend this predatory, violent binary of agent and object.

---

[1] Lindsey Kent, David Perry and Joanne Evans. (1998) Autism in Down’s syndrome: Three case reports. Autism 2 (3), 259

[2] Gary Mesibov, Joaquin Fuentes, Margot Prior and Lorna Wing. (2006) The past decade. Autism 10 (1), 7

[3] https://www.cdc.gov/ncbddd/autism/data.html

[4] Michelle Dawson and Sue Fletcher-Watson. (2022) When autism researchers disregard harms: A commentary. Autism 26 (2), 564

[5] When autism researchers disregard harms, 565

[6] Rebecca Poulsen, Charlotte Brownlow, Wenn Lawson, and Elizabeth Pellicano. (2022) Meaningful research for autistic people? Ask autistics! Autism 26 (1), 4

[7] Alinda Gillott, Fred Furniss, and Ann Walter. (2001) Anxiety in high-functioning children with autism. Autism 5 (3), 278

[8] Giorgio Celani. (2002) Human Beings, Animals, and Inanimate Objects. Autism 6 (1), 94

[9] Ilse Noens and Ina van Berckelaer-Onnes. (2004) Making Sense in a Fragmentary World. Autism 8 (2), 205

[10] Dinah Murray, Mike Lesser, and Wendy Lawson. (2005) Attention, monotropism, and the diagnostic criteria for autism. Autism 9 (2), 142

[11] Lorna Wing. (1997) The History of Ideas on Autism: Legends, Myths and Reality. Autism 1 (1), 20

[12] Dougal Julian Hare. (1998) The use of cognitive behavioural therapy with people with Asperger syndrome. Autism 1 (2), 223

[13] Ashleigh Hillier and Lesley Allinson. (2002) Beyond Expectations: Autism, Understanding Embarrassment, and the Relationship with Theory of Mind. Autism 6 (3), 303

[14] Beyond Expectations, 304

[15] Beyond Expectations, 302

[16] Beyond Expectations, 305

[17] Beyond Expectations, 309

[18] Beyond Expectations, 309

[19] Abigail Werth, Michael Perkins, and Jill Boucher. (2001) ‘Here’s the Weavery Looming Up’: Verbal Humour in a Woman with High-Functioning Autism. Autism 5 (2), 113

[20] Here’s the Weavery, 114

[21] Here’s the Weavery, 117 and 119

[22] Here’s the Weavery, 119

[23] Here’s the Weavery, 119

[24] Here’s the Weavery, 121

[25] Here’s the Weavery, 120

[26] Amanda Taboas, Karla Doepke and Corinne Zimmerman. (2023) Preferences for identity-first versus person-first language in a US sample of autism stakeholders. Autism 27 (2), 566

[27] Mark Stokes and Naomi Newton. (2004) Letters to the editor: Autistic spectrum disorders and stalking. Autism 8 (3), 337

[28] Anon. (1997) Family and personal section: Personal accounts. Autism 1 (1), 111