Abstract
In this qualitative study, we analyse autistic women's accounts, drawing on the notion of performativity to examine the role of diagnostic identities, gender, and the interplay between them in the storying of women's self-hood. We explore how their accounts both utilise and challenge the binaries around diagnosis and gender. We analysed data from 32 autistic women, either interview transcripts (n = 9) or their own publicly available testimonies (n = 23). We drew on thematic and discourse analysis approaches to interpret their accounts, which we organised into three themes: ‘Autistic Authenticity’, ‘Gender is Fluid’, and ‘Conflation of Autism and Gender’. We found that although women considered their ‘autistic’ selves as essential and core to their identity, they considered their gender to be fluid and socially framed (performative), but still within binary understandings of gender. We situate this work as part of the critical diagnosis literature, arguing it may be productive to disassociate the concept of autistic identity from autism neurotype and diagnosis, just as gender has been disassociated with biological sex.
Lay Abstract
Quite a few academic articles have been written about autistic women, and from the perspectives from different disciplines. Some feminists have written about how gender is ‘performed’ but autism instead of being performed is usually considered to be masked. In this study, we were interested in how self-identified autistic women described themselves both in terms of gender and autism. To do this we analysed autistic women's writing and conducted interviews. We found although women considered their ‘autistic’ selves as core; in contrast, they considered their gender to be fluid and socially framed (performative). We are arguing it may be productive to disassociate the concepts of diagnostic identity, biology/neurotype and performance, just as performativity theory casts aspects of gender identity as distinct from biological sex, and performance. Such a disassociation could widen the framework of possibilities for neurodiversity.
Introduction
‘The final blow to feminism will be autism’ declares noted life coach of autistic women, Penelope Trunk (2020). This is because, Trunk reasons, women who think like men have Asperger's syndrome (she states): they are like men because they are on the autism spectrum. Trunk uses a diagnosis, in this case, Asperger's, described as ‘an extreme male brain’ (Baron-Cohen, 2002) to account for women's’ deviance from a hegemonic version of femininity. By using a diagnostic category in this way, Trunk mobilises the biological as a causal, explanatory framework for gendered behaviour. For her, ‘normal’ girls are not great at maths, instead, a girl who is great at maths has autism; the biological, neurological makeup can explain women's gender non-conformity; highly systematic autistic thinking is an aspect of a ‘male brain’ that lacks empathy. Autism, according to this narrative, accounts for how brains can be more male or female, thus gender differences are partially reducible to an aspect of neurological make-up. Needless to say, this is contested territory, for example, see recent papers on sex differences in brains (DeCasien et al., 2022; Eliot et al., 2021). ‘Our obsession with feminism is leading to a healthcare crisis for women’ Trunk continues, resulting in women undiagnosed, with ‘misdiagnosis’ common.
Why Autism Needs a Feminist Lens
Feminist scholars have long challenged the preconception that women should act a certain way and the patriarchal norms underpinning gendered social expectations. Medical and psychiatric constructs have historically been a way to document and discount women's non-traditional gender differences as being a result of biological aberration or madness (e.g. ‘hysteria’): thus diagnosis may be used as a way to police women's behaviour (Micale, 1994).
A significant tradition within feminist thought has sought to destabilise the notion that that fe/males are fixed as a definite way of being. In her classic and updated Gender Trouble, Butler (2006) introduces the theory of performativity: instead of a biological reality, it is the performance of gender, its practices and signifiers that give rise to the illusion of its stable essence. Feminist theorists note how personality traits and aspects of appearance, gait, and so on have been traditionally considered part of the hegemonic feminine. Empathy, caring, a communicative nature, and passivity are all considered feminine traits as opposed to the masculine traits of systematising, assertiveness, and physicality.
The psychological traits and signifiers associated with being male are considered to have a higher value than those associated with being female – both a reflection and reproduction of gendered power dynamics (Schippers, 2007). Gender signifiers include conveying that appearance is the most important measure of a female's worth, assigning high value to types of dress, grooming, decoration (jewellery and make-up) and body shape, for example. Such signs provide females with a form of power – what has been termed ‘erotic capital’ – but only in the context of patriarchy (Hakim, 2010). This and other mechanisms through which gender is performed, signified, or positioned, enactment of gendered psychological traits, and so on, legitimises men's dominance over women, and, paired with highly socially valued hegemonic masculine traits such as assertiveness, physical strength and self-promotion, translates into action, for example, in harassment and discriminatory micro-interaction and institutionalised sexism. So hegemonic feminine traits act to subordinate women.
We argue that as with gender, neurodivergence and specifically autism as a category can benefit from a feminist perspective. Feminist scholarship, especially of a social constructionist orientation, engages with and challenges power dynamics – the dynamics of gendered power, and its intersections with other forms of power.
Increasingly, in the politics of neurodiversity, patients challenge medical power through their mobilised activity and lobbying around the extension or contraction, medicalisation or de-medicalisation, and nature of diagnosis, with patients’ accounts often acting to maintain a dichotomy between groups with and without health conditions (Epstein, 1995; Giles, 2014; Scott, 1990). In Butler's work, gender performance – epitomised by drag – dislocates gender from sex, emphasising its fluid and constructed character. Is such a deconstruction possible for a neurodevelopmental diagnosis? Medical sociologists view diagnosis, like gender, as required to maintain social order. Hedgecoe (2004) has written about how medical authority rests on the claim that clinicians are ‘certain’ about a diagnosis. Jutel (2021) shows that binary diagnostic frameworks are integral to the medical system and require an ‘epistemic defence’ when in doubt. In her example, the common term ‘diagnostic uncertainty’ is a defence of the diagnostic system because as it implies diagnosis is always achievable, thereby reifying the diagnostic binary as the explanatory framework for people's physical and mental ailments. Through diagnosis, medicine has gained the authority to define what is real (Whooley, 2010, 2016). Aspiring to a fluid deconstruction and reimagining of existing identity categories (Egner, 2019), neuroqueer theorists have sought to ‘subvert, disrupt, and deviate from the embodied performance of being neurocognitively “normal”’ (Walker, 2021), and to destigmatise and re-conceptualise diagnoses, and separating them from patriarchal and medical contexts (Johnson, 2021).
Autism and Gender: Previous Literature
Reading feminist and queer theory centring on gender performance is particularly interesting in the context of autism, as autistic ‘masking’, particularly autistic women's ability to mask, has been discussed in depth in the autism literature, reading gender/performativity into autism. Indeed, there is a sub-field of autism literature concerned with gendered differences and masking (Hull et al., 2020). Masking, also described as camouflaging, acting, compensating, or passing, is thought of as the way that an autistic person passes as socially competent (Miller et al., 2021). A research narrative is that women with autism are not diagnosed, or are ‘missed’, because they are able to effectively mask their autism. Autistic women who do not have any intellectual disability, it is thought, may be better at hiding their autism by imitating non-autistic social interactions (Brugha et al., 2016). Autism is diagnosed primarily in males, with an oft-quoted gender ratio of four males to every one female. This figure has been questioned with suggestions the true gender ratio in the general population is more evenly balanced; the discourse in autism literature is that clinicians (and diagnostic tools) fail to pick up females for diagnosis partly due to their ability to mask (Loomes et al., 2017). A proposed ‘female autism phenotype’ (Allely, 2018) has become enmeshed with (socially constructed) gendered behaviour. According to scales and tools to identify females with autism, the female autism phenotype includes not displaying the correct signifiers of gender, such as a preference for boys’ clothes (Russell, 2020). Gendered ‘male' behaviour has thus become counted as a sign of autism (for females). Thus, again, a biologically deterministic model of autism as brain difference is used to explain gender diversity (Moore et al., 2022). Here, invoking autism is very much a performative act, in sociological terms, the ‘sanctioned device’ afforded by autism diagnosis exempts and explains an inability to fulfil gendered expectations.
Autism is a good case study to examine the intersection of gender and diagnostic category. Since the 1990s, autism has increasingly been operated as an identity, as well as a diagnosis. The autistic identity, it is argued by disability scholars and autistic activists, is parallel to gender, ethnicity or sexuality or other category of identity (Kapp, 2020). ‘Autistic’ has been reclaimed by autistic adults as an affirmative identity allowing them to mobilise as part of the neurodiversity movement and challenge their own oppression in a similar way to ‘gay’ or ‘D/deaf’ (Dyck & Russell, 2020). Autistic adults argue autism is a fundamental part of their selfhood, not a disease or condition that should or could be cured. Autistic individuals draw attention to their subordination via disabling practices of society. Just as feminists have used the category of female to defend women's rights and create a politically mobilised movement, so autistic individuals have created the neurodiversity movement rallying around an autistic identity to lobby for change, creating an identity category of the non-autistic majority, known as ‘neurotypical’, and ‘neurodivergent’ to represent in more general terms in and out groups. Neuro-majority politics and demands to normalise to neuro-majority behaviour are resisted by the movement (Ellis, 2023). By contrast, medical texts continue to view autism primarily as a diagnostic category of neurodevelopmental difference, predominantly defined in terms of deficit, not an aspect of identity.
Moore et al. (2022) provided a systematic review of papers published on the intersection of autism and gender identities. They found that dominant autism discourses (e.g. the ‘extreme male brain’ and ‘masking’) acted to restrict gendered possibilities, and highlighted how an essentialised/medicalised construction of autism is operationalised to explain gender non-conformity and gender diversity, both by professionals and autistic individuals themselves. The thematic synthesis of the reviewed papers further outlined the frequent positioning of intersectional gendered autistic identities as ‘other’, less acceptable and subordinate ways of being within social power hierarchies of hegemonic gender and neurotypicality. Autistic individuals reported experiencing pressure to perform normativity, or risk punishment in the form of bullying and discrimination, and the painful impact of this in terms of self-blame and distress (Barnett, 2017; Coleman-Smith et al., 2020; Hillier et al., 2020). Participants in the reviewed studies also expressed the possibilities that their intersectional autistic and gender identities offered for finding belonging and community and resisting oppressive social norms. Similar themes were found by Seers and Hogg (2023), who explored from a feminist disability framework the intersection of gender, autism, difference, and marginalisation in the experiences of a sample of late-diagnosed women. Their findings further highlighted the pressures to perform normativity through masking and the punitive consequences of doing gender ‘wrong’ (Butler, 2006), detailing the toll psychologically and emotionally of masking.
Aims
We wished to develop stronger links between intersectional, feminist and neurodiversity theory and the lived experiences of autistic women. Drawing on Hacking's ideas in Making Up People (1990): a woman who has a late diagnosis is a new type of autistic person – one that has not been identified as autistic in previous generations. We therefore analysed a sample of autistic women's accounts drawing on our understanding of performativity, to explore how they experienced and operationalised the idea of masking, and whether binary or fluid notions of identity were adopted, both in terms of autism and of gender. Our research question was: what is the performative function of autistic women's’ speech and writings about their autism, their gender, and diagnosis?
Methods
Design
To address this question, we adopted a discourse analytic approach drawing on thematic analysis which was considered the best option to examine the function of talk and text. To maximise the type of data that was available to be analysed we (a) conducted interviews with nine autistic women and (b) in tandem analysed publicly available testimonies written by 23 autistic women. This allowed us to strike a balance between exploring accounts (a) where we asked more specific questions and (b) accounts that were unprompted by our questions.
Recruitment/Sample Description
Semi-Structured Interviews (I)
The study initially recruited nine UK-based women who had received a diagnosis of autism. The women who were interviewed were recruited via a Twitter appeal through previous contacts involved in a wider Social Science autism project. The appeal was to women diagnosed late in life as later diagnosis is associated with greater intellectual ability (Hosozawa et al., 2020). According to autism literature, the ability to compensate for autistic traits (masking/passing, etc.) is thought to be particularly apparent in women of typical and above-average ability (Brugha et al., 2016) and masking and performance was a key theme in our research question.
In the interviewed group, ages ranged from 28 to 50 years old and autism diagnoses were given when women were between 26 and 48 years old. Most were very high educational achievers; six either had or were conducting postgraduate qualifications. Eight of the nine were living with long-term partners. Table 1 shows the demographic information available for participants. The average interview length was just under one hour, ranging from 45 minutes to over 2 hours in duration. The initial topic guide is available in the Supplemental Appendix.
Demographics of Primary Interview Data (I) and Secondary Data (SD) Participants.
Secondary Data: First-Person Accounts (SD)
We also coded and analysed data from 23 autistic women whose accounts were in publicly available secondary sources published in commercial media. Most of these secondary accounts were from the UK, with six from the US. They ranged from 2000 word media articles to 4650 word first person autobiographical accounts. In eleven of these accounts, the age of diagnosis was specified, ranging from age 22 to 59, providing a similar demographic of late-diagnosed women. Again, the sample included many high achievers with six women educated to postgraduate level. Of those who reported it, all were married or lived with their own families. All demographic information about the authors of the public testimonies that were provided is included in Table 1.
Data Collection
The primary data in this study was collected using two semi-structured interview schedules at two time points by two researchers, following the topic guides (see Supplemental Appendix). The initial interaction was conducted via email and the participants were offered the interview either face to face in a location of their choosing (usually the participants home), over the phone or video call, or via email. All were recorded and transcribed verbatim. After initial analysis, a second set of interviews was conducted by email. Based upon additional themes emerging from the secondary data analysis, the original participants answered some follow-up questions to explore emergent themes. The second round of interviews took place after some initial coding and analysis. The aim was to provide iterative analysis – to back-check themes that had been identified with the participants.
The secondary data was chosen based on being collected in an interview style or first-person accounts, and exploring adult woman's lived experience of autism, and the experience of being diagnosed as an adult. All the accounts were written by the women themselves in the first person. Publication platforms included a public service broadcaster and medical newsletters, and autobiographical essay collections. Ethical protocol precludes us from naming the sources precisely.
Analysis
The data were initially analysed loosely based on a discourse analysis which used some of the tools of thematic analysis: coding, grouping into sub-themes, as a starting point, and themes were used to group our discourse interpretation. Identified discourses were refined in the second phase of analysis, which took place after themes and codes were initially identified, where a second researcher drew on this discursive approach to draw out further meanings from the data. A discursive approach considers discourse as social action, and considers how reports are constructed by way of discursive techniques (Edwards & Potter, 1992; Wiggins, 2016). After this, extracts were reported to represent the thematic construct underpinning the discursive interpretation.
Ethical approval for this study was granted by the School of Social Sciences Ethics Committee at the University of Exeter. Identifying features were removed from data on transcription and pseudonyms were used to preserve anonymity.
Authors’ Positionality
The authors all identify as women, and we include feminists and critical neurodiversity/diagnosis researchers interested in promoting fluidity and troubling the binary. Of course, the analysis is itself positioned by our own experiences and understandings. As a team, we also include persons who consider themselves neurodivergent as well as those with clinical training and expertise, and from sociological and anthropological disciplines. We aimed for a rich interdisciplinary approach to provide multiple viewpoints with the need to navigate tensions.
Findings
We identified three overarching themes. The first overarching theme, we named ‘Autistic Authenticity’, which comprised a sense of the ‘real’, often coupled with a description of an autistic identity. This contrasted with ‘Gender is Fluid’ an understanding of gender as a constructed attribute, the essentialisation of which these women resisted. A final theme was the intersection of these two categories: we called this ‘Conflation’.
Theme 1: Autistic Authenticity
The women often described the act of masking, acting, camouflaging, or passing as neurotypical. It was in opposition to this enacted and effortful persona that the idea of authenticity, a true authentic, autistic, self was defined. ‘I realize how much time and effort we spent on passing for normal’, one interviewee; Jodie, commented. Passing thus revealed a flip-side, that there was a real autistic identity underneath, considered to be authentic. Hence a sense of true self was narrated in juxtaposition to the performance/enactment of passing. Many of the women cleaved strongly to this sense that an autistic identity represented their identity; an aspect of themselves in line with their ‘true’ self, their core essence. That self was innate to who they were: ‘I always felt different, and I was different’ (Jo), ‘being one's self’ (Lisa) or ‘the real thing’ (Zin), as they variously put it. This was in contrast to and revealed by the opposing ‘false self’ (Zin) that had to be enacted in order to suit social requirements: I know I often appear to be performing social functions … I have learned to act out certain rituals of socializing … hoping … pantomime would pass as real. Stacey, Secondary data (SD) … put on a personality I’ve constructed, as though rehearsing for a play, then getting up on stage Lisa, (SD)
Unsurprisingly given the recruitment methods and use of secondary data from collections that were specifically concerned with autistic identity, the notion of ‘Authenticity’ was associated with an autistic identity: ‘the single most important thing that's ever happened in my life was for someone to say to me, you're right, you are autistic, now go out in the world and be the best autistic you can be’ (Bron). The ‘trueness’ of this identity was stated through describing a lack of needing to act/perform, and instead to just exist/be as you are-according to Lisa: ‘I like being Aspie. My encounters with the world fit a model I understand’. Women recounted a process of actively seeking the authentic self-being so hidden under layers of falseness and performance that it could be hard to find. Conversely being authentic meant paying a price: Being myself means always explaining, always being judged, always being different, never fitting in. Tori, interview (I)
In tandem, there was a feeling of relief or even ‘seeing the light’ at finding the autistic identity which appeared to be akin to religious conversion for some. In this case, for the women, the idea of an autistic identity was embraced as their true nature or their essence. It was revealed by coming into contact with other members of the autistic community: Nothing explained me so completely and so well as what was written by those on the autism spectrum. Margaret, (SD) I feel I am seeing a miracle in action as we discover in others words and descriptions (sometimes just chance comments) insights that illuminate vast vistas of comprehension that have remained so long in darkness for so many of us. Beccy, (SD)
Furthermore, the authentic identity was equated with the physical, neurological or material through hard-wiring computer analogies, or talk of operating systems, and genetics which effectively essentialise the condition (Dar-Nimrod & Heine, 2011). Descriptions of autistic differences as differences in ‘wiring’ were rife. These solidified and stabilised the idea of the autistic difference; the ‘hard-wire’ metaphor de-emphasising developmental aspects, the possibility that autism can change through the life-course; de-accentuating aspects which come into being only through relational interaction; and de-emphasising the historical contingency of the category as if it existed independent of normalising pressures.
To summarise, in the case of women with an autistic identity, it was in the act of masking, for the effort of passing, that an underlying identity that was suppressed was revealed as authentic. Thus, the existence of the true authentic nature was a product of social demands to pass as allistic (non-autistic).
Theme 2: Gender is Fluid
For many women in this sample, there was a sense that aspects of their gender did not fit in with their female sex, they were either ‘genderless’ and/or they felt more comfortable in the company of boys. I was rather boyish and did better with boy games. I never could relate to other females, as a child or a teen … So much of what they say and do is foreign to me … I loved to climb trees and do other boy things. Patty (SD)
This was not because the women viewed themselves as masculine, or male, rather, it had led them to problematise the very categories of masculine and feminine, male and female. To most of them, gender identity was entangled with performance, shaped by societal ideals and differential treatment of which of the two binary genders one belongs. In some of the autistic women's accounts, male peers overtly painted female gender as subordinate, their descriptions dripping with male superiority: My gender came in question at that time – the boys would say ‘You aren't like other girls. You don't cry when you get hurt, so you are better than other girls, but you aren't a boy’. Hilda (SD)
For the majority of the women, the expectation to be feminine or perform the ‘correct’ gendered role played a clear role in the identity-seeking journey and their evolving notion of an authentic (autistic) self. We asked about gendered expectations; of what constituted female and male, which interviewees readily provided: Males should be strong, rather unemotional, practical, and good at things such as DIY, cars, dealing with financial issues. Females should be attractive, witty but emotional, good mothers, taking care of family and children with skills: cooking, housekeeping, nursing of ill and infirm. Anna (I) Girls are expected to be polite, well behaved, nurturing, and prosocial. Boys are expected to be tough, physical, practical. They can get away with being socially aloof if they want to without repercussions. Girls behaving like this [are] completely ostracised. Alise (I)
The qualifier ‘should’ shows how gender as an essential characteristic was questioned due to its signification of the subordination of women. For example, ‘it is “natural” I should take on a caring role for elderly sick relatives. Nothing can be further from the truth’. This resistance to the essentialising of gender often informed the basis for each woman’ sense of identity, which in some cases was described as ‘genderless’, in this sense gender was fluid and not attached to the male/female binary. However, there was some contradiction in that the binary framework was applied too: some accounts read as if gender signifiers were fixed to female/male identity: I wasn't like a girl. I didn't understand how they could be interested in clothes, make-up, dances, dates, and boys. Patty (SD)
Here, instead of an alternative way of being a girl, involving different interests and behaviours, the female or male gender is secured by signifiers. Interestingly, where this fixed gender frame was used and women described their own traits as stereotypically male, it was sometimes read as a sign of autism. There was general resentment at gendered expectations and the implications they had for not following the mould. One of the things that being a woman involves is the role of caregiver: the one who responds to needs, who nurtures, who extends affection and acceptance … Women who do not click with kids are portrayed in the literature and cinema as cold and uncaring, or even downright selfish or cruel. Yuan (I)
Because their personal profiles did not align with traditional gender expectations, this mismatch became a source of conflict during both childhood and adulthood – another aspect of life where they ‘just don't fit in’ (Lisa). The women recounted their lived experiences of being othered due to their non-conformity to conventional feminine traits and the efforts made to control or correct these differences. He said I should be laughing and gathering flowers and making wreaths and drinking wine like a real Latvian young woman. Anna (SD) All my life there has been a feeling of isolation … What's more I was blamed for this. People would say ‘If you could just enjoy the same things that other little girls enjoy, you would be much happier’. It taught me to hate my gender. Anna (SD)
Here, again, the function of the word ‘real’ is ideological in that it represents the real female, and delineates the acceptable behaviours that a real, an authentic, female may perform, and this constitutes a means of subordination. Note the mixing in of nationalism as a further appeal to control, risking exclusion from nationality unless the desired behaviours are enacted: ‘a real Latvian young woman’.
Most of the women's accounts featured uncomfortable uneasiness with, and problematising of, female gendered expectations, but one woman expressed sentiments counter to the trend, in that she felt perfectly accepting of her female identity: During puberty, I felt comfortable adopting clothes that made me appear female and this is the same now. However, this is no longer due to societal expectations, but rather, due to me liking makeup and dresses. I am not confused about my identity at all, I identify as a female and look like one too. At this moment in time, I do not think I would want to appear masculine. I like who I am. Lilian (I)
However, this quote is aligned to most of the others in that it describes gender as performative. In the above quote she acknowledges the ‘appearance’ of female, that it is adopted, enacted rather than predetermined – ‘I am content with appearing female’ she concludes, (as opposed to content with being female).
In contrasting ways, our analysis suggested whilst autism was reified as real, authentic, predetermined, and solid, gender was understood as fluid, de-natured and untethered from any biological sex that may predetermine its essence. Ironically, the women's’ discourses were underpinned by a binary male/female understanding of gendered type behaviours which they railed against.
Theme 3: Conflation
Some women's accounts cleaved strongly to another trope of the autism literature: the idea that autism is the main explanatory factor for the experience of gender dysphoria and mental health issues. Not identifying with gender and having depression (and anxiety) ARE more common among ASers but with reversed logic: AS-ness is the cause, not the result. Olga (SD)
Here it is ‘stable’ biological autistic differences that underpin other types of gender troubles which were constructed as merely a by-product of the fundamental autistic nature. This is consistent with Trunks’ (2020) account and the autism literature that casts the autistic person's lack of understanding of social roles as an explanatory device for gender troubles within a medical framing. By this account, as in much of the literature, non-traditional gendered behaviours are somewhat demeaned and invalidated as a by-product of a lack of autistic understanding of appropriate gendered role performance. They knew what they were doing. But I sure didn't. Like a ship at night with no radar, I was sailing blind through a world full of gender signals invisible to my genderless self. Anna (SD)
The quote above shows Anna opting out of gender altogether (gender doesn't apply to her as she has no ‘radar’) reinforcing the binary system: for other people rather than challenging/destabilising it. If Anna had a radar; she implies, she might successfully perform female gender. It is also unsaid but implied such a ‘radar’ is acquired by non-autistic women.
The attempts as above to fit a ‘square peg into a round hole’ (Lisa) or elicit ‘the correct’ social performance of gender, reveal how participants considered gender expectations to be ‘oppressive’. The ‘othering’ that women recounted, just not fitting in, being ‘not quite right’ (Lisa) seemed at least partially due to the expectations placed on them for ‘appropriate’ gendered behaviour.
This led women to both outing gender as a fluid performance but in tandem seeking a stable authentic identity which was counter to the notion of performance, highlighting autistic as something fixed. In a nutshell, the problem of feeling ‘othered’ was sometimes framed and located as one of an individual woman's autistic experience, rather than the problem being external, that is, located in oppressive gender or neurotypical norms and expectations.
Discussion
Substantive Findings
In this study we thematically analysed autistic women's accounts, to examine the intersections between diagnostic identities (autism/autistic) and gender identities, and how these are operationalised and accounted for in terms of the binaries around diagnosis. Drawing on the notion of performativity, we identified three themes: autistic authenticity, gender fluidity and the conflation between autism and gender. Overall gender involved, ‘Performativity’: the notion of theatre and performance characterised this dataset, descriptions of the performance of neurotypicality and in the realm of gender, with both female and male identity seen as a performance to be played, a role to be acted. ‘Autistic’ on the other hand, was characterised as a more fixed and stable and authentic entity: although ‘non-autistic’ needed to be performed.
While gender was problematised and denatured, autism remained largely untroubled and essentialised. We found although women generally considered their ‘autistic’ selves as essential and core; in contrast, they considered their gender to be fluid and socially framed and performative. Our interpretation was that whilst gender was allowed to vary, this was generally within the constraints of the binary diagnostic frame: you have autism, or you do not. Diagnostic identity was generally seen as stable or ‘true’ within the context of our dataset, although the social, repetitive, and communicative behaviours that characterise the autism diagnosis varied within individuals. Both neurotypical behaviours as performed by an autistic woman through masking, and stereotypically feminine behaviour were seen as performance. Thus, masking took place to pass for ‘normal’, or an acceptable version of the identity – as seen in previous qualitative studies (e.g. Hull et al., 2017).
Overall, our findings exposed the juxtaposition between a when a continuum and binary is applied, in terms of the categories of diagnostic identity (autistic/not autistic) and gender (male/female) vis a vis the fluidity, variability and diversity of these categories, a finding aligned to Moore et al. (2022).
Categories of biological sex, discrete gender, and specific diagnosis or diagnostic identity have constituted the stable point of reference for disability theory (Thomas & Boellstorff, 2017). We do not argue that there are no stable psychological traits, nor dismiss the autistic behaviours listed as a performative set of acts, rather, we find that by highlighting certain aspects of their experience, women made sense of their experience within the framework of culturally accessible and prescribed set of acceptable narratives about autism, authenticity, and gender.
In contrast to feminist thought where performativity creates an illusion of a stable reality underpinning and constructing gender, in the autism literature, compensation techniques are performative but mask, not create, the underlying stable reality, as the discrepancy between an internal state and external mask. This ability to mask autistic symptoms has become so accepted, that masking is considered a sign of autism in women and is used in clinical diagnostic assessment as evidence of the condition (Hayes et al., 2021). The performance neurotypicality invokes an alternative of Autism as an authentic identity to retreat to.
In Butler's theorising of performativity – the presentation of the self creates as well as reflects an impression that is inseparable from the self. There is no underlying ‘authentic self’, only a biology divorced from the performance of gender. A performance which creates the illusion of a permanent gender. In our interpretation of these data, masking was not just a reflection of autism, masking was actively used to construct the autistic identity, in a way similar to that described by Morgan (2023) in her study of how adolescent autistic girls construct self-concept and social identity.
Discussion of Key Finding 2: The Discourse that Autism Explains a Lack of Gender Conformity
Sometimes women's’ discourse constructed autism as an explanatory factor in their self-perceived lack of gender conformity. In our view, this discourse creates problematic exceptionalism similar to Trunk's argument about the death of feminism (2020): women's gender non-conformity is not valid in its own terms unless it is explained by their autism. Here, not displaying the correct signifiers of gender, has become counted as a sign of autism for females: an autism diagnosis explains an inability to fulfil gendered expectations and provides an exemption.
Autistic identity enabled the women to reject or distance themselves from what is considered ‘female’ but based on very narrow and stereotypical constructions of femininity. Paradoxically, by employing discourses that were rejecting the gender stereotype, the effect was to reduce what it means to be female to only the stereotype, reifying the binary. The impact may be to bolster autistic/neurotypical as binary constructs, but also hegemonic masculinity and femininity.
In our reflexive society, individuals’ search for the authentic self is considered paramount. The positioning of autistic authenticity can be situated in the context of this wider cultural fetishisation of the ‘authentic self’. Discursively women were framing their difficulties by utilising contemporary diagnostic categories and understandings of the authentic self (and organising identity narratives around that) which weren't available to previous generations of women. Ultimately defining a ‘true self’ as autistic enables a collective identity that is acceptable (and excuses, pardons, contextualises, etc.) non-traditional gender, for women.
Possibilities for Neurodiversity
Feminist thought, and specifically the theory of performativity, can enlighten neurodiversity studies. Perhaps through a theoretical dissociation of diagnostic identity, neurotype/biology (assumed via clinical diagnosis) and performance, interdependent dimensions of reality could be usefully applied, just as Butler dissociates gender, sex and gender performance (Butler, 2006). Perhaps autism, autistic identity, diagnosis of autism and masking/performance could be usefully separated. Such a disassociation could allow both a freer expression of gender and greater possibilities for both neurodiversity and for studies of diagnosis, without the constant demand for a binary, for ‘authenticity’ and a true explanation rooted in biological neurotype.
Ellis (2023) describes an imagined future, positioning neurotypical not as a real person or a state of being but as a word which serves a political purpose symbolic of pandering to the demands of neo-liberal society to be productive and police social expectations. Perhaps a disassociation between neurotype and identity vis-a-vis autism and other neuro-conditions could open up the possibilities for authentic neurodiversity to flourish in a more fluid, less concrete, framework, within medical and institutional settings, freeing up expectations and negating the need for the explanatory binary diagnostic framing.
Strengths and Limitations
Using more than one type of data source in qualitative research – often referred to as data triangulation – can enhance the richness and credibility of findings, but it also introduces several limitations. One key challenge is the potential for inconsistency across data sources, which can complicate analysis and interpretation. For example, interviews, observations, and document analysis may each reflect different aspects of a phenomenon, making it difficult to synthesise findings into a coherent narrative. Additionally, managing and analysing diverse data types requires is time-consuming, and strained our research timelines and resources. Finally, integrating multiple forms of data can obscure rather than clarify meanings if not handled carefully, leading to ambiguity in conclusions rather than enhanced validity.
Replicating qualitative research is often considered challenging and, in some contexts, not entirely pragmatic. This is because qualitative studies are deeply rooted in specific social, cultural, and temporal contexts, making exact replication difficult. Unlike quantitative research, which emphasises standardisation and reproducibility, qualitative research values depth, interpretation, and subjectivity. However, while exact replication of our findings may not be feasible, conceptual or theoretical replication – where similar questions are explored in different settings – can be valuable for testing the robustness of findings across contexts.
Our analysis resists the idea that autism is a harbinger of the ‘death of feminism’. Instead, we argue feminist theory of performativity can open a freer interpretation of neurodiversity, to lessen the tyranny of diagnosis (Rosenberg, 2002) and widen the ways in which binary categories are used, and that autistic, neurotypical, male, and female persons are expected to behave.
Supplemental Material
sj-docx-1-ndy-10.1177_27546330251348554 - Supplemental material for Diagnosis and Diversity: Feminism, Autistic Identity, and the Possibilities for Neurodiversity
Supplemental material, sj-docx-1-ndy-10.1177_27546330251348554 for Diagnosis and Diversity: Feminism, Autistic Identity, and the Possibilities for Neurodiversity by Ginny Russell, Isobel Moore, Shelley Norman and Jean Harrington in Neurodiversity
Footnotes
Acknowledgements
We would like to thank the women interviewees who generously gave their time to be a part of this study.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Wellcome Trust (grant number 108676/Z/15/Z).
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
