“Peas in a pod”: Oral History Reflections on Autistic Identity in Family and Community by Late-Diagnosed Adults

Since autism was first described as a clinical condition (Kanner, 1943), it has transformed from being conceptualised as a narrowly defined, rare disorder with childhood onset to a widely recognised and heterogeneous lifelong condition (Happé & Frith, 2020; Lord et al., 2018). The rise in autism prevalence, in part prompted by an expansion of the diagnostic criteria to include those without a clinically significant delay in language, cognitive development or adaptive behavior (American Psychiatric Association (APA), 1994), has led to an identification of previously undiagnosed adults as autistic,¹ who were either misdiagnosed in childhood or not diagnosed at all. For this so-called “lost generation” of autistic adults (Lai & Baron-Cohen, 2015), a diagnosis of autism may result in profound changes not only to self-identity but also to their sense of collective identity as Autistic.² The current study explores how these adults create a sense of shared or collective autistic identity with others, and the impact this had on their lives following their diagnosis.

Previous research has described adult autism diagnosis as a process of “biographical illumination” (Tan, 2018), potentially enriching self-identity and social relationships. Scholars of this process have noted that selves are relational achievements (Davidson & Henderson, 2010), in part because identity narratives are produced in and draw on interactions with others (Parsloe, 2015). Our study, co-produced by autistic and non-autistic researchers, adopts a relational framework to investigate perceptions of shared autistic identity in the everyday contexts of both family and community.

Overall, there is very little research investigating the formation of autistic identity within families or between family members, including autistic parents of autistic children. Kanner’s (1943) foundational description of autism as a “unique syndrome” specifically drew attention to potential trait similarities, such as “obsessiveness” and a preoccupation with abstractions, between parents and their autistic children. Since then, a large literature has emerged investigating the “Broader Autism Phenotype” (BAP), that is, the presence of autistic traits in undiagnosed family members of autistic individuals (e.g., Piven et al., 1997a), with attention focused on genetic mechanisms of transmission (e.g., Gerdts & Bernier, 2011; Sucksmith et al., 2011), especially in multi-incidence families (Bernier et al., 2012).

Much of this literature has highlighted negative aspects of familial patterns indicative of autism trait heritability, such as elevated rates of social difficulties (Bolton et al., 1994; Whitehouse et al., 2010) and psychiatric problems (Wolff et al., 1988; Yirmiya & Shaked, 2005) in both first- and second-degree relatives (Ingersoll & Wainer, 2014; Micali et al., 2004). An emphasis on the potential benefits rather than burdens of shared autistic identity within families has only recently emerged in academic literature. In 1988, Ritvo and colleagues first mooted the possibility that autistic children might mature into autistic adults who have their own children. In a later paper (Ritvo et al., 1994), they briefly considered the impact of autistic parenting on familial identity, noting that some parents described how they and their children were “so much alike” (p. 153) and felt that this similarity enhanced their mutual understanding.

A series of recent studies have further explored the potential for a positive sense of shared identity between autistic parents and autistic children. A phenomenological analysis of the experiences of autistic mothers found shared diagnoses helped participants to feel closer and more connected to their autistic children, encompassing an instinctive understanding of how best to meet their needs (Dugdale et al., 2021). Other studies have similarly reported that autistic parents stress the benefits of their own experiential expertise, including heightened empathy towards, and intuitive understanding of, their autistic children (Crane et al., 2021; Heyworth et al., 2022; Winnard et al., 2022). In particular, these parents are concerned to provide support for the ongoing development of a positive autistic identity anchored in the concept of neurodiversity for their children, celebrating autism as an inseparable aspect of self (Fletcher-Randle, 2022; Kapp et al., 2013). Together, these studies highlight ways in which autism diagnosis may contribute to a revised and positive familial autistic identity.

Moving beyond the intimacy of familial contexts, we find that numerous commentators have documented the creation of a shared Autistic identity amongst autistic adults. This identity is usually expressed as part of a joint commitment to self-advocacy situated within a neurodiversity framework (Bertilsdotter Rosqvist et al., 2013; Hurlbutt & Chalmers, 2002). Some of these studies have been authored by autistic scholars (e.g., Botha et al., 2022; Dekker, 1999; Singer, 1999) writing about “the importance of feeling a connection to other ‘like-minded’ people” (Milton & Sims, 2016, p. 529) preferentially fostered in autistic spaces (organised by autistic people for autistic people), which offer mutual acceptance and empathy (Bertilsdotter Rosqvist et al., 2015; Sinclair, 2010). The majority of these studies have focused on shared Autistic identity and friendships in online forums (Bargiela et al., 2016; Brownlow et al., 2015; Davidson, 2008; Parsloe, 2015), including blogs (Seidmann, 2020) and other social media (Mazurek, 2013). Some research has also pointed to the importance of face-to-face interactions between autistic adults as contributing to shared Autistic identity, sometimes described as Autistic community (Idriss, 2021). Botha and colleagues (2022) have recently reported on “autistic community connectedness”, comprising a sense of similarity between autistic people, valued autistic friendships and a shared commitment to political goals informed by neurodiverse understandings.

Such accounts are important because they move us away from longstanding stereotypes of “inborn disturbances of affective contact” (Kanner, 1943, p. 250), which have been highly influential in informing successive iterations of diagnostic criteria as well as the perspectives of autism researchers. As Jaswal and Akhtar (2019) pointed out, the idea that autistic people have atypical social behaviour is part of the current diagnostic criteria for autism (APA, 2013) with diminished social interest remaining central to social motivation accounts of autism (e.g. Abrams et al., 2013; Chevallier et al., 2012; see Clements et al., 2018). Focusing on shared autistic identity in family and community allows us to question these narrow stereotypes, adopting a more nuanced approach recognising heterogeneity in the social abilities of autistic people (Lasgaard et al., 2010; Sinclair, 2010) as well as the potential importance of shared autistic identity to wellbeing (Botha et al., 2022; Milton & Sims, 2016). Investigating the successes of autistic sociality (Ochs & Solomon, 2010), especially between autistic people (Botha et al., 2022; Milton, 2012), suggests that processes of shared identity formation may be pivotal to developing a positive sense of Autistic identity, connecting self-identity to group identity through the sharing of similar experiences and orientations that provide a feeling of belonging “to the same tribe” (Sinclair, 2010, p. 9) or “within neurotype ease” (Crompton et al., 2020, p. 1443).

The present study contributes to a small but gradually expanding body of qualitative research investigating the experiences and perceptions of late-diagnosed autistic adults. In a previous article (Lilley et al., 2021), we analysed perceptions of self-identity among the same group of late-diagnosed autistic adults, drawing on oral history interviews. In oral history, the aim is to record recollections, preserving voices and perspectives, often of those marginalised in conventional histories (Pellicano et al., 2020). The emphasis on the interviewees’³ ownership of their own story as well as the contextualisation of events and feelings in a broader arc of life experiences situated in the specificities of a particular time and place encourages rich and deep accounts. Such an approach, we suggest, elicits detailed reflections on self and others, ideal for investigating processes of identity formation. In the following we extend our previous research focus on self-identity by asking: How do late-diagnosed autistic adults create a sense of shared autistic identity with others, and what are some of the challenges to that process?

To be included in this study, interviewees were required to (i) have been born before 1975, prior to the inclusion of autism in the Diagnostic and Statistical Manual – 3^rd edition (APA, 1980) and before there was much awareness of autism, (ii) have received an independent clinical diagnosis of an autism spectrum condition according to DSM-IV (APA, 2000) or DSM-5 (APA, 2013) in mid adulthood, after the age of 35 years; (iii) be English speaking; and (4) have spent most of their childhood and adulthood in Australia.

Twenty-six adults met these criteria, ranging in age from 45 to 72 years, with 14 identifying as female, 10 as male, one as non-binary and another “preferred not to say”. They had received their autism (n = 19) or Asperger’s (n = 7) diagnosis at, on average, 49 years of age. Some had received their diagnosis very recently (three months earlier), while others had received it 10 years prior to participation in this study (M = 3.14 years). All identified as white European ethnic background and one person also identified as Aboriginal. Most interviewees reported being well educated and currently employed (see Table 1). More than half (n = 14; 54%) reported having additional mental health and/or neurodevelopmental diagnoses, most commonly anxiety disorder (n = 12; 50%).

We identified three themes relating to shared Autistic identity (see Fig. 1). The first two, ‘Conceptualising the Autistic family’ and ‘Creating Autistic community’, refer to the ways in which interviewees construct a shared identity, building a sense of belonging. The third theme, ‘Contesting Autistic identity’, discusses disruptions to that shared identity in the form of diagnostic scepticism and rejection. Readers are advised that some of this material may be distressing with discussion of mental health issues, including negative responses to diagnostic disclosure and suicide.

Taking an oral history approach, we recorded and thematically analysed the life stories of 26 autistic adults living in Australia, all diagnosed after age 35, verbally fluent and reported at least average intellectual functioning. Our study, coproduced by autistic and non-autistic researchers, highlights the active creation of shared Autistic identity, especially within families, and the importance of Autistic community to these late-diagnosed autistic adults.

The relationship between parent and child autism diagnoses has been mainly examined in the literature in terms of the presence of a BAP within families (e.g. Rubenstein & Chawla, 2018; Sasson et al., 2013). For example, elevated autistic traits in parents have been associated with greater parenting difficulties in relation to non-autistic children (Dissanayake et al., 2020). One Australian study of 22 adults (15 mothers and 7 fathers) and their children, all with a clinically confirmed diagnosis of Asperger syndrome, found reduced parental satisfaction among these adults compared to matched control parents of typically developing children (Lau & Peterson, 2011). In contrast, however, our interviewees described the development of a positive sense of shared Autistic identity with their autistic children. For some, this sense of deep connection was thought of as a “gift” that precipitated their own autism diagnosis relatively late in life. Recognition of shared autistic modes of being between (undiagnosed) parents and their autistic children was first suggested by Ritvo and colleagues (1994). Our findings echo this suggestion as well as more recent work documenting how a child’s autism diagnosis can prompt greater learning about autism and a process of comparison between self and child, which in turn can precipitate self-identification as autistic and formal adult autism diagnosis (Huang et al., 2021a; Powell & Acker, 2016; Ryan, 2013; Stagg & Belcher, 2019).

Our interviewees stated that they work hard to be positive Autistic role models and that they saw themselves as more efficacious parents because of their understanding and acceptance of their autistic children. This sense of intimate connection based on experiential knowledge, grounded in a recognition of the full personhood of their child (Lilley, 2011) and bonds of similarity that go “beyond the normal affinity of family” (Navon & Eyal, 2016, p. 1428), has been previously raised in relation to non-autistic parents of autistic children. While there is little literature on autistic parenting, more recent studies have highlighted autistic parents’ perception of a heightened empathy and understanding of their autistic children and their support for the ongoing development of a positive sense of Autistic identity grounded in valuing neurodiversity (Crane et al., 2021; Dugdale et al., 2021; Heyworth et al., 2022). Such findings provide empirical support for Milton’s (2012) model of the “double empathy problem” positing a disjuncture between autistic and non-autistic social actors leading to difficulties in social relationships. The parents in our study felt that being autistic allowed greater insight into the needs and disposition of their autistic children. It remains to be seen whether this relatively recent development of an explicitly conceptualised positive Autistic identity shared between parents and children positively impacts wellbeing over the longer term.

Researchers have described how genetic beliefs shape parent perceptions of autism causation and prognosis (Gray, 1994; Reiff et al., 2017) but the lay identification of diverse family members as autistic has not, to our knowledge, been previously discussed. Our interviewees identified their own siblings, parents, grandparents and/or spouses as (potentially) autistic. In so doing, they mobilised a primarily genetic understanding of autism, turning the diagnostic spotlight on both natal and affinal family members. Interviewees expressed a range of views – positive and negative – about their (potentially) autistic relatives, similar to existing research with autistic young people and adults (Cooper et al., 2021; Cribb et al., 2019; MacLeod et al., 2013; Mogensen & Mason, 2015). For some of our interviewees, seeing family members as possibly autistic provided an overarching explanation of mental health difficulties including tragic events such as suicide. It is important to note here that the self-knowledge of being autistic was presented as a protective factor in the context of mental health struggles and, conversely, not knowing one is autistic as a substantial risk factor. This position broadly supports research findings that autistic identity and greater personal autism acceptance can act as a protective factor against mental health difficulties (Cage et al., 2018; Cooper et al., 2017; Corden et al., 2021).

As Rapp and Ginsburg, (2011), writing about the broader context of disability narratives, have eloquently expressed it, “a diagnosis reverberates across a genealogy, revising understandings of the family tree” (p. 387). When autistic adults identify other family members, past and present, as (potentially) autistic or identify themselves as insightful autistic parents of valued autistic children they engage in a creative process of conceptualising the Autistic family. Most clinical constructs of the BAP focus on negative attributes using the language of “deficits” and “genetic liability” (e.g., Lainhart et al., 2002; Piven et al., 1997b), especially in multi-incidence families (Bernier et al., 2012). Our interviewees also highlighted the presence of autistic traits in multiple family members but they perceived the significance of this largely in terms of a recognition of a similar sensibility that formed part of their own active construction of a positively valued shared Autistic identity.

Our interviewees also highlighted how this shared sensibility extended beyond family members to other like-minded people within the Autistic community. The process of constructing Autistic community in online spaces was first noted by Blume (1997) and has been extensively documented as providing both a safe space for meeting other autistic adults (Seidmann, 2020), developing friendships (Bargiela et al., 2016; Brownlow et al., 2015) and for claiming a positive autistic identity, centred on being different not deficient, often linked to self-advocacy movements (Brownlow & O’Dell, 2002; Parsloe, 2015). Foundational Autistic activist Jim Sinclair (2010) described the importance of both the internet and face-to-face interactions in forging “shared autistic space” in which people have a sense of “belonging to the same tribe” as they form connections and create community. More recently autistic community has been conceptualised as grounded in “within-neurotype ease” (Crompton et al., 2020) or “belongingness” allowing a sense of instant connection encouraging the expression of their authentic autistic selves.

Interviewees stressed the importance of both face-to-face and online interactions in building friendships and peer support networks with other autistic adults, sometimes linking this to a discourse of autistic normalcy (being similar to other autistic adults) (Bertilsdotter Rosqvist, 2012; Hickey et al., 2018). Crompton and colleagues (2022, p. 9) have also noted the “relational and emotional benefits of autistic-autistic interaction” in a recent study of post-diagnostic peer support in adulthood. Interactions with other autistic adults provided our interviewees with self-validation through comparison with like others (“that’s me and that’s also me”), sometimes referred to using the metaphor of belonging to a (neuro)“tribe” (see Silberman, 2015). For some, that sense of shared community was part of a continuing identification with Asperger’s, which may still function as an identity grouping even though it is no longer a diagnostic label (Botha et al., 2022; Giles, 2014). Others were more inclusive in their orientation, explicitly acknowledging the heterogeneity of the autism spectrum and a common identity for everyone diagnosed autistic based on shared ways of thinking (“the autistic mind”) and perceiving the world. Learning to advocate for Autistic rights via online forums was explicitly mentioned as a vital part of online community (Bagatell, 2010; Dekker, 1999).

Interviewees sometimes suggested that they had a “good radar” for identifying undiagnosed others as autistic. In doing so, they emphasised the value of their experience-based knowledge of autism (Pellicano et al., 2019). For most, the identification of others as autistic was perceived in positive terms and as a basis for friendship or romantic attachment. This informal lay diagnostic practice extends the potential membership of Autistic community, building a sense of connection with significant others based on empathetic recognition. This has implications for the idea that social motivation deficits are central to understanding autism (Chevallier et al, 2012). A desire to identify others as autistic, thereby enlarging Autistic community, suggests that autistic adults may sometimes be preferentially socially motivated to engage with other autistic people, whether as family members, friends or sexual partners.

Nevertheless, these active creations of social connections based on Autistic identity were not always successful or trouble free. Some of our interviewees reported that their late-in-life autism diagnosis was met by scepticism or rejection or that they sometimes questioned their own diagnostic status or the claims of others to be autistic. Autistic identity may be contested even when it is based on a formal clinical diagnosis. Interviewees suggested a number of reasons others might reject their diagnosis including i) the perception of autism as a stigmatising category (especially compared to Asperger’s disorder, removed from the DSM in 2013 (APA, 2013; see Giles, 2014)), ii) their ability to successfully “mask” or “camouflage” their autistic traits (see Cook et al., 2021; Hull et al., 2017; Lai et al., 2017; Lawson, 2020) and iii) their normative life trajectories and achievements such as marriage, raising children, postgraduate study or employment (see Atherton et al., 2021; Huang et al, 2021a). In these ways, our late-diagnosed interviewees felt they did not always match people’s pre-existing stereotypes of what an autistic person is capable of or behaves like (Leedham et al., 2020). Diagnostic scepticism or rejection by family members and partners may be partially motivated by an effort to avoid affiliate, or courtesy, stigma (Turnock et al., 2022). It may also be that a late diagnosis of autism substantially disrupts the ways in which consociates have previously thought about, and related to, an individual, making them resistant to reframing their perspective on that person. Scepticism about autistic individuals being “really” autistic, often based on a view that autism is overdiagnosed (Kapp & Ne’eman, 2020), has been widely reported in the literature, especially in relation to girls and women (Anderson, 2020; Bargiela et al., 2016; Bertilsdotter Rosqvist, 2012; Milner, 2019) and late-diagnosed adults (Crane et al., 2018; Finch et al., 2022; Huang et al., 2021b, 2022; Lewis, 2016). Diagnostic scepticism from healthcare providers about adult autism diagnosis has also been previously noted (Bargiela et al, 2016; Finch et al., 2022; Huang et al., 2021b; Jones et al, 2014) as well as autistic females being underserved by clinical criteria and diagnostic processes (Estrin et al., 2020).

Diagnostic questioning of self as autistic has been conceptualised as part of a “reflexive project of the self” (Giddens, 1992), involving “reflection, doubt, evaluation and uncertainty” (Bertilsdotter Rosqvist, 2012, p. 124). One of our interviewees described this diagnostic doubt as a form of autistic “imposter syndrome”. This diagnostic self-doubt has received limited analytic attention. Punshon and colleagues (2009, p. 274) reported that one of their late-diagnosed participants felt constantly anxious that his Asperger’s diagnosis would be revoked while Autistic activist Jim Sinclair (2010) noted that autistic heterogeneity can lead some autistic people to question their own or other people’s status as autistic. While autistic people may often reject self-understandings framed within a model of pathology (Botha et al., 2020), some of our interviewees suggested that negative framings and perceptions of autism can make them doubt their own diagnostic status as well as that of others as either too successful or too “normal” to be autistic. This internalisation of stigma sometimes coexisted with a rejection of stigmatising attitudes towards autism as individuals navigated complex feelings and reactions to late-diagnosis.

Given findings that a stronger autistic identity is related to positive self-esteem (Cooper et al., 2021), reports of both diagnostic rejection by others and diagnostic questioning in relation to self and others are concerning. Certainly, our interviewees expressed considerable distress at diagnostic rejection, especially by family members and partners. Previous research has noted the complex emotional reactions of parents (Legg et al., 2022; Raymond-Barker et al., 2018) and of non-autistic partners in neuro-mixed couples to adult autism diagnosis (Lewis, 2017a) but has not discussed instances of diagnostic rejection experienced by late-diagnosed autistic adults.

Our study contributes to the emerging literature on the complexities of autistic sociality (Ochs & Solomon, 2010), based on strong feelings of social identification with other autistic people (Maitland et al., 2021), and within-neurotype ease (Crompton et al., 2020), also described as “neurodivergent intersubjectivity” (Heasman & Gillespie, 2019). Our findings suggest that late-diagnosed Autistic identity narratives are inherently social, being produced through interaction with and reference to others, drawing on discourses that frame autism as a shared way of being in the world and that explicitly value Autistic community. Previously we have argued that, contra research suggesting autistic impairments in self-awareness, our late-diagnosed interviewees demonstrate a deep capacity for self-reflection (Lilley et al., 2021). The current analysis strengthens and extends this argument. Our interviewees highlighted the importance of shared Autistic social identity in family and community and their active role in shaping that shared identity. While these processes of shared identity building may show some similarities with the identity-based politics of other marginalised groups (Davidson & Henderson, 2010), an important difference is that, historically, autism has been thought of as primarily a social disorder, and this entrenched paradigm continues to inform clinical and scholarly views (Jaswal & Akhtar, 2019). Current diagnostic criteria for autism specify persistent deficits in social interaction exemplified by deficits in developing and maintaining relationships (APA, 2013; World Health Organization, 2019). In stark contrast to these views, our late-diagnosed interviewees, by engaging in kin- and community-based Autistic identity building, demonstrated both marked social identification with and social motivation towards others who are autistic or believed to be so. This suggests a potential disconnect between the experiences of at least some autistic adults and current diagnostic criteria relating to impairments in social interaction (see also Pellicano et al., 2021).

The late-diagnosed interviewees who generously shared their life stories for this study generally embraced their Autistic identity and saw themselves as part of a broader Autistic community. They were motivated to participate in this oral history study and felt competent to share aspects of their autobiographical journeys, in part because they were intellectually able and verbally articulate. Many explicitly aligned themselves with a neurodiversity approach. Their views may not apply to all late-diagnosed autistic adults or autistic people. Neither thematic analysis nor oral history claim to be representative or generalisable (Braun & Clarke, 2019). Rather, these qualitative methods provide nuanced accounts of particular people in a particular place and time. Given the heterogeneity of autism, and the great diversity of lived autistic experience (Dinishak, 2021), that particularity is appropriate to the study of autistic lives.

This Australian project opens up many areas for future research, including the extent to which late-diagnosed autistic adults in other cultural contexts actively build a shared Autistic identity based in family and community and whether or not that identity is, at times, contested. Given the growing phenomenon of self-diagnosis (Sarrett, 2016), in part prompted by numerous barriers to formal adult autism diagnosis, both within Australia (de Broize et al., 2021) and elsewhere (Lewis, 2017b), further research with these adults may illuminate aspects of the active creation of neurocommunities in the absence of clinical identification. Our focus on the processes of shared Autistic identity building highlights the value of participatory approaches, co-designed and co-produced by autistic and non-autistic researchers (see den Houting et al., 2021; Pellicano & den Houting, 2021), that privilege autistic perspectives and experience. Such approaches, we suggest, enable a move away from deficit-focused accounts of impaired autistic sociality towards a recognition of autistic adults as active agents creating a unique sense of family and community grounded in a positively conceptualised shared Autistic identity.