Performative acts of autism

Introduction

Since its initial construction in the 1940s (Asperger, 1944/1991; Kanner, 1943/1985) autism has been extensively researched from a variety of perspectives (e.g. psychiatric, educational, etc.), with the primary focus being on identifying, describing, and treating children who, as Kanner noted, display ‘fascinating peculiarities’ (p. 11). The prevailing models of research surrounding autism have been situated within deficit and medical models of representation (Glynne-Owen, 2010). Research has focused on targeting the (presumed) core deficits, etiology, and treatments associated with the construct of autism (Biklen et al., 2005). Even much of the research on the discursive practices of children with autism labels has focused on describing their peculiar and deficient linguistic characteristics (e.g. Hobson et al., 2010), as well as the various methods by which to correct such presumed deficiencies (e.g. Charlop-Christy and Kelso, 2003) or ‘abnormalities’. With a focus on etic and highly medicalized representations of autism, minimal research has aimed to understand autism from an emic perspective.

There is, however, a growing corpus of qualitative studies focused on, for example, the life stories of individuals with autism labels (e.g. Ashby, 2010; Ashby and Causton-Theoharis, 2009), everyday interactional practices of ‘high-functioning children with autism and Asperger syndrome’ (Ochs et al., 2004: 147), the role of narrative in the talk of children with autism and Asperger’s diagnoses (Sirota, 2010), and identity construction of individuals with autism labels (Bagatell, 2007). Although some conversation analysis studies exist examining the organization of the talk of individuals with autism labels (e.g. Stribling et al., 2007; Wootton, 1999), the majority of such studies presume that ‘autism organizes discourse’ (Ochs and Solomon, 2004: 139) rather than examine ways in which talk itself constructs the very notion of autism. Relatively little research has applied discursive approaches to understanding autism, particularly with a critical orientation to the construct of disability. Yet there is some discursive research that attends to a variety of disabilities, orienting to such constructs as situated and locally produced (e.g. O’Reilly, 2005; Rapley et al., 1998). For instance, some research has reported that for very young children with very little speech (Ochs, 1979) and adults with aphasia (Goodwin, 1995), caretakers and peers often work to collaboratively establish what the young child or adult with aphasia was attempting to communicate. In this way, the communicative capacities of the participants were often oriented to as competent by those who knew the individual best.

Drawing from a larger ethnographic study (Lester, 2011, 2012), in this article we explore the varied meanings of autism performed in and through the talk of the participants, attending closely to notions of ‘normality’ and ‘abnormality’. We report on the findings generated from the analysis of 14 interviews with parents of children with autism labels and eight interviews with the children’s physical, speech and occupational therapists. We situated this study within a discursive psychology framework (Edwards and Potter, 1993), and drew upon critical notions of disability (Thomas, 2004), post-structural understandings of discourse (Derrida, 1981; Foucault, 1971), and certain aspects of conversation analysis (Sacks, 1992).

Performances of normality/abnormality

Historically, disabilities have been constructed as biological truths, with the medicalization of bodies resulting in problems being viewed as discrete diseases that only legitimated agents (e.g. health professionals) are capable of discovering, naming, and treating (Nadesan, 2005). Within this study, we oriented to disability, and autism more particularly, as inseparable from the cultural models that define it, recognizing that: ‘The ideology of ability makes able-bodiedness compulsory, enforcing it as the baseline of almost every perception of human intention, action, and condition, and tolerating exceptions only with difficulty’ (Siebers, 2008: 103). The construct of ‘normal’ is only made possible by comparing the object of one’s study to something or someone else; that is, the embodied performance of another is ‘read’ as normal, abnormal, autistic, non-autistic in and through the process of cultural and discursive enactment. Performances, as Jaworski and Coupland (1999) noted, are ritualized processes through which people ‘construct and project desirable versions of identities’, quite often designed for ‘specific audiences’ (p. 407). In this way, rather than orienting to and/or taking up autism as a pre-existing natural entity, it can be understood as a performance, a performance that is made visible symbolically and discursively, as bodies and minds are positioned as ordered or disordered in relation to able-bodied assumptions. It is only within this comparative framework that the ‘pathological’ or ‘deviant’ body materializes.

Canguilhem (1989) suggested that ‘normal’ is always compared to that which is constructed as ‘pathological’, with the two constructs being mutually constituted. He stated that ‘every conception of pathology must be based on prior knowledge of the corresponding normal state’ (p. 51), for ‘this normal or physiological state is no longer simply a disposition which can be revealed and explained as fact, but a manifestation of an attachment to some value’ (p. 57). Therefore, through a priori standards that are situated within culturally and historically contingent values and practices, that which counts as ‘abnormal’, ‘disabled’, or ‘pathological’ is constituted and continues to be re-constituted discursively.

The 19th-century psychiatric standards and nosological frameworks (Nadesan, 2005) greatly influenced the construction of autism and the very notion of disability in general, and all that, still today, gets named ‘normal’ and ‘abnormal’. Broader psychological and educational discourses have also functioned to build interpretive frameworks that make possible the naming of certain bodily performances as normal or abnormal, autistic or non-autistic, etc. In contrast to research that assumes a deficit model and results in certain bodies being read as inherently abnormal, Rossetti et al. (2008) conducted an interpretivist study in which they explored the actions and/or performances of individuals labeled with autism during communication ‘training’ sessions with professionals. The authors challenged the view that communicative ‘oddities’ (e.g. laughing loudly in response to questions) are signs of incompetence, and suggested that such performances can be ‘read’ instead as signs of a different type of competence. Competence, then, is defined in situated and local ways, most often within a particular community of practice. Rossetti et al. (2008) reported that psychological examiners often failed to ‘read’ the performances of individuals with autism in ways that allowed for alternative interpretations as to why an individual was responding in the way that they were, e.g. with an inability to share ideas through speech. Instead, this action by those with an autism label resulted in constructions of incompetence.

Within this study, we were informed by the social relational model of disability (Thomas, 2004), viewing disabilities as always being positioned in relation to the social environment. We, like Siebers (2008), argue ‘that disabling environments produce disability in bodies and require interventions at the level of social justice’ (p. 25). Further, by assuming that a disability category is an interactional resource, we epistemologically oriented to the able-bodiedness or disabled-bodiedness as being produced by and existing in discourse (Coupland and Gwyn, 2003). Like McSwite (2001), we critique the reification of the body as located outside of discourse, and suggest instead that the body is located in discourse. In other words, the ‘world of words’ which ‘creates the worlds of things’ (Lacan, 1977: 65) includes the body.

Methods

Our study was guided by discursive psychology (DP) as theory and method. As a broad framework, DP attends to how ‘“psychology” and “reality” are produced, dealt with and made relevant by participants in and through interaction’ (Hepburn and Wiggins, 2005: 595). In lieu of giving accounts of research participants’ physiological or psychological states as an assumed underlying explanation for why people say, think or act as they do, DP views physiological states, mental states, and even thoughts to be features which are situated in discursive practices. DP asserts three fundamental principles related to the way in which discourse is defined (Potter, 2004). First, discourse is understood as being action-oriented, leading the analyst to primarily focus on what discourse is doing in the specific context of the interactional event. Second, discourse is viewed as constructed, with the analyst considering how the words and rhetorical devices employed within interactions construct and stabilize certain versions of the world. Finally, discourse is considered situated, being bound up and embedded within a given interaction. More particularly, DP builds on these three principles and the method of discourse analysis to explore how discursive ‘devices’ (e.g. lexical choices) might be used to carry out and manage a variety of interactional activities (e.g. stake management). Edwards (2005) identified three major strands of DP: (1) respecification of psychological constructs; (2) explorations of how everyday psychological categories might be used in talk and text; and (3) examinations of how ‘psychological business’, such as motivation and perception, might be managed in talk and text (p. 259). Within a DP approach, the data analysis process is often grounded in principles of conversation analysis.

Within this study, then, we positioned our examination of the varied meanings and performances of autism within a perspective that presumed that the social world is built in and through talk. To some extent, we also grounded our discourse analysis in conversation analysis principles. Thus, as we attended carefully to the ways in which the therapists and parents performed ‘normality’ and, at times, ‘abnormality’, on behalf of the child, we considered the ways in which the shifting meanings of autism were coupled with performances of normality, abnormality, giftedness, exceptionality, etc.

This study reports findings from a larger ethnographic study of a pediatric therapy clinic in the Midwestern region of the United States, The Green Room, ¹ which began in October 2009. Upon Institutional Review Board approval, interview data was collected from 14 parents (11 mothers and three fathers) of children with a clinical diagnosis of autism and eight therapists (three speech-language pathologists, two occupational therapists, one physical therapist, one autism specialist, and one support group facilitator). This interview data was part of a larger corpus of data which included 175 hours of audio and video data (e.g. from group and individual therapy sessions, interviews and waiting room conversations), 650 pages of observational field notes, and a corpus of e-mail correspondence and artifacts/documents used during therapy sessions and administrative tasks. Face-to-face meetings were also recorded in which our interpretations of the data were shared with the participants.

This article focuses on the parent and therapist interview data. A semi-structured interview protocol was followed, with each interview being audio-recorded and transcribed using Jeffersonian notations (2004). The parent interviews ranged from 22 minutes to 84 minutes and averaged 41 minutes. The therapist interviews ranged from 10 minutes to 42 minutes and averaged 22 minutes. While only families who had a child with a clinical or educational diagnosis of autism were invited to participate in the study, we presumed that a diagnostic label did not automatically indicate that the therapist or child’s parent oriented to autism as a biological truth. Rather, we acknowledged that ‘identity ascription of any kind, and by academics as much as by anyone else, is always occasioned by some interactional or institutional circumstance’ (Rapley et al., 1998: 825). During the interviews, then, when the word ‘autism’ was made relevant as a diagnostic label by the participants, clarification and explanation was elicited. We recognize that both the interviewer and interviewee were active participants within each conversation. In this way, we viewed all of the interviews as discursive events (Potter and Wetherell, 1987), and believed that by ‘. . . asking participants about particular topics’ we would learn ‘. . . about the culturally circulating discourses that are invoked in producing answers’ (Benwell and Stokoe, 2005: 124).

Discourse analysis from a discursive psychology perspective was used to analyze the data (Edwards and Potter, 1993), drawing upon conversation analytic techniques (Sacks, 1992). Analysis moved between the micro-level of the conversation (interview talk) and macro-level discourses. Six broad phases of the analysis included: (1) intensive listening; (2) transcription; (3) repeated reading and listening; (4) selection, identification, organization, and further analysis of patterns across the discourse segments; (5) generation of explanations/interpretations; (6) reflexive and transparent sharing of findings. The first author maintained a research reflexivity journal in which she documented decisions as they were being made to enhance the trustworthiness of the findings. These decisions were reviewed by the second author and discussed as the study progressed.

Findings

Early on in the data collection process, we began to consider the ways in which the shifting meanings of autism were linked to performances of ‘normality’, ‘abnormality’, and even ‘exceptionality’. As we took note of the performative component of autism in the interview data, our interpretation of the parents’ and therapists’ accounts became increasingly complicated. No longer did we orient to the parents’ and therapists’ accounts as solely performing ‘normality’, ‘abnormality’, ‘disability’, or ‘ability’. Instead, autism was performed in shifting and even at times contradictory ways. One of the initial understandings of this particular aspect of the data is illustrated in the following research reflexivity journal entry:

June 2, 2010 8:30 pm

In a parent interview today, I (Jessica) was struck by how ‘autism’ was not separated from the spaces in which the child and parent traversed. ‘Autistic’ was not how the mom first described her son. Autism was made relevant for a reason. It only came up in the context of certain spaces and events. When discussing the neighborhood in which they recently moved, the mother said that the neighbors noticed right away that something was ‘different’ about her son Saturn. She continued, ‘. . . all we had to do was say, “Okay look Saturn has autism and this is the way it is and he’s not ignoring you. He just doesn’t know how to address you or to identify with you just yet. But if you give him some time he will”.’ About school, she said, ‘They had never worked with an autistic child. So, I had to tell them what that meant and how to teach him and that he is a very smart boy.’ So, as she and her son navigated these spaces, autism was performed in varied ways, for different audiences.

One of the ways in which all of the participating parents and therapists performed autism discursively was to reframe the children’s (non-normative) behaviors for diverse audiences. While many researchers cast framing and reframing events as a cognitive entity or task (Shmueli et al., 2006), we orient to this concept discursively and suggest that speakers (re)frame events as they work to make sense of them in coherent and believable ways. Thus, much like Edwards and Potter (1993) speak of building accounts in ways that are believable to others, we view the idea of reframing as the way in which the participants re-accounted for an event or behavior, offering an alternative explanation or account that framed anew an event or behavior assumed to be problematic.

In order to illustrate the primary ways in which the therapists and parents, drawn from the analysis of the interview data, oriented to the performative components of autism, we present three excerpts that capture the contradictory, fluid ways in which notions of autism were framed for diverse audiences. The first excerpt focuses on what many of the therapists referred to as ‘looking autistic’. The second excerpt highlights the elaborate work of performing normality. The final excerpt presents an account in which the deployment of a disability category is made in relation to others’ potential (mis)interpretations.

When talking about the meanings and performances of autism, the therapists, in particular, made relevant the subjective nature of being diagnosed with an autism label. In Excerpt 1 (see Appendix for transcription symbols), Megan, a therapist, pointed to the ‘look’ of autism and questioned the veracity of the expert’s decisions to diagnose Chance, one of the participating children, with an autism label.

Excerpt 1

In the above excerpt, Megan began by referencing Chance, whose ‘pediatrician diagnosed him’ (line 1). Her move to name who (‘pediatrician’) diagnosed Chance is particularly relevant in that across the interview data, both the participating therapists and parents spoke often about the need to educate general pediatricians about the needs, or impairment effects, of children with autism labels. While an autism diagnosis can typically be made by a range of healthcare providers (Allen et al., 2008), specialists in developmental disabilities, such as developmental neurologists or child psychologists, are oriented to in the broader and publicly available discourse as the most skilled autism diagnosticians (Center for Disease Control (CDC), 2010). Thus, Megan’s move to make the ‘pediatrician’ relevant is telling (line 1), and points to both the local and broader discourse about who is and is not qualified to name someone autistic.

Megan next stated: ‘I’m sure to the pediatrician he looks very autistic’ (lines 1–2), implying that there are specific, visible characteristics, physical markers, or performances that increase the chances of being named autistic. In many ways, Megan’s emphasis on the ‘look’ of autism points to the performative aspects of the body (line 2), highlighting the way in which a body might be oriented to and ‘read’ by experts as autistic or non-autistic, normal or abnormal, etc. She then named what it is that ‘looks’ autistic to the pediatrician, including ‘sensory things’, ‘some of those behavior things’, and ‘language that hasn’t developed’ (lines 2–4). As Garland Thomson (1997) noted, within discourse, particular identities are produced and located ‘within a hierarchy of bodily traits that determines the distribution of privilege, status, and power’ (p. 6). So, it is not surprising that Megan’s list (‘sensory thing’, ‘behavior things’, and ‘language that hasn’t developed’, lines 2–4) matched, to some extent, the common ways of talking about, producing, and locating autism, with particular ‘bodily traits’ being noted, including (1) social, (2) communicative, and (3) behavioral deficits. Further, three-part lists often work to discursively create a sense of completeness and representativeness (Bowker and Tuffin, 2007; Edwards and Potter, 1993). Thus, Megan’s three-part list functioned to construct autism as something that is prescriptively determined, yet just as ambiguous and open to interpretation as the ‘official’ diagnostic (and open to interpretation) criteria found within the diagnostic manual (DSM-IV) used by professional diagnosticians (e.g. psychologists).

Although displaying initial tentativeness about offering any concrete judgment of her own, particularly about an individual (psychologist) who typically holds diagnostic clout, Megan initially stated ‘I don’t know’ (Edwards and Potter, 2005; Potter, 1996), and then introduced the idea that ‘even psychologists . . . don’t . . . always know’ (lines 4–6). While psychologists are typically positioned as the key diagnosticians of autism (Allen et al., 2008), Megan oriented to ‘even’ their diagnostic abilities as being questionable on the basis of only one visit (lines 5–6). She undermined the validity of the diagnostic process, suggesting that the psychologist likely cannot really ‘always know’ because ‘. . . they will see a child for one evaluation visit’ (line 5). All of the participants who offered critiques of the diagnostic process for disability labels pointed to the minimal time the diagnostician spent with a given child. Such critiques imply that as an ‘expert’ spends more time with a child, s/he may discover how to more accurately interpret or read the ‘look’ or ‘characteristics’ of autism, eventually being better able to determine whether the child is really autistic or not.

Megan concluded with an if/then conditional statement: ‘Like if they look autistic and they meet the criteria (.) then they’re gunna get a diagnosis basically’ (lines 8–9). Conditional statements often function to distance speakers from being directly accountable for their statements, as such statements are often constructed and taken up as being factual (Potter, 1996). In this case, her final if/then statement pointed again to what happens when a child ‘looks autistic’. Megan oriented to this ‘look’ or performance of autism in the presence of a diagnostician as inevitably resulting in a diagnosis of autism.

In other words, if Chance or any other child ‘looks autistic’ enough to a diagnostician, he will be discovered as such. As many disability scholars have noted, ‘the nondisabled gaze is the product of a specific way of seeing which actually constructs the world it claims to have discovered’ (Hughes, 1999: 155). Thus, as was made visible in Megan’s talk, autism remains ‘. . . a list of symptoms or behaviors or representations that can be studied and discussed, but it is not knowable as a truth. It must always be interpreted’ (Biklen et al., 2005: 3). So, it seems that Megan’s account leaves open the possibility that autism is always already open to interpretation and contestation by those ‘experts’ most often deemed responsible for determining, somewhat subjectively, whether a child really is autistic, as well as those individuals who live or work closely with the child with an autism label. The construct of autism, never located as inherent to the child, is only made real when it is negotiated between the key social actors (i.e. diagnosticians) and the child’s very performance of the ‘autistic look’.

The next excerpt illustrates the ways in which some parents went about performing their child’s normality despite a medical diagnosis of autism. In this excerpt, Alisha constructed her son as ‘a very typical little boy’, locating who he is in relation to a normative backdrop, which is itself a discursive construction (Locke and Edwards, 2003). Excerpt 2, while highlighting the common pattern of performing ‘normality’, shows that few parents went about performing ‘normal’ as elaborately as Alisha did. The excerpt begins just after the interviewer (Jessica) asked Alisha, ‘if others were to meet your son what would be some things that you’d want them to know about him?’

Excerpt 2

In response to the first interview question (‘What things might you want someone who has just met your child to know about him?’), Alisha, like the majority of parents, began by making relevant all that Picasso could do. She began by describing Picasso, her son, as ‘affectionate’ and having ‘feelings’ (line 1); the majority of parents, and all of the therapists, made similar statements about the children they spoke about. Making emotional qualities relevant was particularly significant in that one of the prevailing assumptions about autism is that people with autism labels inherently lack the ability to relate to others, possessing ‘no affective tie to people’ (Kanner, 1943/1985: 24). Thus, with Picasso constructed as affectionate and having feelings, Alisha resisted or at least distanced herself and her son’s identity from the dominating discourses of autism that would likely cast Picasso as possessing a ‘genuine defect’ in the ‘understanding of the other person’ (Asperger, 1944/1991: 81). Instead, she constructed her son as ‘normal’, bolstering this claim of normality by describing all that made Picasso a ‘very typical little boy’ (lines 6–15). Further, she worked to normalize and reframe that which could be named an impairment effect – Picasso’s communication difference. While Picasso was described by his mother and therapists as minimally verbal, here Alisha reframed his communication as ‘in many ways . . . very typical’ (line 6).

Her use of the phrase ‘even though it sometimes doesn’t always seem like it’ (lines 5–6) implied that there were communicative differences that other people might name as atypical. Sacks (1984) suggested that ‘doing being ordinary’ is a common feature in talk, a recurring pattern in everyday social life. To present one’s self or the identity of another as an ordinary social participant implies normalcy; thus, Alisha’s descriptions of Picasso as enjoying being outside or liking to eat chocolate-chip cookies functioned to discursively define all of the specific actions that identified her son as typical and normal (lines 8–10) (Locke and Edwards, 2003). Further, Alisha’s examples of what makes Picasso typical pointed to the cultural assumption that there is a natural set of practices, actions, and behaviors common to all normal children.

With disability categories often positioned as being synonymous with incompetence, particularly when an individual is considered minimally verbal (Biklen et al., 2005), Alisha’s description of Picasso as an ‘above’ par student functioned to counter the presumption of incompetence (lines 11–12). While Alisha constructed Picasso as competent, she did not avoid naming what he could not do, stating ‘even though we can’t have a conversation’. Yet she continued by linking what Picasso could not do to what he could do: ‘he (.) he reads wonderfully’ (line 15). By emphasizing what Picasso can do (i.e. ‘reads wonderfully’), Alisha minimized what some would name Picasso’s primary impairment effect (i.e. ‘can’t have a conversation’, line 13).

Some disability studies scholars have critiqued the ways in which ‘most popular representations of autism . . . impose neurotypical formulae or conventions’ on people with autism labels, ignoring social differences through the validation of ‘neurotypical experiences’ (Osteen, 2008: 9). Indeed, we acknowledge this tendency and the overarching pattern in the broader discourse to impose and privilege certain social conventions. Certainly, one might interpret Alisha’s account as imposing neurotypicality on Picasso, to the demise of his differences. Yet the very task of negotiating normality, abnormality, autistic, or non-autistic identities is filled with contradiction and is a rather fragile endeavor, far more complicated than the binary constructions of typical versus non-typical or normal versus abnormal, notions which are often critiqued by disability studies scholars (Thomas, 2004). Alisha, like all of the participating parents and therapists, engaged in the fragile and contradictory task of performing difference, normality, and even autistic or non-autistic identities in fluid and shifting ways. Much like Butler (1990, 2004) argued about the performance of femininity, all aspects of autism are constantly being negotiated, taken up, and reproduced as performative acts that function to reinforce certain discourses and ideologies surrounding autism and disability in the culture at large. These performances often act to position the very notion of an autistic or non-autistic child as being a natural truth, something that is really achievable; yet to achieve typicality or normality is impossible, as it too is a discursive construction, always shifting across time and place.

Across the parent interviews, we noted that each parent shared a story or account of a time or place in which they felt compelled to make their child’s diagnostic label relevant. We noted that many of these retellings were linked to notions of disability and normative and non-normative ways of behaving and speaking. For the parents, making autism relevant to outsiders was almost always linked to the ways in which the participating children were oriented to by outsiders as troubled or even ‘unruly’ and ‘misbehaving’. At times, autism and other disability labels were used to reframe discursively the child’s way of being, accounting for why the children screamed or moved their bodies instead of using words, for example. The label of autism, then, functioned to explain the participating child’s non-normative ways of speaking and behaving, even making the non-normative more reasonable and justifiable.

In the third and final excerpt, Nicole offers one such account, illustrating how specific events or encounters in public places were linked to the performance of autism as a disability, with the disability category itself functioning to explain her child’s non-normative behavior.

Excerpt 3

Nicole, responding to the first interview question (‘What things might you want someone who has just met your child to know about him?’), emphasized that outsiders or ‘strangers’ orient to her son George in ways that compel her to ‘offer up’ an explanation (lines 1–2). Nicole’s use of the word ‘sees’ (line 1) pointed to this idea of an autistic ‘look’ or body, implying that there is something about her son’s actions or way of being in the world that automatically positions him as different or non-normative to outsiders. She then shared an example of an event that resulted in George’s differences being noted by outsiders. George’s meltdown at a business was apparently heard by everyone (lines 6–9). While Nicole stated that the bathroom door locks were simply ‘upsetting’ to her son, she implied that George’s meltdown was quite troubling to the store employees (line 7). Later in the interview, Nicole returned to this ‘incident’ and further explained that the business employees came to the bathroom to determine whether they needed to call for further assistance or security.

This event, a ‘meltdown’ and the response of the business employees, was used by Nicole to account for making relevant and even naming her son’s disability (line 2 and line 11). Further, as broad categories that are both open to interpretation, the terms ‘autism’ and ‘disability’ functioned to reframe the ‘meltdown’ as explainable, even justifiable. If a child is constructed as autistic and has a meltdown, the meltdown is oriented to as more justifiable, as autism is then positioned as a disabling entity that caused the meltdown. However, if a child is constructed as non-autistic and has a meltdown, the child may simply be positioned as ‘unruly’ and ‘misbehaving,’ as one father spoke about in his interview. Sacks (1992) reported that a common rhetorical move in discourse is to resist being accountable for an action or situation by defining that action as commonplace, normal, or expected (Edwards, 1997). It is possible then that Nicole’s deployment of the terms ‘autism’ and ‘disability’ functioned to distance herself and her son from being responsible for the meltdown, with autism or a disability positioned in that role instead.

A body named ‘a little bit different’, ‘autistic’, or ‘disabled’, much like other bodies, is always already an object of discourse caught up in multiple systems of meanings and representation (Bordo, 1992; Foucault, 1971; Turner, 1984). Normative behaviors in a business space do not typically (if ever) include meltdowns or looking ‘a little bit different’. So, in the above excerpt, Nicole’s construction of George’s behavior as ‘a little bit different’ (line 2) also illustrated the ways in which his body might be constructed and oriented to by others as troubled.

In this way, children’s bodies or actions/behaviors themselves are sites of discourse, with their bodies/behaviors often being constructed as autistic, disabled, abled, normal, abnormal, etc. Foucault (1972) suggested that discourses constitute and even regulate the body in certain ways, ‘exercising upon it a subtle coercion . . . obtaining holds upon it at the level of mechanism itself – movements, gestures, attitudes, rapidity: an infinitesimal power over the active body’ (p. 137).

Discussion

In Nadesan’s (2005) historical account of the making and remaking of autism, she spoke of the performative aspect of autism, stating:

. . . autism has a performative component, as known by every parent who struggled to meet the criteria for government and educational services for their children. For the social services agent, I [as a parent of a child diagnosed with autism] must stress (and even exaggerate) Kamal’s [her son] maladaptive behaviors. For his teachers, I stress Kamal’s high intellect in order to avoid having him labeled as ‘mentally retarded’. For his peers, Kamal performs ‘normality’ in the context of the school playground by stifling his odd interests and masking social awkwardness in order to ‘fit in’ with the other children. (p. 2)

Nadesan’s words capture the social, political, and economic constraints that shape the performative aspects of autism, pointing to how parents of children with autism labels make real varied versions of autism for particular audiences. Similar to Nadesan’s (2005) historical account of the making and the remaking of autism, the findings of this study make visible the performative aspects of autism as the participants made real varied versions of autism for particular audiences. We found that the therapists and parents both oriented to a diagnosis of autism as a subjective matter, as definitions of autism vary and change both officially and unofficially. Rather than equating a diagnosis of autism as a diagnosis of incompetence, we noted that in some situations parents and therapists performed a child’s ‘normality’ (competence) and in other situations they performed the child’s ‘abnormality’ (incompetence).

Building upon Rosetti et al.’s work, this study’s findings point to the ways in which ‘behaviors and actions traditionally linked with incompetence’ can and even should be reframed and reinterpreted (p. 364). For example, the idea that a child’s look ‘or way of moving their body or expressing themselves is read’ as plainly pathological, often results in misinterpretation by outsiders (e.g. the child is constructed as ‘threatening’ or ‘incompetent’), something which was made relevant by the participants in this study over and over again. The body, as a site of discourse and performance, can resist and re-invent identities. As Carlson (1996) noted, ‘performance can work within society precisely to undermine tradition, to provide a site for the exploration of fresh and alternative structures and patterns of behaviors’ (p. 15). Yet few parents worked to perform autism in ways that positioned their children’s non-normative behaviors as ‘alternative structures and patterns’, which could function to redefine and rework all that is named typical or normal development, while many of the therapists did so. Nevertheless, the parents did speak of the ways in which the participating therapists assisted them in making sense of their children’s behaviors/actions and communication styles, helping them and their children to redefine all that counted as successful functioning. The therapists, who often explicitly taught the children how their very bodies were oriented to by outsiders, seemed to go about the work of co-constructing and reframing with the children, in particular all that might be oriented to by others as problematic. This discursive work of reframing most often occurred in the one-on-one therapy sessions of the participating children and therapists analyzed, and these are explored in a future study (Lester, 2011).

Suggesting that performative acts of normality elide differences (Osteen, 2008) ignores the challenge of living in a society that is politically, socially, and economically structured to privilege those who speak and act in normative ways. So, while some of the parents’ discourse functioned to construct their children as ‘normal’ in many ways, they produced such constructions within a social and political context that privileges highly verbal people who behave in normative ways (e.g. people should not flap their hands to communicate excitement, but instead should use words or widen their eyes with surprise, expressing themselves in more ‘acceptable’ or normative ways). If the parents had constructed their children as minimally verbal or quite inept, perhaps the competency and even humanity of the child would be jeopardized or at least questioned by outsiders, particularly individuals who are unfamiliar with the particular communicative style (Rocque, 2010). As Lewiecki-Wilson (2003) noted, ‘we [the majoritarian culture] often demand some verbal response from another as proof of their humanness’ (p. 157). Nelson’s (2004) official statement to the United Nations pointed to the ways in which individuals with autism labels are misunderstood and excluded, stating that ‘we [people with autism labels] experience discrimination in various forms, often because of our different use of language and communication . . .’ (p. 1). Thus, it is important to interpret the discursive practices of the participating parents and therapists in relation to the exclusionary practices and policies that they and their children encounter daily.