In this short paper, we draw on recent developments within critical autism studies where dominant knowledges about autism are scrutinized and autism is considered as a neurological difference rather than a deficit (see for example Armstrong 2010; Davidson and Orsini 2013). From a neurodiversity perspective, autism is viewed as both a possible strength and as a nuanced difference rather than a one-sided shortcoming or deficit. The theoretical position that we are taking of neurodiversity does not seek to identify particular neurological make-up of individuals, but rather seeks to question the status accorded to individuals who have a presumed ‘typical’ or ‘atypical’ neurological make-up. The aim is to interrogate how dominant knowledge makes assumptions about neurologically typical and different people which position people with autism as less able. One such dominant knowledge about autism is the assumption of deficits in forming friendships and developing sociality for people with autism.

Ochs and Solomon (2010) conceptualize sociality as a range of possibilities rather than a unitary dimension of social behaviour, in which sociality is enacted differently across different social situations and differently by different groups. The skills needed to make and maintain friendships, are seen to be ‘a range of possibilities for social co-ordination with others, and autistic sociality is one of these possible co-ordinations’ (Ochs and Solomon 2010, 70). Following Ochs and Solomon, we are interested in the beginning to think differently about the concept of friendship and how high-functioning people with autism experience their emotional lives and friendships. This perspective is combined here with developments in critical developmental psychology (Burman 2008), through questioning ideas implicit within understandings of the development of sociality to consider the concept of ‘autistic friendship’. In this paper, we specifically focus on how developments in online social networks may provide a context through which many people with autism can develop and maintain friendships.

Deconstructing dominant understandings of (autistic) friendship

There is a long-established body of developmental psychological research that has identified and explored characteristics and qualities of relationships and the skills that are required to develop friendships. Within the mainstream approach to understanding relationships, friendship is seen as an intimate relationship providing ‘companionship, mutual support, and affection’ (Freeman and Kasari 1998, 341). Psychological theories of the development of friendships assume a developmental trajectory in which children learn skills needed to make friends (Whitehouse et al. 2009) in which the quality and function of friends map onto broader developmental stages and proficiencies of the child. It is clear from the developmental psychological literature on friendship that social relationships, and in particular friendships, play a vital role in a child's development (Berndt 1999; Hartup and Stevens 1999).

There is a limited body of research that has examined the development of friendships for children with autism (Chamberlain, Kasari, and Rotheram-Fuller 2007). The construction of autism as deficit is drawn on in discussions of the impairment in social interaction and communication. In line with the general focus in the literature on the ‘impairments’ of people with autism, there is a dominant view focusing on the ‘failure’ of autistic individuals to form ‘developmentally appropriate’ relationships with their peers. The proposition of deficits such as theory of mind provides theorists with a framework for viewing children and adults with autism as being disabled in their ability to make friends. For example, Orsmond, Wyngaarden Krauss, and Mailick Seltzer (2004) conclude that their data ‘… indicate that the impairments associated with autism are impediments to the development of peer relationships: the more severe the impairments in social skills, the less likely it is for an individual to form peer relationships’ (Orsmond, Wyngaarden Krauss, and Mailick Seltzer 2004, 253). From within this perspective, researchers have called for more research in order to understand how such ‘deficits’ may affect the formation and development of close relationships for people with autism (Howard, Cohn, and Osmond, 2006; Locke et al. 2010). Research suggests a lack of the establishment of relationships that engage in social and emotional reciprocity, non-verbal communicative behaviours and ‘deficits’ in the development and maintenance of relationships that are ‘appropriate’ for the individual's developmental level (see for example APA 2013).

It is therefore unsurprising that research that has addressed the development and role of friendship for people with autism is very limited. Whitehouse et al. (2009) have noted that within academic research, early studies argued that individuals on the autism spectrum were content with social isolation, or even that such individuals preferred this. However, Whitehouse et al. observe that more recent studies have begun to question these earlier assumptions. Sinclair (2010), an autistic activist, argues that despite a lifetime of ‘distressing and discouraging’ experiences, many people with autism continue to engage with the neurotypical (NT) world, and desire to form close relationships. Indeed a desire to form close friendships and other relationships is a common theme running through personal accounts of people with autism (Jones and Meldal 2001). Writings by people with autism such as Gunilla Gerland and Wendy Lawson provide us with a rich and insightful picture as to the negotiation of relationships by people with autism (Gerland 2004; Lawson 2006). Both of these authors (and many others who have written about their lives as people with autism) highlight that much of the dominant literature suggests that people with autism do not seek to engage in friendships. They dispute this claim, arguing instead that many people on the autistic spectrum do in fact want to have friends but find the management of such relationships problematic at times or have alternative ways of performing or experiencing friendships.

We, and others, argue that dominant understandings of the concept of ‘friendship’ sets up norms for the behaviours and skills needed to have friends (Kasari et al. 2011). Implicit within this approach are assumptions about emotional expression, which are naturalized and are drawn into measurements of appropriate friendship against which people with autism are seen to be deficient. For example, Locke et al. (2010) note that while the quality of friendship between people with autism may not be considered high if we measure it against an NT model of friendship, such relationships are considered to be of a high quality by the persons engaged, and therefore provide the benefits of having friends but in ways that may not be evident in neurotypical friendships. Similarly, when looking specifically at developmental stages of play, the notion of parallel play is a characteristic of early childhood which children are assumed to grow out of. This does not apply readily to children with autism for whom solitary play is often highly valued. Bagatell (2007) noted that, in her fieldwork, socializing was often considered by her participants to be evidence of proximity rather than social interaction and discussion. She noted in her fieldwork that socializing ‘involved practices that I (sic) did not originally perceive as social’ Bagatell (2007, 421).

Proposing alternative understandings of (autistic) friendship

Research and activism drawing on a view of neurodiversity which values autistic traits and acknowledges difference rather than deficit, has begun to question some of the previously held beliefs about the abilities of people with autism. For example, challenging the assumption that people with autism lack the ability to form meaningful friendships with their peers, and to provide evidence that the notions of NT friendship are not the only way in which friendship can be experienced. Autistic activists and other advocates have highlighted that research on peer friendships of autistic people frequently has the goal of studying friendship as a means to improve the social skills training of people with autism, and therefore enable them to develop more neurotypical relationships (see for example, Bauminger, Solomon, and Rogers 2010; Frankel and Whitham 2011). Activists seek to highlight that children and adults with autism make active efforts to make friends; to reformulate the meaning of friendship and what it means to have a friendship on their own terms. A significant development in recent years has been the opportunities to interact with friends via online media.

There has been considerable media and research attention paid to the opportunities and risks for (NT) young people interacting online (see for example, Reich, Subrahmanyan, and Espinoza 2012). The potential for social networking to generate large numbers of ‘friends’ has been seen as a concern for the psychological well-being of children and young people where boundaries of intimacy are seen to be eroded by online communication. However, research suggests benefits for young people interacting online including feelings of being connected to others, emotional and social support. There are additional benefits for young people with autism due to the difference in communicational styles drawn on by online social networking and face-to-face exchanges. Online communication offers opportunities for making and maintaining friendships in ways that can be supportive for people with autism. For example, Chen, an autistic activist, discussed the potential benefits for online social networking for some autistic people in his reflections on his website:

A few years ago, I believed that social networking sites were nuisances which I was forced to join when too many friends send me invitations. In December 2009, I suddenly realized that: With photos and names listed, I can keep track of my friends even when I have trouble with names and faces. All the background information of my friends are available for me to reference. Friends provide updates with which I can use as a context to re-establish contact with them where necessary. All my online communications are recorded and available for me to plan my next social exchange. Social exchanges in real life are simplified and emulated on Facebook. For instance, pokes can be likened to a brief touch on the shoulder. Unlike real life, I can maintain total control of the impression I create via my own profile and what I choose to share (Chen 2010).

Chen's reflections on the use of social networking for people with autism demonstrate the ways in which friendship can be managed in ways different to the assumptions held by NT researchers whose focus has been primarily face-to-face friendships. Lawson (2006) also discussed the possibility for technologies such as email to play a large role in the finding and maintaining of friendships for people with autism. This observation has also been made in research that examines the role of online discussion groups for adults with autism (see for example Brownlow 1997).

Our previous work has indicated that spaces that respect neurodiversity can be created and managed by people with autism in order to facilitate friendships and close relationships (see for example Bertilsdotter Rosqvist, Brownlow, and O'Dell 2013). Given the possibilities for the formation of neurodiverse spaces online, there is a need for research that documents the alternative forms that ‘friendship’ may take for people with autism in contexts that feel safe and supportive. This may involve an alternative understanding of the concept of ‘friendship’; one that does not necessarily rely on neurotypical definitions and expectations surrounding the concept of what it means to be a ‘friend’, where a range of possibilities for being social and having friends is acknowledged and promoted.

Drawing on Ryan and Räisänen's (2008) notion of ‘safe places’, we postulate that social networking sites such as FaceBook or off-line spaces such as camps like Autreat adapted to the needs of people with autism could provide a ‘safe space’ in which people with autism and also NTs can form, develop and maintain friendships. Neurodiverse spaces may be online communities or face-to-face environments that are designed to be accessible to both NTs and people with autism (Bertilsdotter Rosqvist, Brownlow, and O'Dell 2013). An alternative version of friendships, accessed often through the narratives of people supposedly lacking the ability to have and make friends provides us with a valuable insight into a richer and more diverse understanding of friendships. Combining these with a theoretical perspective that acknowledges neurodiversity and challenges taken for granted assumptions about autistic friendships serves to highlight the need for further research into neurodiverse spaces within which people with autism can develop friendships with each other and with neurotypicals; and what those friendships look like and provide for them.