Dicing with Deception: People with Disabilities' Strategies for Managing Safety and Identity Online

Authors

  • Natilene Bowker,

    Corresponding author
    1. Social science cyber-researcher with experience in both quantitative and qualitative methodologies for collecting and analyzing online data. An article on understanding online communities through multiple methodologies combined under a postmodern research endeavor has been published in Forum: Qualitative Social Research. Bowker is currently completing a doctorate within discursive psychology at Massey University, New Zealand, analyzing what it means to be online for people with disabilities. Her previous research experience includes quantitative and ethnographic investigations into the influence of gender on IRC users' capacity to explore their own identities. Her most recent articles are published in CyberPsychology and Behavior, and Disability and Society.
    Search for more papers by this author
  • Keith Tuffin

    Corresponding author
    1. Senior lecturer in the School of Psychology at Massey University where he teaches critical social psychology and discursive psychology. His research interests include political discourse and constructionist approaches to emotion, culture and identity. He is currently working on Understanding Critical Social Psychology (Sage).
    Search for more papers by this author

Address: School of Psychology, Massey University, Private Bag 11-222, Palmerston North, New Zealand.

Abstract

The visual anonymity associated with online interaction offers people with disabilities the potential to participate in social interaction beyond the stigma of a disabled identity. In problematizing traditional notions of reality, however, the online medium also has the potential to become a deceptive social space where people with disabilities become victims of malevolent acts. Considering the dilemma surrounding the choice to participate, this study investigates how people with disabilities are managing issues of deception and harm in online contexts. A discursive psychology framework is utilized. The research was conducted in New Zealand where 21 participants with physical and sensory disabilities volunteered to participate in an online interview. Two different repertoires enabled people with disabilities to manage the dilemma of engaging in a medium where there is potential for benefit and harm. A keeping safe repertoire deployed three safety strategies to protect participants from deceptive acts. Data from several participants was also categorized under a qualified deception repertoire. This allowed participants to access new subjective experiences outside of a disabled identity and to extend their online engagement beyond keeping safe. Both repertoires maintained participants' integrity as online users.

Introduction

This investigation is informed by a larger piece of work researching the online experiences of people with disabilities (c.f. Bowker & Tuffin, 2002). The visual anonymity afforded by computer-mediated interaction neutralizes the physical differences, which have such profound effects on face-to-face interactions for people with disabilities. Online communication offers the possibility of experiencing non-stigmatized identities while also offering the capacity to create harmful and dangerous outcomes, which may jeopardize participants' safety. Considering the dilemmas surrounding online participation, this study investigates how people with disabilities are managing issues of deception and harm online, using a discursive psychological approach. In reviewing the contribution of the social sciences toward understanding the impact of the Internet on social life, Bargh (2002) makes the important point that any effects depend on how the mode of communication interacts with the unique characteristics of users. This study seeks to investigate the online experiences of people with disabilities, who constitute a unique group who have traditionally been identified as powerless in society. This powerlessness stems from the failure of the physical and social environment to cater to their needs. However, within computer-mediated environments, people with disabilities may have much to gain as physical barriers to participation are broken down. Moreover, the online medium's capacity to conceal physical difference brings forth the opportunity for people with disabilities to access a social space for experiencing alternate subjectivities, which operate outside the stigma often associated with disabled identities. To introduce this study we first highlight features of the medium which have particular implications for the social interactions of people with disabilities. Next, we emphasize the potential for both creativity and harm which the Internet offers. In both cases there are important psychological implications for human social interaction. Finally, we discuss the fundamental epistemological orientation on which the study rests by a brief introduction to discursive psychology.

While the Internet offers possibilities, which resemble face-to-face encounters, there are also important differences, which promise new forms of social interaction. Illustrating this, Van Gelder (1991) reported a famous incident occurring on a computer conferencing system during the early 80s where a male psychiatrist posed as Julie, a female psychologist with multiple disabilities including deafness, blindness, and serious facial disfigurement. Julie endeared herself to the computer conferencing community, finding psychological and emotional support from many members. The psychiatrist's choice to present differently was sustained by drawing upon the unbearable stigma attached to Julie's multiple disabilities as justification for not meeting face-to-face. Lack of visual cues allowed the identity transformation to continue, with the psychiatrist also assuming the identity of Julie's husband, who adamantly refused to allow anyone to visit Julie when she claimed to be seriously ill. This example highlights the ease with which identity may be constructed, transformed, and sustained by the textual nature of online interaction and the visual anonymity it affords. The creative possibilities here are enormous and, as Turkle (1995) notes, the Internet may be treated as a virtual social laboratory, which enables the exploration of alternative identities. Similarly, Bargh, McKenna, and Fitzsimons (2002) conclude that the Internet provides a medium through which alternative forms of self may arise. The psychiatrist was able to engage in multiple dimensions of identity creation as he assumed gendered and disabled characteristics. Online, users have access to any form of self-presentation as identity becomes a matter of choice, independent of bodily form, physical gesture, dress, and vocal characteristics (Poster, 1997). The proliferation of gender switching online attests to this creative freedom as reflected in numerous investigations (Curtis, 1996; Bruckman, 1996; Kendall, 1996; McCormick & Leonard, 1996; Pomeroy, 1996; Serpentelli, 1993; Suler, 1997).

Research has previously established the profound importance of aspects of physical appearance in the formation of first impressions and the subsequent determination of whether friendships will develop (Hatfield & Sprecher, 1986). The textual nature of online forums means that visual cues remain inaccessible to others. This visual anonymity creates a unique context for people with disabilities to participate in the social world outside the prejudicial and stereotypic barriers based purely on physical appearance. People with disabilities have the opportunity to interact without automatically, or necessarily, exposing a stigmatized identity. Spears, Postmes, Lea, and Wolbert (2002) refer to this as a form of ‘liberation’ from the inequalities and restrictions of our usual, everyday identities. This applies especially to the disabled community, with research indicating that those within this community regard the online medium as a social space where they can express their ideas without being judged on the basis of their impairment (Hyde & Todd, 1996; Roulstone, 1998; Bowker & Tuffin, 2002).

Online communication ensures the visual cues, which provide a basis for prejudiced judgements, are rendered invisible. Thus, the medium's communicative expedience removes physical barriers that may preclude social engagement. Irrespective of mobile, verbal, or co-ordination abilities, the range of assistive technology devices supporting online communication provides people with disabilities an opportunity to participate in social interaction. Real-time limitations associated with the manual operation of a keyboard are removed with asynchronous forums like email and Usenet newsgroups where responses can be made at the convenience of individual users. Furthermore, the medium's capacity to problematize traditional notions of reality elevates the online context to a tool for political and social empowerment. Removing the correspondence between representation and reality within the context of computer-mediated environments (Poster, 1997), and within the construction of digital information, recalibrates traditional meaning-making apparatus. Uncertainty between the text and its referent (Poster, 1995) undermines fixed notions of reality as simulation makes ‘the real’ redundant (Vasseleu, 1997). This leads to a greater potential for knowledge creation as the medium itself encompasses all means of creating and communicating reality (c.f. McLuhan, 1987). The digital production of information also contributes to a greater potential for knowledge construction.

Traditionally, information has been transmitted through analogue representations requiring a parallel connection between a sign's physical form and its cultural meaning. In contrast, digital code, which provides the building blocks for online communication, reduces information down to the level of binary bits, allowing possibilities to become realities (Chesher, 1997). The ability to re-present the social world through digital code enables other subjectivities to gain realization in interaction. Knowledge is constructed according to how individual participants choose to present and perceive their social world. The medium's capacity to reconstruct traditional forms of knowledge construction offers people with disabilities (and other marginalized groups) a chance to gain political and social empowerment.

In reviewing recent work in the area of human-internet interaction, Bargh (2002) points to the thematic of the medium lending itself to both beneficial and destructive uses. This theme is especially pertinent for people with disabilities. While opportunities and creative potentials may be opened up through the privileging of textual communication, these constructive possibilities may also be used for destructive and harmful purposes. Van Gelder (1991) reports that in the case of the psychiatrist and his alter ego, Julie, those who had developed close relationships with Julie felt deeply betrayed when the ruse was uncovered. While it is possible that harmful consequences may occur unknowingly rather than as result of malice or deceit, people with disabilities form a vulnerable group who must be wary of the possibility of exploitation and manipulation. This is, of course, a controversial area and we remain agnostic about whether the Internet provides a more, or less, trusting environment. Other researchers have firmer beliefs, with some claiming that without the ability to establish trust by looking into the eyes of the other, ethical principles may be threatened. Heim (1991, p. 76) suggests that the outcome of interaction via the electronic interface may lead to “an amoral indifference to human relationships.” The virtual rape and mutilation of two players in a multi-user dungeon community provides an extreme example of the inherent danger associated with engaging in a medium, where other participants can easily manipulate the environment for deviant purposes (Dibbell, 1994). Equally, there are others who claim that Internet communication leads to increased levels of trust, sensitivity, and respect. Walther (1996) is one who holds this position in proposing that Web-based communication may produce ‘hyperpersonal’ interaction. Such relationships are characterized by increased levels of intimacy and social desirability. This work is supported by Joinson's (2001) research, which documents increased self- disclosure among participants using computer-mediated-communication (CMC), and the suggestion by McKenna, Green, and Gleason (2002) that the increased anonymity of Internet communications fosters greater intimacy and closeness. The online experiences of people with disabilities present an interesting context in which to study further the paradox between engaging in an environment where there is simultaneously the potential for beneficial and harmful outcomes.

Discursive Psychology

Discursive psychology is an emerging orientation to research which has grown out of the traditions of semiotics, ethnomethodology, and linguistic philosophy (Edwards & Potter, 1992; Potter & Wetherell, 1987). Prime among these influences is the work of the linguistic philosophers Wittgenstein (1953) and Austin (1962), who regard language as constitutive rather than representational. The representational view of language assumes passivity and neutrality and holds that language merely describes the reality which it claims to portray. This view suggests that language operates in a way which is able to capture the absolute truth and transport this neutrally to an audience (Reddy, 1979). This abstract and rather disconnected view of language overlooks both the subtlety and power of language. In contrast, the constitutive view holds language to be constructive and inextricably involved in a huge range of social achievements. This orientation to language, also known as social constructionism (Gergen, 1985), assumes that language is actively involved in the construction of our experiences, our subjectivities, and our realities. This action orientation is a basic philosophical tenet for discursive psychologists, with the examination of how language works as an underlying rationale for discursive research (Harré & Gillett, 1994).

As Turner (1994) notes, a key advantage of discursive work is that it provides access to the very heart of our conceptual practices. In focusing on the linguistic, rather than what has formerly been regarded as the psychological (Wetherell, 1995), understandings of the world are revealed in linguistic practices, or discourses organized around metaphors and repertoires. With respect to the present study, our interest has been in how people with disabilities manage issues of harm and deception within the context of computer-mediated environments. Working in such environments, it seemed important to utilize an approach to research which was sensitive to the unique aspects of computer-mediated talk. In this regard, discursive psychology has championed a contextualized approach to the analysis of talk and text (Edwards & Potter, 1992; Potter & Wetherell, 1987).

Central to discursive analysis is an emphasis on how language functions as people use language to achieve particular social ends (Edwards & Potter, 1992). People's experiences are constructed within relations of power according to social practices, which both constrain and enable people's actions (Davies & Harré, 1990). As soon as we give voice to a particular version of understanding, we empower that version to be practiced rather over other versions (Foucault 1972, 1977, 1979). In this regard, particular versions can position people's subjective experience in different ways. A discursive analysis can be utilized to demonstrate how oppressive subjectivities can be challenged and replaced, leading to more socially rewarding strategies for positioning experience. Understanding the function of language leads to an examination of the kinds of strategies used to challenge and resist oppressive subjective experiences.

To summarize, discursive psychology involves fine-grained analysis of the detail of text with a view to examining the particular discursive resources which have been deployed, and the ends to which they have been put. Potter (1996) identifies two defining characteristics of discourse analysis: the emphasis placed on discourse as social practice, and the linguistic resources which facilitate those practices. In its broadest sense, this study set out to identify the resources and associated social practices in how people with disabilities talk about managing deception and harm in the context of CMC.

Method

This study was conducted in New Zealand where 21 people with physical and sensory disabilities volunteered to be interviewed online via email or another online communication program of their choice. There was an even distribution by gender, with 10 females and 11 males. Participants ranged in age from 15 to 59 years and came from different walks of life with varying experiences in the workforce. Each participant used the online medium inclusive of Internet and email facilities either daily or at least several times per week.

The study was conducted within strict ethical guidelines, with participants providing fully informed consent prior to being interviewed and being given assurances regarding confidentiality and anonymity. Participants were recruited from various disability organizations in New Zealand, which advertised the study in their newsletters or by word of mouth. To affirm the legitimacy of the researcher, participants were met in person prior to their participation. This afforded a degree of rapport building as well as the opportunity to acknowledge participants' commitment to the research. On occasion, due to geographical distance, this was not possible. Consequently, within each interview the researcher also initiated self-disclosure by beginning with a brief overview of their life history. Incorporating opportunities for closer connection between both parties functioned to reduce the inherent power hierarchy between researcher and participants (c.f. Harding, 1987).

Most participants engaged in email interviews, while two chose synchronous online communication programs using either Internet Relay Chat or ICQ (a creative abbreviation for “I seek you”). Online interviews provide a number of flexible advantages for disabled people, especially with regard to barriers of geographical distance. This was particularly helpful when interviewing participants experiencing mobility and communicative disabilities. A methodological strength of interviewing online was that participants were integrally engaged with the environment where the topic of the interviews was located, facilitating immediate engagement with issues which were covered in the interviews.

Participants who were Deaf were given the choice of participating in a face-to-face interview recorded on audiotape. This option arose at the request of one Deaf participant who felt unable to express himself fully online. In contrast to English, the dominant language of CMC, Deaf participants' first medium of communication is Sign language. Consequently, in some online contexts where the English literacy level surpasses a Deaf person's proficiency, they may be disadvantaged, highlighting the kinds of discriminatory outcomes inherent online with respect to participation. Inadvertently, this also provides further evidence in support of the argument that disability remains visually apparent online. However, while this presented a potential theoretical inconsistency in carrying out the research, Deaf participants interviewed were able to participate regularly in Deaf-related forums, as well as other online facilities open to Deaf and non-Deaf users. Three other interviews were conducted face-to-face and recorded on audio tape due to unpredictability in participants' life circumstances and difficulties with obtaining regularity in response feedback for questions sent via email. While electronic interviews extended over weeks and months, tape-recorded interviews offered more controlled outcomes lasting on average two hours.

The interview schedule was comprised six sections. The first four sections focused on the topic of online experiences, while the remaining interview sections considered demographic and disability-related details. As most interviews were conducted online, extracts retained their original formatting including spelling, grammatical errors, capitalization of words for emphasis, and ellipses (although words in square brackets have been added to provide explanation of any abbreviations or identifiable names participants used). Tape-recorded interviews were transcribed verbatim. Identifying information was removed, substantially altered, or replaced with pseudonyms.

Considering the importance of language use in discursive psychology, it is worth discussing the implications of conducting interviews across different media. Theoretically, conducting interviews across different media may lead participants to form their ideas within divergent cultural structures as dictated by the protocols of the specific medium. Some may even argue that this may bring about contextually different kinds of discursive resources, which may radically alter the nature of analysis, in addition to causing unnecessary perplexity. However, as pointed out by discursive experts (c.f. Parker, 1992; Potter & Wetherell, 1987), discursive constructions deployed by individuals are products of a particular social climate. Although nuances in media may alter constructions, the social ideologies of the historical era dictate the kinds of linguistic repertoires expressed. This does not deny the possibility that on- and offline interviews with the same person will be different. Rather, the primary aim of a discursive approach is to access the broad range of discursive constructions available to participants (not replicating interview conditions). Therefore, conducting interviews within on- and offline arenas may extend the available discursive positions people with disabilities use to manage their engagement in a medium where there is potential for deception.

Analysis

Preliminary coding involved reading extracts from the interviews multiple times to look for patterns in the data (Tuffin & Howard, 2001). Extracts which talked about specific issues in a similar way were grouped together. Continual refining of this process led to data being placed within identifiable categories. Data which did not seem to fit into any categories was left aside for later appropriation. Some items were removed from the analysis after a final refinement of categories because they were outside the scope of the topic of this investigation.

In talking about the central categories identified in the data, we have chosen to adopt Gilbert and Mulkay's (1984) concept of an ‘interpretative repertoire’, which emphasizes the flexibility in linguistic resources available to speakers (Burr, 1995). Interpretative repertoires function to summarize the kinds of explanations available in culture which speakers draw upon to make sense of everyday interactions (Wetherell & Potter, 1988). Although not consciously defined by the speakers themselves, they evolve through the analysts' window of interpretation.

It is important to qualify the notion of a repertoire. Indeed, while repertoires appear as discrete units, theoretically they remain connected with the text (Frewin & Tuffin, 1999). Data does not come ready-made in pre-defined packages of repertoires but becomes conceptualized in this way during the process of analysis. Ideas present in data typically run together in complex and convoluted forms. Analysis of data involves identifying and distinguishing the separate structures that form repertoires. Discourse analysis, in particular, offers a strategy for achieving this.

At a very broad conceptual level, discourse clusters around the idea that participants could operate without becoming victims of harmful and deceptive acts by employing a number of strategies for keeping themselves safe. For the purpose of analysis, these kinds of constructions were defined under the category of a keeping safe repertoire. Interestingly, at other points in the interview the talk from a select number of participants' focused on the appropriation and acceptance of their acts of deception, which varied in consequential importance according to the nature of the situation. In these circumstances, participants worked to justify their actions in ways which showed that they operated outside actions deemed harmful and deceptive. In contrast to the keeping safe repertoire, such discursive constructions were defined under the category of a qualified deception repertoire.

The following analysis aims to address the issue surrounding management of deception online by people with disabilities. Particular emphasis is placed on looking at the kinds of linguistic resources deployed to maintain participants' protection from harm and deception, as well as those resources used to justify situations where participants themselves engage in (minor) acts of deception. The extracts presented have been chosen for illustrative purposes as they offer clear examples of the linguistic resources which our analysis aims to highlight. Each extract is supported by our analysis, which may be regarded as simply a form of reading of which many versions may be possible. In this regard, the partiality of our interpretation is fully acknowledged. The first part of the analysis focuses on the linguistic resources identified within the keeping safe repertoire.

Keeping Safe Repertoire

Within the repertoire of keeping safe, three key resources were identified: happens to others, stranger danger, and personal judgement. The ‘happens to others’ resource is based on the idea that others are susceptible to harmful consequences when they participate online because they do not engage in safe practices. In contrast, the ‘stranger danger’ resource deploys other people as the protagonists of harmful situations. Personal safety is ensured by exercising caution before interacting with strangers. ‘Personal judgement’ is about developing strategies for evaluating the character of other online users. The knowledge gained can then be used to judge the safety of an online interaction. The following section provides an example of the first of these three resources.

Happens to Others Resource

In this extract Daphne is responding to a question about whether there are disadvantages for people with disabilities using online technology.

There is always a danger of “less-worldly” individuals being scammed by dishonest people, but I guess that would apply to the population at large. I heard a story recently of a young disabled man who had a woman he had met on-line come and stay with him (from overseas) and it was a disaster. I guess some pwds [people with disabilities] may be more susceptible to “chat up” lines. I'm not, I have enough people in my “real” world to keep me occupied! (Daphne)

This account highlights how harmful consequences associated with engaging in online interactions happen to other people: namely, those who are “less-worldly.” Daphne provides an example of a young disabled man's experience, which ended unpleasantly after he decided to extend his online relationship into an offline context. Although no reason is given for why the relationship ended in “disaster,” there is an assumption that the man is a victim of another's manipulation involving “chat-up” lines. The account positions the young disabled man as innocent, naïve, and vulnerable because he is willing to open his “real” world to someone he has met in an online setting. This lack of worldliness exposes him to being taken in by dishonest people.

In contrast to the young man's gullibility, Daphne remains untouched by such harmful consequences because she chooses not to extend relationships beyond the online context. Having enough people in Daphne's “real” world strengthens her reason for not engaging in offline relationships. She does not rely on the medium to foster social connectedness as she is satisfied with her offline networks. The account suggests Daphne holds a conceptual distinction between interactions occurring offline (which may possess some kind of realist grounding) and interactions occurring online (which are somehow less real). This leads to another assumption that those who are not grounded in reality are also vulnerable to scams and other dishonest practices. In this case deploying the online medium to foster relationships in the “real” world is considered naïve as indicated by the example of the young disabled man.

A further assumption can be deduced from Daphne's account, which positions those who choose not to be taken in by “chat-up” lines or dishonest actions as worldly. By default, Daphne's behavior places her outside the category of “less-worldly” individuals in which her decision to separate online interactions from offline relationships positions her as worldly. This trait of worldliness removes her from harmful acts because she chooses not to develop offline expectations about online relationships. By differentiating between on- and offline interactions, Daphne maintains her personal safety. Choosing not to meet online acquaintances in person shields her from becoming a victim of others' manipulation. Engaging in a strategy which ensures Daphne's safety presents her as a responsible and competent online user without fear of being taken in.

Interestingly, Daphne's account constructs the proportion of “less-worldly” individuals online as no different from “the population at large.” This indicates that people with disabilities are no more susceptible to dishonest practices than any other group, which reduces the risk surrounding Daphne's online participation; having a disability does not disadvantage her as being a more vulnerable target of others' dishonest practices.

The following extract provides another example of the ‘happens to others’ resource embedded in the keeping safe repertoire. The extract comes from an interview conducted on ICQ and leads on from a response to a question about whether it is better not to mention disability online.

There are some people out there in the internet world who are really nice and others who aren't the sad thing is, some people with or without disabilities can open their heart out to the wrong person, I don't take it to that stage

Consistent with the previous extract, negative outcomes happen to other people, not Sarah. Mentioning that there are some people who are nice and others who are not helps counterbalance the possibility of becoming a victim. The danger associated with online interaction is positioned as equal to the benefits. This may function to diminish the risk attached to online interaction, placing the activity within a neutral zone. It also highlights the potential for good relationships and hence offers some justification for participating in the medium.

Sarah's account argues that it is the degree of engagement which is crucial to maintaining safe experiences online. Sarah's strategy involves not taking “it to that stage.” Presumably, taking it to that stage crosses a threshold over which it is no longer possible to operate safely. Like Daphne's account, Sarah's account indicates a distinction between the “Internet world” and offline contexts in the sense that online Sarah chooses not to engage in intimate disclosure. Not opening her “heart out” protects Sarah from being treated in an insensitive, or negative way.

Also consistent with the previous extract, this account precludes disability from being a factor in becoming a victim. Sarah indicates that the potential for danger happens to disabled and non-disabled alike. This neutralizes the risk of harmful consequences happening solely to disabled online users including Sarah, and may reduce the danger associated with her engagement in the medium as a person with a disability.

Both examples of the ‘happens to others’ resource construct a strategy for managing safety online. By maintaining a distinction between online and offline worlds in relation to the disclosure of intimate and personal details and the development of relationships, Sarah and Daphne protect themselves from experiencing harmful consequences. This safety mechanism also functions to present participants (including people with disabilities) as responsible and competent online users, who possess the knowledge and skills required to maintain their safety.

Stranger Danger Resource

While the ‘happens to others’ resource positions others as victims of dangerous situations, the ‘stranger danger’ resource positions others as having the potential to invoke danger. In the following extract, Bridgette highlights the degree of caution exercised when interacting online. She is responding to a question about positive experiences online. The ellipses contained within brackets signify the interviewer's request for clarification about Bridgette's choice not to visit chatrooms.

Most of my experiences are good … I don't visit chat rooms, I'm a little weary! (sic) […] Lets just say I wouldn't enter a chatroom full of strangers … it's the stranger danger concept … I don't want to be part of anyones porn fantasy… (Bridgette)

The account begins with the statement that Bridgette's online experiences are good because she does not visit chatrooms. She is aware of the potential for bad experiences and hence exercises caution in this regard. Bridgette elaborates on her reason for exercising caution in relation to chatrooms. Beyond the mere location, it is the other people inhabiting these places who are deemed dangerous. Chatrooms full of strangers present a risk to personal safety. There is an assumption that people who frequent chatrooms may indulge in deviant acts of fantasy and will prey on newcomers to be the victims of their pornographic activities. Bridgette is vigilant with respect to these other users, who have the potential to create deceptive and harmful situations. By avoiding the environments strangers occupy for deviant activities, Bridgette constructs her safety. The account functions to warrant Bridgette's participation in the online medium, because she avoids places where others could deceive her.

I don't really want to e-mail people that I don't know it's just nice to meet them first to know what they're like you know yeah like sometimes what I've heard like you know in the chatrooms and that they're not all your age some people might say oh um fourteen they're really fifty I want to meet you you know yeah I just I only e-mail people that I know yeah (Janine)

Janine talks about the danger involved in conversing with strangers online. She deploys a chatroom situation to emphasize the need to be cautious when emailing people she does not know. Consistent with the previous extract, chatrooms are constructed as environments where the potential for deception is high. There is an assumption that other chatroom users may be there to wield deliberate acts of deception. Janine exercises caution by only interacting with people she knows. This removes her from any risk of being deceived and constructs her safety when participating online.

The last example illustrating the ‘stranger danger’ resource focuses on the vulnerability associated with revealing a disabled identity to others as justification for avoiding specific kinds of online interactions. Sally-Ann is responding to a question about how she feels having access to other networks of people online.

I don't really access other networks of people online. I am cautious by nature and steer away from live two - way conversations on the net. My disability does not allow me to type at so called normal speed. My disability being a point of difference also makes me cautious. I accept it other people may not do. It is a common sence (sic) safety protection thing I suppose. (Sally-Ann)

The extract identifies a compelling need to be cautious of other people online because they may be intolerant of disability. Real-time conversations reveal Sally-Ann's inability to type at “so called normal speed.” This “point of difference” exposes her vulnerability as a disabled online user. There is a suggestion that other people's prejudice about disability will also jeopardize Sally-Ann's opportunity to participate online. As a result of other people's intolerance toward disability, she chooses not to engage in real-time conversations. This is considered a “common sense safety” strategy. Sally-Ann's account functions to protect and secure her vulnerability as a disabled participant by avoiding situations which have the potential to expose her physical difference.

All three examples of the ‘stranger danger’ resource construct a second strategy for managing safety online. By avoiding interactions where the potential for danger is high, inclusive of chatrooms full of strangers, conversations with unknown individuals, and real-time interactions, Janine, Bridgette, and Sally-Ann protect themselves from potentially harmful consequences. This safety strategy may function to defend participants' engagement in a medium where the potential for harm is high.

Personal Judgement Resource

This resource is organized around the idea that users possess an inner sense of knowing, which enables them to judge the character of others online. Developing good judgement constructs participants as protected from malevolent situations. The following extract is taken from an ICQ interview. Sarah is responding to a comment made by the interviewer about keeping safe online.

That is the burning question really! after talking to people online a few times I like to think I've got a sense of who they are and most of the time I can judge that by what we talk about

Sarah's account positions safety online as a serious matter, which she feels strongly about. The phrase “that is the burning question really” transforms the interviewer's comment about the need to maintain safety online into a very difficult question to which there are no clear-cut answers. The question Sarah constructs creates a useful rhetorical effect in which to position what follows. Sarah responds to the question she has posed by developing a strategy, which helps her gain a feeling of security in online interaction. After talking to others “online a few times,” she likes “to think” that she can gain an understanding of their character. The phrase, “I like to think,” shows that Sarah's ability to access an accurate reading of a person's character online is not foolproof. This acknowledges the fallibility involved in deploying personal judgement as a basis for developing trust online. Sarah does, however, justify her claim to gain “a sense of who they are” by looking towards a tangible resource on which to base her judgement, namely, what is talked about in conversation. The opportunity to interact is given as a test for developing trust.

The account does some work to dispel the belief that people cannot judge the character of others with any degree of accuracy. In contrast, Sarah's talk highlights that participants can develop strategies for evaluating others online and hence protect themselves. This distances Sarah from the potential of becoming a victim of deception and enables her to operate safely. The ‘personal judgement’ resource is constructed in another example from a tape-recorded interview. Karen responds to a question about whether being online fosters a sense of feeling more connected.

it's a trust there is there's lots of trust and you normally know the ones that you steer away from there's one particular person that used to call himself [name] and they change their names quite often and they will go in and stir the others up and you tend to know because you know by the words that they use so you just you don't just don't go into areas where you know that that will you know get them going (Karen)

Consistent with the previous extract, Karen draws on tangible resources to judge someone's online character. “You normally know” the ones to avoid suggests there are some very clear signals, which can be deployed as guidelines for identifying dubious online characters. Karen proceeds to list four behaviors which she relies on to identify potential perpetrators and hence protect herself. These behaviors cover (a) the particular names by which perpetrators choose to identify themselves, (b) the frequency of changing their names, (c) the stirring-up of other online users, and (d) the words they use. She goes on to mention that participants can avoid areas where they know these perpetrators are likely to cause trouble. Karen utilizes caution in this regard, which is also deployed in the ‘stranger danger’ resource. There is an assumption that if participants are armed with knowledge about such behaviors they can protect themselves from danger.

The opportunity to interact is deployed as a resource for developing trust online, which is also consistent with the previous extract. The amount of trust Karen can access through her online interactions suggests she is able to engage in secure experiences where she is removed from deception, or the threat of harm and danger. These secure experiences can be sustained by Karen's ability to rely on her own judgement in protecting herself from deception and danger. Protecting herself from danger by developing strategies to identify dubious characters, and the degree of trust engaged in online interaction, both function to affirm (in)directly Karen's online engagement.

To summarize this section, the keeping safe repertoire functions to allow participants a pleasant and secure engagement in the online medium. This is achieved through each of the resources, which construct participants' ability to deploy safety strategies. These strategies operate to protect participants from harm and deception, and hence to maintain their well being when participating online. Furthermore, while each resource functions to affirm the online engagement of people with disabilities, each one also highlights and acknowledges the malicious and deliberately deceptive acts other people engage in online. Ironically, some people with disabilities seem to take part in behaviors not dissimilar to these deceptive activities, which online users are compelled to protect themselves from. Hence, a concern arises surrounding how people with disabilities manage the issue of engaging in the very acts which they are so critical of in others. Their management of this complex situation is demonstrated in the following category of talk identified under the qualified deception repertoire.

Qualified Deception Repertoire

This repertoire is based on the idea that activities involving varying degrees of fabrication, either via a reconstruction of details, or through a deliberate withholding of information to suggest a different presentation of self, are separate from deceptive and harmful acts. Data organized within the qualified deception repertoire deploys two key resources: withholding information and reconstructing information. The ‘withholding information’ resource is built around participants' ability to choose not to disclose revealing personal information. The outcomes associated with making disability information available are drawn on to justify withholding information. ‘Reconstructing information’ is based on participants' active reconstruction of themselves and situations, which are justified through specific strategies. The next extract provides an example of the first of these resources.

Withholding Information Resource

This account is taken from an interview conducted on Internet Relay Chat (IRC). The extract developed out of an open discussion between the interviewer and participant about whether people are able to present a more accomplished identity online.

<DANIEL>of course, but reality soon sets in,

<DANIEL>someone can pretend to be a Bill Gates,

<DANIEL>if they can't do everything they say then they end up looking silly,

<DANIEL>I've been using IRC long enough to know that anyone can pretend to be anything they want, I've used female nicknames to see how others (males) respond to me.

<NATILENE>can you tell me about that?

<DANIEL>oh boy, yes :)

<DANIEL>OK

<DANIEL>Gulp

<NATILENE>no worries, I've interviewed male IRC users, who have talked about doing it, although somewhat reluctantly

<DANIEL>I've read lots of articles

<DANIEL>about how females are very marginalized

<DANIEL>and I've observed it too

<DANIEL>50 guys might be online in a chatroom

<DANIEL>2 females

<DANIEL>most of them (males) are talking to the females

<DANIEL>some flirting actively

<DANIEL>some just chatting

<DANIEL>I think it's interesting, that's all

<NATILENE>ok cool.

<NATILENE>um you also mentioned that people can pretend to be anything they want online, do you have anything else to add to that?

<DANIEL>if you are asking if I have ever pretended, no, but I have maximized my anonymity

<NATILENE>that sounds interesting regarding maximising on opportunities for anonymity - can you elaborate?

<DANIEL>ok…

<DANIEL>I have - very deliberately - withheld information I would usually give out,

<DANIEL>things like - age, sex, sexuality, disability, race, that kind of stuff

<NATILENE>right, and what made you decide to do that?

<DANIEL>mainly to see how I would be responded to, see different lifestyles, sometimes (as in disability) I just get tired of stereotypical responses.

This complex account works to exonerate Daniel from an early admission of deception. In the first part of the account, Daniel is responding to a question about whether people appear more accomplished online. Daniel argues that while this is possible there are limitations. These limitations are talked about in a realist context: “reality soon sets in.” The construction suggests a realist philosophical standpoint in which there is a grounded and measurable threshold situated in a given reality where it is no longer possible to maintain a fabricated identity. An example is illustrated with someone who pretends to be Bill Gates but is caught out when they overstate what they can do. This initial talk indicates that deceiving others about your identity online is foolish and unsustainable.

The focus of the conversation turns towards Daniel's experience and knowledge of IRC as he claims that anyone can engage in pretence. He then goes on to qualify this using his personal experience of pretending to be something he is not, as illustrated in the phrase, “I've used female nicknames,” although (arguably) he does this out of curiosity to find out how others, notably males, respond to him. At the request of the interviewer, Daniel proceeds to divulge his interest in fabricating gender, although with obvious reluctance as indicated by “oh boy” and “gulp.” He draws on the marginalization of females online and the different ways males interact with females as a point of interest. This interest has come from having read numerous articles written on the subject the findings of which can be backed up by his own observations. This constructs a very scholarly and scientific account of his reasons for presenting as female, and functions to remove him from any other motivations. The explanation ends with “I think it's interesting, that's all,” which indicates that there are no other reasons for fabricating his identity. Daniel's engagement with alternate identities is constructed within an experimental framework in which he is merely exploring new possibilities. (In a later section of Daniel's account, identity exploration extends to accessing possibilities to engage in an identity removed from disability.) It is acknowledged here that although this identity fabrication provides an example of the reconstructing information resource, which follows in the next section, it is juxtaposed with an example of the withholding information resource. An understanding of the complexity surrounding the withholding information resource in this dialogue is enhanced by the preceding discursive events.

The third section of Daniel's account works to undo the admission of deliberate deception mentioned at the beginning. It also works to illustrate the beneficial opportunities resulting from engaging in a different subjectivity. Daniel states that he has never pretended, which we interpret to mean that he has never presented a fabricated identity even though earlier he indicated assuming a female identity. He qualifies his actions by talking about using the medium purely as a means to maximize its capacity for creating anonymity, which requires identity details to be withheld. This functions to remove Daniel from engaging in deceptive activities. Instead, he is merely utilizing the online resources available. Daniel deploys an illustrative list of identity details that he can experiment with concealing. His gender (or “sex”) is now positioned as one among several he conceals, thereby undermining the suggestion that he withholds information for the purpose of switching gender, which was present in the initial dialogue.

In the final turn Daniel exonerates himself further from deceptive motivations for withholding information by deploying the stigma linked with his disabled identity as legitimization for concealing personal details (c.f. Bowker & Tuffin, 2002). Daniel's decision deliberately to withhold information about his disability is entirely justified because of the negative reactions he receives when his disability is accessible to others' perceptual fields, as indicated by the phrase, “sometimes (as in disability) I just get tired of stereotypical responses.” Being able to withhold details about his disability allows Daniel to access another subjective space, which he values. This positions his motivation for deliberately concealing identity details outside the realm of deceptive and harmful acts in order to protect his own psychological well-being and access a more enjoyable online experience. The benefits of engaging in an online identity removed from disability are also demonstrated in the next example as Shaun talks about the advantages of being online.

I think just being able to partake in a forum where disability (in terms of access to the environment) is just great. For example, when I email people, they don't know that I type with my feet so they're not distracted by that rather than what I have to say. (Shaun)

Shaun mentions the benefits for disability when participating online. The medium is cast as being facilitative of communication by being free of distractions resulting from the visual display of disability. Rather, the visual anonymity afforded by CMC neutralizes the physical differences, which have such profound effects on face-to-face interactions for people with disabilities. Consequently, the medium's ability to withhold information about disability allows Shaun to be evaluated purely on the context of his words rather than on how he operates the keyboard. We read Shaun's description of disability, “in terms of access to the environment”, as suggesting that participating online with a disability is great because it does not influence social interaction. Extracts supporting the second resource identified under the qualified deception repertoire follow.

Reconstructing Information Resource

In this resource, participants actively reconstruct themselves and situations. Engaging in this deception is justified through specific forms of legitimization. The extract below highlights the use of deception when engaging in online interactions with strangers.

It is nice to be able to sit in the privacy of my own home, go on-line & communicate with a stranger …. if my typing speed is queried I say I am new at this! On-line my disability is HIDDEN if I choose it to be. Why should I reveal such personal info when there is no need to? (Bridgette)

Bridgette is prepared to engage in deception to avoid revealing personal information about her disability. This is apparent when she responds to questions about typing speed in relation to being a novice online. Presumably, this incorporates the suggestion that she is still learning to type, as opposed to exposing her disability as the reason for her slow typing speed. This act of deception is legitimized because the online environment allows Bridgette to choose whether or not to expose disability. Hence, reconstructing her personal details enables Bridgette to exercise that right of choice. In addition, a minor act of deception may be justified because Bridgette is communicating with a stranger. Consequently, engaging in deception maintains her privacy as she converses from home.

The following extract, again, illustrates the way reconstructing information is managed. The example is drawn from a tape-recorded interview. Karen is talking about the support generated from a chatroom for people experiencing a chronic illness.

we'll fool around like I said to him I just get my jeep out of the backyard and [name] will go can I can I can I come where's this jeep you're hiding in our back yard and I just you know it was just a big you know they take the monotonous out of it and and it does help to ease the pain because everyone's fooling around but we're very honest towards each other (Karen)

The phrase “we'll fool around” suggests that the situation Karen describes is a playful engagement that happens in the context of this chatroom. The act of reconstruction provides an outlet for dealing with the monotony and pain of living with disability. Yet, the pretense is qualified because everyone is “very honest towards each other.” While deception is actively occurring online, the participants are able to separate themselves from associated negative connotations by constructing their behavior within a relationship based on honesty. This functions to absolve Karen and her associates from negative evaluations. The creative reconstruction of events is positioned as ethical and justified because it is occurring within the context of having fun and playing around, indicating that this kind of activity is a legitimate contextual space for engaging in fantasy while online.

Discussion

The analysis demonstrates how people with disabilities are managing the dilemma of engaging in a social space where there is potential for experiencing non-stigmatized identities as well as the capacity to create harmful and dangerous outcomes. On the one hand, the keeping safe repertoire enables participants to construct and, hence, access pleasant online experiences by providing strategies which position participants as being protected from harmful and deceptive acts. Participants can acknowledge the risks involved in online interaction without exposing their vulnerability. Identifying and deploying strategies for ensuring personal safety enhances participants' integrity as online users. Simultaneously, the qualified deception repertoire functions to extend participants' ability to operate beyond a keeping safe repertoire, where they can explore freely subjective experiences outside of a stigmatized identity. The justifications deployed to manage participants' engagement in acts of deception exonerate them from harmful evaluations.

The ‘happens to others’ resource identified in the keeping safe repertoire is based on the notion that harmful outcomes happen to those who assume a benevolent online environment. Participants expose their vulnerability when assuming they can manage social interactions online in a similar way to relationships that occur in offline contexts. By drawing a conceptual line between on- and offline interactions, thereby maintaining their exclusivity, participants construct a strategy for protecting their personal safety online.

The ‘stranger danger’ resource constructs another strategy for people with disabilities to participate online safely. In contrast to the previous resource, which positions others as victims of dangerous situations, the ‘stranger danger’ resource positions others as having the potential to invoke danger. This resource functions to affirm participation in a medium where the potential for harm is high. This is achieved by avoiding situations where participants could expose their vulnerability, which includes being deceived and presenting a disabled identity to others.

The final resource identified in the keeping safe repertoire is categorized as ‘personal judgement.’ It operates by utilizing an opportunity for social interaction as a test for developing trust in relationships with other online users. This opportunity to interact brings forth tangible resources for evaluating the character of others. As a result, participants retain their safety as well as an ability to highlight their skills in detecting deceptive users. This presents an interesting paradox with the ‘stranger danger’ resource where social interactions are positioned as the site of potentially harmful outcomes, and are, hence, to be avoided to retain participants' safety. Instead, the ‘personal judgement’ resource moves participants beyond the limitations of avoiding harmful social interactions, by allowing them to operate online with additional skills for scrutinizing others in potentially deceptive situations.

Each of these resources constructs a strategy for people with disabilities to operate safely within a context where there is potential for harm and deception. Each strategy also helps participants manage the dilemma of engaging in a medium where there is potential for social and personal empowerment as well as oppression. The strategies identified work to protect and separate participants from the malevolent activities that exist online. Together, these strategies function to retain the integrity of people with disabilities when choosing to engage in social interactions online. Ironically, however, the qualified deception repertoire also enabled participants to further manage the dilemma surrounding online participation without exposing their vulnerability as disabled people, or associating themselves with malevolent intentions.

The ‘withholding information’ resource embedded in the qualified deception repertoire is organized around participants' choice not to disclose revealing information about their identity (c.f. Bowker & Tuffin, 2002). This allows participants to protect themselves from the prejudice and stereotyping to which they are subject as disabled people in society. The prejudice associated with a stigmatized identity, which negatively influences social interactions, is deployed as justification for withholding information about disability online. This functions to distinguish people with disabilities from those who utilize the medium to withhold information for malevolent purposes.

The ‘reconstructing information’ resource provided another opportunity for participants to manage the dilemma of engaging in a medium where there is potential for personal empowerment and oppression. In this resource, participants deliberately construct themselves and situations in ways which have no basis beyond the online interaction. This is entirely justified because of the negative affects of operating within a stigmatized identity. For instance, Bridgette highlights her vulnerability when exposing her disability to a stranger as justification for a minor act of deception. Likewise, Karen draws on the frustrations of operating within a disabled subjectivity as reason for constructing a fictitious event. In both cases, the ‘reconstructing information’ resource allows participants to engage in deceptive activities (which are evaluated unfavorably in the keeping safe repertoire), without incurring the penalties associated with negative evaluations.

The resources identified under the qualified deception repertoire allow people with disabilities to utilize the medium fully while distancing themselves from deviant motivations. By operating within this repertoire, participants are also able to extend their online engagement beyond the restrictions and limitations of a keeping safe repertoire. This leads people with disabilities to access a wider range of subjective experiences and creative potentials not available in other contexts. Deploying the deceptive attributes of the medium for social and personal empowerment demonstrates how people with disabilities are successfully managing the dilemma of online participation where oppressive and empowering outcomes are possible.

The medium's capacity to problematize traditional notions of reality is highlighted in different ways throughout the data. For instance, much of participants' talk was constructed around issues of being able to judge accurately the integrity of online users and the activities in which these users choose to engage. Paradoxically, the integrity of online users with disabilities is retained through each of the resources identified under the keeping safe repertoire. The keeping safe resources construct participants as responsible and competent online users who are skilled in detecting and protecting themselves from the deceptive pitfalls of online interaction. Interestingly, the integrity of people with disabilities as online users is sustained even further when they appear to engage in the very behaviors of which they are so vigilant and critical. This is evident in the qualified deception repertoire where participants deliberately construct themselves and situations in ways which have no reality beyond the online context, yet are entirely justified because they protect people with disabilities from stereotyped judgements. Each repertoire enables people with disabilities to sustain the integrity of their characters while operating under different philosophical standards for constructing reality.

In the case of the keeping safe repertoire, participants preserved a “realist” belief in maintaining a strict correspondence between the management of identity within on- and offline contexts. Support for this belief may be indicated by participants' focus on ascertaining a realist dimension for constructing relationships and the “true” character of online users. Participants' discourse also reflected a need to protect and dissociate themselves from interactions, which had no basis beyond an online context. Under a keeping safe repertoire, participants demonstrated a strong desire to construct a realist benchmark for judging online behavior and for limiting the extent of their online participation. Jordan's (1999) reflections, involving the male psychiatrist's identity reconstruction, pose serious questions about the danger of retaining a one-to-one correspondence between real life and online identity. Not withstanding the severity of the case and its tragic affects on innocent people, many of the women who had experienced intense relationships with Julie felt deeply betrayed when the “true” nature of Julie's identity was brought to light.

In contrast, the philosophical orientation apparent in the qualified deception repertoire moves away from a realist standpoint, thereby avoiding the ethical dilemmas associated with a realist construction of knowledge. This repertoire transgresses the strict correspondence between constructing a realist standard for managing on- and offline events. Instead, justification for participants' actions turns toward a relativist framework in which the online medium is evaluated according to its ability to operate within an entirely different social context for presenting one's self. This context is removed from the constraints of physical reality, which limit the kinds of subjective experiences people with disabilities can access. Operating within a relativist framework enables people with disabilities to access alternative subjectivities and creative opportunities, which extend beyond a stigmatized identity.

The contrast between realist and relativist frameworks used for operating within the online medium present important implications for social interaction. Further, they offer some interesting implications for those researching online communities. For instance, it may be necessary to consider the way participants conceptualize online activities, and whether these constructions are similar to, or independent of a realist framework. Accordingly, either outcome will impact on the way participants conceptualize trust and honesty, and the degree of importance, or irrelevance placed on these issues in online interaction. Additionally, the kinds of rules governing behavior in many online communities may also be influenced by the realist or relativist standpoints held by that community and, or those in authority.

In addition to discussing the discursive construction of resources deployed by participants, it is worthwhile observing that the categorization of extracts by gender may conform to differences reported by researchers investigating gender issues online (Bruckman, 1996; Curtis, 1996; McCormick & Leonard, 1996; Spender, 1995). The exclusively female representation of extracts categorized under the keeping safe repertoire raises the question of whether disabled women are more likely, compared to their male counterparts, to occupy vulnerable positions online, which require them to construct mechanisms to avoid harm. Conversely, the comparatively high proportion of male respondents whose accounts are positioned under the qualified deception repertoire may strike a chord with the literature, indicating that men are more willing to operate outside conventional identity standards. The potential for inequity illustrated through the gender of the extracts' authors highlights support for the prevalence of structural hegemonies operating online. Technology is shaped by social, political, and economic forces. As many researchers have pointed out, computer technologies are socially constructed within a gendered framework, which advantages men's ways of knowing and understanding the world (Edwards, 1990; Papert, 1990; Perry & Greber, 1990; Turkle & Gersch, 1998).

In conclusion, the keeping safe and qualified deception repertoires presented important psychological implications for social interaction. Together, they enabled people with disabilities to manage successfully the dilemma of participating in a medium where communication offers the possibility of experiencing a non-stigmatized identity while also presenting the capacity to create harmful and dangerous outcomes. People with disabilities manage this dilemma by operating within two subjectivities. The keeping safe repertoire allows people with disabilities to access pleasant and secure experiences through strategies which construct participants as being safe from harmful and deceptive acts. This functions to protect participants' vulnerability as online users as well as affirm their engagement in a potentially insecure and deceptive space, thereby maintaining participants' online integrity. On the other hand, the qualified deception repertoire allows participants to manage the deceptive features of the medium for empowering outcomes. Participants can extend their engagement beyond the restrictions and limitations of a keeping safe repertoire by gaining full access to the constructive capabilities of the medium. The vulnerability associated with exposing a stigmatized identity justifies participants' actions. This allows people with disabilities to access subjective experiences freely outside of a disabled identity, without tarnishing their moral standing as online users.

Acknowledgements

We gratefully acknowledge the critical comments of Mary Breheny and two anonymous reviewers on earlier versions of this paper.

Ancillary