Broscience: Creating trust in online drug communities

Online community trust

Every day, people encounter more information than they can possibly use (Hilligoss and Rieh, 2008). The digitalization of everyday life has revolutionized people’s ability to gather and access information (Flanagin and Metzger, 2008), moving us towards a society based on greater information self-sufficiency and decentralization of authority (Lankes, 2008). Burbules (1998) argues that the decentralization of authority on the Internet creates a levelling effect that puts all information at the same level of accessibility, and thus all authors at the same level of credibility in the minds of Internet users. The impact of such information can be limited to learning, or it can be used to make decisions, and can affect attitudes and behaviour (Wathen and Burkell, 2002). However, the plethora of, and access to, Internet-based information also raise the issue of the credibility or quality of information found online (Metzger, 2007). As search engines have become the most prevalent means for information seeking online (Fallows, 2005), scholars have become concerned with people’s ability to critically approach the content they encounter (Hargittai et al., 2010). This ability, usually referred to as media literacy (Livingstone, 2004), is, however, seldom rigorously put into use (Flanagin and Metzger, 2000). As Hargittai et al. (2010) summarize, ‘many adolescents do not possess the expertise required to search the web efficiently or critically assess the credibility of what they find’ (p. 470). For novice Internet users especially, the lack of such literacy is problematic, as those who might benefit most from verifying online information are doing so the least (Flanagin and Metzger, 2000). These findings suggest that Internet users display a lack of critical sense and that they are overly trustful of online content. This could pose a danger to those seeking drug-related information on the Internet and further stresses the importance of exploring how issues of trust are resolved within online communities.

In the sociological sense, trust can be defined ‘as a state of favourable expectations regarding other people’s actions and intentions’ (Möllering, 2001: 404). In contrast to earlier psychological frameworks, which treated trust as a personal attitude, the sociological interest in trust was inspired by a societal shift towards increasing complexity and opaqueness (Sztompka, 1999). Sztompka argued that individuals faced a growing range of options and uncertainty and became more dependent on persons whom they did not know. Without trust, people would be paralyzed and unable to act. Therefore, trust became a necessary resource (Sztompka, 1999: 12–13). On the basis of these premises, Sztompka defined trust as involving a set of beliefs and expectations, based on a concept of primary and derived trustworthiness. The former concerned reputation, performance and appearance. The latter concerned the accountability of the trustee, pre-commitment and situational features (Sztompka, 1999).

While Sztompka (1999) saw trust as ‘the trait of interpersonal relations, the feature of the socio-individual field in which people operate, the cultural resource utilized by individuals in their actions’ (p. 14), scholars such as Möllering (2001) advocate a greater theorization of the foundations of trust. The empirical quest is usually to explore how a state of favourable expectation regarding other people’s actions and intentions can be reached in reality. Möllering (2001) argues that the transition from ‘good reasons’ to actually favourable expectation represents a proverbial ‘mental leap’ of trust that theories must not overlook (p. 412). Trust is therefore conceptualized as a mental process of the three elements: expectation, interpretation and suspension, where the latter is the mechanism which makes interpretive knowledge momentarily certain and represents ‘a point where our interpretations are accepted and our awareness of the unknown, unknowable and unresolved is suspended’ (Möllering, 2001: 414).

Despite Möllering’s (2001) solicitations, research on how users evaluate and assess online content is usually focused on perceptions of credibility and believability (Hargittai et al., 2010) and has tended to emphasize how individuals assess credibility in isolation, rather than considering group- and social-level processes (Flanagin and Metzger, 2008). The latter should receive renewed interest as the inherent bottom-up technology of the Internet promotes group dynamics, involving interaction and participation from those who contribute (Walsh, 2011). Henderson and Gilding’s (2004) study of trust in online friendships found that the Internet presented opportunities, facilitating communication, self-disclosure and risk-taking. They argued, following Sztompka (1999), that reputation, performance, pre-commitment and situational factors promoted online trust (Henderson and Gilding, 2004). Although Sztompka (1999) argued that the Internet limited trust because of its anonymity, he still cited Luhmann (1979: 33), when claiming that ‘the familiarity of the trustee is undoubtedly a vital factor’. Therefore, estimates of trustworthiness are most likely to occur in circumstances of closeness, intimacy and familiarity (Sztompka, 1999: 81). Boyd’s (2002) study of trust on the online auction site eBay offers the concept of community trust, where the rhetorical construction of community provides a foundation for trust between users. Thus, the sociotechnical factors involved in online communities are of great value when exploring issues of trust in an online-mediated setting.

Online drug-related communities

Although debated, the term ‘community’ is widely ascribed to an array of social formations online. First described by pioneers such as Rheingold (1993) and Turkle (1995), the interactive resources on the Internet arguably provided the basis for relationships increasingly created through online interactions and groups to form based on the free flow and circulation of information (Kahn and Kellner, 2003). As a result, the Internet has become a central place for drug users to interact and to gather user-relevant information, regardless of physical or temporal proximity (Barratt, 2011; Belenko et al., 2009; Griffiths et al., 2010; Murguía et al., 2007; Wax, 2002). Within such communities, drug discussions are found to be characterized by a concern for safety and harm reduction among users (Boothroyd and Lewis, 2016; Soussan and Kjellgren, 2014) and are framed in a way that privileges the pleasures of getting high (Barratt et al., 2014). This highlights the significant shifts in power offered by the emergence of online life as it has moved from a top-down, professionalized discourse of harm reduction (Irwin and Fry, 2007) to one that is peer-generated and user-led (Boothroyd and Lewis, 2016). Especially for activities such as drug use, such communal affiliation is likely to occur because of its subcultural character and need for mutual aid in reducing risk, which entails individuals to rely on, care about and share with others (Fine and Holyfield, 1996). However, the increased Internet surveillance and censorship have reduced the global availability or freedom to use online spaces for such activities (Barratt et al., 2013), thus fostering alternative spaces within the dark web for drug users to congregate, not only to buy drugs but also to form supportive communities and develop political drug-related activism (Maddox et al., 2016).

There is, however, nothing new about drug users creating communities or subcultures where they can share information and learn about drugs (Norman et al., 2014). Becker’s (1963) seminal study of the social construction of deviance provided a useful framework in understanding how groups of ‘outsiders’ offered internal support and thus provided members of deviant subcultures with protection and isolation from societal judgements. He also argued that deviant acts such as drug use were a result of social experiences, in which individuals learned to define the activity as pleasurable and desirable, where the associated meanings and dispositions for drug use were intrinsically social (Becker, 1953). Norman et al. (2014) argue that contemporary drug communities on the Internet facilitate similar group mechanisms such as those previously proposed by Becker. On the Internet, members can learn how to use various drugs, recognize their effects and enjoy their sensations, all within a virtual sphere that provides far more effective interaction and communication between drug users (Norman et al., 2014).

Whereas subcultures traditionally have been seen as contingent on physical space and geography (Cohen, 1955), the Internet offers a means to overcome such geographical barriers and an opportunity for people to seek meaningful social relationships online (Williams and Copes, 2005). This connectivity thus allows a myriad of subcultural groups to propagate and members to construct a wide variety of non-mainstream identities and communicative practices (Kahn and Kellner, 2003). Within such online subcultural communities, members may develop social bonds and gain insight into how their peers perform subcultural selves – promoting culturally bounded networks of people who come to share the meaning of specific ideas and practices through interaction (Williams and Copes, 2005). Online communities devoted to such subcultures may also provide members with a supportive milieu in which to work out personal issues (Clement et al., 2012; Klein, 1993; Melrose et al., 2011). Members adopt informal roles and norms, and they develop a sense of obligation towards the community (Averweg and Leaning, 2012), including symbolic processes of collective experience and cultural meaning (Fernback, 2007). Over time, this interaction develops into a discourse that structures the generation, activation and diffusion of these ideas and practices (Williams and Copes, 2005). Participation in an online community therefore includes participation in the narratives of that culture, a general understanding of its stock of meanings and their interrelationships (Richardson, 1990). As Sandberg and Tutenges (2015) argue, the ‘us’ and ‘them’ in such stories are critical as they constitute the glue that holds the culture together. They separate people into groups and generate feelings of similarity and group membership (Lamont and Molnár, 2002). This may facilitate concepts of shared belonging and provide members with a sense of collective identity (Jenkins, 1996). As such, communal boundaries are shaped by the subculture’s context and the cultural repertoires, traditions and narratives that members have access to – featuring categorization systems to differentiate between insiders and outsiders, and common vocabularies and symbols (Lamont and Molnár, 2002). Within subcultural theory, these differentiations are usually articulated as a struggle for authenticity (Vannini and Williams, 2009), in which levels of trustworthiness may be unequally distributed.

As Fine and Holyfield (1996) argue, voluntary communities provide the environment in which relationships can flourish and experiences and knowledge can be shared. This is magnified by the presence of risk and external dangers. The community therefore represents a cocoon to protect members from risks of the activity. However, group cohesion is a prerequisite, in which the establishment of trustworthiness becomes critical. Fine and Holyfield (1996) continue, arguing that ‘one must be socialized to risk and to competence, and the organization must establish procedures – formal or implicit – by which trustworthiness is created’, emphasizing that what began as sharing of knowledge ‘eventually becomes a basis for establishing tight-knit social connections, linked both to status claims and to emotional ties’ (pp. 24–26). However, the presence of an online arena magnifies these issues, with its peculiar challenges in creating stability and cohesion within a group of anonymous individuals.

While most online trust research has aimed at developing skills needed to critically evaluate online information (for an overview, see Metzger, 2007), few have embarked on a mission to explore how the group dynamics of online social life may affect notions of trust. This promotes a different angle; rather than facilitating a top-down perspective (what Internet users should do), the emphasis should be from a bottom-up perspective and thus explore how Internet users attach meaning to their online practices and how trust is performed and evaluated within online subcultural communities. This perspective views trust as highly interactional and negotiated (Fine and Holyfield, 1996), with emphasis on the interactive and communal processes. Following from this perspective, this article seeks to explore this gap and explore notions of trust among members of two Norwegian Internet drug forums in order to gain an understanding of the group and social-level mechanisms of online community trust.

Broscience, trust and ambivalence

When interviewing the forum members, one recurrent story dealt with their efforts to minimize the potential harms and maximize the positive aspects of drug use. Typically, their involvement on the online forums was fuelled by a wish to gather user-relevant information and thus be better prepared for the drugs they wished to consume. Relevant information usually revolved around dosages, expected drug effects and poly-drug use. In the online context, however, not everything was trustworthy. Although all of them used the online information to guide themselves through the ever-increasing jungle of available substances and their use, the information was not ready-made. As a result, the forum members described an ambivalence towards the value of such user-generated information.

Ole was one of the forum members interviewed. He was in his mid-20s and had spent a lot of time searching for drug-related information on the Internet. When interviewed, he explained his duality towards the sharing of online user-generated drug information:

Ole spoke about the online user-generated information with ambivalence. The duality between reliable and unreliable information made it difficult to manoeuvre in this virtual landscape. Speaking of ‘broscience’ in a sarcastic and almost condescending manner, Ole touched upon the skills he had acquired over the years online, which enabled him to evaluate the quality of the information while also emphasizing the collective effort to minimize incorrect information. Without downplaying the importance of user-relevant information, especially for minimizing the risks involved with drug use, Ole explained how one had to master the necessary skills in order to separate the useful information from the useless and potentially dangerous.

Shared narratives and boundary work

When explaining the benefits and their initial fascination towards the forums, the interviewed members told stories about harm-reduction initiatives based on personal drug experiences. The lived experiences from those providing information were deemed unbiased and trustworthy and offered the type of user-relevant information they searched. On the forums, they received detailed information on how to use specific drugs, what dosages they should use and they got advice on what to avoid. The online discourses were therefore a result of users sharing their experiences, as well as discussions and questions – all encompassed by shared norms of free access to information, in which drug use was made safer. This collective project was also discussed in stark contrast to information provided by public health officials, which was deemed biased and untrustworthy. Eric explained:

I could have read official statistics, numbers or something like that. However, that’s not legit enough. I was looking for concrete user information, you know. Like, how can I use this drug in the safest possible way? If you want that kind of information, you have to listen to someone who’s actually tried it before.

Describing the official drug information as ‘not legit enough’, Eric clearly valued information provided from the bottom-up. The result of the community members’ evaluations and advice represented an alternative discourse, where drug use was reinterpreted into something responsible as long as one possessed sufficient information. Regardless of the substances they used, the interviewed members had more confidence in the information based on other users’ experiences. In contrast, the official information was discussed as biased and solely concentrated on the negative effects, thus weakening its legitimacy within the online scenes. The online communities therefore represented an alternative, based on the shared experiences of the community members, in which their alternative drug discourse generated a collective project, defined in contrast to the dominant, official discourse. This type of boundary work, where the community members defined their online activities as a distinction, was powerful as it provided members with a sense of a collective identity. Their notions of ‘us’ and ‘them’ in the constant struggle over drug-related information and those providing it separated people into groups and helped generate a sense of similarity within their online communities. These boundaries were constantly shaped and negotiated within the forums, and developed into shared narratives that the members had access to. These stories defined their core projects while also defining boundaries towards others. This was articulated more specifically in the interview with Thomas:

I’m confident that an honest community like this can provide just as valuable information on drug use as official sources. I mean, this is first-hand experiences. There is no taboo. You only get the reality. People can ask questions and get advice on how to do things safely. If someone is deemed too young, they are encouraged to wait, you know. People are responsible.

Thomas emphasized members’ responsibility and claimed that ‘you only get the reality’ when dealing with first-hand experiences. Enthusiastically, he described the honesty and collective effort within the forums, which in turn resulted in a safer and more informed drug use. This shared community narrative touched upon the fundamental story of the forums, which basically was a story about trust. The inherent logic within the forums entailed cooperation between members, in order to ensure that the information was trustworthy. This form of trust resonates in Sztompka’s (1999) theory of derived trust, which emphasizes the context in which the trustee operates. These situational features could facilitate increased trust as there are ‘features of the setting in which the relationship takes place, that exert general facilitating or constraining pressure on the trusters to grant or withdraw trust’ (Sztompka, 1999: 93). This also highlights the characteristic social mechanism within groups of ‘outsiders’ (Becker, 1963) – where members offer support and mutual aid to avoid negative outcomes (Fine and Holyfield, 1996).

The shared narratives within the forums encompassed stories on drug-related harm reduction, based on the collective effort of members. This local story was powerful, as it provided members with an alternative drug discourse, defined in contrast to the dominant one, generating a sense of shared belonging and collective identity. These situational features promoted community trust, based on the repetition of community narratives that reinforced its ideals. As such, the narratives involved in the online communities seemed more important than adult-provided drug education, media or any other outside message because of the higher credibility of stories from those one had a social connection with. These narratives highlight the structure of knowledge and importance of boundary work within online drug communities.

Subcultural competence and authenticity

While the shared narratives and boundary work generated a sense of collective identity within the forums, they did not explain the initial basis on which members were deemed trustworthy (or untrustworthy). In order to gain trust within the communities, members had to display an appearance on which to be trusted. However, within the forums, there were limited cues for members to evaluate each other. Apart from profile pictures and online aliases, which it was difficult to assess trust from, language was the key ingredient for evaluating members’ appearance and thus their level of competence. As the drug-related information usually entailed highly technical and medical-like descriptions, specific argots developed within the communities, which the members adopted and mastered to a varying extent. Peter explained how he evaluated members’ trustworthiness based on their writing skills:

It’s sometimes difficult to assess the trustworthiness of what you read online. However, you can clearly see the quality of what is written, I mean if someone writes good or bad. You can look at the forums, some of the posts; it’s as if a doctor had written it. Then you get an idea whether the person is trustworthy or not, even though you don’t necessarily double-check the information.

In order to build a trustworthy appearance within the forums, members had to write in a convincing matter. Although it varied between the forums and the drug use that the members engaged in, the quality of the forum posts was still evaluated on the basis of members’ ability to write in an ethnoscientific language and with the site-specific vocabularies that were characteristic of the community. This usually entailed various medical descriptions covering the drug’s effects, combined with specific user knowledge, in which members displayed their formal expertise as well as their subcultural competence. Following Peter’s arguments, the written words were sufficient to build trust as they communicated a drug-specific expertise, as well as an understanding of the subcultural codes deemed valuable within the forums. This demarcation of authenticity was crucial in members’ ability to evaluate trust.

Sztompka (1999) highlights the value of performance and appearance in evaluating trust when the record of past deeds is suspended. However, it does not allow for a judgement as to whether trustworthy performance is continuous; it rather focuses on the present conduct and the criteria on which it is evaluated. In this respect, the initial criteria on which members evaluated each other were their ability to perform a written language that was within community standards. Their categorization systems differentiated between insiders and outsiders, based on the vocabularies and symbols that were dominant within the communities.

When interviewed, the forum members had no clear definitions of what constituted a ‘good language’ on which they deemed other members to be trustworthy. Rather conversely, they emphasized those whose language did not meet community standards. In their narratives concerning trustworthy conduct, they described performances that were deficient and incidents with forum members who violated norms of trustworthy conduct. Jon fiercely described how those posting ‘bad’ information upset him:

I’m very interested in everything involving chemistry, biology and science. I gradually dedicated myself to it. I’m intellectually interested and on the Internet, one can read everything that exists. There are plenty of forums and I began to read published research articles. Therefore, I knew what it was all about. However, there are many who do not, especially the young, they do it completely wrong. They don’t know anything. When I see their posts on the forum, they obviously haven’t done anything right.

While exalting himself, Jon also criticized other, less competent members. Describing them as unknowing and lacking competence, he emphasized the unilateral distribution of power, which became apparent in their inability to communicate within community norms. This struggle over authenticity conceptualized the social dimension in which Jon claimed insider status. As a marker of status or method of social control, he referred to a set of qualities that members had come to agree represent an ideal while also describing the power relations within the forums. In Peter’s and Jon’s descriptions, the key was to appear trustworthy by mastering a well-written language, in which medical-like descriptions were a central component. Thus, forum members vaguely described those who were ‘in the know’ and with an embodied competence in which their online appearance was deemed trustworthy, facilitating an uneven distribution of power within the forums, which further highlights the complex social mechanism involved in online community trust.

Online reputation

When explaining the organization within the forums, the interviewed members emphasized the evolving social structures and hierarchies between participants. Even though cooperation and exchange of information were common denominators, competition between members was central to the way knowledge was challenged and thus evolved over time. Members constantly negotiated and contested the dominating discourse, based on their resources and positions within the community. Over time, different roles developed, based on previous contributions and feedback. Some even got elevated, on the basis of years of dedication to forum. As Simon explained, some members were ‘glorified because they had contributed with valuable information over a long period of time’. In this way, they had earned a dominant position within the community and become influential in shaping the evolving online discourse. In terms of trust, these forum members had a prominent position. Their level of experience, combined with the feedback from the community, elevated their online profiles. In both forums, there were features that made it possible to react to each individual post, similar to a ‘like’ on Facebook. Whenever a member posted something, their accumulated number of posts and likes were displayed together with their profile pictures and aliases, making it easy to identify their level of involvement on the forum, as well as the ratio between number of posts and likes. In debates and disagreements on the forums, highly ranked members earned respect by virtue of their online profiles. Some even had a celebrity-like status and were often referred to in the interviews as those providing valuable and trustworthy information. When browsing the online discussions, the interviewed members usually held in high regard those with a history of providing valuable information. André explained:

You know, there’s always someone who’s been on the forum for several years, with thousands of posts. If you go online and read, you’ll see people mentioning ‘vets’, then they mean veterans. They usually have a lot of experience and they get a special status on the forum. I usually rely on them … I mean, when they have over a thousand posts, you know that they have a lot of experience and then it’s easier to trust what they say.

Within the online drug forums, different roles developed based on each individual member’s contributions and the community’s feedback on their posts. They were assigned different functions and positions based on the resources they added to the network. In this way, some members earned power that gave them a particular position within the forum’s social space. Their reputation entailed expectations and predictability, and over time and with repeated successful interactions, their trustworthiness grew stronger. As such, the interviewed forum members’ expectations of the value of the information were influenced by the experiences they brought to bear on their expectations. Following André’s reasoning, it was easier to accept the information provided by those who already had earned trust within the community because their previous contributions provided a foundation on which to be trusted.

This level of reasoning among the interviewees is in line with Sztompka’s (1999) concept of primary trustworthiness and reputation. Describing it as the ‘record of past deeds’ (Sztompka, 1999: 71), Sztompka argues that a history of trustworthy conduct makes it easier to trust the person involved. As the forum members are visible to one another in their performances of social roles, trustworthiness accumulates over time. Sztompka also states that this form of trust is contagious, in that people imitate the trust of significant others, thus highlighting the bandwagon effect of trust. This cooperative element facilitates a further understanding of the social mechanisms involved in online community trust and highlights the notions of community and culture, and its influence and importance for trust in an online-mediated setting.