Internet anonymity provides democratic potential through allowing a greater sense of openness, equality, and freedom to express controversial view points. This allows for an ‘unprecedented’ wider public sphere participation from individuals who otherwise have ‘had little opportunity to participate in public debate’ (Rowland 2006: 519). Anonymity means that views can only be considered in terms of rational content: neither property nor privilege can be used to influence weight (Rowland 2006: 532). However, anonymity also allows for disassociation from individual identity and the creation or appropriation of other identities. It is frequently assumed that anonymity breeds incivility, damaging the Internet’s real potential for effective public sphere discourse. In the recent flurry of moral panic over Internet anonymity, trolling has become the mass media’s catch-all term for any unpleasant, hostile, or offensive comments made anonymously on the Internet. However, these attacks on trolling are perhaps misguided. This paper will delineate and examine some potential implications trolling has for effective public sphere discourse online.

The public sphere acts as a mediating social space ‘between the state and society’ (Delanty 2007: 3721). If private liberty is to be valued, then society must permit a plurality of heterogenous views. However, heterogeneity increases opportunities for marginalization of non-normative views, problematising democracy as ‘collective self-government’ (Peters 2008: 39). Ideally, for government to be truly democratic, all individuals must voluntarily recognise and accept the decisions made (Peters 2008). Peters (2008) suggests that through communications of a certain type, the public sphere can reach this position voluntarily. Habermas (1996: 361-2) states that the public sphere does not enact policy changes, rather, it serves to influence the actions of political institutions which are legitimated through serving the democratic common interest of all. Thus, the basic function of the public sphere is to rationally negotiate and locate the common ground for the ‘practical problems of collective life’ that require or are subject to policy (Peters 2008: 37). Through this expression of solidarity, pressure is exerted on political institutions to legitimate policy by making changes that reflect the common interest.

It is important to note that public sphere discourse does not only exert influence on the formal systems of the state, but also upon the informal aspects of everyday society (Peters 2008). A second-person perspective must be taken in order to attempt to frame propositional attitudes in terms of common interest claims and understand the common interest claims of others (Habermas 1996). This unique perspective creates a relationship to the common ground of ‘both nearness and remoteness simultaneously’ (Simmel 1971/1908: 147). As a result, the varying viewpoints create a Simmelian strangeness which forces exposure and reassessment of latent dogmatic assumptions.

For Peters (2008: 37-8), effective public sphere discourse requires three fundamental features: equality and reciprocity; openness and adequate capacity; and a specific discursive structure. To represent the common interest of all, each agent ‘capable of expressing themselves in public’ must be equally free to do so (Peters 2008: 37). Furthermore, each view must be afforded equal consideration. This does not mean each view has equal value, rather that the value of each view is determined only in accordance with its rational weight. If each agent is to have an equal opportunity to both speak and be heard, then they must also reciprocate this right by listening equally to every other agent. Therefore, all views must be openly accepted and only through public debate can their validity be determined. This assumes that public sphere participants have an adequate capacity to rationally consider and equally debate the common interest of each view. The debate must follow a discursive structure. All views, whether disagreements or proposals, must be expressed as falsifiable arguments, rather than opinions, that demonstrate the link to the common ground between all agents. To preserve rational integrity, logical fallacies, such as ad hominem attacks, cannot be allowed. There must then be a ‘mutual respect’ that this rational integrity will be upheld (Peters 2008: 37-8).

Habermas (1996: 367) states that associations help to ‘distill’ the common interest of their members into a form suitable for effective public sphere discourse. However, it is perhaps a daunting task to try to locate and unite a substantial number of likeminded individuals who are dispersed external to personal milieux (Tocqueville 1999/1840: 92-4). Tocqueville (1999/1840: 92-4) argues that the utility of newspapers is of fundamental importance in achieving this goal. Newspapers give the individual the power to present a thought to large number of people simultaneously. In short, newspapers allow likeminded individuals to reach out to each other and realise the interconnect of their common interests so as to form associations (Tocqueville 1999/1840: 92-4). However, tendencies towards political parallelism and clientelism in the Australian media negatively affect its political functions both as marketplace of ideas and fourth estate providing objective and unbiased critique on politically important issues (Jones and Pusey 2010: 456-7). Nevertheless, new media forms, such as the Internet, offer associational utility in ways Tocqueville perhaps could not have imagined. Moreover, it is frequently hoped that the Internet can be utilised to overcome many of the deficiencies and shortcomings of traditional news media, and ‘herald new possibilities for political participation’ (Bohman 2004: 131).

The mass media use of the term “trolling” has come to include, arguably incorrectly, direct threats, intimidation, malice, aggression, and deliberately abusive cyber-bullying (for instance, see Hildebrand & Matheson 2012; The Times 2012; Herald Sun 2012). For the purposes here, I wish to draw a distinction between trolling and these other behaviours. This is not to discount the problematic nature of abusive comments that have seemingly become commonplace on much of the Internet. Nor is this distinction to ignore the significant overlap that often occurs between trolling and these other behaviours. Rather, this distinction is to acknowledge the considerably obvious detriment these other behaviours have for rational discourse both online and offline. In logic terms, these can be dismissed as ad hominem attacks that lack any argumentative purpose. What I wish to explore here is whether “proper” trolling can have any relevance or value to rational and communicative public sphere discourse.

The best way to understand the definition of “proper” trolling, is perhaps by comparison with flaming. Flaming can be defined as making a deliberately offensive or inflammatory remark directed towards a person or group, usually as ‘an impulsively angry response to a previous message or a perceived breach of Internet etiquette’, in which satisfaction is derived directly from making the remark (OED Online 2012a). For instance, an offline equivalent is when I yell an insult at a driver who cuts me off in traffic. This is by no means a comprehensive definition. However, it serves the purpose for the distinction I am attempting to make. Trolling, on the other hand, involves making a remark designed to incite a response. For something to be a troll it must involve a bait. This usage originates in the fishing terminology that means ‘to trail a baited line behind a boat’ (OED Online 2012b). Hence, to troll is ‘to post a deliberately erroneous or antagonistic message … with the intention of eliciting a hostile or corrective response’ (OED Online 2012b). In this sense, trolling is closely connected to flaming in that trolling often provokes “flame wars”. It is the motive of inciting a response which differentiates trolling from flaming: ‘a troll who gets no response has failed’ (Hardaker 2010: 232). The question then is what, if any, latent democratic potential exists in this motive.

On the surface, trolling seems to lack near all of the necessary features for public sphere discourse as outlined by Peters (2008). The very purpose of trolling seems to be to derail the discursive structure of rational debate. Trolls, by definition, intentionally use logical fallacies to incite responses. Their arguments are not presented as falsifiable, rather they are false to begin with. In this way, trolls cannot be seen to be extending the mutual respect the discursive structure requires. Additionally, the intentional irrationality seemingly indicates that there is no reciprocity of equal consideration to claims on the basis of rationality. There is then no demonstration of adequate capacity. Moreover, trolling undermines the mutual respect which preserves rational integrity. The very goal of trolling is to escalate arguments into irrational flaming through antagonistic behaviour whilst appearing sincere. Donath (1999: 45) states that ‘the troll attempts to pass as a legitimate participant, sharing the group’s common interests and concerns’. Dahlberg (2001: n.p.) suggests that this ‘identity deception’ negatively affects trust within the group, causing participants to be suspicious of others and wary of putting their own views forward. In this sense, whilst not actively denying participation, trolling can have a negative effect on freedom to express individual views. Nonetheless, trolling cannot be dismissed entirely on face-value.

Papacharissi (2002: 23) states that online interactions are often dominated by only a small percentage of actual users. Furthermore, there is a degree of Internet access stratification whereby certain people could be precluded from online interaction due to a number of reasons such as technical proficiency, age, disability, income and so on (Buchstein 1997; Papacharissi 2002: 21). Ruiz et al. (2011: 475) observe that on many news websites, users rarely comment more than once and so are not really participating in the necessary back and forth that public discourse requires. Moreover, Papacharissi (2002: 17) argues that making a single political comment on a website often gives users a false sense that they have participated in actual public discourse. Trolling can provoke users to engage in further interactions which force them to back up their claims with arguments demonstrating the common interest. Furthermore, as trolls often take counter positions, they then could inadvertently voice otherwise unspoken concerns.

Internet anonymity allows for disassociation from individual identity and the creation or assumption of another (Rowland 2006). Rowland (2006: 533) states that this sense of ‘deindividuation’ can allow individuals to ‘act in a disinhibited fashion’ and ‘exhibit extremes of behaviour’. Massaro and Stryker (2012: 418) point to a study which suggests ‘that selective exposure to ideologically extreme positions can, in fact, produce extremism’. Thus, trolling often has a self-feeding effect which “fans the flames”, creating increasingly polarised extremist positions among participants. Trolling proves problematic as it easily provokes a continual divergence away from seeking the common ground and towards incivility. Moreover, Papacharissi (2002: 22) notes how anonymity manifests as a lack of accountability. Even if trolling somehow avoided extremism and found the common ground, that common ground would lack legitimacy as one of the vertices of difference across which it is stretched is nonexistent in terms of accountability.

Nonetheless, Massaro and Stryker (2012: 419-20) argue that true anonymity on the Internet is a rare occurrence. Most websites require users to provide certain details to create accounts and users can often be identified according to the IP address assigned to them by their Internet Service Provider. Moreover, a combination of automatic moderation software and real-life moderators allows websites to monitor for comments that attempt to derail discussion or involve incivility (Massaro & Stryker 2012). Ruiz et al. (2011) also note that many news websites allow for user moderation where users can report comments that breach the terms of service. Ruiz et al. (2011) observe that news website terms of service generally seek to encourage intelligent debate and discussion via principles similar to those outlined by Peters (2008) for effective public sphere discourse. If the purpose of trolling comes from inciting a response, then moderation forces trolls to become more enterprising and creative so as not to be detected. It could come to a point where, for trolls to be successful, they must actually be part of civilised political debate. The fundamental difference being that trolls seek to play devil’s advocate rather than expressing authentic claims to common interest.

The lack of authenticity and accountability in the troll’s ersatz position can perhaps prove very problematic for a public sphere seeking to influence policy. If the trolls do not represent real claims, then there must be an identical authentic claim to fill the void left by the troll. However, this is a problem that perhaps applies equally to anonymous online discourse in general rather than being limited specifically to trolling. It is frequently argued that the Internet allows people to try on a number of fragmented virtual selves and liquid identities (Burkart 2010; Turkle 1995). Furthermore, Miller (2011: 177) points to Goffman’s (1959) dramaturgy metaphor to suggest that the self others see, regardless of whether it is online or offline, is always a performance that involves some sense of withholding a back-room identity. Miller (2011: 176-80) argues that the offline/online divide cannot provide a simple distinction of more and less authentic selves. In this regard, anonymous online performances can potentially be equally representative of a stranger’s true self as the impression given in an encounter with the same person on the street.

Massaro and Stryker (2012: 414-9) state that the available research is conflicted on whether Internet use increases homogeneity or heterogeneity of political views users are exposed to. They note that, in general, individuals find divergent viewpoints threatening. Rather than expend the extra cognitive energy required to process different views, individuals instead seek simplicity. However, this does not mean individuals necessarily avoid opposing views or disengage when exposed to them, only that they are unlikely to seek them out. Massaro and Stryker (2012: 415-9) contend that the wide range of ‘discourse options’ on the Internet provides greater opportunity for users to seek digital enclaves where their biases are reproduced. Nonetheless, this diversity also creates the possibility of inadvertent exposure to challenging viewpoints (Massaro and Stryker 2012: 419). Trolls add to this diversity by intentionally seeking to challenge other views.

It is possible that trolls could infiltrate homogenous, populist type groups and internally provoke them to examine dogmatic biases. Populism is an anti-elitist, exclusionary, “we the people” type politics which frequently lays blame for its troubles on weaker minority groups (Wells 1997; Berlet & Lyons 2000). Populism misappropriates free-speech rights to give value to its immediate “gut-feeling” opinions whilst dismissing any criticism as undemocratic and elitist (Wells 1997). Instead of using rational argument, populism uses loudness of voice to discount other views. Populism’s lack of accountability is dangerously undemocratic as it does not reciprocate equal consideration (Wells 1997). Thus, populism undermines public sphere legitimacy by refusing to follow the rules.

Through constructing an identity of seemingly sincere participation, the troll becomes part of the in-group. As a result, trolls could act in a way that provokes debate inside populist minded groups. That is, whilst trolling may not necessarily fulfil the requirements for effective public sphere discourse, it may still provide democratic potential through infiltrating other, politically influential, populist type groups who also refuse to follow the rules of public sphere discourse. By infiltrating the group, trolls avoid the “us and them” type problems where challenging views are dismissed as outsider elitism. Trolls could provoke populist groups to rationally legitimate the common interest of their own claims. By doing so, the claims are unwittingly transformed into following the discursive structure of the communicative public sphere. In a similar way, trolling could also subvert environments of political parallelism. However, trolling may also have the opposite effect whereby, as people become more aware of trolling, the label of “troll” could be used to dismiss authentic in-group dissent. Nonetheless, the most effective and successful trolls are those which avoid detection altogether. So the best trolls are cunning enough to avoid any relegation to out-group status.

Outlined here are some potential implications, both positive and negative, trolling has for Internet public sphere discourse. Research into this area is particularly scarce and much more is needed in order to draw more concrete conclusions. If my argument has been successful, what has been demonstrated here is that trolling should not be immediately dismissed as entirely uncivil and lacking any democratic potential. Rather, the very nature of seeking a response and initiating actual dialogue is something that is conspicuously absent on many parts of the Internet. This is definitely not to suggest that trolling is some saintly panacea. Only that the mass-media focus on trolling as the undemocratic root of all evil is misguided and far overstated. Instead of scapegoating trolling behaviours, more energy should be expended on ways to promote rational public sphere discourse and strategies to deal with anonymously abusive Internet users.

References

Berlet, C. & Lyons, M.N. (2000), ‘Introduction’, in D. Kellner (ed.), Right-Wing Populism in America: Too Close For Comfort, New York, The Guilford Press: pp. 1-18.

Bohman, J. (2004), ‘Expanding Dialogue: The Internet, the public sphere and prospects for transnational democracy’, The Sociological Review, 52: 131-155.

Buchstein, H. (1997), ‘Bytes That Bite: The Internet and deliberative democracy’, Constellations, 4(2): 248-263.

Burkart, G. (2010), ‘When Privacy Goes Public: New media and the transformation of the culture of confession’, in H. Blatterer, P. Johnson and M. Markus (eds.), Modern Privacy: Shifting Boundaries, New Forms, New York, Palgrave MacMillan: pp. 23-38.

Dahlberg, L. (2001), ‘Computer-Mediated Communication and The Public Sphere: A critical analysis’, Journal of Computer-Mediated Communication, 7(1): n.p.

Delanty, G. (2007), ‘Public Sphere’, in G. Ritzer (ed.) The Blackwell Encyclopedia of Sociology, Massachusetts, Blackwell Publishing: pp. 3721-2.

Donath, J.S. (1999), ‘Identity and Deception in the Virtual Community’, in M.A. Smith and P. Kollock (eds.), Communities in Cyberspace, London, Routledge: pp. 29-59.

Goffman, E. (1959), ‘Performances’, The Presentation of Self in Everyday Life, New York, Anchor Books, Doubleday: pp. 22-30.

Hardaker, C. (2010), ‘Trolling in Asynchronous Computer-Mediated Communication: From user discussions to academic definitions’, Journal of Politeness Research, 6: 215-242.

Habermas, J. (1996), ‘Civil Society, Public Opinion and Communicative Power’, Between Facts and Norms: Towards a Discourse Theory of Law and Democracy, Cambridge Massachusetts, The MIT Press: pp. 359-387.

Herald Sun (2012), ‘Charlotte Dawson: How the cyber trolls beat me’, The Herald Sun, 3 Sept., at http://www.heraldsun.com.au/news/charlotte-dawson-how-the-cyber-trolls-beat-me/story-fnbk7kwa-1226463900647, accessed 10 Nov. 2012.

Hildebrand, J. & Matheson, M. (2012), ‘Twitter Makes Moves to Prevent Online Trolls’, The Daily Telegraph, 15 Sept., at http://www.dailytelegraph.com.au/news/sydney-news/twitter-moves-on-trolls/story-e6freuzi-1226474468650, accessed 10 Nov. 2012.

Jones, P.K. & Pusey, M. (2010), ‘Political Communication and ‘Media System’: The Australian canary’, Media Culture Society, 32(3): 451-471.

Massaro, T.M. & Stryker, R. (2012), ‘Freedom of Speech, Liberal Democracy, and Emerging Evidence on Civility and Effective Democratic Engagement’, Arizona Law Review, 54: 375-442.

Miller, D. (2011), ‘Fifteen Theses on What Facebook Might Be’, Tales from Facebook, Cambridge, Polity Press: pp. 164-204.

OED Online (2012a), ‘Flaming, n.’, Oxford English Dictionary Online, Oxford University Press, at http://www.oed.com/view/Entry/71033, accessed 10 Nov. 2012.

OED Online (2012b), ‘Troll, v.’, Oxford English Dictionary Online, Oxford University Press, at http://www.oed.com/view/Entry/206615, accessed 10 Nov. 2012.

Papacharissi, Z. (2002), ‘The Virtual Sphere : The Internet as a public sphere’, New Media Society, 4(1): 9-27.

Peters, B. (2008), ‘The Meaning of the Public Sphere’, in H. Wessler (ed.), Public Deliberation and Public Culture: The Writings of Bernhard Peters 1993-2005, New York, Palgrave: pp. 36-42.

Rowland, D. (2006), ‘Griping, Bitching and Speaking Your Mind: Defamation and free expression on the Internet’, Penn Street Law Review, 110: 519-538.

Ruiz, C., Domingo, D., Micó, J.L., Díaz-Noci, J., Meso, K. & Masip, P. (2011), ‘Public Sphere 2.0? The democratic qualities of citizen debates in online newspapers’, The International Journal of Press/Politics, 16(4): 463-487.

Simmel, G. (1971/1908), ‘The Stranger’, in D. N. Levine (ed.), Georg Simmel: On Individuality and Social Forms, Selected Writings, Chicago, The University of Chicago Press: pp. 143-149.

The Times (2012), ‘Celebrities Advocate Abuse of Trolls’, The Sunday Times UK, 17 Sept., http://www.theaustralian.com.au/news/world/celebrities-advocate-abuse-of-trolls/story-fnb64oi6-1226475194355, accessed 10 Nov. 2012.

Tocqueville, A. de (1999/1840), Democracy in America, Volume II, Champaign Illinois, Project Gutenburg; Boulder Colorado, Net Library.

Turkle, S. (1995), ‘Aspects of the Self’, Life on the Screen: Identity in the Age of the Internet, New York, Simon & Schuster: pp. 177-209.

Wells, D. (1997), ‘One Nation and the Politics of Populism’, in G. Bligh (ed.), Pauline Hanson: One Nation and Australian politics, University of New England Press: pp. 18-28.