How the Internet Has Brought Us Too Close Together (and the Wisdom of Trolls)

A couple of days ago, Scot McKnight posted on the subject of ‘crowdpounding’ and ‘crowdaffirming’, remarking upon the treatment of both Ellen Pao (Reddit’s Former CEO) and Julie Rodgers. McKnight traces these phenomena back to the underlying dynamic of ‘groupthink’: ‘crowdpounding is groupthink in the accusatory and denunciatory mode’, while ‘crowdaffirming is the positive side of group adulation of a person.’ He suggests that these dynamics are illustrative of Girardian mimetic theory and the scapegoat mechanism.

Pao argues that the culture of the Internet has changed from what it first was:

The Internet started as a bastion for free expression. It encouraged broad engagement and a diversity of ideas. Over time, however, that openness has enabled the harassment of people for their views, experiences, appearances or demographic backgrounds. Balancing free expression with privacy and the protection of participants has always been a challenge for open-content platforms on the Internet. But that balancing act is getting harder. The trolls are winning.

McKnight focuses upon the culture of the comment box, placing responsibility at the door of blog owners who exercise poor moderation and with commenters who fail to show civility. Without disagreeing with these points, I think there is a much larger picture that needs to be taken into account and neither Pao’s and McKnight’s accounts give much attention to the systemic factors that produce a dysfunctional online culture and which, to some extent, have changed the Internet from what it once was.

McKnight helpfully brings René Girard into his analysis of the situation, but the significance of Girard’s theory here merits closer attention. Girard focuses upon non-differentiation as something that provokes mimetic crisis and encourages the development of what one could call ‘herd’ dynamics. Girard argues that the more that social differences and hierarchies are broken down and the more people become equal and closely connected to each other, the more at risk a society is of dangerous mimetic contagion.

Relating Girard’s work with that of Edwin Friedman is worthwhile. Both Girard and Friedman identify the immense dangers and potential for violence that lie in a sort of excess of ‘community’, about the danger of our togetherness locking us reactivity and herding dynamics. Their cautions are never more relevant than in a society that over-values the togetherness impulse and a high emotional conductivity (‘empathy’) between its members. What both Girard and Friedman highlight is the extreme importance of the factors that arrest or prevent the movement of impulses and emotions from one person to another, those structures, traits, and practices that enable us to create boundaries and distance between ourselves and others, to resist the pull of empathy, and to establish a well-defined—‘differentiated’—sense of ourselves and our own ‘response-able’ agency (as opposed to reactivity).

I believe reflection on the character of the Internet—and especially on its contemporary form—will reveal that our online media exacerbate the problem of herd dynamics and the mimetic violence that accompanies it, the ‘crowdpounding’ and crowdaffirming’ to which McKnight refers (I hope he won’t mind if I generally circumlocute such ugly neologisms). In its very structure, the Internet tends to bring us too close together in a number of ways that invite dysfunctionality.

Friedman makes an important distinction between ‘organism’ and ‘environment’. McKnight and Pao focus more upon the factors at the level of individual ‘organisms’ within the environment of the Internet. Organisms have the capacity to respond in a ‘differentiated’ manner to their environments, resisting the pressure to fall into certain patterns of behaviour, for instance, or recognizing the agency that they possess to change the circumstances that they find themselves within. We shouldn’t separate organism from environment, however. Environments and organisms, while distinct, are mutually conditioning and it is this interrelationship between environment and organism that this post will attend to in the context of Internet culture. Certain environments tend to strengthen or encourage certain traits or types of persons and the Internet is no exception. Even though it never utterly deprives its organisms of their potential for self-differentiation, it can make it considerably harder for them to exercise this responsibility.

How exactly does the contemporary Internet decrease differentiation? Here are a few ways. The Internet is fast, diminishing the differentiating factor of time. When everything moves at such a pace, we tend to react rather than taking the necessary pause for reflection necessary in order to respond. The Internet is densely social and distractingly stimulating, denying us the differentiating personal and emotional space in which to read and think for ourselves and arrive at our own conclusions without other things or persons constantly intruding upon our consciousness. The intense sociality of the social web makes it a powerful source of peer pressure and an engine of conformity (in preference falsification, for instance). The emotional energy that it takes to say something controversial on Facebook or Twitter, where herding drives can be exceedingly strong, is a powerful disincentive for many and pushes people to bring—or at least in appearance—their views into line with those of the masses. The Internet is an incessant and addictive source of distraction, which decreases attention spans and attenuates any single voice’s hold upon it.

The Internet largely disengages our bodies, removing the differentiating factor of our physicality and leading us to interact with people more immediately without its friction (one reason why people will often share intensely personal things about themselves with strangers online that they wouldn’t readily share with strangers offline). The fact of the deep and ‘full-bodied’ otherness of persons online is less pronounced when they only register through words on a screen. The Internet also tends to decrease the differentiation of our social spaces, especially between the public and the private. As a result, disagreements increasingly intrude into the context of our intimate social spaces and our—no longer so—private viewpoints can cause trouble in our wider or more formal relationships or public lives. Rather than being placed within a dedicated alternative space—a sort of ‘playing field’ upon which we can thrash out our disagreements, before leaving the field, shaking hands, and returning to less fraught private spaces—such a differentiation of space is harder to maintain in many contexts now. The Internet also collapses many of our social spaces and conversations into each other, meaning that we are constantly and inescapably eavesdropping upon conversations that we would previously been excluded from and, even when we are specifically addressing a particular group of persons, will be heard by many outside of that group. The Internet, with its hyper-accessibility and hyper-publicizing, makes it hard for us to differentiate conversations, to restrict conversations to those who are personally, morally, psychologically, intellectually, etc. equipped to have them.

The result of this decreased differentiation is that, unless we are unusually well-defined as ‘organisms’, we are highly vulnerable to cycles of reactivity and emotional contagion. The excessive closeness of the Internet produces a culture driven by ‘virality’ that moves faster than thought and by a herding instinct that is intensely reactive. It leads to fierce cycles of outrage that are only curtailed by the fickle fleetingness of the Internet’s attention. It produces people with shallower reservoirs of independent character, thought, and agency. It produces excessive emotional attachment, vulnerability, and proximity to others.

Ellen Pao laments a change in online culture since the age of the early Internet. Once again, reflection upon the more structural changes to the character of the Internet over the past two and a half decades or so will go some way towards helping to explain this. The early Internet that many of us remember from the early nineties to the early noughties was rather different in character and demographics from the current Internet and its culture was shaped by this. Here are some significant ways in which the Internet and its denizens have changed:

1. The early Internet was a place where we had to make a lot more for ourselves. It was considerably less accessible and open to those who were not technologically aware. The new Internet is largely ‘paved over’, its architecture predominantly provided by big corporations. Earlier users of the Internet, by contrast, were much more likely to create and form their own spaces, rather than just joining sites such as Facebook or Twitter (which didn’t exist until very recently). They were more likely to be crafters of their media, not just the messages those media bore. Early users of the Internet tended to be more active, engaged, enterprising, pioneering, and creative persons on average, and less likely to be more passive media consumers. Users of the Internet today are much less likely to function as active producers who regard themselves as responsible agents within a process of collaborative creation and conversation and much more likely to adopt the habits and modes of action of reactive and hypoagentic consumers, who are more accustomed to expecting other parties to form their environments to their preferences on their behalf, and to complaining when they don’t.

2. The typical early users of the Internet were very heavily skewed in the direction of young, white, well-educated, economically-privileged Western males. For instance, according to this piece of research from around the time I first started going online, 90% of the Internet users were male, 88% were white, 89% had some level of university education, 99% came from the US, Europe, or Australia, and the average user’s age was 31 years. The early online population was heavily skewed in the direction of the open and creative and in the direction of the technologically-minded. All of the dominant traits of early Internet users, in their own ways, tend to correlate with a culture of discourse that values free expression, diversity of ideas, vigorous exchange of opposing viewpoints, a relatively clear distinction of ideas from persons, etc. As the demographics of the Internet changed—and those demographics swiftly shifted as we moved into the 2000s—its culture naturally changed too. The early forum and blogging communities that I participated in had relatively similar demographics to those described by the research above and enjoyed the vigorous yet collegial culture of discourse that tends to come with that. These communities worked so well in large part because they weren’t very fraught by gender differences in cultures of discourse or by racial or class tensions, and because almost every participant had some socialization into the standards of argumentation and discourse that one expects of those with higher education. The demographics of contexts and communities are very significant in determining the sorts of discourses that they can sustain and the demographics of the early Internet were demographics that encouraged the ‘broad engagement and diversity of ideas’ that Pao mentions. The widening of access to the Internet is obviously a good thing in many respects, but we should not think that we can radically change demographics without radically changing the character of our discourse. Unfortunately, more inclusive and diverse discourse is not always more effective or illuminating discourse.

3. Communication in the early Internet was less immediate, but much less socially fraught than communication in the contemporary Internet. Anonymity and our greater degree of obscurity enabled us to have discussions, to express and explore viewpoints and ideas in a way that was considerably less likely to intrude upon or prove costly our offline relationships. For many of us, this afforded an avenue of escape from contexts that might otherwise have proved claustrophobic. The fact our online communities of discourse were also typically considerably more detached from the offline communities and relationships within which we exist gave a stronger sense of the heterotopy of Internet discourse, of the fact that Internet discourse occurs in an alternative space. The Internet wasn’t a realm where we ‘lived’, but a weird and wonderful frontier where we interacted with strangers. When people gradually stopped thinking about the Internet like pioneers, explorers, and prospectors (or even as settlers) and started thinking more like village and town inhabitants, genuine differences and opposition became a lot more threatening.

4. The early Internet was not monetized to the degree that the modern Internet is. Communication was often costly and offered limited rewards. The effort and cost in time and money that it took to communicate and the lower rewards tended to encourage conversations dominated by people who were deeply invested in what they were talking about, independently of personal tensions involved or social or financial rewards or reinforcement for speaking (I recall the considerable effort of creating, designing, and publishing essays on a site before I started blogging in 2003). The culture of the early Internet was powered by the passion and generous culture of hobbyists, obsessives, and people who have a natural drive to create and share something new with others, even if it never registers in the public consciousness. For instance, a site like Wikipedia developed out of the work of such people. The ease of communication and access—commenting being a clear example—in the contemporary Internet lowers standards of communication, providing few disincentives to people who aren’t really invested in the truth.

5. The early Internet was a wilder and less known place, with a lot of obscure corners. The obscureness of much of the Internet meant that things seldom came to us: we had to invest time in looking for like-minded people and interesting conversations—or creating contexts where they didn’t previously exist—much like explorers venturing into previously uncharted territory. The obscureness of many corners of the Internet meant like-minded people would often find each other and interact largely undisturbed by people who were mere troublemakers, with no genuine investment in the conversation. It also meant that conversations were less likely to be flooded by the uninformed and unqualified and that such persons could be much more easily recognized and removed.

6. The early Internet did not lend itself to publicizing in the same way. Information spread less rapidly and it didn’t provide the opportunities for the firestorms of outrage and the viral movements of emotional charged reactions that we encounter in the dense human forests of the contemporary Internet and its social media. Together with the previous factor, this meant that distinct conversations were less likely to bump into each other. The early Internet didn’t have the ‘mass’ culture that the Internet has today.

7. The early Internet was much less intimate and was more socially differentiating. It was much less emotionally charged as a result. As differences occurred within more neutral space, things were less likely to become personal (the proximity of so much of our online discourse to the personal profile isn’t always a good thing). Some of my favourite interlocutors over the years have been pseudonymous and anonymous, because such persons often seek to retain such a separation between private identity and ideas. On the other hand, the early Internet was more social in some respects. In the early Internet, we were more likely to function as mindful and intentional ‘community-builders’ than as passive consumers experiencing community. Our blogs weren’t generally set up as private means for self-publication—or even in order to form or host communities in our private space—but in order to participate in, contribute to, and collaborate in the formation of a wider conversation. My most worthwhile interactions online typically occur in contexts that date from this period of the Internet, or that share its characteristics—private e-mail discussion lists, obscure interactions in less known quarters of the Internet, e-mail correspondence, etc. Sites like Facebook encourage a collapse of differentiated social interactions into a much less differentiated social space. Another of the strengths of the older Internet is the way that it encouraged more differentiated social interactions. For instance, I have created over a dozen blogs or websites in my time, devoted to a variety of different matters of interest, speaking to specialized communities, enabling far richer interactions as a result.

Many more such changes could be listed. At this point, however, I would like to conclude this post with some reflections upon another element that comes up within this discussion: the figure of the troll.

The term ‘troll’ is significantly overused. It is—quite mistakenly, I believe—treated as interchangeable with terms such as ‘cyberbully’ or—the older term—‘flamer’. It is also overused in the poorly differentiated context of the Internet. Many people are hyper-sensitive to opposition in the context of the contemporary Internet precisely because the non-differentiated character of the environment and their personalities heightens reactivity and sensed vulnerability to others. They feel a need to be surrounded by people who affirm them and think in similar ways to them and need to be shielded from all ‘threatening’ viewpoints. They need to be part of a herd. Such people frequently mislabel critics as ‘haters’, for instance. They have very thin skins and tend to take opposition in an unnecessarily personal manner.

The ‘crowdpounders’ and ‘crowdaffirmers’ that McKnight refers to illustrate the dynamics of the herd. Trolls, however, represent something else. They function as agents of chaos or irritation, preying upon the emotional dysfunction and lack of self-consciousness of highly reactive communities, exposing it to others or merely deriving amusement from stirring things up themselves. In contrast to ‘crowdpounders’, trolls tend to be far more independent personalities, with a deep antipathy for groupthink, who love to antagonize people and exploit the dynamics of groupthink, people’s emotional reactivity, and lack of differentiation to get a rise from them. Trolls disrupt communities, and especially communities that are dysfunctionally reactive or non-differentiated.

Trolls also differ from flamers and cyberbullies. Unlike flamers and cyberbullies, trolls aren’t merely trying to hurt people, nor are they merely hurling abuse. While flamers and cyberbullies directly attack, trolls are all about the carefully laid bait (some trolls are so subtle as to be indiscernible to all but the most observant). Trolling thus often involves a degree of deception. For instance, one type of troll might feign to be a naïf expressing a position that they sincerely hold, when in fact they do not truly believe anything of the kind. The point of the bait is to provoke the targeted community or persons to a response that reveals their stupidity, dysfunction, or reactivity (generally for the perverse amusement of the troll and others—the ‘lulz’).

Trolls are more often than not highly dysfunctional—though often extremely intelligent—people themselves (other forms of trolling are shrewd uses of communities’ reactivity for a somewhat worthier purpose, such as the Apostle Paul’s trolling of the Jewish council in Acts 23). Their pleasure in pressing people’s buttons is typically a perverse one (someone like Katie Hopkins is a good British example), albeit one often arising out of a deep annoyance with the obvious unhealthiness and perverseness of the dynamics of reactive and non-differentiated communities. Nonetheless, they can—generally inadvertently—perform positive social good in the course of their trolling. By exposing dysfunctions and making non-differentiated communities and reactive persons look ridiculous, they can save us all the trouble of taking these persons and groups as seriously as they would like to be taken, even if the troll themself appals us.

The troll can often be a deep thinking and independent-minded person—a less elevated version of Socrates’ gadfly—who likes to incite people. The humour of the troll typically rests upon an ironic distance dependent upon the naivety and lack of self-awareness of the troll’s victims and upon the troll’s ability to read and exploit the dynamics of a community in a way that it cannot read itself. This is why trolls prey upon communities that tend to operate with herd dynamics, rather than upon self-aware communities. Their wry appreciation of triggering herd dynamics is often a coping mechanism for or a means of venting their deep frustration with the prevalence of such dynamics more generally and the ways that these dynamics impinge upon their lives.

People’s particularly strong reaction to unwelcome truths is also an important weapon in the troll’s arsenal. That certain facts and truths aren’t voiced because people find them offensive or uncomfortable is one of the characteristics of the reactive and non-differentiated community that most encourages the troll to antagonize them. Trolls tend to have an exceedingly high tolerance for other people’s discomfort, but they also often have a very low tolerance for society’s resistance to truths that make people feel uncomfortable. They will use these truths to antagonize society, not chiefly in order to hurt genuinely vulnerable individuals who are supposedly being protected from the truth, but rather to attack the social resistance to truth and the overly thin skins that are being created by this. To put it crudely, trolls can sometimes be the a**holes that societies need to expel the s**t that they otherwise can become full of. As trolls are particularly attuned and attentive to the unhealthy patterns of societies’ reaction and communication, and to the naivety of groupthink, trolls can be especially well-equipped and driven to expose the falsehoods and misconceptions that thrive in such contexts. It is not surprising, for instance, that the founder of the famous myth-busting site, Snopes.com, David Mikkelson, started off as a troll on Usenet forums. The disruption and incitement of trolls is not uncommonly motivated by an instinctive antipathy towards the lies, the bull, the misconceptions, and the half-truths that reactive and non-differentiated societies often run on.

Trolling is also often a guerrilla tactic used by those with a natural affinity to something closer to the more anarchic culture of the earlier Internet against the groupthink that often arises in the mass corporate culture of the contemporary Internet. These sorts of trolls are often intelligent, self-aware, independent, long-term Internet users, intensely well-versed in Internet culture, who dislike the way that online culture is being reshaped to make it a friendly and ‘safe’ place for entitled and hyper-sensitive passive online consumers, curated and controlled by corporations, squeezing out the diversity, unpredictability, confrontation, vigorousness, independence, creativity, and agency that they highly value in the Internet culture that was once more accommodating to, because it was largely formed by, people not unlike themselves. As this post observes, the troll represents a threat to and disruption of the corporate vision of the Internet, where everyone is clearly and neatly defined, where all keep in line, and interact in a predictable fashion.

Trolls can take many different forms and some trolls are decidedly unpleasant personalities. However, at their best, trolls can greatly enrich our online world, ensuring that the Internet never fully succumbs to the state of the sleepy settlement or to corporate colonization, but always retains something of the strangeness and unpredictability of the frontier, where people need to keep their wits about them, where they must develop thicker skins and take responsibility for themselves, and where startling and illuminating discovery can still occur. At their very best, trolls, like Socratic gadflies or biblical prophets, can serve to unsettle societies’ and individuals’ groupthink and their complacent relation to the truth. Such irritants can be some of the most important members of society.

Some of the most stimulating conversation partners I have ever had have been people that many would classify as ‘trolls’. My experience with them has revealed a more complicated picture. It has revealed that such people are often deeply concerned about truth and that their trolling is frequently provoked by their chosen victims’ proclivity for taking emotional refuge in falsehood. Their abrasive forms of engagement are often calculated to smoke out reactive persons. Such individuals are often anonymous or pseudonymous. This seems to be in large part because they have a sense of the heterotopy of cyberspace, that the Internet often works best when we understand it as an alternative space to those of our personal relations and identities, that the Internet is a frontier where we can intrepidly leave behind the safety, the certainties, and the intimacies of home and aggressively explore and discover new things, foreign ideas, and engage with strange people who might feel threatening if they moved into our ‘neighbourhood’. They don’t come onto the Internet to form close personal relationships but to have the sort of intense and stimulating interactions that aren’t so possible in contexts closer to home. They often exemplify the value of a thick skin for substantial discussion; they are wonderfully non-reactive, so one can speak forthrightly with them, without tiptoeing around sensitivities. They exhibit a strong differentiation that enables them to be more playful in their interactions. While I may disagree with their occasionally uncivil manner of engagement and their appetite for offending people in many respects, I find their implicit vision of what the Internet could (once again?) be very attractive and it is one I largely share.

Drawing these points together, the sort of troll that I describe is often an—albeit somewhat dysfunctional—advocate of the sort of Internet that we need to encourage if we are to have fruitful and productive conversations. They represent a more differentiated mode of interaction, one that sharply contrasts with the non-differentiated herd dynamics of the masses of more passive online consumers in the contemporary social web. The sort of trolls I am describing are typically detached from the herd dynamics and it is at the herd dynamics that they take aim—not just to cause hurt for hurt’s sake as many suppose (that is the characteristic of vicious cyber bullies).

If we are to encourage a healthier Internet culture of discourse, I submit that we may need to make a half-turn towards the trolls, away from the non-differentiated culture of the reactive masses in the social Internet the corporations and the massification of the Internet have formed for us. We need contexts that are less ‘safe’ and which demand more from us. We need to push towards the creation of more differentiated environments of discourse. We need to recognize that healthy conversation may require a greater degree of exclusivity and even exclusion, something which existed more organically in the earlier Internet. Most people are not equipped for such conversation, not without considerably more formation. Vigorous and fruitful exchange of diverse ideas is only possible where a certain culture exists and this culture requires particular types of persons and contexts to sustain it, people who regard themselves as self-defined collaborative architects of a conversation and contexts that are more capable of sustaining confrontational and more differentiated interactions.