The trouble with "us"

The blurring of social roles and the consensus illusion

Consensus among online communities may all too often prove fragile if not illusory. But, writes Kathrin Passig, as long as Internet users can adapt to groups that actually agree on only a select few issues, there is no need to lose faith in social media.

“Everybody on Facebook and Twitter always writes such an awful lot of nonsense” is a frequently heard complaint, even from people who have been on the Internet long enough to know better. “You choose who you follow”, I tell them. “The Internet I use is full of people saying interesting, intelligent things.” Which makes me look clever and the others rather silly. Unfortunately, it’s not entirely true. My friends, and the people I follow on Twitter, have some clever things to say, but they also spout a lot of unfounded and untenable opinions. Every day, I feel like thumping my digital broom on the digital ceiling to discourage them from doing so. One possible reaction, most frequent among those who make such complaints, is to assume that Facebook has dumbed down our otherwise smart friends in some mysterious, probably American, way. Another reaction is to suspect that these friends always had such dubious views and prejudices, and that the social networks simply bring them out in the open.

Photo: Orkhan Aslanov. Source:Shutterstock

Controversial topics only rarely come up in direct conversation, and even then they are discussed cautiously. It’s more fun to find common ground, and shared opinions strengthen a friendship. Recently, however, several social spaces have emerged on the web where we can no longer tailor our comments to suit the company we keep. Private conversation is carried out in public. Friends and acquaintances can all see what we say, what opinions we hold, what we like and what we disapprove of – and they can see all this even when we’re not directly talking to them. This happens in all social networks that are at all widely used and not restricted to particular topics – and, at the moment, above all, on Facebook and Twitter.

Despite the grand promises made about the “filter bubble”, which, thanks to homogenous circles of acquaintances is supposed to reliably exclude controversial topics, the fact is that we can’t rely on anything in social networks. No matter how self-evident your opinion on gay marriage, religion or bicycle helmets may seem to you, there is always someone on the fringes of your circle of friends – or even right in the middle – who doesn’t share your opinion, and who says so straight out. We work hard to persuade ourselves that we are part of a group where “we” all agree on the most important matters; and then all of a sudden something happens to show that half of us are actually, somehow, them, the others.

Pessimists proclaim that it is safest to appeal in social networks to the lowest common denominator, so as not to draw unwelcome attention to ourselves. However, this only solves the problem from the writer’s point of view. From the reader’s point of view, it is obvious that even the most anodyne post or message can unleash despair at the state of the world – if the reader has fundamental objections to food pics, LOLcats or emoticons. Controversial statements are often made in the mistaken belief that they are harmless or entirely uncontroversial – and readers are often more offended by clueless indifference than they would be by a deliberate attack on their values. Even in the most like-minded group of people, we offend and irritate someone every day and are offended or irritated by others in turn.

Until the Internet came along, community was a much rarer commodity than it is today. In the 1980s, the sexologist Eberhard Schorsch wrote that “except for homosexuality […] sexual perversion goes hand in hand with complete isolation. It is in principle impossible for e.g. fetishists, exhibitionists or voyeurs to form a group, or to come together to create their own subculture”. This changed just a few years later, and changed not just for those with divergent sexual appetites but also for narcissists,1 Raelians, adult My Little Pony fans, psychogeographers, fan fiction authors, those afflicted by rare disorders and anyone else you could care to mention. Even hikikomori, who avoid social contact by never leaving their homes, have their own online communities.

But in the first few years, Internet “us” groups had very clear boundaries. When two different communities met at a party because the host was a member of both, the two factions tended to regard one another with mistrust. This is hard to understand today, rather like the pointless battles between neighbouring villages in Louis Pergaud’s War of the Buttons.

For a long time it was normal to spend time on the web in the company of strangers. This is because it was simply too difficult to persuade our existing circle of friends to get online. Right up until the early 2000s, Internet communities simply did not offer the technical or conceptual support to cater to existing friendships. The common elements were topics of interest or certain forms of social behaviour, rather than geographical proximity or existing relationships. In an intermediary step, it was possible to meet online friends face to face – if the distances involved were not too great – and to drop offline friends who did not share common interests so that, over time, the online and offline social groups came to be one and the same. Then, as social networks spread, circles of acquaintances came together that had originally formed in the context of wholly discrete online fora.

In 2007, the social media scholar danah boyd published an essay on the imminent schism in social networking activity; this was the moment when highly networked young people at college were beginning to join Facebook, while minorities, migrants, subcultures and working-class children were left behind on MySpace. Over the next few years, however, everybody joined Facebook anyway. MySpace dwindled and was shut down in mid-2013. The same thing happened to many social networks associated with particular countries, such as the Google-owned Orkut, which was very strong in Brazil and India for a few years. About one-third of the population of Germany now regularly uses Facebook, while in the Scandinavian countries and the United States the proportion is one half and in Iceland it is about three-quarters. This makes Facebook the place where the mass of early-adopters meets people who were previously never online at all. Rivers springing from many different life contexts flow into the great Facebook sea.

At the same time, other well-defined communities continue to exist and are central to the identities of those involved. New sites and communities spring up on the web, just as they always have done. Anyone can unwrap whatever aspects of their personality they care to share online, and can choose to do so anonymously or under a pseudonym too. In the social networks, however, the edges of these communities blur. There are no discrete groups at all on Twitter, and although Facebook offers the option of creating groups, it is essentially structured along the lines of intertwining, overlapping individual networks. Thus, in many sites, there is no longer any such thing as a clear inside and outside, or as belonging and remaining an outsider.

This disappearance of clearly delineated groups is part of a larger trend connected to the waning importance of physical presence (“meatspace”). The media theorist Joshua Meyrowitz described in No Sense of Place (1985) how television – and radio before it – dissolved the previously impermeable boundaries of social groups, allowing outsiders to come in, and thereby unleashed social upheaval in the 1960s.2 Television was a “secret exposing” machine that blurred the distinction between various areas of public life and between public and private behaviour. Social status was no longer tightly bound to physical presence. Meyrowitz writes that films on television show children how adults behave when the children are not around, and offer adults insights into how the informational worlds of others – other age groups, social classes, genders or inhabitants of other regions. “By bringing many different types of people to the same ‘place’, electronic media have fostered a blurring of many formerly distinct social roles. Electronic media affect us, then, not primarily through their content, but by changing the ‘situational geography’ of social life.” As already indicated, this was written in 1985, when electronic media meant television. Social networks now fulfil the same function of exposing secrets. However, they broadcast the lives not of other groups and other classes, but of our own friends.

The phenomenon of otherwise unconnected social contexts blurring together is older than social networks. Once the private homepage appeared on the scene in the 1990s, it became impossible to keep track. Unlike discussion threads in forums or newsgroups, these pages aimed their messages at no clearly defined audience. Furthermore, by now, there was no longer any guarantee that everybody online was a university-educated technophile. A homepage had to present its subject to the world at large, which resulted in journalists repeatedly scoffing at homepages that they perceived as ugly or pointless and criticizing the whole idea of “representing oneself” before an ill-defined public.

In early social networks such as Friendster or Orkut, profiles were in principle built along the same lines as a private homepage. The owner alone decided which content appeared and was free to choose how often to update the profile, be it once a year or never. Step by step, however, the principle by which Facebook now operates crept in. The static profile was replaced by a stream of constantly updated status reports, with friends’ actions also finding their way onto network profile pages. These days, a profile on a social network is only to a very limited extent the result of careful decisions by the owner. Interaction with family, friends and acquaintances determines our behaviour, on the Net as elsewhere. Among other things, this means that it has become harder to embellish one’s image on these profiles. Profiles on dating and singles websites tend to function along the former lines of individual, non-networked self-description, which probably has less to do with the user wishing to conduct their search in private than with the opportunity to present the facts in the best possible light – without being contradicted.

Our social groups overlap ever more closely, our “(self-)portraits” are now created collaboratively, and active use of the Internet has become more and more normal. All these factors combine to cause various problems in adapting our behaviour. One of the most common and specific worries is that when someone applies for a job, the human resources manager will find pictures on Facebook of the applicant fooling around at a party; another concern comes from a 2009 study which showed that a Facebook user’s sexual orientation can be reliably deduced from the number of openly gay friends. While these difficulties have been widely debated, including from a legal point of view, there are many more of which we are only vaguely aware and that have yet to be properly defined.

Thus employees are generally reluctant to use social media in a work-related context, because they correctly intuit that their own private concerns do not mesh neatly with the interests of the company they work for. Using social networks for work demands that we present ourselves in one and the same way toward colleagues, customers and friends, where previously we could behave in three distinctly different ways. In many situations in the workplace, keeping silent may be the better option when the alternative is to say something that lays us open to attack from one – or two – of these groups. Not everyone believes whole-heartedly in their employer’s values, and we can quickly get into a situation where we either defend the company’s decisions and give our friends good reason to laugh at us when they read our comments. Either that, or we are honest with our friends but make ourselves unpopular at work. Even if we take up a new job and find that we can neatly incorporate our new role into our digital lives, this does not always last. It seems to be the rule rather than the exception that even the most open and forthright users of social networks tend to hold back and tone down as soon as they put their skills at the disposal of a company.

Earlier, when websites were neatly categorized by topic, it was harder to stumble into a debate unintentionally. Whoever found their way to a literature forum wanted to discuss literature, and whoever posted in a thread about genital mutilation or circumcision knew what they were doing. Twitter, however, has no “Subtwitters” for particular topics the way the news aggregator Reddit has “Subreddits”. Users see a feed determined solely by whose tweets they choose to follow. Readers are confronted with further unpredictability in the form of retweets3 and hashtags.4 Most outrage on Twitter is caused by unwanted retweets concerning sensitive topics. Recent examples on German Twitter include the circumcision debate in 2012 and the #aufschrei (“outcry”) campaign in 2013, which collected everyday examples of sexual harassment and provoked heated discussion. But nobody follows complete strangers on Twitter: we follow our friends and we choose to follow other people whose opinions we value, and when they retweet a message it is because they think it worth disseminating. Even though Twitter does allow the user to withhold the option to retweet, the opinions that our friends express themselves own can cause arguments and lead us to doubt their soundness of mind.

It has been known since the 1930s that we tend to overestimate how far other people, especially our friends, share our own opinions, and subsequent empirical research has backed this up. The phenomenon is known nowadays as the false consensus effect, ever since a study published in 1976 brought together the results of various experiments. The psychologists Lee Ross, David Greene and Pamela House asked students at Stanford University to walk up and down campus with a sandwich board proclaiming “Eat at Joe’s” – or, in a variant of the experiment, “Repent!” Students who agreed to do so believed that around two-thirds of those asked would take the job. But students who declined to wear the sandwich board reckoned that two-thirds of people would refuse. (In fact the proportions in the experiment were 50:50.) The study aroused great interest and nearly fifty papers over the next ten years added to the data on the effect.

The consensus illusion is everywhere, affecting our perceptions about the frequency of certain personal characteristics, or of mental problems, eye colours, tastes in food or films, religious attendance or participation in sporting events. The effect occurs most often when those affected find one another attractive and can reasonably assume that they will stay in touch in future. The effect is even present when respondents know for sure that the majority does not share their opinion; they consistently overestimate the size of the minority to which they belong. False consensus is more prevalent within a group than it is between group members and the outside world.

Shared tastes in consumer culture – or the lack thereof – among groups of friends has become the subject of study in the last fifteen years, as a side-effect of work on marketing algorithms. Nothing in the research findings indicated any significant correlation between certain circles of friends and shared tastes in books, films or music.5 Nevertheless many people will insist, privately and in public, that they share their friends’ tastes in most matters or, at the very least, that they would be able to recommend media to suit their tastes.

As soon as one examines it more closely, consensus among friends proves to be an illusion in other questions of opinion too. On the singles site OkCupid, registered users can answer and comment upon multiple choice questions. Here the spectrum ranges from political, ethical and religious matters through sexual preferences right up to the question of whether atomic war might be a turn-on. The OkCupid algorithm then uses the results to suggest possible partners. The user can look at other profile pages to see which members give similar answers to which questions, and where they disagree. The results are especially interesting where these other users are not strangers but friends who have also registered as members. Thanks to having made such comparisons myself, I now know that, at least in my own circle of friends, my secret fascination with atomic war is very widely shared, whereas nobody else agrees with the statement that “no” does not always mean “no” – which I had always thought was obvious.

And if our friends’ opinions do not always neatly match our own, this is all the more true of the friends of friends. Twitter only shows all responses to a tweet to followers of both participants in the thread, so that in the normal course of events a subscriber will not see comments left by the friends of friends. On Facebook, on the other hand, all comments are in principle visible to anyone who can see the original post as well. This means that we now see daily evidence of what previously we only encountered from time to time (most often when a friend started dating a particularly obtuse new partner). We realise that even one degree of separation is enough to shatter shared opinions on fundamental matters. The Pirate Party of Germany promotes concepts of “liquid democracy” and “delegated voting”, whereby voters do not hand over their vote to a delegate in any and all matters but can rather nominate the most suitable person to represent them on a topic-by-topic basis. In principle, these ideas reflect the insight, familiar to everyone who has been on the web for a while, that we are never going to agree with anyone for long on all matters. As it happens, though, our nominated delegate can also pass the vote along to someone else. This can only work if we assume that the first and last members of the chain have a relationship of trust only because such trust exists in the links along the way. Comment threads, and the situation on Facebook, have however shown us that friends-of-friends are barely any different from randomly chosen strangers. There are various ways to react to the ever-broadening, blurring shape of our social contexts and the disappearing consensus illusion. The first reaction – the gut reaction – is to decide that everyone else is stupid. “Dreamt I was getting ready to insult all 80,000 Tweeters on the #aufschrei tag in alphabetical order, then my telephone rang.”6 After a brief flash of anger, we return to the illusion of consensus. Perhaps everybody else had just suffered a temporary spell of blindness.

The second possible reaction is that the offended party threatens to purge his circle of friends: “Can you please ban #annewill #lanz #sterntv #Sexismus and #Aufschrei from the TL or do I have to stop following you all?”7 He might even make good on his threat. Such behaviour has its costs however, since for one thing there were always good reasons to be friends with someone or follow their stream, and those we criticize today will nevertheless have interesting and thought-provoking things to say tomorrow. For another thing, it is simply impossible to purge your circle of friends until everyone left is always of one mind. Just as with the Communist purges, new suspects will always come to light.

This dilemma then leads to the third coping strategy: a temporary or permanent aversion to the Internet. Sometimes this manifests itself in a decision to turn our back on “bogus” online communities. It turns out however that, though we may imagine our offline relationships to be more authentic, more reliable and more harmonious, this is not necessarily the case. It may be that this too is another case of wishful thinking about “us”.

Rather than lose faith in our own group of friends or in Internet users generally, we should instead give up believing in groups where lasting consensus can be reached on more than a very few key points. We need a more realistic assessment of the general consensus about our own views. One of the central challenges of the networked society is the realization that “other” people really are different.

Cf. Kathrin Passig, "Internetkolumne. Unsere Daten, unser Leben", Merkur 05/2012

Joshua Meyrowitz, No Sense of Place: The Impact of Electronic Media on Social Behavior, Oxford University Press, 1985

Retweets (or on Facebook, shares): friends add comments into our timeline which originated from people we do not ourselves follow.

Hashtags: comments on a particular topic are identified by a keyword with the hashtag. For instance we can follow comments related to the German Literature Days (Tage der deutschsprachigen Literatur) under the hashtag #tddl.

Cf. Kathrin Passig, "Internetkolumne. Abschied vom Besten", Merkur, 5/2010

The author, on Twitter, 4 February 2013

Andreas Witte, on Twitter, 30 January 2013. "TL" is the Twitter timeline, "unfollow" is the equivalent to cancelling one's subscription to a newspaper.

Published 7 March 2014
Original in German
Translated by Samuel Willcocks
First published by Merkur 10-11/2013 (German version); Eurozine (English version)

Contributed by Merkur © Kathrin Passig / Merkur / Eurozine



Subscribe to know what’s worth thinking about.

Related Articles

Space debris

COP26’s mandate focuses on averting the loss and damage caused by climate change. And technological solutions hold much promise. But high-tech and the environment aren’t always best matched: largely unregulated mega satellite projects are on the increase with space debris a real near-space threat.

Cover for: The common good or the market?

Under the commercial imperative of ratings and likes, public service broadcasters are moving complex and quality content to online only. As the example of Germany’s Westdeutsche Rundfunk shows, this fragments audiences, thereby undermining a core principle of public service.