Pros and Cons of Political Polarization on the Internet

For years, political science has held up a golden ideal: the good citizen.  This benevolent prototype embodies all the qualities presumed to be necessary for participation in a liberal democracy: making voting decisions based on genuinely felt values, actively sought-out information, and reason.

In the complex real world, there are many ways people fail to be good citizens, but one of the most common factors talked about in modern society is political polarization: the ideological separation of parties or individuals.  It’s not uncommon to hear breathless pronouncements that polarization is “destroying America” or blaming it as the reason why “politics is broken.”  This is an understandable narrative: particularly on the internet, it’s easy to imagine a hyperpolarized person stewing in an echo-chamber, taking in limited information, refusing to engage with alternate perspectives, and reaching conclusions emotionally rather than rationally.  Such a person is clearly far from the ideal of good citizenship.

However, the reality of polarization is more complicated: it is a nuanced phenomenon that can manifest in a variety of ways and can have both positive and negative outcomes.  In fact, the current social attitudes against polarization are in stark contrast to the prevailing academic view of the mid-20th century, which warned against a political system where the parties weren’t polarized enough.  Ideologically divided parties provide clear, policy-related reasons to prefer one candidate over another; without enough of a values-driven difference between political parties, voters tended to revert to geographically based habits.

Furthermore, despite the stereotype of the shrieking, frothing extremist, there is evidence that political extremists are not, by nature, unthinking and irrational: Brandt and colleagues found that extremists were in fact better than political moderates at avoiding certain kinds of cognitive bias.  Also, some forms of extremism are plainly morally superior to their alternatives: I do not really want to spend much time with someone who lacks an extreme attitude against genocide.

So, the issues with polarization are not that people are going around with extreme and divergent attitudes and beliefs, per se.  Rather, there are problems that occur as a result of these extreme attitudes… particularly in an internet context designed to feed some of people’s worst instincts.

Political polarization and false consensus

The false consensus effect, first discovered decades ago and observed in countless contexts since, is the tendency to assume other people share your own (or your community’s) viewpoints, attitudes, knowledge, or beliefs.  People believe themselves to be normal; they believe that their own experiences are accurate representations of the whole world.  If I believe something, and if everyone around me seems to believe that same thing, then it only makes sense to conclude that it’s what everyone else thinks, too.

False consensus is magnified by the presence of ideological echo-chambers.  The internet contains both massive amounts of information and also increasingly effective ways of filtering that information; as a result, people are able to set up ‘bunkers’ where they primarily encounter information that already agrees with them.  People are naturally inclined to prefer information that supports their pre-existing attitudes: it’s upsetting to think you might be wrong or foolish, and we consider our ideological allies to be better and more accurate sources of information.

A researcher named Magdalena Wojcieszak has spearheaded multiple studies about the way that online echo-chambers facilitate false consensus.  She has found that time spent in ideologically homogenous spaces increases the extent to which people assume others share their views… even among people with fringe beliefs (i.e. hanging out at Stormfront makes you think a higher proportion of the population is white nationalists).

Furthermore, people use their assessments of the way ‘normal people think’ to set their own standards for what makes a view biased.  This plays out in the hostile media effect, the tendency for people to see news sources that support their own views as neutral…. and therefore, to see everything else as biased.  This can result in situations where a majority of people see a truly moderate, neutral news source as biased, but half see it as skewed-liberal, and the other half see it as skewed-conservative.  Even worse, this can feed back to influence the news sources themselves.  The path to being seen as trustworthy, unbiased, and respectable might paradoxically be to actively chase a partisan audience.  That way, half of readers will think you’re unbiased, and the other half would scorn you even if you were unbiased.

Hostility and polarization

Beyond its customization capabilities, internet communication tends to have another feature associated with polarization: it tends to be text-based and anonymous.  The Social Identity model of Disinhibition Effects (SIDE) synthesizes a wide variety of research around the idea that people have multiple ways they can think about themselves.  For instance, I am myself, but I am also a psychologist and a Democrat and a Duke fan and so forth.  Each of these identities has a distinct set of ideas about the appropriate way to act, which occasionally conflict.  People tend to take on the norms of whatever aspect of their identities is most salient: personally, I value respecting people from all walks of life, but at a basketball game, surrounded by a sea of royal blue, I wouldn’t piss on a UNC fan if they were on fire.

Anonymity and text-based communication can both have the effect of drawing people’s attention away from their own, personal identities.  Instead, they focus on a salient group identity.  For instance, a specific individual might personally be ambivalent about the role of the government in health care, but that same person, anonymous on the internet and hanging out in a conservative subreddit, is likely to prioritize the group-level, conservative norm that government health-care should be fought.

One belief which is common to many groups is: people outside this group are bad.  In-group preference and out-group disfavor appear to happen automatically, even in groups formed randomly.  They are even stronger in groups that are in competition with one another, and they are stronger still in groups defined by differing moral values, both of which are true for politics.  As a result, political polarization is often characterized not only by the groups disagreeing, but also by the groups actually loathing one another.

Because of all this, interacting online in politically polarized environments often leads people to increasingly dislike political opponents, even to the point of feeling disgust for them.  Disgust, in turn, leads to dehumanization: the tendency to see a person or group as less sophisticated or worthy of respect.  People who dehumanize political outgroup members may be more willing to engage in bullying or hostile online behaviors against them, and they will have less sophisticated theories about the motivations and arguments underlying the other side’s positions.

A positive side of polarization

In the midst of all this negativity, it’s important to remember that not all outcomes of political polarization are harmful.  Most importantly: polarized people are also likely to be high in political engagement: they are more likely to talk about politics, to care about politics, to be informed about politics, to donate money or time to political causes, and, most importantly of all, to vote.

The relationship between engagement and polarization appears to go both ways.  The people who aren’t polarized tend to be the people who simply don’t care enough about politics to bother getting polarized.  Simultaneously, exposing yourself to opposing points of view leads to confusion and ambivalence, which in turn leads to disengagement.  This phenomenon is especially important among groups of people that disproportionately feel disaffected by politics, such as racial minorities, the working poor, or young people.  Online political activity, and its subsequent polarizing effect, can reduce political inequality by chipping away at the extent to which political motivation is tied to demographics.

Finally, I once again should note that political polarization, free of all its negative trappings, is simply another name for strong political diversity.  Mere disagreement is not something that is damaging to good citizenship!  Also, our particularly human cocktail of biases and motivations means that, in many cases, the alternative to polarization and echo-chambers is apathy, which is hardly a solution.  And, as I said before: sometimes the extreme view is just correct.

Therefore, my focus is not on ending political polarization or smashing ideological echo-chambers, but instead on being mindful of the harmful, potential consequences.  Are you starting to feel disgusted by someone you disagree with?  Have you caught yourself assuming that everyone must agree with your point of view?  There’s the time to intervene.

Partial bibliography:

American Political Science Association (1950). Toward a More Responsible Two-Party System: A Report of the Committee on Political Parties of the American Political Science Association. New York, New York: Rinehart

Barberá, P., Jost, J. T., Nagler, J., Tucker, J. A., & Bonneau, R. (2015). Tweeting from left to right: Is online political communication more than an echo chamber? Psychological Science, 26(10), 1531-1542.

Brandt, M. J., Evans, A. M., & Crawford, J. T. (2015). The unthinking or confident extremist? Political extremists are more likely than moderates to reject experimenter-generated anchors. Psychological Science, 26(2), 189-202.

Dvir-Gvirsman, S., Garrett, R. K., & Tsfati, Y. (2015). Why do partisan audience participate? Perceived public opinion as the mediating mechanism. Communication Research, 18(5), 1-25.

Gentzkow, M., & Shapiro, J. M. (2006). Media bias and reputation. Journal of Political Economy, 114(2), 280-316.

Holt, K., Shehata, A., Strömbäck, J., & Ljungberg, E. (2013). Age and the effects of news media attention and social media use on political interest and participation: Do social media function as leveller? European Journal of Communication, 28(1), 19-34.

Iyengar, S., Sood, G., & Lelkes, Y. (2012). Affect, Not Ideology: A Social Identity Perspective on Polarization. Public Opinion Quarterly, 76(3), 405-431.

Holt, K., Shehata, A., Strömbäck, J., & Ljungberg, E. (2013). Age and the effects of news media attention and social media use on political interest and participation: Do social media function as leveller? European Journal of Communication, 28(1), 19-34.

Knobloch-Westerwick, S., Johnson, B. K., & Westerwick, A. (2014). Selective exposure for better or worse: Its mediating role for online news’ impact on political participation. Journal of Computer-Mediated Communication19(2), 184-196.

Lea, M., & Spears, R. (1991). Computer-mediated communication, de-individuation and group decision-making. International Journal of Man-machine Studies, 34(2), 283-301.

Mutz, D. C. (2002). The consequences of cross-cutting networks for political participation. American Journal of Political Science, 46, 838-855.

Ross, L., Greene, D., & House, P. (1977). The “false consensus effect”: An egocentric bias in social perception and attribution processes. Journal of Experimental Social Psychology, 13(3), 279-301.

Stroud, N. J. (2010). Polarization and partisan selective exposure. Journal of communication, 60(3), 556-576.

Wojcieszak, M. (2008). False consensus goes online: Impact of ideologically homogeneous groups on false consensus. Public Opinion Quarterly, 72(4), 781-791.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s