Sign In

Communications of the ACM

The blogosphere

Democracy and Filtering

View as: Print Mobile App ACM Digital Library Full Text (PDF) Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook

Is the Internet a wonderful development for democracy? In many ways it is. As a result of the Internet, we can learn far more than we could before, and learn it much faster. If you want to get information to a wide range of people, you can do it quickly, via email and Web sites, and is another sense in which the Internet is a great boon for democracy. In particular, the rise of specialized sites and blogs increases the opportunity for people to read and write on an extraordinary array of topics. If you have an opinion and want to express it in public or want to find an opinion of almost any kind, chances are you can, at trivial cost.

But in the midst of the celebration, I want to raise a note of caution. I do so by emphasizing one of the most striking benefits provided by emerging technologies: the growing power of consumers to "filter" what they see. As a result of the Internet and other technological developments, many people are increasingly engaged in the process of personalization, limiting their exposure to topics and points of view of their own choosing. They filter in, and out, with unprecedented powers of precision. These developments make life much more convenient and in some ways much better; we all seek to reduce our exposure to uninvited noise, and many of us like to read opinions we find congenial.

But from the standpoint of democracy, filtering is a mixed blessing. Above all, I urge that in a heterogeneous society, such a system would require something other than free, or publicly unrestricted, individual choices. On the contrary, it imposes a distinctive requirement: People should be exposed to materials they would not have chosen in advance. Unanticipated encounters, involving topics and points of view we have not sought out and perhaps find irritating, are central to democracy and even to freedom itself.

Back to Top

Complete Individuation

To explore this issue, consider the following thought experiment. It involves an apparently utopian dream, that of complete individuation, in which consumers can entirely personalize (or customize) their own communications universe. Imagine a system of communications in which each person has unlimited power of individual design. If people want to watch news all the time, they are entirely free to do so. If they dislike news and want to watch football in the morning and situation comedies at night, that is fine, too. If they care only about America and want to avoid international issues entirely, it is simple indeed; so too if they care only about New York or Chicago or California. If they want to restrict themselves to certain points of view, say, conservative, moderate, liberal, vegetarian, or Nazi, it is entirely feasible with a simple point and click. If they want to isolate themselves and speak only with like-minded others, that is feasible, too. If they seek to read only those authors who agree with them and support the political candidates they favor, they are perfectly able to do so.

The implication is that groups of people, especially if they are like-minded, will end up thinking the same thing they thought beforebut in more extreme form, and sometimes in a much more extreme form.

At least as a matter of technological feasibility, and with the rise of countless options, the U.S. communications market is moving quickly toward this apparently utopian picture. It is not entirely different from what has come before. People who read newspapers do not all read the same newspaper, and some people do not read any newspaper at all. But in the emerging environment, there is a difference of degree if not of kind. What is different is a dramatic increase in individual control over content, along with a corresponding decrease in the power of general-interest intermediaries, including newspapers, magazines, and broadcasters. For all their problems, and their unmistakable limitations and biases, these intermediaries have performed some important democratic functions.

People who rely on such intermediaries experience a range of chance encounters with diverse others, as well as exposure to material they did not specifically choose. You might, for example, read a city newspaper and in the process come across stories you would not have selected if you had the power to control what you see. You might watch a particular television channel, and when your favorite program ends, you might see the beginning of another show, one you would not have chosen in advance.

In fact, a risk with a system of perfect individual control is that it can reduce the importance of the "public sphere" and of common spaces in general. One of the important features of these spaces is that they tend to ensure that people will encounter materials on important issues, whether or not they have specifically chosen the encounter. And when people see material they have not chosen, their interests and even their views might change as a result. At the very least, they will know a bit more about what their fellow citizens are thinking.

We can sharpen our understanding of this problem if we attend to the phenomenon of group polarization. Found in many settings, it involves like-minded people going to extremes. More precisely, group polarization means that after deliberating with one another, people are likely to move toward a more extreme point of view in the direction to which they were already inclined. With respect to the Internet, the implication is that groups of people, especially if they are like-minded, will end up thinking the same thing they thought beforebut in more extreme form, and sometimes in a much more extreme form.

Consider some examples of the basic phenomenon, as studied in more than a dozen nations. (a) After discussion, the citizens of France become more critical of the U.S. and its intentions with respect to economic aid. (b) After discussion, whites predisposed to show racial prejudice offer more negative responses to the question of whether white racism is responsible for certain conditions faced by African-Americans. (c) After discussion, whites predisposed not to show racial prejudice offer more positive responses to the same question. (d) After discussion, a group of moderately pro-feminist women become more strongly pro-feminist. (e) Republican appointees to the federal judiciary show far more conservative voting patterns when they are sitting on a panel consisting solely of Republican appointees; and Democratic appointees show far more liberal voting patterns when they are sitting on a panel consisting solely of Democratic appointees.

It follows that, for example, after discussion with one another, those who tend to dislike President Bush and the war in Iraq will come to dislike him and the war intensely; that those inclined to favor more aggressive affirmative-action programs will become extreme on the issue; and that those who believe tax rates are too high will come to think that large, immediate tax reductions are an extremely good idea.

The phenomenon of group polarization has conspicuous importance to the U.S. communications market, where groups with distinctive identities increasingly engage in within-group discussion. Customization makes this possible; specialized Web sites and blogs compound the problem. If the public is balkanized, and if different groups design their own preferred communications packages, the consequence will be further balkanization, as group members move one another toward more extreme points of view in line with their initial tendencies. At the same time, different deliberating groups, each consisting of like-minded people, will be driven increasingly far apart, simply because most of their discussions will be with one another. Extremist groups will often become even more extreme.

We cannot say, from the mere fact of polarization, that there has been a movement in the wrong direction. Perhaps the more extreme tendency is better; indeed, group polarization is likely to have fueled many movements of great value, including, for example, the ones for civil rights, abolishing slavery, and gender equality. All were extreme in their time, and within-group discussion bred greater extremism. Still, extremism need not be a word of opprobrium. If greater communications choices produce greater extremism, society may, in many cases, be better off as a result. But when group discussion tends to lead people to more strongly held versions of the same view with which they began, and if social influences and limited argument pools are responsible, there is legitimate reason for concern about sensible self-government.

Back to Top

Dangers to Democracy

Emerging technologies, including the Internet, are hardly an enemy here. They hold out at least as much promise as risk, especially because they allow us all to widen our horizons. We can certainly use them to learn more, rather than to live in an echo chamber. But to the extent they weaken the power of general-interest intermediaries and increase our ability to wall ourselves off from topics and opinions we would prefer to avoid, they create serious dangers to democracy. And if we believe that a system of free expression calls for unrestricted choice by individual consumers, we will not even understand the dangers as such.

Whether such dangers materialize ultimately depends on the aspirations, for freedom and democracy alike, by whose light we evaluate our practices. What I have sought to establish here is that in a free society, citizens aspire to a system that provides a range of experiencewith people, topics, and ideasthey would not have selected in advance.

Back to Top


Cass R. Sunstein ( is the Karl N. Llewellyn Distinguished Service Professor of Jurisprudence at the University of Chicago Law School.

©2004 ACM  0001-0782/04/1200  $5.00

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.

The Digital Library is published by the Association for Computing Machinery. Copyright © 2004 ACM, Inc.


No entries found