Cass Sunstein correctly points out that digital technology provides consumers with unprecedented choice and control over their intake of news. What is far less clear, however, is whether the replacement of editorial selectivity by user selectivity will have the consequences he foresees. Should we really fear that personal selectivity will reinforce biased and uninformed opinions, reduce chance encounters with unknown voices, exacerbate group polarization, and strain the institutions that currently function to aggregate group interests into public policy? My own view is that the enhanced ability to self-select into news audiences facilitates the democratic process by empowering individuals.

To start at the beginning, I disagree with Sunstein’s premise, namely, that the present media marketplace—the one that is threatened by the Internet—resembles a “public sphere” and provides citizens with regular exposure to diverse perspectives. In reality, most users get their news from local television newscasts, where viewers encounter precious little “perspective.” If there is a perspective, it is that the world outside is a Hobbesian state of nature characterized by rampant threats to public order. By covering violent crime and little else despite declining crime rates nationwide, the conventional media reinforces this viewpoint. By frequently associating violent crime with nonwhite perpetrators, media coverage of crime has the effect of strengthening primitive racial stereotypes.

Even in the more prestigious journalistic forums, news content is characteristically shallow. During political campaigns, coverage is largely non-substantive. In place of hard information on candidates’ positions and past performance, media coverage gravitates toward the more “entertaining” facets of the campaign—the horse race, the strategy, and, whenever possible, instances of scandalous or unethical behavior. The “rhetoric of personal destruction” turns off voters and probably depresses participation.1

The professional culture of journalism further impedes the public’s interest in elections. While candidates have become savvy about utilizing the media to their advantage, journalists have resisted these efforts. Rather than depicting candidates as principled agents of the political parties who are committed to implementing their campaign pledges, reporters today emphasize the scripted and typically manipulative aspects of candidate behavior. As carefully controlled studies demonstrate, this “strategic” frame makes citizens distrust the individuals who seek elective office and feel cynical about the entire process.2 Media-based campaigns do not, in fact, function constructively to enhance the “public sphere”; rather, they breed apathy and negativism.

Against this backdrop, the Internet offers a promising avenue for renewing direct communications between candidates and voters. The proliferation of candidate and other political websites makes it possible for voters to bypass or supplement media treatment of the campaign. The 2000 election provided the first opportunity to test the potential of such unmediated campaign communication. Through the generous support of the Pew Charitable Trusts and the Carnegie Corporation of New York, the Political Communication Laboratory at Stanford University produced an extensive and easily searchable election database, which was distributed to any voter who requested it. Compiled on a compact disk, the database included the two major presidential candidates’ campaign speeches, televised ads, and party platforms. Of course, we did not expect that voters would read through this voluminous compilation from beginning to end. Instead, we expected them to seek out information that matched their political sensitivities. Rather than waiting passively, and most likely in vain, for the media to provide coverage of relevant issues, voters could, on their own, obtain information that was personally meaningful.

The CD initiative also included a controlled study of voter attitudes and behaviors. The CD was sent to a randomly selected sample of adults. The CDs produced for study participants were programmed to enable “usage tracking.” That is, each time the CD was accessed, the user’s computer activated a log of the specific pages visited and the length of time each page remained on the screen.3 Following the election, study participants returned their tracking data. They were also interviewed and asked questions about their levels of participation, interest in the campaign, and other measures of political engagement.

We found that use of the campaign CD consistently and powerfully strengthened interest in the campaign, voter turnout, and perception of the self as politically influential. Participants in the study also expressed somewhat more enthusiastic views of American-style elections. That is, use of the CD served to reduce their level of political cynicism.

The CD study also sheds light on selective exposure, the phenomenon Sunstein believes is the principal liability of the Internet. In suggesting that voters build “gated communities” in order to nurture their partisan preferences, Sunstein is following in the footsteps of the founding fathers of modern political communication research. Here is how Lazarsfeld and his collaborators described the problem of partisan selectivity in their classic study of the 1940 presidential campaign:

In recent years there has been a good deal of talk by men of good will about the desirability and necessity of guaranteeing the free exchange of ideas in the marketplace of public opinion. Such talk has centered upon the problem of keeping free the channels of expression and communication. Now we find that consumers of ideas, if they have made a decision on the issue, erect high tariff walls against alien notions.

Subsequent research into the partisan selectivity hypothesis has demonstrated that American voters are not especially motivated to tune out dissenting voices or sources of information. Based on their assessment of the relevant literature, David Sears and Jonathan Freedman in 1967 concluded that de facto selectivity—the process by which people decide about the extent of their political engagement—rather than motivated or partisan selectivity—the process by which people decide which information to pay attention to—was the norm.4 Given what we know about the level of information among the mass public, de facto selectivity implies that candidates should worry less about their ability to recruit from the ranks of the opposition and more about their ability to reach anyone at all.

Rather than screening information on the basis of their partisan values or ideology, voters are more inclined to employ a relevance, or utility-based, criterion that prompts them to tune in more carefully to news reports about issues that affect them. Thus citizens do not confine their attention to information they agree with; instead, they give special attention to information on topics that interest them. As candidates and the media discuss particular issues, the composition of their audience changes, so that voters personally affected by the “target” issue (for instance, unemployment) join the audience, while others for whom the issue is of less consequence, depart. This form of selectivity is hardly an impediment to deliberation; for a voter who places a high priority on healthcare, and who thus pays more attention to both candidates’ positions on the issue, selectively facilitates issue-oriented voting.

In keeping with previous research, our CD study detected only slight traces of partisan selectivity, but substantial evidence of issue-based selectivity. For example, voters who were more susceptible to out of pocket health-related costs (those with sick and elderly parents, or whose insurance coverage had been sporadic) devoted significantly more attention to both candidates’ statements and positions on health-related issues, regardless of the candidate’s party. On the other hand, Democrats and Republicans were only marginally more attentive to their candidates’ messages.5

Thus, Sunstein’s assessment of the political role of the Internet is suspect on two independent grounds. As a comparison of new media versus traditional media, his analysis suffers from a somewhat idealistic premise that traditional media provide a meaningful and accessible marketplace of political thought. In fact, little information is provided, and the information that is available turns voters off. Nor is there any serious ground for concern that online sources will only attract users who already share their points of view. The available evidence suggests the contrary. There is every reason to hope that the Internet will provide a better way to inform and engage voters.



1 Stephen Ansolabehere and Shanto Iyengar, Going Negative (New York: Free Press, 1995).

2 Joseph N. Cappella and Kathleen Hall Jamieson, Spiral of Cynicism: The Press and the Public Good (New York: Oxford University Press, 1997).

33 Of course, study participants were fully informed of this feature.

4 David O. Sears and Jonathan Freedman, “Selective Exposure to Information: A Critical Review,” Public Opinion Quarterly 31 (1967): 194-213.

5 My point here is not that partisan selectivity is absent; rather, the effects of partisan selectivity are much weaker than the comparable effects of issue-based selectivity.