How Might Cyberspace Change American Politics?


Eugene Volokh*


(34 Loy. L.A. L. Rev. 1213 (2001))



How might cyberspace technology change American politics? Any answer to this question is guesswork, but the participants in this conference have provided some extremely intriguing guesses. Rather than trying to respond to their thoughts point by point (a project that would probably be of little interest to them or to the reader), let me try to make two guesses of my own-more to broaden the range of possible futures that we're discussing rather than to deny the possibility of one future or another.


I.  Electronically Guided Voting

Most people have little idea about which way they should vote on many matters, especially as to nonpartisan races and many initiatives and referenda. Being rational consumers of political information, voters don't spend the many hours needed to educate themselves on every race; rather, they rely on proxies, such as party affiliation, endorsements, or the identities of the people signing the arguments for or against an initiative. But often even this rough data isn't easy to gather, and if you forget your cheat sheet at home, you might just not vote on some issues, or make a very rough guess.

Say, though, that you're voting from home on the Internet; and you know that going to will not only show you the National Organization for Women's recommendations, but will also let you-with one or two clicks-download a small program, that, when executed, places the recommended votes right into your electronic ballot form. Of course, NOW might not have views on all the races, but you could also go to the site and download its recommendations; the software could even flag for you the discrepancies between the two recommendation packages.

If you'd like, you could even go to a central site (say,, fill in a form indicating which groups' views you respect, and with a few clicks have their recommendations merged and placed right into your ballot.[1] Then one more click by you, and your filled-in ballot gets sent to the elections board. (The idea so far borrows heavily from Jerry Kang's extremely thoughtful comments at this very conference.[2])

A lot of people, of course, might not want to delegate so many of their voting decisions to groups, even groups that they mostly trust. On the other hand, most of us already delegate these decisions to newspaper editorialists, friends, and so on: We realize that, after long study, we might be able to better decide which vote better comports with our own true views, but few of us are willing to undertake the long study.

If people can fill in their ballot in a few keystrokes, relying on the downloaded suggestions of trusted political parties, public interest groups, and commentators, I suspect that many (though by no means all) voters will take advantage of this. True, voters could already do something like this today by bringing several printed recommendations into the ballot booth with them, one thing the Internet teaches us is that little differences in convenience can translate into big differences in behavior: When people can instantly use the recommendations, rather than having to file away and then sort through every mailer or editorial they read, they'll be much likelier to act on those instant recommendations.

Of course, the downloaded recommendation program would have to come from a credible source; if I want to vote as the Libertarian Party and the Logging Industry Council suggest, I want to make sure that I'm indeed following their suggestions and not someone else's. But if becomes a trusted name-perhaps it might be put up by CNN or the League of Women Voters or some other reputable institution-people will be quite willing to believe that the recommendations recorded on it are authentic. Certainly voters should be much more comfortable following those recommendations than the recommendations on some slate mailer that they get a few days before the election.

So Internet voting might increase the power of trusted interest groups, because their recommendations will make more of a difference; but (and here is where I try to build on Jerry Kang's thoughts) think how much it can increase this power if the groups can measure and report the number of times their ballot-filling-out program was downloaded and executed. If the NRA or the ACLU can go to a legislator and say "Last election, 30,000 voters in your district downloaded and acted on our recommendations," the legislator is likely to take notice, and is more likely to do what it takes to get the group's endorsement.

People know how many readers a newspaper has or how many members a group has, but they don't know how many people act on the newspaper's or group's printed advice. But when recommendations have to be individually downloaded and executed, each use of the recommendations can be reliably counted.[3] Nor is it necessary that all the group's sympathizers use this program; if even a large minority does so, that could still substantially increase the group's leverage with politicians whom it's lobbying.

My sense is that such a system would evolve quickly after Internet voting is introduced. It shouldn't be technically hard to implement; at most, it would require users to download a program that finds the ballot window on the computer, identifies the boxes, and fills them in properly.[4] Interest groups should have a strong incentive to provide such programs, both because such programs will increase the vote for their favored candidates and measures, and because they will give the groups more influence with the candidates. And voters who sympathize with those groups, and want to increase their favorite interest group's power, should likewise have an incentive to use these programs: After all, these programs give each voter a cheap extra opportunity to express his voice (by supporting a particular interest group) in a way that politicians should notice.

Will there, should there, and can there be a legal reaction to such programs? One could imagine the government trying to ban such programs on the grounds that they are a sort of ballot tampering. Just as many states try to ban electioneering near physical polling places,[5] so the government might try to ban the distribution of recommendations into a ballot window on a person's computer.

But, first, there is at least a credible argument that such proposals might violate the First Amendment; after all, their very point is to prevent organizations from giving a certain kind of voluntarily sought-after advice about the election.[6] Second, it's not clear that the public will support a law that effectively bars people from using, in their own homes, a computer program that makes it easier for them to vote and to express support for their favorite interest groups. And third, many of the interest groups whose power would be increased by electronically guided voting are already pretty powerful; they may well do a good job of fighting any attempts to block such guided voting. (Of course, some other interest groups that might see these programs as diminishing their influence or strengthening their opponents' influence might support bans on such programs.)

The government might also try to stop these sorts of programs technologically. The easiest way to implement Internet voting, and the way used in the 2000 Democratic Party primary in Arizona,[7] is through a standard Web interface that's set up as just another window on your computer; and if that's the interface that the voting system uses, it's easy enough to write a program that pastes data into that window or clicks the proper window buttons. To prevent this from happening, the government could stop using a standard Windows interface and instead require people to vote using a special program that's run on their computers and that disables any concurrently running processes that might tamper with its data.

Ultimately, though, such unusual interfaces might be less appealing to the voting public (and also more error-prone) than the more standard window-based interfaces. In the computer industry, going with well-established, well-tested, familiar user interfaces is usually the best bet, especially given that many consumers are uncomfortable with computers in the first place. Thus, the government may thus be driven back to the standard approach, notwithstanding the fact that such an approach makes guided voting quite simple.

So I suspect that guided Internet voting won't be easy to stop (though it might in fact be stoppable); and I have no conclusion on whether people should try to stop it. But I do want to flag the possibility that guided Internet voting could, if it isn't stopped, become a tremendously important part of the political landscape, both because of the way it affects how people vote and of the way it affects the power of those who make voting recommendations. Any investigation of the merits of Internet voting thus must include consideration of the merits of guided Internet voting.

II.  Cyberspace and Geographical Consciousness

On a different note, let me suggest that cyberspace may in some ways increase people's sense of connection with their geographical community-and thus might influence their judgment on how our political system should treat geographical communities-at the same time that it in other ways decreases this connection.

Right now, Americans move often, which probably diminishes their connections to their home communities. What's more, many of their moves are to places where they need to work; I suspect that in such situations, many newcomers may feel especially little attachment to their new home community, because they didn't voluntarily select it but rather had it thrust upon them by economic necessity.

If e-mail and, more importantly, videoconferencing make it easier for people to work from home, then more and more people will live where they like living (say, Monterey or Montana or their family's home town) rather than where their jobs require them to live. People might thus move less often than they do now, and even if they do move, they may feel that they've chosen their new home town, and thus might feel more affection and concern for it.

This possibility sheds some light on Eben Moglen's and Pam Karlan's conjecture that the Internet, by fostering the creation of online communities united by common interests rather than physical proximity, will diminish people's fixation on geography and thus the American insistence on geographically defined legislative districting: "[A]s individuals grow comfortable with the idea that important communities may be entirely non-geographic and that individu­als may affiliate for important purposes along dimensions that have little or nothing to do with where they live, they may become both more skeptical of a purely geographic way of defining political constituencies and more open to election systems that allow voters to affiliate along dimensions other than residence."[8]

Moglen & Karlan's theory is intriguing, and it may well be right. Still, I'm somewhat skeptical, for three reasons. First, as Michael Froomkin put it, "No-one lives in cyberspace." Even if people make plenty of friends online, many of the main reasons they care about their government-control of violent crime, environmental protection, and the like-will still be focused on where they happen to live.

Second, Americans have long been participating in self-conscious communities defined by common interest (religion, politics, ethnicity, hobbies, consumption patterns[9]), but such participation has not appreciably diminished of support for geographical districting. Such communities may have led Americans to care somewhat more about national politics than they otherwise might have: The more personal connections people have to people in other states, the more they'll feel like Americans rather than just Californians or New Yorkers, and the more willing they'll be to see federal rather than state solutions. But these national communities don't seem to have led people to focus on nongeographical representation systems.

Third, if the conjecture with which I began this section is correct, and if Moglen & Karlan are right that the level of connection to one's geographical community influences one's support for geographical districting, then it follows that the Internet might make geographical districting more appealing-for at least some people and for some reasons-rather than less. True, this effect may be present, at least at first, mostly for a small, disproportionately professional and intellectual group of people whose jobs make telecommuting easier. But this is also the group that because of its relative wealth, articulateness, and political engagement is probably most likely to disproportionately influence public policy.

What will be the relative magnitude of the effect that Moglen & Karlan posit and the effect that I suggest? That's anybody's guess, but we should remember that the influence of the Internet on geographical consciousness is not entirely in one direction.

* Professor of Law, UCLA School of Law ( Many thanks to Todd Gaziano, Orin Kerr, Gary Leff, and Gene Meyer for their very helpful suggestions.

[1] Compare, which asks you for the names of groups or people that you trust and shows you their recommendations on various questions. This site isn't integrated with Internet voting software (since Internet voting is still so rare), but it easily could be when Internet voting becomes commonplace.

[2] Jerry Kang, E-Racing E-Lections, ___ Loy. L.A. L. Rev. ___ ¶¶ 28-29 (2001) (Nov. 29, 2000 draft).

[3] The count will be most reliable if the recommendation program (1) fills in the ballot with the requested recommendations, (2) asks the voter whether he wants to submit the e-ballot, (3) triggers the sending of the ballot to the elections board, and (4) only then updates the count for those organizations whose recommendations have been used. Some voters may be reluctant to use such programs, since they may fear that the programs will jeopardize the privacy of their vote (given that the programs will have to send a message back to the central computer after the vote is cast, in order to update the counts for each recommender). Cf. Paul M. Schwartz, and Internet Politics, ___ Loy. L.A. L. Rev. ___, at 11-13 (forthcoming 2001) (Nov. 16, 2000 draft). Still, if the site becomes credible enough, this reluctance may diminish; recall that already people are willing to give up a considerable amount of privacy to vote, for instance making their party affiliation a matter of public record in order to vote in most primaries.

On the other hand, if the site only updates the counters when it downloads the recommendation program, then it'll be possible for people to visit the site many times, download the program many times, but only vote once (or not at all), thus artificially boosting their favorite organizations' recommendation counts; this will make the counts much less accurate. Still, the counters would probably remain more accurate than the less reliable proxies that are used now, such as a newspaper's readership or an organization's membership.

[4] When I say "fill in the boxes," I mean to include both actually filling in the empty spaces in forms that call for such action, and "pushing" the buttons (which is to say marking the buttons as pushed) in forms that operate by asking users to click on a particular button.

While I'm not an expert on Windows, and while of course the details of the program I describe would depend on the details of the particular computerized voting system that's being used, my 12 years of experience as a computer programmer (1980-1992) lead me to think that this should be a pretty simple program to implement.

[5] Cf. Burson v. Freeman, 504 U.S. 191, 215 n.1 (1992) (Scalia, J., concurring in the judgment) (collecting statutes).

[6] Such a law would differ from the one upheld by a divided Court in Burson, because it wouldn't raise any concern about intimidation of voters caused by the physical presence of demonstrators or other political speakers.

[7] Cf. (providing a demonstration of the Arizona system).

[8] Eben Moglen & Pamela S. Karlan, The Soul of A New Political Machine: The Online, the Color Line and Electronic Democracy, ___ Loy. L.A. L. Rev. ___ (forthcoming 2001).

[9] See, e.g., Daniel J. Boorstin, The Americans: The Democratic Experience (1973) (describing the emergence over the last two centuries of "consumption communities," "invisible new communities . . . created and preserved by how and what men consumed").