Steve Ansolabehere reports an awkward encounter with Italian visitors shortly after the most recent presidential election. At that time the election was still unresolved and it was clear that there had been a number of problems in the casting and counting of ballots. The Italian visitors, he says, posed many difficult questions. Ansolabehere might be relieved, however, to know that it could have been even worse if the meeting had taken place a little later. After all the recounts and court decisions had been concluded and the new President had taken office, I spoke at a conference in Madrid; by that time the puzzlement overseas had been replaced by cynicism and scorn. This was the nation that so liked to lecture the rest of the world about democracy?

It is clear, regardless of one's political affiliation or how one might have voted in that election, that the credibility of America's election system and confidence in it—among both foreign observers and, more importantly, American voters—demands serious and prompt attention. In all such cases, however, two kinds of error may occur: inaction (or insufficient action) on the one hand, and excess, or overreaction, on the other. It is important to remember that what is at stake here is more than a question of perception: how our elections are managed is an issue at the very heart of American democracy. What is called for is a response that is both prompt (political sores such as this cannot be allowed to fester) and cautious (there is, as Ansolabehere implies, a danger that important principles will be neglected in the rush to "fix" the problem).

Having endorsed the essential determination that remedies are required, I would like to focus my attention on the twin dangers alluded to earlier: inadequate response and lack of restraint. Chernyshevsky, in the late nineteenth century, posed the famous question, "What is to be done?" That is, indeed, the essential question, and it carries with it the obverse: what is it that must notbe done?

One danger here is not that the issues will be treated too cavalierly—that their complexities will not be fully appreciated—but rather that they will be made to appear more difficult than necessary. Striving for a perfect form of representative government is a never-ending process, and it is unlikely that even the most ardent defender of the status quo would dare suggest that we have yet achieved an ideal system. A good many questions remain difficult to answer. Why, for example, are elections held mid-week, during normal work hours, when it is difficult for many to get to the polls? But if they were held on weekends, or if special holidays were created for elections, wouldn't many less-committed potential voters, as recent studies have indicated, simply choose to use the time for recreation rather than civic participation? Doesn't the electoral college undermine the one-person, one-vote ideal of a political democracy? But doesn't the organization of the United States Senate—two senators from every state, whatever the population—do the same thing, suggesting that perhaps pure "democracy" is not quite what the Founding Fathers had in mind? Is it not desirable to maximize the number of people voting? But is good government advanced by increasing the participation of citizens who have given scant attention to the issues at stake? And so on; there are lots of legitimate and important questions that have affixed themselves to the current controversies over technologies, ballot design, vote counting, and recount procedures. What can be done to strengthen civil rights protections? Can the voter registration process be made simpler and more accessible? Can more be done to protect the rights of voters with various forms of physical disability?

The problem is that as the attempt to "fix" the system is broadened to include suggested remedies for all evident or arguable shortcomings in American democracy (we can all agree that such shortcomings do exist), there is a danger that the inevitable disputes—partisan, ideological, and otherwise—will produce, as the saying goes, more heat than light; more recrimination and less solution.

The Caltech/MIT study, as well as other studies undertaken by Harvard, the Constitution Project (in which I participated), and the Ford-Carter Commission, produced a remarkable amount of agreement on a short list of improvements that can be undertaken fairly quickly. These include federal support for the purchase of improved technology (both for vote-casting and for sharing voter information among state agencies), and providing means by which voters can double-check their choices and cast provisional ballots if their eligibility is questioned. All of these proposed actions fall within the deliberately narrow formulation that framed the Constitution Project's work to forge a bipartisan consensus in support of election reform, to ensure that all registered voters who wished to vote were able to vote and to have their votes counted accurately. Admittedly, it is not a formulation that encompasses all possible improvements to the political system. But it is one that addresses the most glaring problems of the moment, is susceptible to the development of quick and relatively easy solutions, and, if met, would go far to rebuild public confidence that our election results accurately reflect the preferences of the voters.

The second great danger in the current debate is that the desire to craft a "national" solution to the problem will undermine the third important principle underlined in Steve Ansolabehere's paper: America's constitutional commitment to decentralized authority. There is no need to elaborate on Ansolabehere's points: state and local administration of elections rests not only on a philosophical preference for constrained federal power but on quite practical concerns as well; Americans get to choose their own leaders at all levels of government, and federal elections (for president and for members of Congress) occur simultaneously with the selection of thousands of other officials.

Just dealing with the voting machinery part of the solution is sufficient to create a month's worth of headaches for technology experts. The news media's impatient insistence on real-time results (including the development of exit polling so one need not wait until the votes have actually been counted) makes a widespread return to easy-to-read, easy-to-count paper ballots impractical. So we're faced with questions of how to tweak existing technology to allow voters to more easily follow instructions, double-check their votes, and ensure that no chads are allowed to clog up the works. As Ansolabehere points out, determining the "right" machinery is not as easy as it appears, and the development of new technology is highly expensive. That alone is a sufficient problem to occupy reformers.

One can understand the desire to address, now, the entire range of problems inherent in the implementation of American democracy, as well as the desire of some to reopen old debates (the electoral college, "motor voter" registration, on-line voting, etc.). In the political world, where it is often difficult to force action, it is tempting to hitch a ride on the next legislative train leaving the station. If not now, when? Once an election "reform" bill is passed, how long will it be before there is another chance to discuss these more difficult issues? The danger, however, is that in the desire to fix all perceived wrongs simultaneously, the whole effort will come to naught, the "good" falling victim once more to the pursuit of the perfect.

Developing new technologies is an important part of fixing what is wrong with American elections, and providing states with funds to buy those new technologies will go a long way toward restoring confidence in the system by which we choose the men and women who govern our collective community. Funding assistance, along with the establishment of a minimum number of mandates for voter access, vote-checking, and provisional voting, should be at the top of the political priority list. If not now… Well, the next presidential elections will be here soon, and none of us wants to face the Italians, the Spanish, or the voters in our own communities without having acted on this problem—the existence of which is all too obvious and the solutions to which are so surely within our grasp.