With Election Day (finally) near at hand, the nation is braced for a long count. It could hardly be otherwise—given not only our 36-day adventure in Florida four years ago but also the armies of poll watchers and lawyers now reporting for duty in the swing states. Indeed, the government is sufficiently alarmed by the specter of Florida 2000 that, just days before the election, it announced the assignment of a corps of more than 1,000 Justice Department marshals to observe voting procedures in “trouble spots” among the nation’s more than 192,000 polling districts. Although this column is ill-timed for predictions, awkwardly appearing the Monday before Election Day, it nonetheless seems a safe bet that few Americans will go to bed early on Tuesday night knowing who their next president will be.
The expectation of another nail-biter is also fueled by the polls, which tell us that this will be a close election. They also tell us that President Bush is running slightly ahead of Senator Kerry, but the gap is not significant, politically or statistically. Then, too, the national polls don’t actually mean anything, thanks to the splendors of the Electoral College. The polls that really matter are in Wisconsin and Ohio and Florida and New Hampshire—and they’re close, too.
The polls may also be wrong—which compounds the uncertainty. Pollsters (as is now well known) have not been calling cell-phone numbers: This certainly means that the young are underrepresented in the polls—since few young people (who favor Kerry by a wide margin) have phones that do not fit easily in their pockets for travel to bars and movies.
A deeper problem with the polls is the “likely voter” issue. To be accurate predictors of the election, polls have to capture the opinions of people who are really going to vote. This can be a tricky business because of one of the most fundamental and troubling facts about American political life: Half of all Americans eligible to vote usually do not. Voters and nonvoters, moreover, are not identical populations: In the recent past, turnout has been relatively high among the well-educated and the well-off, lowest among the poor and the young.
“Our creaky and failure-prone election process has become a magnet for conflict and manipulation.”
If the polls turn out to be wrong—which may mean that the election is not a cliffhanger—it will be because of a surge in turnout somewhere in the electorate. A huge showing of young people, poor people, and minorities could sweep Kerry into office; lackluster turnout from those groups, coupled with even-higher-than-expected turnout from Evangelical Christians, will give Bush four more years. Part of the charm of American politics (which drives political professionals nuts) is that no one can be sure which electorate is going to show up.
It would be comforting to think that the fog will lift on Tuesday night when we begin to count real votes. But one thing we certainly learned from Election 2000 is that “real votes” are not always easy to identify, and counting them can be a strangely challenging task. Vote counting, like the right to vote itself, is a haphazard affair. Even in the face of hard-won constitutional protections and great technological advances, our democracy uses no single standard to define legitimate ballots and tally them up. Although Florida has now gotten rid of its punch-card machines, three quarters of the districts in Ohio are still using them, auguring a possible revival of the Battle of the Hanging Chad. Some of the new electronic voting machines in use throughout the country will surely malfunction, and many of them—including the machines used by roughly half of Florida’s voters—do not have paper trails that would even permit manual recounts.
There will also be problems identifying voters. Most states have delayed modernizing their registration lists (as the misnamed Help America Vote Act instructed them to do), which means that numerous voters will show up at the polls only to be told that they are not registered. This will happen because of either errors or dirty tricks (there are reports from Nevada and Oregon that a Republican-linked firm registered new voters and destroyed the forms submitted by those who said they were Democrats). Other voters could have their legitimacy challenged at the polls, for example by the thousands of poll watchers hired by the Republicans to “prevent fraud” in Ohio’s cities. In theory, the post-2000 requirement that states offer such voters “provisional ballots” (which would be counted only after registrations were confirmed) should alleviate these problems, but the procedures governing provisional ballots are already complex enough to have spawned lawsuits. Counting provisional ballots, moreover, will take forever (since they have to be individually verified), and there may be millions of them.
The palpable messiness of our electoral system has deep historical roots: It can be traced, in part, to flaws embedded in our Constitution. If there is an original sin (other than slavery) in our political order, it is the absence of a constitutionally guaranteed right to vote. The Founding Fathers, for reasons having less to do with political theory than with the pragmatic politics of the 1780s, left nearly all control of voting rights to the states; they also left it to state legislatures to determine the “manner” in which presidential electors would be selected. Those constitutional decisions left citizens with something less—often far less—than ironclad assurances that they could cast their ballots and have their votes counted. Entrusting electoral matters to the states permitted the disenfranchisement of millions of Americans, including African-Americans in the South, Native Americans everywhere, and non-English speakers in New York City (until the mid-sixties). It also permitted the luxuriant growth of state laws and regulations that may keep hundreds of lawyers busy in coming weeks.
So it could be a long night on Tuesday—or a long week. History doesn’t repeat itself in detail, and it is not likely that we will again find ourselves riveted to recount debates in Florida. But our electoral apparatus—with dozens of different technologies, laws, rules, and procedures—seems ill-equipped to produce a smooth process and an uncontested result; creaky and failure-prone, it has become a magnet for conflict and manipulation. The system hardly seems designed to promote the performance of democracy’s most sacred ritual. After this week’s tumult has subsided—and however it all turns out—we might want to do something about that.