Curtis Gans, the director of the Committee for the Study of the American Electorate, expects between 118 million and 121 million Americans to vote next week, up from the 105 million who voted in 2000. The larger figure would represent a turnout of 60% of eligible voters, the highest since 1968, and is one indicator of the passions aroused by this campaign. But regardless of turnout and your preferred candidate, we all ought to hope that the victor wins by a big enough margin to avoid an aftershock of contested ballots, lawsuits, and court decisions.

We don’t need a repetition of Florida on a grander scale. The danger is not simply a delay in knowing who the next president is, or the prospect that he’ll be hampered in governing or the probable fury of the loser’s supporters that the election was “stolen.” The more unsettling danger is that, having engaged in two rounds of post-election combat, party warlords will make this a permanent part of the political process.

Election by litigation is a sensationally bad idea. Undertaken piously to guarantee voters’ “rights” or to prevent “fraud,” it would erode popular confidence in elections’ integrity. We’d be bombarded (as we already are) by complaints about how corrupt voting practices have become.

Sooner or later, many Americans might cynically conclude that the side with the busiest poll watchers, cleverest lawyers, and friendliest judges had secured an unfair advantage.

We aren’t there yet. A new poll from the Pew Research Center finds that only 11% of registered voters fear that their votes won’t be accurately counted – 62% are “very confident” that votes will be correctly counted, 26% “somewhat confident.” But anything that taints the voting process corrodes trust because voting is so symbolic. We see it as an inalienable right. This is a psychological truth and, unfortunately, a historical half-truth.

Voting was long a limited privilege, as Harvard historian Alexander Keyssar shows in his book “The Right to Vote.” In 1790, most of the original 13 states restricted the vote to white males who either owned property or paid local taxes. Perhaps 60% to 70% of white men could vote, Mr. Keyssar says. By the Civil War, most property restrictions had disappeared. But there was a constant tension between demands to expand the electorate and more restrictive pressures: selfish fears by ruling elites that more voters would threaten their power; and principled worries that too much democracy, that is, letting “undesirables” vote, would destroy democracy.

Even in 1824, not all states allowed citizens to vote directly for president; some state legislatures still selected presidential electors (the Constitution lets states decide how to pick electors). Later, war often served to expand voting. Congress passed the 19th Amendment guaranteeing women’s right to vote in 1919, partly because women had helped so much in World War I. They “feverishly sold bonds…knitted clothes, and gave gifts to soldiers,” writes Mr. Keyssar. In the Vietnam War, Congress lowered the voting age to 18; it was hard to ask soldiers to die if they couldn’t vote. After the Civil War, blacks received the right to vote via the 15th Amendment, ratified in 1870.

Of course, the 15th Amendment was often nullified. Physical intimidation, literacy tests, and poll taxes kept huge numbers of blacks in the South from voting. But that shameful chapter is now past. Mr. Keyssar judges that the Voting Rights Act of 1965 and various court cases constitute a “revolution’ in voting access. Indeed, it is because voting rights are now greater than ever that the potential for disillusion is so great.

We know from Florida and from history that, when votes hang in doubt, all heck breaks loose. In a four-man race, Andrew Jackson won the popular vote in 1824 but not a majority of electoral votes. The House of Representatives selected John Quincy Adams; Jackson and his enraged followers saw this as theft. In 1876, Democrat Samuel Tilden won the popular vote, but amid charges of fraud four states submitted competing electoral votes. Congress chose Rutherford B. Hayes, often referred to as “Rutherfraud.”

Against this backdrop, it’s hard to be optimistic about next week. The Washington Post reports that Democrats plan to have 10,000 lawyers “at the polls in battleground states”; Republicans say their lawyers will be “covering 30,000 precincts.”

Campaigns can never be “fair,” but elections should be. People need to have confidence in the honesty of the process. The next president ought to fashion a bipartisan consensus over acceptable election procedures. But this will be hard if there’s a post-election legal slugfest and the outcome is settled by litigation. Here’s what we really need: the winner takes 52% of the popular vote and enough electoral votes so that a few close states don’t matter.