The Democratic candidate was crushed. An urban, ethnic liberal from the Northeast, he had been caught flatfooted by the waves of vitriolic attacks that smeared his background, his years of dedicated public service, the character of his beloved wife, as well as his religious beliefs and cultural values. He lost the heartland, and even the traditionally Democratic South had turned against him in unprecedented numbers, and it looked as though Republicans would continue to control not only the White House but also both houses of Congress and the Supreme Court for a long time to come.

The Democratic candidate I’m referring to was Al Smith. He lost his run for the Presidency in 1928 by a much larger margin than the one that defeated John Kerry, yet within four years, millions of the same Southerners and Midwesterners who had voted against Smith were embracing the New Deal coalition that would dominate American politics for most of the ensuing half-century. The moral of the story is that after their big loss last November, Democrats might be better off sitting on their hands rather than wringing them. In America, shifts in power rarely occur without some significant outside event. Anticipating just what that event will be—war, recession, scandal—is impossible.

“This was the greatest vote, the greatest margin and the greatest percentage (61 percent) that any President had ever drawn from the American people; we shall live long before we see its like again,” the inventor of the modern campaign chronicle, Theodore H. White, wrote after Lyndon Johnson’s lopsided triumph over Barry Goldwater in 1964. In fact we would see its like again twice in the next 20 years, and from the Republican side, as Richard Nixon and Ronald Reagan swamped their opponents in 1972 and 1984, respectively. Nor was White alone in his nearsightedness.

“The result was one of the great landslides of American political history, raising ominous question marks for the future of the Republican Party,” opined that wild-eyed young liberal Robert Novak after Goldwater’s defeat. This is why we historians love having the benefit of perfect hindsight. It saves us from having to propound the sort of hasty, hip-shooting predictions that can be roundly mocked afterward (usually by historians).

Some issues in our history have simmered for years, arising predictably in one campaign after another. Slavery (and later civil rights) was certainly one. The question of “hard” or “soft” money was another, playing a major role again and again in elections from the end of the Civil War into the 1930s. The same could be said, in recent decades, for the Cold War and crime.

But just as often, presidential elections have been decided by issues and events that four years earlier no one dreamed would have been important. How many people, for instance, would have predicted in the wake of LBJ’s romp that Vietnam would dominate the 1968 campaign—and force him out of the race? Who could have guessed, before 1950, that the 1952 election would hinge on a war on the Korean peninsula? Or that the obscure governor of Georgia would rise to the Presidency in 1976 thanks to “a third-rate burglary”?...

Those gripped by either despair or euphoria over the 2004 election might want to reflect on how quickly, and unforeseeably, political fortunes have changed in the past. As to just what the next catalytic shock will be, and how it will affect the 2008 election, I would be happy to make my own predictions. Just come see me 20 or 30 years from now.