Put 'em to the test: for a nation of minimally informed voters.(THE NATION II)

LAST summer, Mark Osterloh of Tucson, Ariz., was the Paris Hilton of the civics newsbeat, receiving lavish coverage for his cockamamie scheme to bribe Americans into voting. The eccentric activist wanted to make Arizona ballots double as lottery tickets, essentially enrolling every voter in a million-dollar drawing.

The Founding Fathers, spinning in their graves, must have paused in their dervish-like contortions to appreciate the almost magisterial stupidity of this idea. Still, Osterloh's scheme reflected the basic assumptions driving the national panic over low voter turnout: Healthy societies have lots of voters, unhealthy ones don't, our public policies are flawed because not enough people vote, and if everyone did vote our problems would be solved.

Of course, the conventional wisdom among left-leaning scholars and most journalists is that, if everyone voted, it'd be a huge boon for the Democrats. Arend Lijphart, a former head of the American Political Science Association, argues for mandatory voting on the assumption that America's social policies would be vastly improved if poor, uneducated, and otherwise disengaged citizens had more power over government priorities. This is the real agenda of the get-out-the-vote industry. Rock the Vote and its various sister organizations are simply dishonest when they say they want people to vote. What they really want is for people to vote for a prepackaged "youth" ideology that includes the usual wish-list of liberal policies, from environmental protection to education subsidies to affirmative action.

In fact, it's far from clear that maximum turnout would help the Democrats. "Simply put," political scientists Benjamin Highton and Raymond Wolfinger wrote in 2001, "[American] voters' preferences differ minimally from those of all citizens; outcomes would not change if everyone voted." Ten years earlier, Stuart Rothenberg had issued his study "What If Nonvoters Voted?" He concluded, "There is no compelling evidence that nonvoters are so distinct from voters that they constitute a bloc ready to alter the fundamental balance of power in this country." The key difference between voters and nonvoters isn't race, age, class, or gender, but laziness. People who've recently moved, for example, vote less than others because they haven't gotten around to registering. If you disproportionately activate liberal nonvoters, they'll probably break Democratic, while conservative nonvoters will probably break Republican.

Left-wingers think this analysis misses an important point--that if a politician aimed his campaign at nonvoters, he would try to activate them with decidedly populist issues. There's some evidence to back this up. Studies show that, if everyone voted, candidates such as Ralph Nader and Ross Perot would do somewhat better. Lijphart argues that opinion polls showing few differences between voters and nonvoters miss the fact that nonvoters "have not given these questions much thought," haven't been "mobilized," and lack "class consciousness." Translation: If you can get the poor and uneducated all riled up, they'll demand a bigger slice of the pie, and America will be a better place.

But what if Osterloh, Lijphart, et al. are wrong? Wrong about what? you ask. About everything. Ever since the Progressive era, America has been working on the assumption that if you have just a little more democracy, a little broader franchise, then--ta-dah!--everything will be fixed. Bryan Caplan, in his spellbinding new book The Myth of the Rational Voter, calls this sort of thinking "democratic fundamentalism." Consider the hoary cliche, attributed to Al Smith in 1928, that "all the ills of democracy can be cured by more democracy." As Caplan notes, this means that no evidence can ever, under any circumstances, be held against ...