In other words, "the people" lied or honestly deluded themselves. Or at least 15 percent of them did. But we don't know which 15 percent and we never will, just as we don't know how many Americans lie, fudge, or mislead pollsters. We know a large number must, because pollsters are constantly asking "the people" about incredibly complex issues and "the people" almost always pretend to know what they think.
I'm sorry, I may not be smart enough to understand why "Anchorman" isn't a classic of American cinema. But I do know a lot of really smart people, and when I ask them the same questions pollsters regularly ask (Should Israel trade land for peace? Is the war in Iraq going well? Is Social Security partial privatization a bad idea? Why is "Charmed" still on TV while "Angel" and "Buffy" were cancelled?) and I usually get six-part answers, festooned with ifs, ands, buts and on-the-other-hands. But "the people" always seem to have a fully-formed opinion handy.
And this leaves out the fact that a big chunk of "the people" are grotesquely ignorant about their government and current events. And I'm not just referring to that running segment on the "Tonight Show" where Jay Leno asks people to figure out what Flag Day celebrates. Just this month, the ABA (hint: the lawyer thing, not the bunch with the red, white and blue basketball), released a poll that found that 22 percent of Americans think the three branches of government are Republican, Democrat and Independent. In 1991 another ABA survey found that one-third of Americans didn't know what the Bill of Rights is. In 1987, 45 percent of Americans thought Karl Marx's dictum "from each according to his ability to each according to his needs" was in the U.S. Constitution. In 1964 (!) only 38 percent of the American people knew the Soviet Union wasn't in NATO.
Now, I don't think the American people are as stupid or confused as all this suggests. But if they aren't, the polls must be. Maybe people panic when they talk to pollsters. Maybe the methodology stinks like that aardvark-sock thing. Maybe respondents are distracted by the pollster's annoying habit of tapping his glass eye with a ballpoint pen while awaiting an answer. Who knows?
What I do know is that, even if the polls were 100 percent accurate, they would still be a stupid way of setting policy. Why? Because "the people" only exist on election day. Before and after that, they're just a bunch of individuals spouting off to strangers.