Should Exit Polls Be Released Before Voting Ends?

Scott Shuger is a Slate senior writer and author of “Today’s Papers.” Jack Shafer is Slate’s deputy editor, “Press Box” columnist, and resident embargo-breaker (click here to catch up on the controversy and to read Shafer’s op-ed that ran in the Wall Street Journal).

Hi Jack:

Because I, too, am amazed at how often journalists have one set of rules for themselves and another for everybody else and because I think polls have gotten way out of hand and generally love anything that takes them down a peg, it feels odd not to be on your side on the exit-polls flap. But I’m not.

In your, you correctly boiled the controversy down to the question “Why should news organizations care about voter turnout?” Your answer is, they shouldn’t. This is where I disagree. Sometimes—as in the Weimar Republic elections of 1933—the press should even care mightily about which candidate wins. That’s because a free press has a stake in freedom. I believe journalists operating in civil society have a stake in keeping civil society possible. For instance, although it’s fun to bring down the mighty, the real reason we should be interested in doing stories that expose bad behavior by the powerful is because when the powerful behave better, society works better. And because a society in which folks feel like stakeholders works better for all, this same sense of civic journalism commits me to the view that almost always, the press should avoid promoting voter apathy. (I am a minimalist on civic journalism: I say it’s not the media’s job to boost turnout, but that one of its jobs is not to depress it.)

And I think the official Slate “let them eat exit polls” position is a bit cavalier about this, primarily because it casts the argument in purely a priori terms rather than looking at some of the facts of elections. In your Journal piece, you were content to cite one study finding that early projections deter less than 3 percent of the voters. But think about it, Jack—the 1960 presidential election was decided by a margin of 1 percent—a difference of one vote per precinct. How exactly would it play into your sense of anti-pundit iconoclasm to be willing to let Walter or Chet help make Nixon president in 1960?

Similarly, our friend and fearless leader Michael Kinsley wondered in a 1992 Time essay that is the locus classicus for your argument, “Why should knowing the outcome discourage voters for the loser more than voters for the winner, or vice versa?” Well, if all voters were as clearheaded as Mike, it wouldn’t, but to assume that is to think of elections as much tidier affairs than they really are. For instance, based on the most recent pre-election polls, in last week’s California primary, Gore voters stayed home in much greater numbers than McCain voters did, even though they all should have known that their individual votes were equally insignificant. And there were consequences: The turnout was unusually conservative, and as a result, the conservative positions on most of the ballot referenda prevailed. That’s not the sort of consequence I want Dan and Tom and Peter to help achieve.

Your position gets its rhetorical force from the paradox that although elections are important, your vote (my vote, her vote) isn’t. And this leads to a dangerous temptation: to reason that since my vote isn’t important, I shouldn’t bother. This is an argument anybody can make with equal force, but because that way lies general apathy, it’s in society’s interest and therefore, by the terms of civic journalism, in the media’s interest, not to make it. Even when a vastly superior army assaults a minimally defended position, the first wave of the charge will suffer a lot of casualties, but it would not be very smart for the officer leading the assault to point this out. Voting, like an infantry charge, is susceptible to the “threshold effect”: It works only if a certain number of people participate. And that’s why it’s counterproductive to remind each person that he/she has excellent reasons not to participate. Broadcasting early exit polls is just such a reminder.

But those reminders—about the dangers of going over the top first, about the pointlessness of your vote—are true. Which gets at the last point I’ll make for now. Your position makes the plausible assumption that the press should always pass along the truth as soon as it gets a hold of it. Plausible but false. Reporters covering a war should often withhold publication of information the dissemination of which might cause a military operation to fail. If a police reporter learns of a single bit of information about a killing that only the cops and the real murderer know, he should hold off on publishing it until the cops have had a chance see if the suspect will provide it without being able to say that he read it in the newspaper. Of course, there are value choices beneath such decisions to suppress the truth. But the view that journalists don’t have to truck with values, that they’re mere conduits of the truth, just doesn’t fit what we do. A workable society depends on sometimes trumping truth with other values. You still laugh at the boss’s jokes, brides still wear white, and funerals still keep mum on the matter of worms. One of those other values is the false feeling that my vote counts.