Go negative!

IN POLITICS, THE SEASON of attack ads is fast approaching-and with it come the predictable complaints about negative advertising. Tom Daschle, the former senator from South Dakota, once called attack ads ''the crack cocaine of politics," while The Washington Post's magisterial David Broder has lamented that most ads today contain ''never a hint as to why a voter should support the person paying for the TV spot." And it's not just pundits who hate negative ads. Four in five voters, according to a 2004 poll, said negative campaigns bothered them ''somewhat" or ''a lot."

Contrarians see consensus and sense an opening. And that's where John G. Geer, a professor of political science at Vanderbilt, comes in. In a new book, ''In Defense of Negativity: Attack Ads in Presidential Campaigns" (Chicago), Geer puckishly argues that negative ads are more nutritious for democracy than sunnier, Morning-in-America-style spots. ''You just don't benefit," he said in an interview, ''from seeing candidates running around in a grassy field with their families." People hate negative ads, but they hate broccoli, too, and both are good for you.

Not one to hold back, Geer writes, ''If negativity ever happened to disappear from our electoral battles, we can safely assume that so would our freedoms."

Many people have pointed out that American political campaigns have always been brawls. Andrew Jackson was called a murderer and a cannibal. Geer alludes to such historical arguments, but bases his own case on a statistical analysis of every commercial aired by a presidential candidate from 1960 to 2000: some 800 ads, featuring 9,600 claims and charges. (Geer discusses the 2004 campaign, but ads from that year are not in his main data set.)

Like other political scientists, Geer finds that there has, indeed, been an uptick in negativity. In 1960, the Kennedy and Nixon campaigns were atypically civil, with only 10 percent of the statements in the ads negative comments on the other guy. In 1964, the year that saw perhaps the most famous attack ad of all time-the Johnson campaign's anti-Goldwater ''daisy" ad (pretty little girl, mushroom cloud)-the figure jumped to around 30 percent, where it lingered until 1984. By 2004, Geer estimates, half of the candidates' claims were negative.

Sounds bad. But Geer says that the proportion of attacks on candidate ''traits" and ''values," the ones that seem most offensive to many people, is quite small-and hasn't changed over time. Instead, 72 percent of the claims in negative presidential ads speak to policy issues, a far higher figure than is the case with positive ads (49 percent).

The point, Geer says, is that campaigns should provide information about the differences between candidates on issues, and attack ads do a good job of this. Furthermore, the claims made in negative spots are far more likely to be documented onscreen than those in the positive spots. And of all categories of ads, the ones whose content most closely correlates with the issues voters say are most important to them are negative ads-specifically, attack ads by the challenger.

If we think negative ads have become overwhelming since the 1988 campaign, it's for two reasons, Geer writes: One, the political parties are simply further apart than they used to be. And two, the press, as it has grown ever more obsessed with the political process instead of policy, has relentlessly played up attacks (even as it deplores the negative trend).

Geer even questions how much voters are really turned off by the ads: Political scientists' latest estimates suggest that voter turnout, which is surprisingly hard to measure, has remained stable nationally for the last 30 years-and it jumped up in 2004.

Unsurprisingly, other political scientists raise questions big and small about Geer's argument. On the turnout issue, Shanto Iyengar, a Stanford political scientist, says that overall national turnout is a poor measure of the effects of negative ads: In controlled studies in laboratories, Iyengar has shown that exposure to even mildly negative ads makes independents, especially, say they are less likely to vote.

Thomas Patterson, a professor of government at Harvard's Kennedy School, questions two of Geer's fundamental premises: First, that negative ads tend to take up issues on the public mind. (The ACLU? The Pledge of Allegiance?) And, second, that attacks are almost never made up out of whole cloth. (Geer says the attacks on Senator John Kerry as a vacillator and President Bush as incurious got at something real.) ''So much of this negative advertising finds a wart and tries to make it stand for the whole," Patterson says. Knee-jerk bashing of ''negative" ads may be stupid, he adds, but numerous studies show that misleading negative ads corrode trust in democracy. Factcheck.org, a service of the Annenberg Public Policy Center of the University of Pennsylvania, found inaccuracies in 88 percent of the attack ads it examined from the 2004 campaign.

Geer says this is partly because negative ads simply include lots of facts to quibble with, and partly because all advertising, by its nature, includes a bit of exaggeration. Which gives me an idea for an attack ad should Geer ever run for office: ''Professor endorsed lying." A little truth-stretching never hurt anyone, you know.

Christopher Shea's column appears biweekly in Ideas. E-mail critical.faculties@verizon.net.