The marketing of fear for political advantage has become so ubiquitous that the phrase “the politics of fear” is almost a cliché, but still many doubt the power of fearful messages to influence voters. “Despite the best of intentions, election campaigns can quickly turn into a competition about who can most effectively frighten voters,” complained Dr. Rowan Williams, the Archbishop of Canterbury, in a letter sent to Britain’s party leaders prior to the 2005 election. Don’t let this happen, the archbishop asked. It’s unethical and destructive. And besides, it doesn’t work. “Like a lot of other people, I suspect that voters don’t make up their minds primarily on the grounds of fear, and that this aspect of campaigning, while it certainly grabs headlines, may not be especially decisive. The technique is a bit too transparent and usually too over-the-top to be taken wholly seriously.”
Academic research on this point is surprisingly limited. Even the role of emotions in campaign ads—the most explicit and quantifiable means of deploying emotion in politics—has been little studied. But still, among campaign consultants and political journalists, there is “widespread agreement about the practices of selling candidates,” writes Ted Brader, a political scientist at the University of Michigan, in Campaigning for Hearts and Minds. “These political observers believe that emotional appeals are powerful staples of campaign advertising, embedded in music and images, that manipulate uneducated and uninformed voters.” To put this in psychological terms, when an educated and informed voter sees an ad, Gut may react but Head corrects and adjusts; when an uneducated and uninformed voter sees the same ad, Gut reacts but Head does not step in, and so that person is swayed by the emotional pitch. Or so the political experts assume.
Brader first tested these assumptions by conducting a large-scale analysis of ad content. “Almost all campaign ads appealed to the emotions of viewers and yet a substantial majority, 79 percent, also appealed to the viewers’ capacity to reason by encouraging them to draw conclusions from evidence,” Brader wrote. “Nevertheless, the reputation of political advertising as primarily emotional is well founded: in nearly 72 percent of ads, the appeal to emotions dominated the appeal to logic.” Only 10 percent of ads targeted a single emotion. Three-quarters included at least one enthusiastic appeal—Fred Jones for a brighter future!—while “nearly half of all ads include some sort of appeal to fear, anger and pride.” The ads Brader analyzed came from elections in 1999 and 2000, including the notably low-key presidential contest between Al Gore and George W. Bush. The 2004 presidential election was a far nastier affair, and it’s safe to say that fear was more widely deployed in that campaign. So were TV ads. “Americans in 2004 were collectively exposed to over a million campaign ads on television, ” Brader writes. “Candidates, parties and groups spent over $1 billion on advertising.”
Brader found that political ads use fairly predictable sounds and images to enhance emotional content. Enthusiastic appeals are drenched in bright colors, sunshine, smiling children, and soaring or sentimental music. Fear ads are often shot in black and white or very dark colors. They are “rich in visual cues associated with death, decay and desolation” such as old people and barren landscapes. Music is either tense or somber, or it’s simply discordant noise.
Although ads usually have a dominant emotional theme, one-third “contain appeals to both positive and negative emotions.” Brader notes this fits with how Rutgers professor Montague Kern described political advertisingin the 1980s: It is “the ‘get ’em sick, get ’em well’ advertising concept, in which advertisers try to create anxieties and then reassure people they have the solution.” This should sound familiar. It is the same message used in those cholesterol ads, home-security pitches, and the banner featuring a little boy playing soccer behind an electrified fence: You are threatened by something bad or scary, but if you buy our product—also known as the “candidate”—life will be delightful.
To see precisely who is influenced by these emotional appeals, Brader devised an ingenious series of experiments. Simply having people sit down and look at pretend ads for fake candidates wouldn’t do, he reasoned. People come to ads with prior beliefs and feelings. And they don’t see ads in isolation—they pop up amid news stories and McDonald’s commercials. Sometimes people notice them and pay attention, sometimes they don’t. To simulate this and produce reliable data is a challenge, but Brader found a way: In 1998, he recruited 286 volunteers through community-service announcements and flyers in eleven communities in Massachusetts. At the time, a primary election race was underway, with two leading candidates vying to be the Democratic nominee to be governor of Massachusetts. It was an “impressively lackluster” contest, Brader writes, with no major controversies or hot issues in a time of peace and prosperity. That was perfect for Brader’s purposes, because it made it a “fairly demanding test case for the ability of campaign ads to elicit enthusiasm, fear or any other emotion.”
As they arrived at the libraries, meeting halls, and churches where the experiments were held, people were asked to have a seat and watch a videotape of a newscast. The purpose of the experiment, they were told, was to figure out what people take away from the news. They then watched half an hour of a news show, including the commercials. One of the commercials was, of course, an ad for one of the two leading candidates for the Democratic nomination. These ads weren’t real, however. They were created by Brader using scripts he wrote, along with video clips and music taken from past political ads. There were four in all. The first ad featured a voice-over that was “enthusiastic” and positive, but the images and music were bland; the second ad used the same voice-over, but the words were matched with soaring music and images of sunny skies and grinning children. A third ad featured a fearful script about crime and drugs, but again it used bland pictures and music, unlike the fourth ad, which had the same script but paired it with ominous music and harsh images of guns, criminals, and drugs. The idea was to separate the effects of negative and positive information from those of negative and positive emotion—since both versions of the ads had the information but only the second versions were “juiced” with emotion. When the screening ended, participants answered a series of written questions about the newscast, the commercials, and the upcoming elections. The results were startling: People who saw the “juiced” version of the enthusiastic ad were more likely to say they would volunteer for a campaign, vote in the primary election, and vote in the general election than were those who saw the bland version of the same ad. Note that this was the result of a single, casual viewing of one short ad.
Fear seemed to be much less influential, however, as there was little difference between the answers of those who saw the fear-drenched ad and those who saw the neutral version. But Brader had also asked people to answer factual questions about the election, and he used that information to divide them into those who knew more about politics and those who knew less. That changed everything. It turned out that the effect of the emotional “enthusiasm” ad was universal—it influenced everybody, whether they knew anything about politics or not. But the effect of the fear-based ad was divided. It did not boost the rate at which those who knew less about politics said they would get involved. But it did significantly influence those who knew more—making them much more likely to say they would volunteer and vote.