News Release

Anger drives support for wartime presidents

Peer-Reviewed Publication

Washington University in St. Louis

Alan Lambert, Washington University in St. Louis

image: Alan Lambert is an associate professor of psychology at Washington University in St. Louis. view more 

Credit: WUSTL Photo

It's no secret that Americans tend to throw their support behind a sitting U.S. president when the nation is thrust into a war or other potentially violent conflict with a foreign foe – a phenomenon known as the "rally 'round the flag effect."

But new experimental psychology research from Washington University in St. Louis is the first to offer compelling evidence that these wartime surges in presidential support represent a collective reaction to a specific human emotion.

"It's about anger, not anxiety," says Alan Lambert, PhD, professor of psychology in Arts & Sciences and lead author of the study published this month in the Journal of Personality and Social Psychology. "Anger is the engine that drives these sudden spikes in presidential approval ratings."

His findings, to be presented July 7 at the 33rd annual meeting of the International Society of Political Psychology in San Francisco, Calif., show that anger – not anxiety – is the dominant emotion that both triggers and feeds the rally effect.

While there are many competing theories about why rally effects occur, their existence is well documented. Rally effects contributed to dramatic and sustained spikes in popularity for John F. Kennedy after the Cuban Mission Crisis, for George Herbert Walker Bush during Operation Desert Storm and for George W. Bush after the 9/11 terror attacks on the World Trade Center. After 9/11, George W. Bush's popularity surged by almost 40 points, reaching as high as 90 percent and remaining much higher than normal for nearly a year, according to the Gallup Poll.

Contrary to popular opinion and previous speculations among psychologists, Lambert's study shows that the impulse to support the president in times of war has little to do with feelings of anxiety or uncertainty or needing a president to somehow make us feel safe. Nor do pre-existing political ideologies and values prevent us from being pushed at least a bit further down the militaristic path.

"Whereever you start on the ideological spectrum, your support for the president is pushed in an upward direction by feelings of anger," he says. "It's not a rationale thought process. It's a very primitive, almost knee-jerk response."

Lambert's findings are based on a five-year study that began in 2003 with experiments measuring shifts in pro-military attitudes among college students who watched eight minutes of a CNN "America Remembers" documentary on the 9/11 terror attacks.

He and a team of Washington University psychology student co-investigators surveyed a group of 136 college students to establish a baseline score of the participants' attitudes toward George W. Bush, pro-war policies, patriotic symbols and conservative views.

In some of the earlier studies, half the students were randomly assigned to watch the CNN terror attack video, while the other half completed simple anagram word games. Participants' moods were then assessed to measure current levels of anger, unhappiness, anxiety and other emotions; each then rated their own favorableness toward a list of prominent political figures and controversial political issues, including iconic examples from both ends of the liberal and conservative spectrum.

As expected, those exposed to the CNN 9/11 video scored much higher on measures of anger than those completing the mundane word game, and regardless of their initial political leanings, most video viewers showed a marked increase in their support both for George W. Bush and his militaristic policies.

"Keep in mind," says Lambert, "that this was a group of typical college students, many of whom didn't like the war and for the most part, didn't like George Bush at the beginning of the study. But, if you make them angry and you remind them about the war, they still end up showing increased support for George Bush."

While existing theories tend to frame rally effects as part of a general shift to the right, Lambert's study found no evidence for an across-the-board increase in support for conservative social policies, such as restrictions on abortion and prayer in schools.

"Anger clearly increased support for the president, but that effect was very narrowly confined to the president's military policies," he says. "It didn't affect attitudes toward gays. It didn't affect attitudes toward abortion. It had absolutely no effect at all on any other political ideology apart from militaristic attitudes. It's absolutely a rifle shot, something that pushes just this one button."

As part of the study, Lambert and his team conducted three more experiments designed to show that findings from the first experiment were not confined to scenarios involving George W. Bush and the 9/11 attacks. Nor were the findings skewed by some underlying patriotic bias embedded within the CNN documentary.

In one follow-up experiment, the CNN video was replaced with a session in which participants were asked to write about their own personal recollections of the 9/11 attack. In another, participants were asked to recall some event in their past about which they still felt angry, an exercise that prompted recollections ranging from broken romances, lost jobs and fights with sports coaches.

"Our last experiments showed that you can cause increased support for military action even when you make people angry for reasons that are completely unrelated to politics," Lambert says. "In our last experiment, we came up with this hypothetical Bush-like hawkish politician and we got the same effects."

As Lambert explains it, most types of threats -- especially outside threats -- tend to trigger off two types of emotion: anger and anxiety. His study used various experimental and statistical methods to tease apart and isolate the consequences of anxiety from anger, showing that feelings of anxiety may actually diminish support for a president's war plans.

"When you isolated the anger from anxiety, the anger led to more support for the president and the anxiety led to somewhat more negative reactions," he says. "Our data suggests that when people are anxious, they don't really like risky political policies. Going to war or continuing a war is a risky sort of activity, and when people are anxious, they tend to be risk averse."

Making people angry, on the other hand, sets into motion a much different cluster of behavioral tendencies.

"When people are angry," Lambert says, "they start preferring and liking things that they would not normally prefer. They start liking aggressive politicians more than they otherwise would, and they start disliking passive or dovish politicians. So, in a sense, once the emotion is activated, it has a life of its own, and starts directing your thoughts and attitudes, independent of how you might have felt before you became angry.

"If I kick you, you're going to look for some way to kick me back; but when something happens to the entire country, you need to look for someone more powerful than you to carry out that role, and in least in our culture, it's the president that does that."

Lambert suspects that the psychology behind the rally effect may explain why leaders in countries such as North Korea and Iran are so quick to rattle sabers in their dealings with adversaries around the globe – the threat of conflict abroad can shore up support for politicians at home.

These findings also have clear implications for America's support for Barack Obama. Presidential approval ratings for Obama, while still relatively good, are not nearly as high as they were at the beginning of his presidency. Lambert's research suggests that Obama's popularity would be most likely increased in the context of a military, as opposed to a non-military conflict. For example, even if Obama had managed to quickly resolve the problem with the oil spill in the gulf, this would have been unlikely to result in a sustained boost in approval.

While findings from this study may be limited to an American context, Lambert recently returned from a year as a Fulbright Scholar in the former Soviet republic of Georgia, where he conducted similar research and found much the same results.

"Essentially, a rally effect is a function of collective remembering and collective emotion," Lambert says. "There's a difference between one isolated person feeling something, or even just a couple thousand. Here, we're talking about millions of people all remembering and all feeling the same thing at more or less the same time. That doesn't happen very often, but when it does, it's an incredibly powerful thing."

###

Lambert's co-authors include current graduate students Laura Scherer and John Paul Schott along with former undergraduates Kristina Olson, Rick Andrews, Thomas Obrien and Alison Zisser, all then in the Department of Psychology in Arts & Sciences at Washington University. (Since this research has been conducted, Kristina Olson has gone on Yale University, where she is a member of the psychology faculty; the other co-authors (Rick Andrews, Thomas O'Brien, and Alison Zisser) are now enrolled in graduate programs in the social sciences.)


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.