29 November 2004
Dying To Be A Nicer Person
by Kate Melville

The phrase "don't speak ill of the dead" appears to be more than just good manners according to an upcoming article in Human Nature. Professor Jesse Bering of the University of Arkansas has been working on a series of experiments that show people tend to upgrade their valuations of another person when they think that person has died.

For one experiment, Bering presented a group of subjects with facial photos of several different people. The subjects made assumptions about the people's personality traits - were they friendly, kind, hardworking, outgoing, - based only on the photos. At a later date, the subjects were again presented with the photos, but this time they were told that some of the people had died since the first experiment. Bering found the subjects tended to upgrade their guesses about the people who they thought were dead, providing the first scientific evidence that people really don't like to speak ill of the dead.

"It's like we want to remind these people who've died how good and nice they are, so they won't hurt us," Bering said. "None of this is conscious, of course, but it doesn't have to be, so long as it works in nature."

As part of his research, Bering examined how subjects were affected by the perceived presence of a dead agent. He had three separate groups of subjects take a computerized test where the highest score earned a $50 prize. The subjects were told that the computer program had a glitch in it - sometimes the correct answer to a question would pop up on the screen with the question. This presented them with the opportunity to "cheat" at the task. By pressing the space bar, the subjects were told, they could delete the correct answer before they read it so they could respond truthfully. One of the groups read a made-up memoriam before the test, stating that the fictional graduate student who developed the test had died. Another group also read the memoriam, and were casually told that the graduate student's ghost had been spotted in the very room where they were taking his test. Those assigned to the third group heard nothing about a dead graduate student. Bering found that the subjects who were told about the ghost were generally less likely to cheat, because the subjects hit the space bar more rapidly than those in the other two groups.

"Cross-culturally, we're seeing evidence of a belief in the absolute causal power of people who've died," Bering said. "If I perceive that someone is evaluating me, whether it be my dead grandmother or God, I'm less likely to commit a transgression, possibly because I'm afraid of the consequences." Bering's findings led him to one of the questions he'd like to be able to answer: Is that belief an odd byproduct of regular social cognition or an adaptive trait that serves a moral policing function?