Communal Bereavement

Anyone who has ever mourned the death of a loved one knows that grief can be a curiously somatic experience -- that the body can register sorrow as sensitively and as involuntarily as a seismograph, that sorrow can make you sick. Scientists have documented insomnia, depressed immune function, greater susceptibility to heart attacks and elevated levels of corticosteroids, the so-called stress hormones, in recently bereaved people. But a bereaved person is by definition someone who has lost an intimate. Is it even possible that vicarious grief -- grief for people you have never even known -- can afflict the body as well?

Two public-health researchers, Ralph Catalano and Terry Hartig, believe it can. They have introduced a term -- communal bereavement" -- for what they call "the widespread experience of distress among persons who never met the deceased." Catalano, a professor of public health at the University of California at Berkeley, and Hartig, a lecturer in the psychology department at Uppsala University in Sweden, were not conducting their empirical research in the aftermath of Sept. 11, but the applications to it are obvious. In the weeks after the terror attack, it was clear that sadness had become, among other things, a public-health problem: pharmacies reported filling more prescriptions for antidepressants; doctors observed that chronic-pain patients were suddenly complaining of flare-ups. In a poll conducted by the Pew Research Center for the People and the Press the week after Sept. 11, 71 percent of Americans said they felt depressed, and 23 percent said they had been suffering from insomnia. These were numbers that could move you with what they seemed to say about the scope of empathy and frighten you with what they seemed to say about the depth of vulnerability. The measure of communal bereavement that Catalano and Hartig use in their research is the prevalence of very low birth weight babies -- babies born small as a result of premature labor, which can be induced by psychological stress. They wanted to know, as Catalano puts it, "How far away from the individuals who died can you be and still experience the bereavement response at a biological level?" Studying the effects of two national traumas in Sweden, and controlling for other variables, they found a 21 percent increase in very low birth weight babies after the assassination of Prime Minister Olof Palme, and a 15 percent increase after the ferry Estonia sank, killing hundreds of passengers. Though Catalano and Hartig, who are publishing their results in The Journal of Health and Social Behavior this month, admit that other explanations might prevail, they say that their work strongly suggests that communal grief can exert an effect on women in their third trimester of pregnancy. They are planning to study its effects in the first trimester too, when they hypothesize that it might lead to more miscarriages.

But Catalano and Hartig say that not every public loss triggers communal bereavement. "All well-liked national leaders will die at some point," as Hartig says, "yet a massive outpouring of grief, fear and anxiety does not necessarily always attend their deaths. Did Harry Truman's death at 88 have the impact on Americans that the assassination of J.F.K. did?" Similarly, celebrity deaths produce their share of tears but seldom the feeling that the world has been fundamentally altered.

The number of people killed is surely relevant -- the fact that, according to one estimate, 20 percent of Americans either know someone who was killed or injured on Sept. 11 or know someone else who does makes for an extraordinary ripple effect. But numbers aren't the only salient factor. Natural disasters might kill as many people as terrorist attacks and yet not be as emotionally destabilizing. Some awful fates may be rare, but their peculiar horror makes them threatening in a way that a more quotidian death is not. Which is why, though the dogged rationalists among us (God bless them) are forever pointing out that we are more likely to be killed walking across the street than in a biowarfare attack, people keep blithely walking across the street, though they might stare at their unopened mail in terror.

Hartig and Catalano surmise that to trigger communal bereavement, an event has to shake people's confidence in safeguards and institutions "essential to the normal functioning of the community," so Sept. 11 would seem to be the textbook case. Hartig and Catalano plan to study it closely in the months to come.

False-Identification Prevention

Short of DNA evidence, an eyewitness who can view a police lineup and say, "Yep, that's the guy; I'd never forget that face" seems like the surest way to bring the right perp to justice. The problem is, eyewitnesses forget or misremember faces all the time. Gary Wells, a psychology professor at Iowa State who has studied why so many eyewitnesses point to innocent men and women, thinks that one big reason is the way standard police lineups are conducted, with six people displayed simultaneously. He advocates a radically different approach: showing eyewitnesses one person at a time and allowing them to say yes or no to each one.

The trouble with traditional lineups, Wells argues, is that witnesses who view all the possible offenders at the same time tend to make a relative judgment -- that is, they pick the person who looks most like the suspect they remember relative to the other people in the lineup, and then they confuse this decision with actual recognition. In one study of 63 wrongful convictions, for example, 53 were based on ID's made by eyewitnesses. In nearly all of these, the real criminal had not appeared in the lineups. On the other hand, if witnesses are shown one person at a time -- the so-called sequential method -- the rate of mistaken identification goes down, while the rate of accurate identification holds steady. In one of Wells's first studies, in which 243 college students witnessed staged thefts, 43 percent of those who saw a simultaneous lineup picked the wrong thief, compared with 17 percent of those who saw sequential lineups. "All the science since shows that you don't get fewer correct identifications, just fewer incorrect ones," says David Feige, a Bronx attorney who has championed the sequential method. "It's as if they invented the magic machine that screens out the innocent."

In recent years, DNA evidence has come along to confirm what lawyers have long known -- that eyewitness memory can be fragile and spotty. Of the first 40 cases in the United States in which DNA was used to exonerate people convicted of crimes, 90 percent involved mistaken identifications by one or more eyewitnesses. But DNA evidence isn't always available; indeed it's almost never available in certain crimes, like drive-by shootings or robberies. That's why accurate eyewitness testimony is still crucial. "My approach is not to give up on eyewitness evidence but to make it better," Wells says. "It should be treated like fragile physical evidence -- and collected just as carefully as blood or fiber or semen."

The Justice Department now recommends Wells's sequential method as significantly less error-prone than the traditional police lineup. And in October, New Jersey became the first state to adopt the sequential method as standard procedure. Other police departments may follow New Jersey's example. "The police there like it," Wells says. "They can trust their results better and be much more confident going forward with prosecution. They don't like getting the wrong guy, either."