November of the Soul

by George Howe Colt

  The debate reached its climax at the turn of the century. After the New York World published the article “Is Suicide a Sin?” in 1894, the New York Times accused the World of provoking an unprecedented number of suicides. A rash of Cleveland, Ohio, suicides in 1910 was attributed to press coverage, and in 1911 the National Association of Retail Druggists protested newspaper reports publishing the dosages of poisons used in suicide attempts as “inducing morbid people and criminals to use these poisons.” That same year, at the annual meeting of the American Academy of Medicine, statistician Edward Bunnell Phelps denounced “the pernicious influence of neurotic books and newspapers” as “an accomplice in crimes against the person.” He singled out morbid literature and plays, newspaper accounts of suicides, and lurid tales in the Sunday supplements—a “literary chamber of horrors” from which he culled a few examples: “The City of the Suicide Germ,” “Chain of Suicides Strangely Arise from Love Match,” and “The Pathetic Mystery of Suicide on the Eve of Marriage—What Secret Hides Behind the Recurring Tragedies of Self-Destruction at the Brink of Nuptial Union, Even Where Every Known Promise Is for a Happy Future.”

  Did morbid literature and sensational newspaper accounts of suicide really increase the number of suicides? Or were the suicides that were blamed on the printed word the deaths of troubled people who would have killed themselves anyway? The sociologist Émile Durkheim believed the latter. In his landmark 1897 study, Le Suicide, he reviewed the research linking suicide and suggestion and concluded that the effect of imitation on the national level of suicides was minimal. Those few suicides that might be triggered by suggestion, he said, would eventually have occurred in any case; a book like Werther or a sensational newspaper report merely hastened the timing.

  Seventy-seven years later, David Phillips, a thirty-year-old Princeton-trained sociologist, disputed this conclusion. Checking the vital statistics of the United States against the New York Times index for front-page stories on suicide since World War II, he found that suicides increase significantly in the month after a highly publicized suicide story. The greater the publicity, the greater the increase. For instance, the suicide of Marilyn Monroe in 1962 spurred a 12 percent jump (197 more suicides than would have been expected in the month following her death). Phillips also found that the increase occurs primarily in the geographic area in which the story is published. Finding that there was no matching “dip” in the rate after the publicity had died down, Phillips concluded that these deaths were “extra” suicides, not inevitable suicides that would otherwise have taken place a little later. Phillips called this phenomenon “the Werther effect.”

  While his work offers compelling evidence linking suggestion and suicide, Phillips is careful to emphasize that the media story does not itself cause suicide. “The factors that drive a person to suicide may build up over many years,” he says. “I’ve been investigating only one aspect of it. I’m studying the trigger and not what loaded the gun.” Phillips goes on to speculate on how that trigger might be squeezed: “A suicide story in the newspaper may be a sort of natural advertisement. Just as, suppose, watching television, I have this strange vague feeling inside me, but I can’t put a label on it. Then I see an ad for McDonald’s, and I say to myself, ‘Come to think of it, I think that strange vague feeling I have is hunger. And there are various ways to assuage my hunger, but now I’ve seen this option suggested, I think I’ll go to McDonald’s.’

  “Now it’s possible that unhappy people out there, who may or may not realize they’re unhappy, read the story about Marilyn Monroe’s suicide and become aware that they are unhappy and maybe also become aware of an option to end their unhappiness. They may feel they have been given permission because they see another person has done it. Maybe they say, ‘Gee, if even Marilyn Monroe is feeling bad enough to do this, shouldn’t I do it, too?’”

  Over the years, numerous other studies have found evidence for the Werther effect. Researchers agree that the magnitude of the effect is related to the amount, duration, and prominence of the coverage; it is especially strong if the story is placed on the front page, with large headlines. Not surprisingly, it is more pronounced in those whose demographic characteristics—age, gender, nationality, and so on—resemble those of the victim. In nearly every study, however, teenagers have been shown to be especially susceptible.

  The likability of the victim also makes a difference. The 1978 mass suicide at Jonestown, for instance, did not trigger a spike in the suicide rate. Similarly, news accounts of murder followed by suicide have had no effect, perhaps, researchers suggest, because of the “nonattractiveness” of the victims. After the highly publicized 1999 tragedy at Columbine High School in Littleton, Colorado, in which two teenage boys shot and killed twelve students and one teacher before killing themselves, one might have expected an increase in adolescent suicidal ideation. But an ongoing, yearlong CDC survey of high school students from fifty states allowed researchers to compare those who filled out the survey before Columbine with those who filled it out afterward. The percentage of students considering suicide decreased significantly following the tragedy, a result attributed by the survey’s authors to the nature of the media coverage. Although stories about the homicide victims focused on the outpouring of grief among their families and friends, stories about the suicide victims focused on the social problems they’d experienced at school.

  While Phillips himself has provided evidence that the contagion effect extends to television news coverage of suicide, studies analyzing the effect of fictional television suicides have been less conclusive. But there is evidence to suspect that TV portrayals of self-destruction may lead to increased rates of suicide and suicide attempts, often using the same methods depicted in the shows. A 1999 English survey of patients presenting in forty-nine emergency rooms during the week following an episode of a popular TV series in which a teenager had overdosed on paracetamol found cases of self-poisoning up 17 percent. Twenty percent of the attempters said the broadcast had influenced their decision to take an overdose, while 17 percent said that the show had influenced their choice of drug. While the impact of film portrayals of suicide on the rate remains unproven, it is clear that the rate of cinematic suicide is rising. An examination of American films from 1917 to 1997 found that the portrayal of suicide in films has increased dramatically; nearly one in ten films now depicts a suicide or a suicide attempt. Whether this is a case of art imitating life or life imitating art is not yet apparent. (At least one study asks whether cyberspace—in the form of suicide-themed Web sites, where young people gather to discuss the subject and argue the merits of potential methods—may be fertile ground for the contagion effect.)

  “If the mass media were to reduce the publicity of suicide stories, it’s pretty clear that the number of suicides would go down,” says Phillips, pointing out that several studies have documented a decrease in suicide deaths during newspaper strikes. Nevertheless, he opposes censorship. “I grew up in South Africa, where the press was controlled and individual freedom substantially limited. I think it would be extremely unfortunate if my studies were used as ammunition to pressure the media to change their coverage. If the media want to do this voluntarily, it’s up to them.” In 1987, following a highly publicized three-year rash of suicides by jumping in front of subway trains in Vienna, the Austrian Association for Suicide Prevention alerted journalists to the possible negative effects of excessive coverage and suggested alternative, less sensational strategies. When the amount of reporting on the deaths dropped, subway suicides and nonfatal attempts by that method fell more than 80 percent in the next six months, while the total number of Viennese suicides declined as well. In 1994, the CDC, acknowledging that “it is not news coverage of suicide per se, but certain types of news coverage, that promote contagion,” issued recommendations for the media. They urged journalists to avoid referring to suicide in the headline, to minimalize morbid details, to avoid publishing photographs of the death site or the funeral, to om
it technical information about the suicide method, to avoid glorifying the victim with descriptions of grieving relatives and classmates, to avoid simplistic explanations (“Boy, 10, Kills Himself Over Bad Grades”), and to acknowledge that suicide is never caused by a single factor but by a complex interaction of many factors, usually involving a history of psychosocial problems.

  Was anybody listening? Four years later, the Annenberg Public Policy Center reviewed a year’s worth of suicide coverage in nine of this country’s highest-circulation newspapers, as well as three years’ worth in the New York Times. Seven of the nine papers featured the word suicide in at least half their stories’ headlines, often in sensationalized fashion (“Eighth-Grade Sweethearts in a Love Suicide” was one 1995 New York Times headline). In 60 percent of the New York Times articles, either the fact of suicide or the method was mentioned in the headline, yet only 8 percent cited depression as a possible factor. Many of the stories included explicit details about the method used. It was hardly surprising that the newspapers had ignored the CDC guidelines; interviewing sixty-one reporters and fifteen editors who had reported on acts of suicide, the Annenberg Center found that the journalists had never heard of the guidelines. Some were unaware that certain types of coverage could increase copycat suicide, and even those who were familiar with the phenomenon expressed doubts about its validity.

  On at least one occasion in this country, voluntary media restraint has had encouraging results. In 1994, when charismatic grunge rock star Kurt Cobain shot himself, experts expected a wave of copycat suicides. Rather than portraying Cobain as a misunderstood latter-day Werther, however, most news stories highlighted Cobain’s depression and substance abuse problems, while providing names and places to turn to if readers needed counseling. Cobain’s widow publicly and rather vituperatively expressed not only her grief but her anger at the senselessness of his death and at his abandonment of their nineteen-month-old daughter. A study published two years later found Cobain’s death had no effect on the adolescent suicide rate.

  In 1984, however, the media could not restrain itself. Suicide and suicide clusters were a hot story. Ironically, the focus on clusters may allow us to overlook the extent of youth suicide generally. While spotlighting “affluent suburbs” such as Plano, for instance, the Dallas papers virtually ignored the fact that the teen suicide rate was far higher in their own city. And Westchester County alone would have a total of six teenage suicides in 1984, compared with an average of five over the previous eight years and fewer than the high of seven in 1979. In that case, the 1984 suicides barely qualified as an “epidemic,” which is defined as “more than would normally be expected,” leading some to suggest that the “Westchester cluster” was a statistical mirage.

  On February 16, 1985, seventeen-year-old David Balogh of Tarrytown was found in a car parked at a landfill, dead from carbon monoxide poisoning. In the following three weeks there were three more adolescent suicides in Westchester County, the same number as in 1984 when the suicides had ignited media and public hysteria. “Last year at this time it was chaos,” a high school counselor told me. “This year there hasn’t been a peep. No calls from the media. Nobody mentions the word cluster.” Said a local psychologist, “I’d like to think it’s because the media is more sensitive to the issue, but the more cynical side of me wonders if it isn’t just old news.”

  With or without media attention, adolescent suicides continued in the tricounty area. But gradually the suicides became just a part of the overall rate of adolescent suicide in the United States. Some people say, in fact, that the attention showered on the Westchester suicides of 1984 was, in fact, ultimately beneficial because it focused attention on the problem of adolescent suicide after many years of silence. “I think the way the media dealt with that particular episode had some very positive results,” Westchester pediatrician Kenneth Schonberg told me a few years later. “Despite the fact that, in all honesty, that was not a very exceptional year. Unfortunately, adolescents are dying at a rather constant rate from suicide, and it wasn’t restricted to the late winter and early spring of 1984. Eighty-four or eighty-three . . . which was it? I forget already.”

  V

  DANA

  “I USED TO LIE in bed and imagine what my life could be like. I’d set up these great scenes in my mind. In my favorite I’d be driving down the road and there’d be a car accident, and I’d jump out and save the guy’s life, and it would be in all the newspapers—‘Hero Saves the Day.’ Or I’d go to college, become a doctor, and every hospital in the country would want me because I’d pioneered some new operation. Or I’d win the Olympics in swimming—every distance, every stroke imaginable. Or I’d be out at a romantic dinner, the kind you see on TV, where the camera focuses on the couple staring into each other’s eyes. The man would be tall, with a great body. Rugged good looks. Like Tom Selleck. And he couldn’t live without me.”

  At seven on a Tuesday morning in March, Dana Evans heard her mother leave for work. She rolled over and went back to sleep. It was a school day, but she didn’t feel like going. When she woke again, it was nine-thirty. She turned on the television set at the foot of her bed. Then she got up, shuffled downstairs to the living room, and poured herself a mug of vodka—straight, as always. She knew that if she added tonic, she wouldn’t get drunk as fast. Then she climbed back into bed and stared at the TV. Dana paid little attention to the game show. She sipped her vodka and thought about how miserable her life was.

  Although Dana, a high school senior, had often been told she was smart, she rarely did her homework, skipped school at least once a week, and was failing several classes. Although she was a promising swimmer, she had quit the team at the end of her junior year. Although she had a cute face and brown, curly hair, she was convinced she was ugly. She had never had a boyfriend and was certain she never would. She had no close friends. Her parents were divorced. Her father lived in Texas, and her mother, who worked two jobs, was rarely home. Her older sister was away at college. While her classmates were obsessed with boyfriends and college plans, Dana spent most of her time sleeping, watching television, drinking, or getting high.

  Suddenly Dana slammed the mug of vodka down on her bedside table, pulled on some jeans and a T-shirt, and stomped angrily through the apartment, pausing occasionally to punch the wall. Her head rang with imagined insults:

  Here you are again, hanging out, doing nothing.

  Well, that’s cuz no one wants to be with you.

  That’s cuz you’re an asshole and you’re ugly.

  Well, so, what do you expect?

  Well, you’re stupid.

  Well, you’re ugly.

  Well, you’re a fucking asshole.

  Well, then fuck it. Today’s the day.

  It was eleven-thirty. Dana realized that she would have to hurry. Tammy, a classmate who was staying with Dana and her mother, was due home at five-fifteen. Dana had always imagined that she’d leave a note telling her mother, her sister, and Tammy that she loved them. Now she decided against it—let them wonder, she told herself. She thought about where to do it. “I decided I’d better use the bathroom because from what I’d read, that’s where everyone kills themselves,” she says. “Plus if I got the bedroom rug dirty”—there is no trace of irony in her voice—“my mother would kill me.”

  At noon Dana went to her room and fished a razor blade from her purse. She walked into the bathroom and locked the door. “And then I just started,” she says. “I cut my wrists. I watched the blood go into the sink, and because the sink was already wet, the blood spread out and I realized how neat it looked.” She made vertical and horizontal slices on both wrists, lightly at first, and then, after taking a breath, deeply. “It hurt, but I didn’t mind,” she says. “It was almost as if I wanted it to hurt because I wanted to be tough.” Every so often she paused. “Everything in the bathroom is white,” she says. “White tile walls and white tile floors. I had to keep stopping and cleaning up because I didn’t want to m
ake a mess. I poured water in the sink so the blood would go down the drain. I’d rest my arms on my pants to soak up the blood so it wouldn’t get on the floor.” Dana felt herself grow weak, but she kept cutting. Then she stopped for a moment, looked at herself in the mirror, and said aloud, “Good-bye.”

  While attention to adolescent suicide has focused primarily on completed suicide, for every adolescent who dies, there are at least twenty—some say as many as one hundred—who make an attempt. Although no official statistics are compiled for suicide attempts, in 2002 an estimated 125,000 visits to emergency rooms in this country were made by adolescents who had attempted suicide. Many more made attempts that didn’t require medical attention.

  For years attempted and completed suicide were regarded as psychologically similar acts, differing only in their outcome. It is for this reason, in part, that until a few decades ago, people who attempted suicide and survived were called “unsuccessful” or “failed” suicides; those who died were called “successful.” The suicide attempter was regarded as a double failure—not only at life but at death. Most suicide research, in fact, was based on attempters—who were available to be interviewed—and then generalized to include completers. Clinicians now maintain, however, that attempters and completers form two different though overlapping groups, each with its own goals and motivations.