Lingua Franca, October 1997
Good Grief!
The case for repression. By Emily Nussbaum

A bereavement conference hardly sounds like a fun time, but the group of scholars and therapists who have gathered in the nation’s capital on a hazy July morning are cheerfully buzzing over coffee and bagels, readying themselves for panels on everything from “When Mother Kills Father” to “Demystifying the Role of Coroner.” Meanwhile, George Bonanno, a professor of psychology at Washington, D.C.’s Catholic University, stands among them with a sardonic smile on his face, looking a bit like the Grim Reaper himself, albeit in tanned, rested form.

Bonanno has come to the conference, the annual meeting of the Association of Death Education and Counseling (ADEC), not to praise the work of grief counselors but to analyze it. His goal: to place the so-called grief-work hypothesis—the widely held assumption that venting negative emotions and “telling the tale” are necessary for regaining mental health—in the pincers of the scientific method. So far, the forty-one-year-old psychologist’s experiments have yielded intriguingly counterintuitive results, at least to those who believe in the benefits of letting it all hang out. Instead of working through their pain, Bonanno’s studies suggest, grief-stricken individuals who express intense negative emotions when discussing their loss appear to do worse in the long term, while so-called repressors recover more successfully.

So much for the talking cure? Bonanno is, in fact, part of a larger effort to figure out whether therapy works, as dozens of skeptical psychologists and their colleagues in neuropsychology attempt to break down the therapeutic relationship into neatly objective, quantifiable terms. His brand of “grief studies” is a microcosm of this movement—and it could hardly be better timed, given the blossoming of what Bonanno dismissively calls “the bereavement industry,” a welter of professional bereavement therapists and support groups intended to help people through the grieving process. “We’ve got all these therapists treating people for grief,” says Bonanno. “And we haven’t even defined what it is! We don’t know what ‘normal’ grief is, what ‘pathological’ grief is. It’s all very undefined, very unrigorous.”

Many in Bonanno’s camp blame Freud, who, in his theory of repression, claimed that painful memories get shoved down into the unconscious, where they are the source of neurotic symptoms and physical expressions like hysteria. Psychoanalytic practice is based on the idea that spilling the beans is the key to renewed health; as the patient releases his denied feelings—especially negative ones, such as hatred for a parent—the neurotic symptoms dissipate and the patient is able to move on. Grief, Freud theorized in “Mourning and Melancholia,” is normal, but when the mourner refuses to acknowledge ambivalent or hostile feelings toward the lost person, the result is a narcissistic self-loathing pathology derived from the mourner’s continued ego-identification with the deceased.

Like many of Freud’s ideas, these concepts have seeped into American culture in ways that would have boggled the mind of the staid Viennese theorist. A Girl Scout troop in New York, for instance, instituted a “grief patch” in 1993; troop members could earn this medal by sharing painful feelings with one another, writing stories and poems about death and loss, and meeting with bereavement counselors. Last year’s Miss New Jersey chose bereavement counseling as her special issue, in response to the loss of her own father. And at the Tiny Hearts Bereavement Group in Ronkonkoma, Long Island, grieving children participate in an “If Onlys and Farewell Day,” during which they write down guilty or regretful feelings about deceased friends and relatives, tie them to balloons, and release them into the sky. As for the professional field of bereavement counselors, it has swelled—ADEC now boasts more than 2,000 members.

But even as bereavement support has become both increasingly professionalized and increasingly democratized (via the self-help movement and the growth of the Internet, where discussion groups abound), an academic countermovement has slowly crystallized to pick apart its assumptions. In 1989, SUNY Stony Brook’s Camille Wortman and UC-Irvine’s Roxane Cohen Silver published “The Myths of Coping With Loss” in the Journal of Consulting and Clinical Psychology. The paper concluded that there was little evidence that “those who initially show minimal distress following loss are likely to become significantly depressed at a later point.” And in 1993, Margaret and Wolfgang Stroebe, researchers at the University of Utrecht in the Netherlands, edited an anthology of the latest research on grief, the Handbook of Bereavement: Theory, Research, and Intervention (Cambridge), establishing a hard-line statistical stance toward the subject: “In our view, bereavement is an issue that needs to be understood from a sound base of theoretically oriented and empirically derived knowledge and not purely on subjective, descriptive accounts.”

The Stroebe contributors (a multidisciplinary squad of empiricists from various fields, including anthropology, epidemiology, and sociology) depart radically from the influential “five stage” process originally described by Swiss psychologist Elisabeth Kübler-Ross: denial, anger, bargaining, depression, and acceptance. In one chapter, Wortman and Silver follow up on their earlier essay with findings suggesting that individuals confronted with drastic life events (such as paralysis or the loss of a loved one in a car accident) experience positive emotions far more often than expected. They add that victims of a trauma generally “shape meaning” about their experience shortly after the event. Their conclusion: Depending on how well bereaved individuals can reconcile the traumatic event with their philosophical belief systems, a complete “working through” of painful feelings may or may not be necessary. After all, a fundamentalist Christian and an atheist aren’t likely to follow the same emotional path when a loved one dies.

Such arguments seem not to have influenced the presentations by clinicians at the bereavement conference, where “subjective” and “descriptive” accounts abound. It’s all a bit touchy-feely for Bonanno’s taste: Tense attendees can enjoy “complimentary massages” in one of the conference rooms or visit an odd mourning mini-mall offering books like Why Are the Casseroles Always Tuna? A Loving Look at the Lighter Side of Grief. Of course, many of the sessions are undeniably affecting. During one, an earnest support-group leader shows a videotape of one of her sessions—jittery documentary crosscuts of parents of homicide victims sharing their most painful feelings. It is wrenching to watch. During the question-and-answer period, audience members offer up their own painful stories. Then Bonanno stands and attempts to change the terms of the discussion. “That tape was undeniably moving,” he says cautiously, cocking his head. “But I want to ask you: Does that kind of support group work? And if it works, why do you think it works?” The counselor—sweet faced, makeup free, with a shiny blond bowl cut—smiles at him across the semantic divide: “A couple of words,” she says. “One of them is caring. And love. That becomes the focus.”

There’s nothing new about the mutual miscomprehension between researchers and clinicians, of course: It’s the Spock-Bones archetype writ large, number-crunching statistical Vulcans versus gut-instinct Empaths. And psychology itself has always existed in what British psychologist Colin Murray Parkes, one of the most venerable of the grief researchers and author of Death and Bereavement Across Cultures (Routledge), calls a “no-man’s-land between art and science.” Still, research on such a sensitive subject is bound to raise hackles. “People who are grieving think that researchers are full of crap—and part of me says, I’m with you,” remarks Tom Golden, a prominent bereavement therapist who specializes in male grief. “We don’t have the tools to measure it yet, there’s no grieve-o-meter. We need to develop a sense of not knowing.”

Bonanno understandably disagrees. “I think that’s a ridiculous statement!” he says heatedly. “You can measure grief. People want to take a magical, mystical perspective, but it’s very dangerous to assume that they have access to a sacred realm that research can’t touch, relying only on their own observations, feelings, and thoughts—things that are very unreliable.” This sort of clinical criticism, he argues, is the result of a simplistic attachment to individual patients’ life stories. “The criticism I most often hear is, ‘Your research is very bad, because I have a patient who feels such and such.’ Well, I’ve studied hundreds of people.” A former clinician himself, Bonanno says he eventually grew “very jaded about psychotherapy,” disillusioned with the careerist aspects of the field and determined to hold it to a higher statistical standard with studies that were double-blind, longitudinal, and controlled.

One of Bonanno’s bereavement studies, published two years ago in the Journal of Personality and Social Psychology, has attracted much attention. Using newspaper ads, Bonanno and his colleagues recruited middle-aged men and women who had lost their spouses within the last six months. After filling out numerous psychological-testing questionnaires, the subjects told the story of their loss to an interviewer while being filmed and having their vital signs taken. The goal: to get a sense of how the bereaved individual discussed his loss in a social context, thus approximating the “laboratory equivalent of the ‘working through’ process.”

Using this data, Bonanno identified a group of “repressors”: those who smiled and expressed positive emotions despite showing physical indications of stress and arousal. In other words, these subjects seemed most alienated from their “true” feelings—and weren’t even aware of the disassociation. According to the traditional assumptions of grief counselors, repressors should have had all of their submerged anguish bubble up later on, along with a whole set of physiological complaints dogging them in the long term.

Instead, while those demonstrating “verbal-autonomic response dissociation” had elevated physical complaints at six months, they were back to low levels at fourteen months. And at twenty-five months—when they should have been experiencing “delayed grief”—they had significantly lower levels of suffering than expressive subjects according to several measures, including the independent assessment of outside therapists who didn’t know what the study was about. Bonanno concluded that the inhibition of negative emotion did not have a cumulative cost; rather, it might be an undervalued coping skill. For that reason, laughter, optimism, and inappropriate-seeming positive emotions should not be simply stigmatized as “denial.”

Bonanno’s research, although tentative, has been bolstered by a few other studies. One study found that exaggerated optimism among HIV-infected patients was correlated with better psychological health. And in “Dream Repression in Adjusted Holocaust Survivors,” published in the Journal of Nervous and Mental Disease, Hanna Kaminer and Peretz Lavie of Haifa’s Israel Institute of Technology found that the psychologically healthier survivors blocked out painful memories, restrained troubling thoughts about their experience, and were unable to remember their dreams. “Repression has been understood as a pathological phenomenon,” write Kaminer and Lavie. “Our findings contradict this assumption…. The survivors’ ability to close off, repress, and prevent memories of past atrocity from reentering their consciousness has adaptive value and suggests that the treatment approach should be different.” Their suggestion for an improved therapeutic method? One that “helps survivors to seal off the atrocities that they experienced.”

This suggestion that therapists should encourage their patients to block out painful memories is certainly provocative: Are venting sessions in self-help groups dangerous, discouraging “healthy repression”? At the same time, there’s a highly disturbing undercurrent to Kaminer and Lavie’s arguments. What are the social and political ramifications of a therapeutic culture that pushes genocide victims to forget what happened to them? Should counselors really teach skills of disassociation rather than, say, encouraging patients to know the truth about their own lives? Is this the Sybil cure?

Kenneth Doka, professor of gerontology at the College of New Rochelle and former president of ADEC, supports the trend toward increasingly rigorous research design but is cautious about how well it can be applied to clinical practice. “What I don’t want to see happen is what’s happened in sociology—a lost notion of meaningful causality,” he explains. “We do stuff with elaborate path analyses, and at the end it’s a very elegant example of statistical process, but it doesn’t necessarily explain much about human behavior.” Doka has a point: Quantitative research may never yield the kind of hands-on knowledge that therapists acquire in their work. Furthermore, even if repressive coping mechanisms do help suffering individuals to get on with their lives, must we assume this has to be the case?

In fact, further research is likely to explore the way particular social systems heal or aggravate the experience of bereavement. Why do the expressive, anguished mourners do so poorly in the long run, anyway? Perhaps the reason has less to do with the individual psyche than with society: People who truly vent their grief, suggests Bonanno, may drive away support systems around them. “We showed the films of the research subjects to a panel, with the sound turned off,” he explains. “The people who laughed evoked positive feelings, while the people who talked about a lot of painful stuff made viewers want to flee.” If that’s so, one might conclude, the psychological problem lies less with the angry, expressive mourner than with the inability of listeners to bear the intensity of such emotion.

Bonanno responds by pointing out that quantitative psychological research is concerned with practical matters—“how people can live in the culture we have,” as he puts it—rather than with such philosophical best-of-all-possible-worlds musings. In the here and now, he believes, there’s plenty of work to do. “I’m not even that invested in whether the grief-work hypothesis holds up,” he muses. “But it’s as though I’m hiking through the mountains and stumble into a valley ripe for exploration. There’s research for years—and a chance to ask questions no one has asked before!”

Bonanno has come to the conference expecting boos and hisses—at previous conferences, clinicians have heckled him during lectures. He says that when he contacted the St. Francis Center in Washington, D.C., to help with his study, they responded: “I’m sure your research is just fine, but you’re wrong.” (Robert Washington, the head of St. Francis, was unable to confirm this, but he says with bemusement: “I’m sure most people here would say repressing emotions is not a good thing. I don’t argue with his findings, I just think that human beings are more complex than what you can capture in a research design.”)

In fact, Bonanno’s lively lecture is greeted respectfully, except by two women who roll their eyes at his conclusions. Ironically, his highly expressive personal style—gesticulating, cracking jokes, nearly jumping off the ground in excitement at times—may make his research more palatable to the emotion-friendly, numbers-doubting therapists in the audience. The only truly peculiar moment comes when he makes an offhand remark about the ethical concerns of such research: “We can’t very well take one hundred people, kill half their spouses, then track how they deal with it for the next five years.” There is a nervous laugh. This may be the wrong crowd for that joke.

At least on the surface, this conference may represent a turn toward the quantitative researchers. One of the scholarly speakers is Margaret Stroebe, and her speech is a surprise. Tiny and soft-spoken, she adopts an unexpectedly conciliatory approach, presenting a new “dual process” model—one in which expression and repression, coping skills and breakdown, are regarded as necessary for full recovery from mourning. Is this a genuine theoretical turn away from the quantitative researchers’ “anti-grief-work” vendetta or simply a spoonful of sugar to help the research go down? Bonanno is skeptical of Stroebe’s Big Tent approach: “In my view, there just isn’t very much research that backs up the grief-work assumption.”

What is the future of grief studies? Some of the most intriguing challenges to Western assumptions may come from looking at other cultures. Balinese mourners, for example, smile and laugh during funerals. In the rituals of the Saramaka of Suriname, weeping and wailing are monitored and strongly discouraged—even during the funeral; the mourners are told to “get on with your life, forget your sorrow.” The Hopi Indians, too, discourage overt expressions of grief: If the bereaved feel compelled to cry, they are encouraged to do so alone. (Of course, some cultures go to the opposite extreme: Hindi families may have an “experienced mourner,” who encourages the bereaved family to weep, shriek, wail, and talk openly of the deceased.)

Bonanno himself is preparing the first, tentative studies of grieving in China. Designing an experiment will be difficult. To begin with, Chinese mourners rarely speak about their memories of the deceased. And how does one define “emotional well-being” in a society that tends to value group loyalty over personal happiness? “It’s going to be a real challenge,” acknowledges Bonanno. “I’m dubious about questionnaires, but we’ve got to come up with something. We’re thinking of using beepers—you beep the person, and they write down what they were thinking about.” Despite these difficulties, Bonanno remains undaunted: “This is what a scientist lives for.”