Born to Be Good: The Science of a Meaningful Life

As George planned the next stages of this longitudinal study (he has assessed the well-being of these individuals for several years), he sent me the videotapes of these conversations. For an entire summer, locked up in my laboratory video coding room in the basement of my department, I coded these six-minute conversations with Ekman and Friesen’s Facial Action Coding System. Each conversation took about six hours to code. Spending eight hours each day listening to stories of dying and coding such deep emotion left me exhausted and humbled. Almost all of our participants showed numerous displays of negative emotion, such as anger, sadness, fear, and, less frequently, disgust.

Our question was a simple one that had never been addressed before: What emotions predict healthy adjustment to the death of a spouse, as assessed with clinically sound measures of anxiety and depression, as well as measures of prolonged bereavement, which captures the individual’s continuing longing for the deceased and inability to reenter into daily living? And which emotions predict poor adjustment during bereavement?

Traditional bereavement theories offer two clear predictions. This thinking is based on Freudian notions of “working through” the emotional pain of loss and the cathartic release of anger. It predicts that recovery from bereavement depends on the increased expression of negative emotions, such as anger and sadness. A second prediction is that the expression of positive emotion is in actuality a pathological sign of denial, of an intentional turning away from the existential facts of trauma, and impedes grief resolution. Our thinking was just the opposite, that laughter would allow our bereaved participants to distance themselves momentarily from the pain of the loss, to gain perspective, to look upon their lives in a more detached way, to find a moment of peace, to take a deep breath, so to speak.

Our first finding lent support to this view of laughter. Measures of laughter (and smiling) predicted reduced grief as assessed at six, fourteen, and twenty-five months postloss. Those participants who showed pleasurable, Duchenne laughter while talking about their deceased spouses were less anxious and depressed, and more engaged in their daily living, for the next two years. Just as important, people who showed more anger were observed to be experiencing more anxiety, depression, and disengagement from daily living for the next two years.

A first objection one might raise with respect to these findings concerns the nature of the death. Perhaps those individuals who laughed had partners who experienced easier deaths and thereby felt less initial grief and, as a result, were better able to adjust to this difficult loss. We know from empirical studies of bereavement that the nature of the death matters—sudden deaths, and deaths that produce greater financial demands upon the spouse, lead to prolonged grief and difficulty readjusting. We also know that the severity of the individual’s initial grief powerfully predicts the degree of difficulty in adjusting that that person will experience later. These possibilities did not explain away the benefits of laughter: Those individuals who showed pleasurable laughter compared to those who did not did not differ in the nature of their spouse’s death (its unexpectedness or financial impact) nor in their initial levels of grief.

One might likewise argue that perhaps our individuals who laughed at death were just happier individuals to begin with. Perhaps our results linking laughter to adjustment were simply the products of the temperamental happiness of the individual and not the emotional dynamics and perspective shifts accompanying laughter. This alternative too proved untenable—our people who laughed did not differ on any conventional measure of dispositional happiness from our individuals who did not laugh.

Buoyed by these findings, George and I went on a search for further evidence in support of the benefits of laughter. Why did laughing while talking about the deceased partner relate to increased personal adjustment? What we observed were findings very much in keeping with the laughter as vacation hypothesis. Our first analysis looked at how bereaved individuals’ experience of distress tracked one physiological index of arousal—elevated heart rate. The bereaved individuals who laughed showed similar heart rate arousal as those who did not laugh. But whereas our nonlaughers’ feelings of distress closely tracked increases in their heart rate, our laughers’ feelings of distress were decoupled from this physiological index of stress. Metaphorically, laughers were taking a vacation from the stress of their partners’ deaths, freed from the tension of stress-related physiology.