THE STUDY OF MARRIAGE THEN AND NOW
Over the past three decades, rising divorce rates and other dramatic changes in sexual norms, gender roles, and family life awakened a new
Self-reported ratings of happiness or satisfaction to study marriage were widely criticized. Some researchers argued that the concept “marital happiness” was hopelessly vague. Others questioned the validity of asking people to rate their own marriages, particularly since the high ratings most people gave to their own marriages seemed at odds with high divorce rates.
There were also deeper problems with these earlier studies. Even the best self-report measure can hardly capture what goes on in the private, psychosocial theater of married life. In the 1970s, a new wave of marital research began to breach the wall of marital privacy. Psychologists, clinicians, and social scientists began to observe families interacting with one another in laboratories and clinics, usually through one-way mirrors. The new technology of videotaping made it possible to preserve these interactions for later analysis. Behavioral therapists and researchers began to produce a literature on the kinds of statements and actions that differentiated between happy and unhappy couples. At the same time, social psychologists began to study close relationships of various kinds.
During this period I began my own research on marriage, using couples who had taken part in the longitudinal studies carried out at the Institute of Human Development (IHD) at the University of California at Berkeley. One member of the couple had been part of the study since childhood and had been born in either 1921 or 1928. Each spouse had been interviewed in depth in 1958, when the study members were thirty or thirty-seven years old. They were interviewed again in 1970 and 1982. The interviews were carried out by experienced clinical psychologists, who later rated each spouse on a large number of personality items and five aspects of marital adjustment. The participants also supplied a great deal of other information about themselves, including selfreports of marital satisfaction.
Despite the richness of the longitudinal data, it did not include observations
The newwave of research has revealed a great deal about the complex emotional dynamics of marriage and, perhaps most usefully, showed that some widespread assumptions about couple relations are incorrect. But there is still a great deal more to learn. There is as yet no grand theory of marriage, no singular road to understanding marriage, no “one size fits all” prescription for marital success. But we have gained some important insights into marital (and marriage-like) relationships. And there seems to be a striking convergence of findings emerging from different approaches to studying couples. Here are some of these insights.
For Better And For Worse
The sociologist Jesse Bernard argued that every marriage contains two marriages—the husband's and the wife's (1972), and that his is better than hers. Bernard's claims have been controversial, but in general, her idea that husbands and wives have different perspectives on their marriage has held up over time. And on a variety of subtle measures of marital satisfaction—for example, would you like your children to have a marriage like yours?—men do seem more content with their relationships than their wives (Skolnick 1993).
But apart from gender differences, marital relationships also seem to divide in another way: Every marriage contains within it both a good marriage and a bad marriage. Early studies of marital quality assumed that all marriages could be lined up along a single dimension of satisfaction, adjustment, or happiness—happy couples would be at one end of the scale, unhappy ones at the other, and most couples would fall somewhere in between. More recently, marriage researchers have found that two separate dimensions are needed to capture the quality of a marriage—a positive dimension and a negative one. In effect, the ultimate
In my own research, I came across this same “good marriage/bad marriage” phenomenon among the Berkeley longitudinal couples (Skolnick 1981). First, we identified couples ranging from high to low in marital satisfaction based on ratings of the marriage each spouse had made, combined with ratings made by clinical interviewers who had seen each separately. Later we examined transcripts of the clinical interviews to see how people who had scored high or low on measures of marital quality described their marriages.
In the course of the interview, each person was asked about his or her satisfactions and dissatisfactions in the relationship. Surprisingly, it was hard to tell the happily married from their unhappy counterparts by looking only at statements about dissatisfaction. None of the happy spouses were without some complaints or irritations. One husband went on at length at what a terrible homemaker his wife was. The wife in one of the most highly rated marriages reported having “silent arguments” —periods of not speaking to one another—which lasted about a week. “People always say you should talk over your differences,” the wife said, “but it doesn't work in our family.”
What did differentiate happier marriages from the unhappy ones could be found in the descriptions of satisfactions. The happy couples described close, affectionate, and often romantic relationships. One man remarked after almost thirty years of marriage, “I still have stars in my eyes.” A woman said, “I just can't wait for him to get home every night; just having him around is terrific.”
The most systematic evidence for this good marriage/bad marriage model emerges from the extensive program of studies of marital interaction carried out by Gottman and Levenson (1992) and Gottman et al. (1998). Their research is based on videotaped observations of couple discussions in a laboratory setting. These intensive studies not only record facial expressions, gestures, and tone of voice but also monitor heart rates and other physiological indicators of stress. The partners also watched a videotape of their conversation and were asked to use a rating dial to provide a continuous report of how they felt during their interaction.
Surprisingly, these studies do not confirm the widespread notion that anger is the great destroyer of marital relationships. Gottman and his colleagues found that anger in couple interaction does not predict
Ultimately, the key factor in the success of a marriage is not the amount of anger or other negative emotion in the relationship—no marriage always runs smoothly and cheerfully—but the balance between positive and negative feelings and actions. Indeed, Gottman gives a precise estimate of this ratio in successful marriages—five to one. In other words, the “good” marriage has to be five times better than the “bad” marriage is bad.
It seems as if the “good” marriage acts like a reservoir of positive feelings that can defuse tensions and keep arguments from escalating out of control. In virtually every marriage and family, “emotional brushfires” are constantly breaking out. Whether these flare-ups develop into major bonfires depends on the balance within the marriage that makes it a good one or bad one.
Gottman identifies a set of four behavioral patterns—he calls them “the four horsemen of the apocalypse” —that constitutes a series of escalating signs of marital breakdown. These include criticism (not just complaining about a specific act but denouncing the spouse's whole character); which may be followed by contempt (insults, name calling, mockery); then defensiveness (each spouse feeling hurt, mistreated, and misunderstood by the other); and finally, stonewalling (one or both partners withdraws into silence and avoidance). Eventually, the couple may end up leading separate lives under the same roof.
Tolstoy Was Wrong: Happy Marriages Are Not All Alike
The most common approach to understanding marriage, as I have shown, is to compare or correlate ratings of marital happiness with other variables. However, some studies over the years have looked at differences among marriages at a given level of satisfaction. Among the first was a widely cited study published in 1965. John Cuber and Peggy Harroff interviewed 437 successful upper-middle-class men and women about their lives and marriages. These people had been married for at least fifteen years to their original spouses and reported themselves as
Only one out of six marriages in the sample conformed to the image of what marriage is supposed to be—that is, a relationship based on strong emotional bonds of love and friendship. The majority of others were what Cuber and Harroff called “utilitarian” relationships. These fell into three types: first, “conflict habituated” couples resembled the bickering, battling spouses often portrayed in comedy and tragedy—the kind of relationship that would today be called “dysfunctional.” Yet these couples were content with their marriages and did not define their fighting as a problem.
A second group of couples were in “devitalized” marriages—they had started out in close, loving relationships but had drifted apart over the years. In the third “passive congenial” type of relationship, the partners were never romantically in love or emotionally close. They regarded marriage as a comfortable and convenient way to live while they devoted their energy to their careers or other social commitments.
The most recent studies of marital types come from the research of John Gottman and his colleagues, described earlier. Along with identifying early warning signs of later marital trouble and divorce, Gottman also observed that happy, successful marriages were not all alike. He also found that much of the conventional wisdom about marriage is misguided. Many marital therapists and much of the public believe that a couple who argue and fight a lot have an obviously “dysfunctional” relationship. On the other hand, those spouses who sweep their conflicts under the rug and avoid communicating with one another about troublesome issues are also in a dysfunctional marriage.
Many marital counselors and popular writings on marriage advocate what Gottman calls a “validation” or “active-listening” model. They recommend that when couples have a disagreement, they should speak to one another as a therapist speaks to a client. One spouse, for example, the wife, is supposed to state her complaints directly to the husband, in the form of “I” statements; for example, “I feel you're not doing your share of the housework.” Then he is supposed to calmly respond by paraphrasing what she has said and empathize with her feelings: “Sounds like you're upset about this.”
To their surprise, Gottman and his colleagues found that very few couples actually fit this therapeutically approved, “validating” model of marriage. Gottman et al. found, like Cuber and Harroff, that people can
In my own study, I too found a great deal of variation among the longitudinal couples. Apart from the deep friendship that all the happy couples shared, they differed in other ways. Some spent virtually twentyfour hours a day together, others went their own ways, going off to parties or weekends alone. Some were very traditional in their gender patterns, others egalitarian. Some were emotionally close to their relatives, some were distant. Some had a wide circle of friends, some were virtual hermits.
They could come from happy or unhappy families. The wife in one of the happiest marriages had a very difficult relationship with her father; she grew up “hating men” and planned never to marry. Her husband also grew up in an unhappy home where the parents eventually divorced. In short, if the emotional core of marriage is good, it seems to matter very little what kind of lifestyle the couple chooses to follow.
Marriage Is a Movie, Not a Snapshot
Heroclitis once said that you can never step into a river twice because the river is always in motion. The same is true of marriage. A variety of studies show that over a relatively short period of time, families can change in the ways they interact and in their emotional atmosphere. In studies of police officer couples, described in more detail below, the same marriage could look very different from one laboratory session to the next, depending on how much stress the officer had experienced on each day. The IHD longitudinal studies made it possible to follow the same couples over several decades.
Consider the following examples, based on the first two adult followups around 1960 and the early 1970s (Skolnick 1981).
Seen in 1960, when they were in their early thirties, the marriage of Jack and Ellen did not look promising. Jack was an aloof husband and uninvolved father. Ellen was overwhelmed by caring for three small children. She had a variety of physical ailments and needed a steady dose of tranquilizers to calm her anxieties. Ten years later, she was in good health and enjoying life. She and Jack had become a close couple.
Martin and Julia were a happy couple in 1960. They had two children, an active social life, and were fixing up a new home they had
Perhaps the most striking impression from following these marriages through long periods of time is the great potential for change in intimate relationships. Those early interviews suggest that many couples had “dysfunctional” marriages. It seemed at the time to spouses, as well as to the interviewers, that the source of the marital difficulties was psychological problems in the husband or wife or both, or else that they were incompatible.
For some couples, these explanations were valid. Later interviews revealed the same emotional or personality difficulties and incompatibilities. Some people had divorced and married again to people with whom they were a better fit. One man who had seemed emotionally immature all his life finally found happiness in his third marriage. He married a younger woman who was both nurturing to him and yet a “psychological age mate,” as he put it.
Although close to a third of the IHD marriages eventually did end in divorce, all of those couples were married years before the divorce revolution of the 1970s made divorce legally easier to obtain as well as more common and socially acceptable. Many unhappy couples remained married long enough to outgrow their earlier difficulties or to advance past the circumstances that were causing the difficulties in the first place. Viewed from a later time, marital distress at one period or stage in life seemed to be rooted in situational factors: problems at work, trouble with inlaws or money, bad housing, too many babies too close together. In the midst of these strains, however, it was easy to blame problems on the basic character of the husband or wife or on their incompatibility. Only later, when the situation had changed, did it seem that there was nothing inherently wrong with the couple's relationship.
The Critical Events of a Marriage May Not Be Inside the Marriage
The longitudinal data, as noted, revealed a striking amount of change for better or worse depending on a large variety of life circumstances. While the impact of such external factors remains a relatively unacknowledged source of marital distress, there has been growing interest in the impact of work and working conditions—especially job stress—
Briefly, Levenson's part of the study looked at the impact of stress on couple interaction in the laboratory. His procedures called for each spouse to keep a stress diary every day for thirty days. Once a week for four weeks, the couples went to the laboratory at the end of the work day, after eight hours of being apart. Their interaction was videotaped, and the physiological responses of each spouse were monitored continuously.
In my part of the project, we used an adaptation of the IHD clinical interview with officers and their wives in their homes. (The sample did not include female officers or police couples.) The aim was to examine their perceptions of police work and its impact on their marriages, their general life circumstances, and the sources of stress and support in their lives. I discovered that these officers and their wives were making heroic efforts to do well, against enormous odds, in their work and family. The obvious dangers and disasters police must deal with are only part of the story; sleep deprivation, frustration with the department bureaucracy, and inadequate equipment were some of the other factors adding up to enormous stress.
Moreover, some of the ways people in other occupations relieve stress were not easily available to these officers. The department offered counseling, but there was a stigma attached to using it. The officers feared any troubles they had would get back to their superiors, and they would be “hung out to dry.” Socializing with other officers was difficult because of an aspect of police culture described by the police novelist Joseph Wambaugh as “choir practice” —that is, drinking and womanizing. The men we interviewed felt they had to steer clear of this culture to protect their marriages.
Despite their difficult lives, these couples seemed to have good, wellfunctioning marriages, at home and in the laboratory—except on highstress days. Levenson's study was able to examine the direct effects of different levels of stress on the face-to-face interaction of these couples—something that had not been done before. The findings were striking.
More surprising, it was not just the police officer who showed evidence of stress but the partner as well. Even before either partner had said a word, while they were just sitting quietly, both the officer and the spouse showed signs of physiological arousal. In particular, there was a kind of “paralysis of the positive emotion system” in both partners (Levenson, Roberts, and Bellows 1998). Looking at the videotapes, you didn't need the physiological measures to see what was going on. The husband's restless agitation was clear, as was the wife's tense and wary response to it. The wives seemed frozen in their seats, barely able to move. In fact, just watching the couples on videotape is enough to make a viewer also feel tense and uneasy.
Recall that these couples did not look or act this way on the days they were not under high job stress. However, in these sessions, the couples' behavior revealed the same warning signs that Gottman and Levenson had found in their earlier studies to be predictors of who was likely to be divorced. The “paralysis of the positive emotion system” means that the “good” aspects of the marriage were unavailable just when they were most needed. Repeated often enough, such moments can strain even a good marriage; they create an emotional climate in which tempers can easily flare, hurtful things may be said, and problems go unsolved. Police work may be an extreme example of a high-pressure occupation, but it is far from the only one. “What's the difference between a stressed-out business executive and a stressed-out police officer?” asked a New York columnist not long ago, after a terrible case of domestic violence in a police family. The officer, he went on, “brings home a loaded gun.”