From the May/June 1994 issue

EVERY THERAPIST KNOWS HOW MYSTERIOUS MARRIAGE CAN be witness the apparently incompatible couples who seem to have more fights in a week than most spouses average in a year but still stay together for a lifetime, and even seem to be happy with each other. Or, at the other extreme, consider those couples who approach potential conflict like a fatal virus, dodging and hedging around disagreeable subjects, unable to openly discuss, let alone resolve, what would seem to be critical issues in their marriage. And yet, these same couples frequently labelled “in denial” or “repressed” by the marital therapy profession not infrequently raise families and merrily celebrate 30th and 40th wedding anniversaries in spite of doomsday prognoses.

The truth is that, for all our theorizing about how a good marriage coheres and a bad one unravels, few of us can claim to know very much about the mysterious inner workings of this most intimate relationship. The actual daily stuff of marriage the slow accretion over the years of countless small, subtle but deeply telling patterns of interaction that make or break individual marriages, is still terra incognita to most of us, married couples and therapists included.

This news may come as a shock, considering the plethora of psychological prescriptions for fixing broken marriages, but until very recently, we did not know much more about the emotional and behavioral processes within marriage than we did about the physiology of sexual behavior before Masters and Johnson. And most of what we still claim to know comes from personal musings based on the idiosyncratic practice of individual therapists; almost none of our theory and practice is founded on empirical scientific research. This is not to say that therapists have not helped countless couples save their own marriages, or have not themselves accumulated a wealth of informal wisdom about marriage over the years. But the reasons that individual marriages succeed or fail remain mysterious, and much of the marital advice whether it works or doesn’t has only the frailest of empirical foundations.

Does it matter that marital therapy is, by and large, uninformed by empirical research? After all, marital therapy works pretty well without science, doesn’t it? Unfortunately, not as much as we’d like to think. In spite of the large numbers of marriages probably improved or even “saved” by various forms of therapy, studies of marital therapy outcomes overall suggest that treatment is too often ineffective or useless. A 1993 study by my colleagues at the University of Washington, Neil Jacobson and M.E. Addis, for example, notes that about 50 percent of couples who have received any kind of marital therapy do not improve their marriages. My own research with couples who have had marital therapy also demonstrated that the correlation between marital therapy and divorce is consistently about 50 percent in other words, it appears that many couples use therapy as a conduit to divorce.

These facts are not very encouraging to professionals deeply committed to actually making a difference in their clients’ lives, nor do they make a great case for marital therapy with the bean counters of managed care. As if the ethical obligations to provide demonstrated and replicable results in therapy weren’t enough, the professional and financial incentives the growing demand of insurers for provable results ought to make every marital therapist a devotee of empirical research.

And in spite of the promulgation of the conventional psychological approaches in thousands of marital enrichment seminars, marital self-help books, mass-circulation magazines and TV talk shows, the general population doesn’t seem to have benefited much, if the depressing state of American marriage is any indication. While divorce rates appeared to be leveling off during the 1980s, more recent data suggests the opposite. Currently, more than half of all first marriages and 60 percent of second ones end in divorce, while a full 67 percent of all recent first marriages dissolve.

In spite of vast rivers of ink poured into identifying various social culprits including women’s emergence into the workforce, the population shift from small-scale agrarian communities to large, industrial, urban environments, increasing social violence, the decline of religion and morality, divorce laws that are too lenient, excessive value placed on personal freedom and individualism, to name a few there is still very little knowledge about what happens inside the privacy of an individual marriage that makes it live or die. Sociological studies that may help explain overall patterns of divorce are of no help in predicting which couples will divorce or stay together, which will be fatally vulnerable to social stresses, and which will exhibit the built-in patterns that act as a kind of immunization against the slings and arrows of society’s bad fortunes.

If the explanation for some of the mysteries of marital stability and disruption is to be found in the private, moment-to-moment messages couples can’t help passing back and forth the way they talk and listen to and look at each other, what they think and feel about each other and their own marriage shouldn’t these factors be systematically observed and examined? The answer, obviously, is yes, and the way to go about it would be to conduct solid experiments that examine both stable and troubled marriages over time, carefully tracing the emotional currents that over the years lead some couples to drift apart and others to build strong unions. Ideally, just as physiological studies have demonstrated that overweight, alcohol abuse, smoking, physical inactivity, high blood pressure and stress are strongly correlated with the eventual development of heart disease years before a coronary crisis actually occurs, so research begun early in marriages, before the final “outcome” of marriage is known, might determine which practices protect couples from divorce and which put them at greater risk.

TWENTY YEARS AGO, I MADE MY own first forays into this new world of marital research. As a young therapist I felt stymied in my attempts to help a particularly troubled couple, whose therapy sessions inevitably disintegrated into bitter personal fights, over which neither they nor I had much control. On a hunch, I made a series of videotapes of the sessions, then had them watch the tapes with me and tell me what they had been feeling and thinking during certain moments of their filmed interaction.

Both the couple and I were astonished by the vividness and clarity on the tape of the pattern of criticism, contempt and defensiveness they repeatedly fell into a pattern of which they had been largely unaware and even I couldn’t see clearly in session. While I still didn’t know what to do to help turn them around, not only did the tape act as a kind of positive catalyst for them it shocked them into working harder at hearing each other and trying to improve the way they spoke to each other it gave me my life’s work. I wanted to develop a science of marital interaction, a body of replicable data about the destructive emotional processes between spouses that, unless they were interrupted fairly early on, could reliably predict the dissolution of the marriage. In short, I wanted to be able to see, identify and intervene in these specific and observable patterns before they achieved critical mass.

From this more or less ad hoc beginning, I gradually developed, along with many colleagues, a complex multimethod research model for studying the interstices of marriage. Our data base has been drawn, over the last two decades, from 20 different studies based on the three videotaped conversations of 2,000 couples overall, correlated with electronically measured physiological responses and backed up by questionnaires and interviews.   We have used a coding system for relating facial expressions to emotion, as well as measuring other indices of emotional expression, including voice and language, to demonstrate levels of affection, interest, amusement and joy as well as anger, sadness, fear, contempt and disgust at different points during the conversation.  We have also watched and listened for specific kinds of verbal and physical behavior that communicated, for example, complaint, blame, criticism, whining, defensiveness, belligerence and domineering. In addition, physiological data heart rate, blood-flow rate, perspiration during stress, gross motor movement and, sometimes, stress-related hormones in urine and blood have been synchronized with observed interactions of the couples. We then have interviewed spouses about what they were thinking and feeling (and what they think their spouses were thinking and feeling) during specific moments of the taping. Finally, we have collected questionnaires and oral histories about the state of their marriages, their feelings of loneliness or togetherness, what they think and feel about the other, and what they think the other feels about them.

While we still have a long way to go before we truly understand the complex processes of marriage, we have gathered enough data about the way individual couples interact in marriages to develop a theory of the factors that, if they are not interrupted, put a couple on a trajectory toward divorce, one that grows steeper and slipperier over time. We think there is evidence to show that if these negative patterns of interaction are not reversed in time, there is a point of no return, after which not much can be done to save the marriage.

Even without being able to fully identify and analyze the thousands of factors that go into the subtle and complex communication patterns between a couple, we believe we are still in a position to predict, based on our data, which couples are most likely to be divorced in the future. Of the 2,000 couples in our data base, we have followed 484 couples, many for as long as 10 years, testing four years after the initial interview for the impact of factors we thought might predict divorce and continuing to monitor them after that. We found such strong linkages between the information we had collected originally and the couples’ marital status four and more years later whether they had divorced or not that we now feel confident about our abilities to predict the potential for divorce in particular kinds of marriages. Indeed, I can now tell from a brief interview with a couple, a few questionnaires and a portion of a videotape what the eventual fate of a particular marriage is likely to be. In fact, from just six variables from our standard Oral History interview, I can predict with 94-percent accuracy which marriages are headed for divorce.

WE ALREADY HAVE ENOUGH evidence to unseat some of the most venerable truisms of marital therapy, including what comprises a “good” marriage. The satisfactorily married couple, according to the conventional therapeutic wisdom, is first of all, deeply compatible: the spouses do not necessarily have to come from the same ethnic, religious and class background (though it helps), but they must agree on important things sex, money, religion, child-rearing and should be able to compromise on about everything else. Not that this couple doesn’t argue they do, but their arguments seldom get lowdown and dirty, or even very heated. When they disagree, they naturally and without prompting do exactly what therapists advise less compatible and more troubled spouses to do: recognize conflicts, acknowledge differences openly but address them honestly and calmly before they degenerate into shouting matches. Conventional wisdom says they listen respectfully and empathize with each other’s point of view; they don’t interrupt much and if neither can persuade the other to this or that side of the issue, they negotiate a workable compromise. Not surprisingly, these couples look and sound a lot like two psychotherapists engaging in a dialogue.

Undoubtedly, this kind of union what we call the validating style of marriage usually works very well. And it is, therefore, not surprising that a wide range of marital theories and therapies insight-focused, behavioral, psychoeducational are geared to getting all troubled marriages to approximate this pattern. Certainly, viewing this style of marriage as the ideal has simplified the careers of marital therapists; their fundamental goal has been to help unhappily married couples get back to the bottom-line compatibility they are all presumed to have started with, if the marriage was ever viable in the first place. It followed then that either a lot of fighting or no fighting at all were both signs of a marriage on the rocks, that both indicated hidden agendas and unrecognized symbolic conflicts, which were undermining the marriage. Whichever it was fighting or no fighting the couple needed to uncover and “hash out” their differences, then come to a compromise so they could achieve the kind of idyllic balance represented by the validating couple.

But from what we see in the laboratory, the idea that the only truly satisfying marriage is cast in the mold of the validating style is wrong. Likewise, the orthodox belief that compatibility is indispensable to marital happiness and the reduction of conflict is critical to saving troubled marriages is a myth.

Our research shows that it isn’t the lack of compatibility that predicts divorce, but the way couples handle their inevitable incompatibilities, not whether they fight all the time or never fight at all, but the way they resolve conflicts and the overall quality of their emotional interactions. Indeed, our research data suggests that it is the balance between positive and negative emotional interactions in a marriage that determines its well being whether the good moments of mutual pleasure, passion, humor, support, kindness and generosity outweigh the bad moments of complaint, criticism, anger, disgust, contempt, defensiveness and coldness. In fact, after studying, tabulating and analyzing probably tens of thousands of marital interactions in our data base, we have concluded that we can actually quantify the ratio of positive to negative interactions needed to maintain a marriage in good shape. And we found that satisfied couples, no matter how their marriages stacked up against the ideal, were those who maintained a five-to-one ratio of positive to negative moments. Whether they fought a lot or not at all, whether they seemed passionately engaged with each other or distinctly distant, and most important, whether or not they were compatible socially, financially, sexually what counted was the overall balance of positive to negative. This claim sounds entirely presumptuous: how can something with the mercurial, idiosyncratic and labyrinthine dynamics of a marriage be reduced to a simple ratio of interactions? But much to our own surprise, we found that certain kinds of marriages that would seem doomed to failure according to standard therapeutic prognoses, were actually quite successful; what set these rule-breaking, good marriages apart from others was their adherence to the five-to-one ratio.

Perhaps the most classic example of the presumably endangered marriage is the volatile type between spouses who apparently live to fight. These Punch-and-Judy couples have intensely emotional marriages, characterized by epic brawls, high levels of jealousy, prickly interactions, petty bickering, sarcastic asides and hair-trigger tempers. Unlike the “compatible” couples that are the marital therapist’s dream, these excitable couples do not fight fairly. When they argue, they go for the jugular, rarely listen to or empathize with their mates during the course of battle, and attempt to steamroller each other to their respective points of view.

And yet, that these couples engage in a lot more sturm und drang than most couples (and many therapists) could tolerate does not, of itself, mean they don’t have good marriages. In a successful marriage of this type, for every nasty swipe, there are five caresses, so to speak. Indeed, far more than other marriages, however solid and satisfying, volatile marriages are inclined to be deeply romantic and frequently dramatic. And because the spouses tend, as one would expect, to be passionate and intense people, their relationship when it is satisfying can be much more exciting and deeply intimate than the marriages of less emotionally engaged people.

Of course, there are pitfalls to these volatile marriages. Neither spouse worries overmuch about hurting the other’s feelings, nor do the two believe that discretion is the better part of valor. Thus, for these spouses, who readily wade into controversy and contumely, the five-to-one ratio is more dangerously vulnerable to shifting downward than for more cautious couples. And when they fly at each other without forethought, as they often do, they can inflict unforgivable wounds. Under external stresses the birth of a baby, for example the normally argumentative style of their marriage can deteriorate into endless bickering and quarreling, even violence.

At the opposite end of the spectrum of presumably “dysfunctional” marriages, at least according to received wisdom, are the imperturbable couples who cannot stand fighting the conflict avoiders. When a potential disagreement raises its serpent’s head, these conflict-avoidant or conflict-minimizing couples are more than likely to step around it, eyes averted. Interviewing these couples, we found it extremely difficult to even find a subject of continuing disagreement between them, and had to settle for relatively trivial problems that may have caused an occasional twinge of discomfort. (One couple admitted that they disagreed once on whether to have chicken or pizza for dinner.) When differences cannot be smoothed over or ignored sexual incompatibility, for example they are likely to resolve them by not resolving them, concluding that although they recognize the conflict exists, they don’t consider it as important as the many areas of common agreement they do share. In a sense, all of their conversations (they couldn’t be called arguments) with us about differences between them ended in standoffs; they agree that they disagreed, but make no attempt to persuade each other, nor haggle out a compromises. They simply agree to continue disagreeing, and then drop the subject.

According to standard theories of marital therapy, the union of this couple, even more than the first, is doomed. Compatibility seems to have become too much of a good thing, as it were, and they are terrified of disagreement. Their unacknowledged conflicts, while “repressed” and “denied,” so goes the standard theory, feed a deep, toxic undercurrent of hostility and rage. The prognosis for these couples is that either the spouses will become quietly antagonistic strangers living separate parallel lives in the same household, or, according to the “volcano” theory of marital interaction, the repressed fury (which, it is often assumed, must be there) may explode into outright violence.

Our research shows these “doomed” marriages survive. The reason is that these couples, just like the volatile spouses, share an interactive ratio of five positive to one negative moment. While they have fewer negative interactions, they also have fewer positive ones. The difference is that their relationships are less emotional than the volatile marriages. They are less likely to fight passionately, but also less likely to love passionately. Instead of a marriage that resembles a raging torrent, theirs looks like a calm lake. They are likely to be truly compatible, as well paired as matching bookends. Often they come from the same social and economic background, hold similar beliefs about religion, values, childrearing practices, financial issues and the like. They share a sense of their marriage as a kind of secure bastion, a solid fortress of “us,” so strong a bond that they can afford to overlook disagreements. Probably for centuries, traditional marriages resembled this type; the married couple, as unbudgeable a social institution as church and state could make them, didn’t require romance, or even active companionship, to shore up a union considered by nature, law and religion undissoluble.

Of course, these marriages also have their weaknesses. Because these couples allow so little negativity into their interactions, they may not be able to deal effectively with disagreements that cannot be ignored or evaded, in which case they may live with a good deal of unresolved misery and frustration. In an effort to avoid any confrontation at all, they also tend to undermine intimacy and, as a result, they may see their marriage become rather cold and distant. While they have fewer negative interactions, they also have fewer positive ones. Nonetheless, like the volatile couples, these conflict-minimizing pairs also have a very good shot at making and keeping a good, solid marriage for life in spite of the fact that they, too, do not follow the marital “rules” of the therapeutic trade.

In fact, we found such consistency of success for all three types of marital unions validating, volatile and conflict avoiding if they maintain the five-to-one ratio that we are inclined to consider it a universal constant. Like any other living thing, the marital relationship must sustain a kind of emotional ecological balance in order to survive. Marriages seem to thrive on, proportionately, a little negativity and a lot of positivity. The total amounts vary substantially from style to style, but the proportion between the pluses and minuses must remain the same. One couple’s successful marriage exhibits a lot of both negative and positive affect, another shows moderate amounts of negativity and positivity and the third, small amounts of each but all will show the same ratio.

Just as striking, however, is the strong possibility emerging from .our studies that only couples from one of these three affective styles seem able to maintain the necessary ratio for a satisfying marriage. It seems, in short, that lasting marriages come in three discrete types, and that there are no in-between types that work well or last very long. The fighting styles of each individual spouse probably reflect deeply entrenched personality traits and worldviews; a conflict between styles may well represent fundamental disagreement on the very constitution of happy versus miserable marriages. So, if a validator  (temperamentally disposed to calmly and rationally work problems out) or a conflict-minimizer (content to let problems remain unresolved) marries a volatile type (who thrives in the heat of passionate battle), serious problems in the marriage are almost foreordained. Typically, the volatile spouse (often the wife) first feels puzzled and impatient, then patronized and frustrated, finally frustrated and maddened by her validator or conflict-minimizing husband’s refusal to go mano a mano with her. The more moderate and reasonable he is, the more irritable, insulting and furious she becomes, driving him to defensive retreat, silent contempt and cold hostility. It seems from our studies that one or the other spouse would have to make a very concerted and probably difficult attempt to change his or her style of fighting, which might mean transforming some very basic personal attitudes, as well.

Paradoxically, successful couples are compatible but not in the way traditionally suggested by marital therapy theory. As it turns out, the spouses within each different style are compatible fighters; they do implicitly agree on the way they will disagree, on how they will traverse the rough terrain they inevitably cross on their trek through marriage.

Our research suggests that while disagreements and fights are not pleasant, and no couples, except the volatile seem to enjoy them, they are necessary in some degree to all good marriages. Recognizing disagreement and engaging in it, even if it is never settled or no compromise is reached, helps couples cope with difficult issues, while enriching and stimulating both of them. We speculate that the function of negativity, including anger, in marriage is to create a dynamic rather than static equilibrium between the spouses; certain forms of negativity are like spice that keep relationships from going flat. Anger, when directed at a particular issue and expressed without contempt or global criticism, is healthy, perhaps even necessary. We have found in our studies that while angry exchanges made both spouses unhappy during the period when they were happening, they correlated with long-term marital satisfaction. Blunt, straightforward anger seems to immunize marriages against deterioration.

NOT ALL FORMS OF NEGATIVITY are equal, however, in the ecology of marriage, and we have observed that some are clearly more dangerous, more toxic than others. The five-to-one ratio is a measure of a satisfying marriage; that couples experience it at one point in their marriage is no guarantee that they can count on it forever. From what we can see, no marriage of any type even one exhibiting a healthy five-to-one ratio of positive to negative interactions can long sustain itself once four particularly corrosive personal exchanges have insinuated themselves into the relationship. These four processes criticism, defensiveness, contempt and stonewalling I call the “four horsemen of the apocalypse” because they seem to have the inherently destructive power of a virus or a cancer; if they are not checked, they can colonize and ultimately destroy a relationship.

Although every couple engages in the terrible four from time to time, they need to be aware lest they begin gradually to occupy a growing proportion of normal fights and disagreements. Therefore, both couples and therapists need to understand the sometimes subtle, but always critical, difference between less damaging forms of negativity and the terrible four. Anger and disagreement, for example, are quite distinct from criticism and contempt. In the former, a husband might say, “I’m upset that you didn’t balance the joint checkbook. The bank called today about two bounced checks, and I was very embarrassed.” In the latter, his remarks would be less specific, more global, aimed less at his spouse’s actions, more toward her very being. “As usual, you screwed up our checkbook and humiliated me. You’re no rocket scientist, but I’d think you could learn to do some simple addition and subtraction.” It is not the anger that makes this attack destructive, it is the derision and gratuitous insult added to it. She not only made a mistake, she always makes mistakes and is kind of stupid, to boot.

Clearly, the walls are already closing in on these two, leaving no room for maneuvering into a more tolerable mutual exchange. Typically, they would trade attacks until one or the other, probably the wife, started screaming and the husband would engage horseman number four stonewalling. He would “remove himself” emotionally or physically refuse to answer or look at her, or storm out of the room.

Stonewalling is a characteristically male thing to do; in one of our samples of couples, we found that 85 percent of our stonewallers were male. And, in the course of our research, we have made some startling discoveries about the physiological differences between men and women that account for this disproportion.

Marital strife has a significant physiological as well as psychological component, which shows up differently between men and women during arguments. At the onset of a fight, men become more intensely upset physiologically than women measured in terms of higher heart rate and blood pressure and they remain distressed for a longer time long after their wives have calmed down. Probably this difference in wiring had evolutionary survival benefits: the prehistoric male of our species, to protect the female and her young, had to be more alert and physiologically responsive to external danger than she did more ready to attack and fight or flee in the face of environmental danger. In modern life, this propensity to higher arousal is not nearly so adaptive; it feels terrible, and to avoid the acute distress it causes, men are likely simply to shut themselves down, refuse to respond, try, as much as possible, to turn themselves into unfeeling stone. But this pain-reducing strategy is terrible for marriage. The stonewalling, as it turns out, increases the woman’s feelings of unpleasant physiological arousal more than anything else her spouse does, much more than shouting back, for example.

This physiological gender difference may help explain some of the truisms about male and female styles of fighting in marriage why women are more likely to be “emotional” during a fight and pursue their mates with complaints, criticisms and demands, while men tend to engage in rationalizations, avoid the subject, withdraw into silence and impassivity, or physically retreat. In fact, our research confirmed that men engage in such maddeningly avoidant behaviors (to their wives, at least) precisely because they are much more unpleasantly physiologically aroused by a fight than their wives.

On the other hand, we also found that in happy marriages, the widely touted theory that men are less emotionally expressive than women was not confirmed in our research, in fact, we found that, by and large, in satisfying marriages, there are no gender differences in emotional expression: men are as likely to share their most intimate emotions as women. Surprisingly, in happy marriages, men are more likely to reveal personal information about themselves dissatisfaction with the self, hurts, dreams, aspirations, reminiscences than their wives. And when these men are angry, they don’t stonewall, but openly let their wives know what they are feeling which, again, is much less stressful for their wives than stubborn withdrawal. Unfortunately, marriage is still about the only outlet for emotional expression in the lives of most men. Whereas wives usually have a fairly wide support network outside the marriage of friends and relatives, husbands, in essence, only disclose to their wives and nobody else. It is not surprising, then, that unhappily married men are deeply lonely.

What is surprising is that we found that men who did housework were likely to be more happily engaged and involved in their marriages than men who did not, and less lonely, less stressed and less likely to be sick four years after the initial meeting with them in the laboratory. Tested as a separate factor in men, doing housework, by itself, was related to lower heart rate, less physiological arousal in general and better health four years later. Clearly, what is measured here is not the fabulous, curative powers of housework, but the mutual and supportive engagement of spouses in good marriage not such a startling fact on its own, but astonishing in that it is expressed so clearly in physiology, in the very life and well being of the body.

Unhappy marriage is not physically good for either spouse, though the actual effects differ according to gender. Men are inclined to withdraw from marital interaction to buffer themselves from physically stressful feelings of arousal to a certain extent, this mechanism protects them. But years spent warding off emotional and physiological feelings of being flooded by their wives’ anger the use of enormous stores of energy for continual stonewalling and withdrawal take a very high toll on men’s physical health.

Conversely, women actually become sick after too many years of figuratively knocking their heads against a stone wall trying to get a response from someone who relentlessly refuses to respond. In fact, we found that the husband’s contempt in marriage predicted, over time, a wife’s susceptibility to illness; for example, by counting the number of a husband’s facial expressions of contempt for his wife, we could correctly estimate the number of infectious diseases she would have over the next four years.

IT IS AN UNPALATABLE, BUT unescapable, truth that some marriages cannot and should not be saved. Not only do patterns of toxic marital interaction keep the body in a state of unhealthy physical arousal, they create a psychological climate of helpless misery neither spouse can surmount the negativity and hostility that have seeped into virtually every shared marital interaction. Our study shows a physiological linkage between spouses in the laboratory, physiological responses of each spouse can be predicted by those of the other. In other words, in negative interactions, something like a complex feedback loop occurs between spouses, which includes the back-and-forth exchange of negative physiological arousal, psychological misery and destructive behavior. The repeated trauma of the marital interactions has not only become, in a sense, “hard-wired” into the physiology, but these bone-deep states of arousal can no longer be willfully controlled. Couples at the end of these marriages are unable to muster the cognitive and social abilities that, in less damaged relationships, could see them through to better times.

In good marriages, couples can readily repair the damage done during fights and the inevitable fallow periods (there are times, probably, when every spouse secretly wonders if the marriage hasn’t been a terrible mistake) just in the mutually soothing exchanges that make any relationship flow. But these repair mechanisms no longer work in badly ailing marriages; there is literally nothing the couple can talk about, no subject, no common interest that is not fully colonized by the all-absorbing state of their mutual contempt and defensiveness. The range of available positive exchanges has so shriveled, and the negativity grown so cancerous, that both spouses have literally lost the ability to breathe and move normally in each other’s presence; their muscles tense, their hearts beat harder, they feel they are suffocating.

At this point, we believe efforts to save the marriage are more likely to be disastrous than helpful. The partners are overwhelmed by a sense of failure, hopelessness and mutual alienation; they have been at war so long that there is no common ground left between them. Not only is it fatuous to suggest that they just “try harder” at this juncture, it may be bad for their health witness our data suggesting that staying in a hostile, distant marriage actually compromises the immune system, increasing susceptibility to illness.

Furthermore, while divorce is never desirable, the research of Andrew Cherlin, Mavis Hetherington and others suggests that it is probably better for children, as long as coparenting tasks are well managed, than a marriage reduced to a vicious intermingling of mutual hostility and loneliness. The evidence from research I am currently doing reinforces this position. It appears that a well-managed divorce, in which both spouses were helped therapeutically to separate with some degree of calm and dignity and make reasonable mutual childcare arrangements, is better for children than forcing them to live in the donnybrook of their parents’ terrible marriage.

So, how can a therapist help preserve and improve any marriage? We suspect that one major difficulty with marital therapy in general is that even unhappy couples who do have a chance at salvaging their marriage may already be so highly aroused physiologically by each other that they actually are unable to intellectually or emotionally absorb the lessons that therapists can teach them. The best therapeutic approach to all couples highly distressed physiologically and emotionally, whether they end up divorcing or putting their marriages back on track, might be to engage them in what I call “minimal marital therapy” a kind of pre-therapy training in soothing skills that both spouses could use to calm themselves and each other. I recommend that during a potentially difficult interaction, couples monitor their heart rates and stop the interaction when their hearts are beating more than 10 beats per minute over baseline. At that point, the couple should engage in a soothing ritual take a scheduled break (at least 20 minutes a shorter time is useless) during which each spouse avoids thinking about the faults of the other. The point of this technique, which might sound like a gimmick, is to slow the couple down and dampen the physiological arousal that makes them unable to hear each other. Along these same lines, spouses can be taught to replace distress-maintaining thoughts of righteous indignation and anger with self-soothing cognitions of the type at which women are generally better: “He doesn’t really mean what he’s saying,” “I know she loves me in spite of her anger,” “Just relax and don’t take it personally.”

At some point in this process, the couple can be taught to soothe each other by trying to listen without giving in to the temptation to interrupt with defensive whines, denials and counter-accusations, for example. Or they can learn to consciously use ordinary “validating” responses look at each other, relax their faces, breathe softly and deeply, answer briefly with comments that indicate that the speaker is at least being heard and understood (“I see,” “Oh, okay,” “Mmni-hmmm”). At a more advanced level, they can learn to say simple, affectionate things to each other, like, “Look, the money problem is not your problem, it’s our problem, and we’ll work it out together as we always have.”

This minimal therapy will probably seem very dull for many therapists, particularly those trained in highly sophisticated systemic, psychodynamic and behavioral marital therapy. But the most brilliant and philosophically subtle therapy in the world will have no impact on a couple not grounded in their own bodies to hear it, let alone put it into practice. The idea here is that if the spouses practice this simple set of skills over and over again until they become second nature, they will be able temporarily to halt the apparently unstoppable physiological overload that makes them unable to control their social interactions. Once they are breathing more normally, once they can think and process information again, they can take advantage of therapy or gain access to their own temporarily lost abilities to repair their relationship. It is important, of course, that the spouses are doing the repair work, not the therapist.

THE CONCLUSIONS THAT HAVE emerged from this large body of research show that “what everybody knows” is not always true attempts to turn perfectly good, conflict-minimizing or volatile marriages into the validating-style marriages that therapists prefer are unsuited to the real spouses in those real relationships.

A good theory of marital dissolution-why it happens must accurately predict which couples are most at risk long before the ultimate slide. Until recently, it would have seemed foolish to think that anything as complex, personal and idiosyncratic as marriage adhered to predictable patterns. And yet, it seems to be so. When we ask enough questions, when we deeply and carefully observe the smallest piece of behavior, we find in the most apparently chaotic marriages an intricate, but ultimately predictable, web of patterns at work. As we study marriage with the same respect and the same attention to detail that natural historians apply to the apparently inchoate confusion of nature, we come up with the same amazing discovery human relationships, like other natural processes, are not random and unknowable, but appear to obey certain laws. Science, which might be called the study of natural laws, will not compromise the fundamental mystery of the heart, any more than science will eliminate the mystery of spring because the processes of germination and photosynthesis are understood. But with every new bit of knowledge about the couples who come into our laboratory, we increase our ability to help them and ourselves fulfill in all of our lives the original promise inherent in love’s beginning.

John Gottman

John Gottman, Ph.D., was voted one of the Top 10 Most Influential Therapists of the past quarter century and was recently honored with the 2021 Lifetime Achievement Award by the Psychotherapy Networker. Professor Emeritus in Psychology at the University of Washington, Dr. Gottman is known for his work on marital stability and relationship analysis through scientific direct observations, self-report and physiology. He is the author or co-author of over 200 published academic articles and more than 45 books, including the bestselling The Seven Principles for Making Marriage Work; What Makes Love Last; The Relationship Cure; and Why Marriages Succeed or Fail. He is the co-founder of The Gottman Institute and Affective Software, Inc., which has created a tele-therapy technology that will live on cell phones, computers, and tablets to assist therapists in couples therapy and offer direct relationship building services directly to couples. Learn more at