John Gottman: what fifty years of watching couples can teach us

Who is John Gottman?

John Gottman is an American psychologist born in 1942. He earned his PhD in clinical psychology at the University of Wisconsin in 1971, then taught for many years at the University of Washington in Seattle, where he is now professor emeritus. It was there that he set up the lab the press nicknamed the “Love Lab”: a furnished apartment where couples spent a weekend, filmed and wired with sensors that measured their heart rate, their sweating, their movements.

His scientific partner was Robert Levenson, a psychophysiologist at the University of California, Berkeley. Starting in 1983, the two of them recruited couples in Bloomington, Indiana, and followed them for years. Those cohorts fed Gottman’s most cited papers. In 1996, he and his wife, clinical psychologist Julie Schwartz Gottman, founded the Gottman Institute, which trains therapists and publishes books for the general public.

Why it matters for your relationship: before Gottman, people theorized a lot about couples and observed them very little. He chose to look at what couples actually do when they argue, minute by minute, and to check years later who is still together. A good part of the vocabulary you come across today (the four horsemen, the positive ratio, repair) comes from this work.

The central idea

Couples who last don’t argue less than the others. What sets them apart is how conflict unfolds and gets repaired: the proportion of positive and negative exchanges, the presence or absence of contempt, the ability to bring the tension down before the body takes over. Gottman also showed that a large share of a couple’s disagreements never get resolved. They come from lasting differences, and the goal isn’t to make them disappear but to learn to talk about them without hurting each other.

Key concepts

The four horsemen

Gottman calls them the “four horsemen of the apocalypse”: four behaviors that, once they settle in, signal a breakup. They are criticism (attacking the person instead of describing a fact), contempt (mockery, sarcasm, eye rolls, a sense of superiority), defensiveness (justifying yourself, shifting the blame) and stonewalling (shutting down, no longer responding). In his data, contempt is the most toxic of the four. Each horseman has an antidote: a targeted complaint, a culture of appreciation, owning your share of responsibility, an announced break.

The 5-to-1 ratio

During an argument, the stable couples in his studies produced about five positive exchanges for every negative one: a smile, a joke, a gesture of affection, a “you’re right about that.” In couples who end up separating, this ratio falls below one to one. It isn’t a rule to apply mechanically. It’s a sign that negativity has to stay buried in positivity to be bearable.

Repair attempts

A repair attempt is any gesture that slows an escalation: a hand reached out, a bit of humor, a “wait, let me start over,” an apology. For Gottman, the deciding question isn’t whether a couple makes them (everyone does) but whether the other person receives them. A couple who lets those offers pass by is in danger, even if their arguments seem moderate.

Flooding

When the heart rate passes a certain threshold, the body switches to alert mode: you no longer hear, you no longer process information, you attack or you flee. Gottman calls this flooding. It is one reason his sensors mattered: they showed that some partners, often the men in his cohorts, stayed physiologically activated long after the conversation was over. Hence the advice to take a real break, at least twenty minutes, before picking things back up.

The Sound Relationship House

In the late 1990s, Gottman pulled his findings together into a model with several floors, the Sound Relationship House: knowing the other person’s inner world well, cultivating fondness and admiration, turning toward each other rather than away in the small moments of daily life, keeping a positive outlook, managing conflict, supporting each person’s dreams, building shared meaning. The two load-bearing walls are called trust and commitment. This is the model that structures the therapy taught by the Gottman Institute.

What the research says

Gottman’s work rests on cohorts followed over time, rather small but observed very closely. Here are the main ones, and what they really show.

• 1992, with Levenson: 73 couples recruited in 1983 and seen again four years later. The couples whose negative exchanges outweighed the positive ones during a conflict discussion were the ones who entered the “cascade” toward separation: dissatisfaction, thoughts of leaving, separation, divorce.
• 1998, with Coan, Carrère and Swanson: 130 newlyweds filmed in the first months of marriage, then followed for six years. Models based on how couples enter conflict, on refusing the other’s influence and on the lack of soothing predicted separation and satisfaction with a reported accuracy of 80 to 83%.
• 2000, with Levenson: 79 couples followed for fourteen years. Two critical periods for divorce: the first seven years, and midlife, when the first child becomes a teenager. Negative emotions during conflict predicted early divorces; the absence of positive emotions predicted late ones.
• Perpetual problems: according to the data gathered in The Seven Principles, about 69% of a couple’s disagreements come back over the whole length of the relationship, because they stem from differences in personality or needs.

What these studies really show: robust associations between certain behaviors during conflict and how the couple fares later. What they don’t show comes right next.

Limits and criticism

The most serious criticism concerns the famous “90% divorce predictions.” These figures come from statistical models built after the fact, once the couples who had divorced were known: the equation is fitted to the very data it is supposed to predict. In 2001, two researchers at New York University showed that an equation of this kind, 90% accurate on its own sample, loses a large part of its accuracy when tested on new couples. Gottman’s samples are also small (a few dozen couples per study), which makes this kind of overfitting even more likely. Predicting, strictly speaking, would mean calling it in advance, on a couple never seen before, and then checking. That hasn’t been done at this scale.

A second reservation: the leap from the lab to therapy. Gottman’s observations are solid. The therapeutic method that follows from them has been less tested by independent trials than its fame suggests, and some of his conclusions (for example, that active listening is useless) were challenged by other teams as early as 2000. Finally, his cohorts are mostly married, heterosexual, American couples from the 1980s and 1990s: how well the findings apply to other couples and other eras remains to be shown.

None of these limits cancels the essentials: the fine-grained description of what happens during an argument, and the idea that repair matters more than how often you fight. They simply invite us not to turn this work into an oracle.

How MyLoveLab uses it

The MyLoveLab program takes four tools from Gottman, without the magic numbers: the four horsemen and their antidotes, practiced right as tension rises in the Conflict management dimension; repair attempts and the break with a time to come back; the logic of the positive ratio in the Communication dimension, where you practice turning toward each other in the small moments; and the Sound Relationship House, whose load-bearing walls, trust and commitment, structure the Trust and support dimension. The test doesn’t claim to predict any divorce. It measures where these mechanisms stand in your relationship, today.

Further reading

• Why Marriages Succeed or Fail · Book, 1994The first book for the general public: the four horsemen, the positive ratio, the three couple styles.
• What Predicts Divorce? · Book, 1994The academic version, dense, for anyone who wants to see the data behind the slogans.
• The Seven Principles for Making Marriage Work (in French: Les couples heureux ont leurs secrets) · Book, 1999The bestseller, with Nan Silver: the Sound Relationship House in seven principles, with exercises.
• The Marriage Clinic · Book, 1999The manual for therapists, where the Sound Relationship House is laid out in detail.
• The Science of Trust · Book, 2011Trust and betrayal seen through game theory and emotional attunement.

Frequently asked questions

References

1. Gottman, J. M. & Levenson, R. W. (1992). Marital processes predictive of later dissolution: Behavior, physiology, and health. Journal of Personality and Social Psychology, 63(2), 221-233.
2. Gottman, J. M. (1994). What Predicts Divorce? The Relationship Between Marital Processes and Marital Outcomes. Erlbaum.
3. Gottman, J. M., Coan, J., Carrère, S. & Swanson, C. (1998). Predicting marital happiness and stability from newlywed interactions. Journal of Marriage and the Family, 60(1), 5-22.
4. Gottman, J. M. & Silver, N. (1999). The Seven Principles for Making Marriage Work. Crown.
5. Gottman, J. M. & Levenson, R. W. (2000). The timing of divorce: Predicting when a couple will divorce over a 14-year period. Journal of Marriage and Family, 62(3), 737-745.
6. Stanley, S. M., Bradbury, T. N. & Markman, H. J. (2000). Structural flaws in the bridge from basic research on marriage to interventions for couples. Journal of Marriage and Family, 62(1), 256-264.
7. Heyman, R. E. & Slep, A. M. S. (2001). The hazards of predicting divorce without crossvalidation. Journal of Marriage and Family, 63(2), 473-479.
8. Gottman, J. M. (2011). The Science of Trust: Emotional Attunement for Couples. Norton.

See the full bibliography