'The Four Horsemen': Why Marriages Fail The second installment of the Day to Day series on marriage education examines the 20 years of research providing the intellectual underpinning of a federally-backed marriage education curriculum. Psychologists have identified four key problems that lead to divorce: criticism, defensiveness, contempt and stonewalling.
NPR logo

'The Four Horsemen': Why Marriages Fail

  • Download
  • <iframe src="https://www.npr.org/player/embed/4823861/4823862" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript
'The Four Horsemen': Why Marriages Fail

'The Four Horsemen': Why Marriages Fail

  • Download
  • <iframe src="https://www.npr.org/player/embed/4823861/4823862" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript

ALEX CHADWICK, host:

This is DAY TO DAY from NPR News. I'm Alex Chadwick.

We talk about marriage all the time. And still we never seem to get it quite right. Half of all marriages fail. Now there's an approach called marriage education. The Bush administration supports it. It wants $1 1/2 billion from Congress to help teach Americans how to make marriages last longer. In our second story on marriage education, NPR's Alix Spiegel explains that this whole idea of marriage education was developed by psychologists who were trying to figure out what it is that makes marriages fail.

ALIX SPIEGEL reporting:

According to John Gottman, one of the leading marriage researchers in the country, it was the psychologist Lewis Terman who pioneered the science of marriage. Terman, it seems, had a talent for reducing the complexity of human behavior and saw his research into relationships as an extension of the work that made him famous. In the early years of the 20th century, Terman helped develop the IQ test.

Mr. JOHN GOTTMAN (Researcher): And he believed that just like there was IQ, there was EQ, that there was emotional intelligence, and that he could actually measure the emotional intelligence of people in relationships, and he tried to develop measures of emotional intelligence. And what he wound up finding out is that people had all these different opinions about what made for a successful and happy and satisfying relationship, and almost none of them were true.

SPIEGEL: For example, Terman discovered that contrary to popular belief, happy couples didn't have more sex than unhappy couples and that there wasn't a specific personality suited to relationships. But perhaps Terman's most troubling finding was that on average, happiness in marriage decreased over time. The longer people stayed together the less content they were. This was in the 1930s, and it wasn't until the mid-1960s that a new generation of psychological researchers, people like John Gottman, began asking themselves if there was something they could do about this disturbing reality.

Mr. GOTTMAN: We said, well, you know, some people are doing real well. Some people really, you know, love their relationships, they stay romantic, they're really very happy about having been together. How are they different?

SPIEGEL: It was this question--What differentiates couples who are happy from couples whose relationships disintegrate?--that became the focus of study. But until the mid-'70s it was difficult to make progress on the question in part because of research technologies. Howard Markman, a psychology professor at Denver University who was part of the generation that changed marriage research, explains.

Professor HOWARD MARKMAN (Denver University): We had interview data. We had self-reports. But people aren't good observers of their own behavior. So interview data and self-report data are inherently limited.

SPIEGEL: The breakthrough for marriage research came--it must be pointed out--from a very unlikely source: the entertainment industry. You see, in the '70s, an early version of home video was introduced into the marketplace, and suddenly marriage researchers began to tape like mad. They would invite ordinary couples into the lab, give them a list of topics to talk about, and then unobtrusively tape their interactions.

(Soundbite of tape)

Unidentified Woman #1: And I make her clean her room and I get mad about her grades and I do all that stuff too. I just don't maybe have...

Unidentified Man #1: You don't make it necessary for her to have to learn from it?

SPIEGEL: This recording comes from the video library of Howard Markman's lab. It is one of literally hundreds of videos that he has recorded over the years. On the tape, a husband and wife discuss one of the central issues of their relationship: how to discipline their children.

(Soundbite of tape)

Unidentified Man #1: It was--What?--four days after she was in trouble that you came to me and said that you think she needs some time out?

Unidentified Woman #1: No, it was like two weeks.

Unidentified Man #1: It was about four days.

Unidentified Woman #1: It was two weeks.

Unidentified Man #1: Come on.

Unidentified Woman #1: But I'm talking--I didn't think it was that short. Maybe it was.

SPIEGEL: It is a banal conversation, the kind of conversation that you overhear of have all the time. But for 20 years, researchers like Markman and Gottman would meticulously code these exchanges, carefully cataloging each eye roll and sigh. Howard Markman.

Prof. MARKMAN: You have 10 or 15 minutes of videotape of a couple talking about their number-one problem area, and then you would transcribe who says what, then you would parse it--in some of our early work parse it into what we would call thought units, and then that would become the coding unit, and then you would train observers to watch the tape to code the verbal and non-verbal behavior, both watching the tape and using the transcript, of the speaker and the listener.

SPIEGEL: John Gottman.

Mr. GOTTMAN: Several of the laboratories around the country just sort of independently started doing the same thing at the same time, and we were coming up with the same findings: that there were systematic differences, and in fact you could measure that and you could describe them.

SPIEGEL: Over the years, this approach yielded many insights, including the surprising finding that one of our most basic cultural assumptions about marriage was misguided. Howard Markman.

Prof. MARKMAN: There was at the time and probably remains today a sense that, you know, love is really the most important feature of a happy marriage, and how much you love each other is really important. And our research has found that that's not the case.

SPIEGEL: Turned out that the intensity of love was not predictive when it came to determining whether or not a couple would stay together. What did appear to predict divorce, and in many cases the level of happiness in the marriage, was how the couples managed conflict.

Prof. MARKMAN: So I think one of the big-picture findings that really, I think, started turning the field in terms of how people understand what a healthy marriage is--kind of taking a focus off the positive side of things and focusing in on when you're angry, when you're upset, when you're frustrated, when you're sad, how do you handle that?

SPIEGEL: It was through coding the tape that the researchers were able to scrutinize very, very specifically the negative side of relationships, what behaviors were most destructive. And though different researchers have slightly different ways of describing the problem patterns, there's general agreement on the primary predictors of divorce. John Gottman.

Mr. GOTTMAN: There were four things that were the best predictors of divorce and continued misery if they stayed together. And we call them the four horsemen of the apocalypse. Criticism--like saying, `You never listen to me, you always talk about yourself,' instead of saying, `You know, I'm upset that you really talked about yourself during dinner. You didn't ask me questions and I need you to ask more questions,' which is a complaint. All right. The second thing--the second horseman of the apocalypse was defensiveness. The third thing was contempt or disrespect. And the fourth was stonewalling.

SPIEGEL: Gottman says some of these behaviors are more toxic than others. For example, contempt, which is especially destructive to wives.

Mr. GOTTMAN: That is sulfuric acid for relationships. Contempt is our best single predictor of divorce. And it not only predicts divorce, but it predicts how many infectious illnesses she's going to have in the next four years.

SPIEGEL: Once the predictors of divorce were identified, a single question remained: Would it be possible to alter the outcome of a marriage by teaching couples to change these extremely destructive behaviors. In order to answer this question, it was necessary to answer yet another question, a version of that age-old conundrum: Which came first, the chicken or the egg?

Mr. GOTTMAN: What was cause and what was effect--you know, was it that they were miserable and so they started being mean to each other, or was it that they started off kind of being mean to each other and became more miserable? And we found, in fact, that it's really not the misery that is determining the interaction. It's the interaction that's determining the misery.

SPIEGEL: This insight, in turn, led to the creation of marriage education. Again, Howard Markman.

Prof. MARKMAN: We knew that these patterns started predicting the future, so let's try to teach the couples while they were still happy some skills that would enable them to talk more constructively and, most importantly, fight less destructively in the face of conflict.

SPIEGEL: Tomorrow, we will hear about how this new approach, marriage education, found its way out of the psychology lab and into the political arena where it was embraced by an unusual coalition of political activists. For NPR News, this is Alix Spiegel in Washington.

CHADWICK: I'm Alex Chadwick. More to come on DAY TO DAY from NPR News.

Copyright © 2005 NPR. All rights reserved. Visit our website terms of use and permissions pages at www.npr.org for further information.

NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.