Are the four horsemen in your relationship?

Which one is in your arguments

John Gottman’s four horsemen are the four communication habits that predicted divorce most reliably in his lab: criticism, contempt, defensiveness and stonewalling. We have written elsewhere about what each of the four actually looks like. This quiz does the other half of the job: not explaining them, but working out which ones are currently in your relationship, because the fix is different for each and almost nobody has all four.

The antidote depends on the horseman

Gottman’s research paired each habit with a specific replacement. They are unglamorous and they are trainable.

• Criticism attacks the person rather than the problem: "you never think about anyone else." The antidote is a gentle start-up, which means saying what you feel and what you need instead of what they are. "I felt alone last night, I’d like us to plan the weekend together."
• Defensiveness is counter-attacking or explaining why the complaint is invalid. The antidote is taking responsibility for some part, even a small part. Not all of it. Some of it.
• Stonewalling is shutting down and going blank, usually because the body is flooded rather than because the person stopped caring. The antidote is a named break with a stated return time, then actually returning.
• Contempt is different from the other three, and worth treating separately.

Why contempt is the one that matters most

Contempt is eye-rolling, mockery, sarcasm aimed at the person, the sense of speaking down from a higher position. In Gottman’s data it was the single strongest predictor of divorce, and it is the horseman that does not respond to communication technique, because it is not a communication problem. It is what fondness looks like once it has eroded.

The antidote is therefore slower: rebuilding admiration and describing your own needs rather than your partner’s failings. If the quiz surfaces contempt, that is the result worth taking seriously, and it is the clearest case for working with a real therapist rather than a self-help routine.

What this quiz cannot tell you

Two honest caveats. First, the famous prediction accuracy came from trained observers coding recorded conversations in a lab, not from people rating themselves, and self-report is a much blunter instrument. Second, every couple does some of this. The horsemen are not evidence that a relationship is failing; their frequency and whether repair happens afterwards are what carry the signal.

That last part is the most encouraging finding in the whole body of work. Couples who do all four but repair well do fine. Repair is a small move: a joke, a hand on an arm, "that came out wrong, let me try again." The Gottman Method on communication covers more of the repair side.

This is a reflective tool, not a diagnosis or a prediction about your relationship. And if what happens in your arguments involves fear or feeling unsafe, that is outside what any quiz is for, and it deserves real support now rather than a technique to practise.