Most advice about arguing is advice about talking: how to make a point, rebut a claim, choose a word. Yet many people come out of a disagreement unable to say what the other person actually believed. The quickest improvement most of us can make is not to argue better but to listen better and to ask better questions.

Looping

The mediators Gary Friedman and Jack Himmelstein spent decades helping divorcing couples, business partners and feuding families settle disputes without going to court. A core practice in their work, later popularized by the journalist Amanda Ripley in her book on intense conflict, is called looping. It has two steps. You summarize what the other person said in your own words, then ask whether you got it right. If they say no, or "sort of," you try again, until they say yes.

Looping sounds trivial and is surprisingly hard. People in conflict tend to hear the first sentence and start preparing a reply. Looping forces them to keep listening. It also does something for the speaker: being accurately understood lowers defenses, because a person who feels heard no longer needs to keep repeating themselves louder.

The illusion of understanding

Questions can do what assertions cannot. In a 2013 study, Philip Fernbach, Steven Sloman and colleagues asked Americans about policies such as a cap-and-trade system for carbon emissions or a national flat tax. Some participants were asked to explain, step by step, how the policy would actually work. After trying, many discovered they knew less than they had thought, and they became less extreme in their positions. Another group was asked only to list their reasons for their view. Listing reasons did not moderate them; if anything, it gave them a chance to rehearse their case.

The finding suggests a practical move. Instead of asking someone why they believe something, which invites a defense, ask them how it would work. "How would the ban be enforced?" "Walk me through what happens to prices." Ask yourself the same questions.

Sounding receptive

Michael Yeomans, Julia Minson and colleagues studied what makes a person seem open to opposing views, a quality they called conversational receptiveness. They found it is signaled less by what people believe than by specific features of how they talk. Minson has summarized these features with the letters HEAR: Hedge your claims, Emphasize agreement, Acknowledge the other view, and Reframe to the positive. "I could be wrong, but" is a hedge. "We both want safer streets" emphasizes agreement. "I understand why you'd worry about cost" acknowledges. And "I'd like to find a plan that works" reframes toward a goal, instead of "Your plan won't work."

In their 2020 paper, people who read receptive responses saw their writers as better teammates and more reasonable. In a study of Wikipedia editors arguing over article changes, receptive editors were less likely to be personally attacked. Receptiveness is learnable: a short training in these features changed how people wrote and how they were judged.

Questions that build connection

Other research by the same circle of scholars, led by Karen Huang in 2017, found that people who asked more questions in conversation were better liked, and that follow-up questions, which build on what the other person just said, had the strongest effect. A follow-up question is proof of listening. It cannot be asked by someone who was only waiting for their turn.

Franklin's vocabulary

Benjamin Franklin described a similar discovery in his autobiography. As a young man he loved to win arguments and made himself unpopular doing it. So he changed his language. He wrote: "I even forbid myself... the use of every word or expression in the language that imported a fix'd opinion, such as certainly, undoubtedly, etc., and I adopted, instead of them, I conceive, I apprehend, or I imagine a thing to be so or so; or it so appears to me at present." His habit of replacing words of certainty with tentative phrases was, he believed, a large part of why he later had so much influence in public life. It is the hedge in HEAR, more than two centuries early.

Worked example: the vaccine conversation

Elena, a nurse, is talking with her neighbor Tom, who has decided not to vaccinate his children. Her instinct is to list the evidence. Instead, she loops: "So you've read about reactions, and you feel like doctors brush off parents who ask questions. Is that right?" Tom says that is mostly it, but the bigger thing is that his son had a bad reaction to a shot once and nobody took it seriously. She loops again with that included, and he says yes.

Then she asks a follow-up: "What happened after the reaction?" Elena hedges ("I may not know your son's history, but"), emphasizes their agreement that his son's safety comes first, and asks how he decides which health sources to trust. She does not change his mind that afternoon. But Tom agrees to bring his questions to a doctor Elena trusts, and he does not feel attacked, which makes the next conversation possible.