why does one of us shut down when the other wants to talk?

when one of you shuts down and the other talks harder, those two moves are feeding each other. researchers call it demand-withdraw: pressure on one side, retreat on the other. the quiet is often an attempt not to make it worse. the chase is often an attempt not to be left in it.

the dishes are still in the sink. one of you says, can we talk about tonight. the other gets very interested in a screen, or walks to the bedroom and leaves the door half closed. the first person follows, because silence feels like the problem getting bigger. the second person goes quieter, because the following feels like the problem. ten minutes later you are fighting about who left the room, and tonight is still sitting there.

what is the demand-withdraw pattern?

andrew christensen and christopher heavey described the loop in a 1990 paper. one person presses for a change through emotional requests, criticism, and complaints. the other retreats through defensiveness and passive inaction. they filmed 31 married couples in two conversations: one about a change the husband wanted, one about a change the wife wanted. husbands, wives, and observers described the same shape.

the paper's shorthand says the demander is usually the woman and the withdrawer is usually the man. the results were narrower than that sentence. wife-demand and husband-withdraw was more likely than the reverse. that gender gap showed up when the couple discussed a change the wife wanted, and the paper did not find the same gap when the husband's change was the topic. looked at separately, both partners demanded more when the change was theirs, and withdrew more when the change was their partner's. men were more withdrawn overall. women were not more demanding overall. thirty-one heterosexual marriages in a lab cannot speak for every couple. what they show is that the chair you sit in can change with the subject on the table.

does the pattern track how a relationship feels?

paul schrodt, paul witt, and jenna shimkowski pooled 74 studies and 14,255 people. across the outcomes in that pool, demand-withdraw had a moderate association, r = .360. wife-demand/husband-withdraw (r = .380) and husband-demand/wife-withdraw (r = .392) were close in size. studies that included distressed or clinical couples came in higher (r = .413) than studies of couples who were not (r = .345). relational outcomes (r = .423) and communicative outcomes (r = .418) sat above well-being (r = .249) and demographic differences (r = .239). the paper describes the pattern as tied to unwanted results, including dissatisfaction and later drops in satisfaction, and it also says the loop can sometimes travel with a positive change. an average across many kinds of result is a map of the research. it is not a forecast for your next tuesday.

why does the quiet person go quiet?

scott woolley, in a gottman institute interview, describes the withdrawer's private logic. talking looks like it will make things worse: blame, a bigger fight, the chance of losing the person. shutting down is an attempt to protect the relationship. from the other side of the room it reads as not caring. woolley says the usual case is the opposite. the withdrawer cares enough to be afraid of the damage. the interview page quotes john gottman in why marriages succeed or fail: "this classical marital impasse is all too common — a wife seeking emotional connection from a withdrawn husband." the filmed study is the correction to the word "always." the impasse is common. the roles are not a law.

the pursuer's logic is the mirror. after a disconnection, talking feels like the way back to safety. say what happened, hear what the other person feels, stay until it is settled. when that fails, the same person can get critical, because the withdrawal hurts, and because any contact, even a fight, can feel closer than none. woolley is plain about the cost. both people often experience the other as deeply important. both end up feeling unloved. the pattern is what does the chewing. he also notes that someone can pursue on one subject and withdraw on another. the label matters less than noticing which threat puts you in which chair.

is shutting down the same thing as stonewalling?

stonewalling is the moment the listener leaves the talk while still in the room. the gottman institute describes it as a wall: tuning out, turning away, acting busy, because the person is flooded. heart rate up, stress hormones in the blood, too stirred to keep the conversation fair. the same page says that 85% of stonewallers studied in the love lab were men, and that stonewalling by women was quite predictive of divorce. the page states those findings and does not reprint the sample. demand-withdraw is the wider loop those moments sit inside. one person presses, the other retreats, and each move recruits the next. a single shutdown in a hard hour is a moment. a shutdown that arrives whenever one of you asks for a change is the pattern. the body state under that wall is its own subject. what flooding is is the surge that makes a fair reply hard, before the wall goes up.

how is this different from having the same fight?

a repeating topic and a repeating shape are different problems. why the same fight keeps returning is about a difference you will probably meet again, and about staying in dialogue with it. this page is about what happens to the dialogue itself. one of you leans in. the other leaves. you can have a perpetual difference and still both stay in the chairs. you can also chase and shut down about a problem that would have stayed small if the talk had stayed a talk.

loneliness beside a partner is the pain of being unknown while the person is still there. a shutdown can feed that pain, and it is a specific scene: the talk ended because one of you left it. supporting a stressed partner is for a hard day that came from outside the two of you. once the heat is between you, "just tell me" lands as more pressure. the apology is a later sentence. how to apologize is what you say once you can name the harm. a sorry aimed at someone still behind the wall often sounds like another knock.

how do you take a break that is not a disappearance?

none of these papers tested a script for your apartment. what they describe is a body that cannot keep listening, and a partner who hears the exit as rejection. the move that fits that description is a break you agreed to before anyone needed it.

the gottman institute's stonewalling page says to pick a signal in advance. a word, a phrase, a hand raised. if one person invents the pause in the heat, the other hears a bigger wall. both of you agree to honor it. then separate, and do something that actually settles a body. the same page says the break should last at least twenty minutes, the time they give for the physiology to come down. during it, skip the speech you are rehearsing. "i don't have to take this" and "why do they always" keep the alarm where it was. a walk, music, a slower breath. a gottman page on timeouts adds the other edge. a pause that stretches past a day starts to feed a colder story, and a stalemate about who has to walk back in first. the page notes that one partner often pursues and one distances, and that the split is not a measure of love. put a clock on the break while you are still polite enough to mean it. "twenty minutes, then the couch" is a plan. "i can't do this," with no return, is the pattern in a nicer coat.

what should you remember so the next shutdown is recognizable?

by the next evening you remember that you meant to be gentler, and you cannot remember what their quiet was for. write four lines once you are both steady.

• the signal, and what the break is. a walk around the block, the bedroom door closed, twenty minutes, no following. "they need space" is your summary. their sentence is the one you can use.
• the need under the chase, in their words. "i need to know we are still on the same side" is usable the next time. "they are dramatic" is what the pursuer sounds like from inside the wall.
• which topics put which person in which chair. the christensen films suggest the presser is often the one who wants the change. your money talk and your plans talk may swap the roles. write both down, or you will swear it is always the same person.
• what coming back sounds like. who speaks first, and the sentence that means "i am back" rather than "here is why i left." a return with a speech attached can restart the loop in the doorway.

date the lines. a true need from a brutal month can be the wrong map in a calmer one. tended is a place to hold the signal, the sentence under the chase, the topic that swaps your roles, and the line that means you have come back. you still have to leave the room kindly and return when you said you would. an app cannot calm a flooded body. it can keep the next shutdown from starting as if you had never learned what their quiet meant.

what can you do this week?

• on a calm day, name the loop without assigning a villain. "when i press, you go quiet, and then i press harder."
• pick the signal and the length. twenty minutes is the floor the institute describes. name the place you will both be when it ends.
• each of you says the need under your move, one sentence, with no rebuttal until both have finished.
• write the four lines that night. mid-argument you will not remember them fairly.

when is shutting down more than a conflict style?

some quiet is a wall built for a different kind of safety. days of silence used as punishment. a partner who frightens you, or who will not let you leave the room. contempt as the climate of the house. that is past a signal and a timer. a couples therapist is the right next room when you cannot stay in an ordinary conversation, or when the loop has been the only way you talk for months. if you are unsafe, safety comes before a better pause. for the ordinary version, two people and a loop, say what the quiet is for, and come back when you said you would.