Skip to content
MyFreud

Stonewalling: Why People Shut Down Mid-Argument

Going blank and silent in a row is usually a body overwhelmed rather than a punishment. What the research on flooding shows, and the break that actually works.

6 min read

A flat illustration of two people at opposite ends of an orange sofa, one looking down at a tablet and the other hugging a cushion and looking away.

Key takeaways

  • Stonewalling is shutting down and withdrawing during conflict: going quiet, looking away, giving one-word answers, or leaving. To the other person it reads as contempt, and it usually is not.
  • Underneath it is often physiological overwhelm. Once the body passes a certain level of arousal, the capacity to take in new information and problem-solve drops sharply, and shutting down is what that looks like from outside.
  • Couples research has found withdrawal during conflict among the behaviours that predict relationships ending, and found it far more common in men than women.
  • The intuitive response, following the person who has withdrawn and pressing harder, makes it worse, because the shutdown is a response to load and pressing adds load.
  • The intervention with the best support is an agreed break of at least twenty minutes, spent genuinely calming down rather than rehearsing the argument, with a firm time to return.

Stonewalling is what it is called when somebody in an argument stops responding: goes quiet, looks past you, answers in single words, or walks out. From the other side of the room it looks like the most contemptuous thing a person can do.

From the inside it is usually closer to a system tripping a breaker. That gap between how it looks and what it is turns out to be the whole problem.

What is actually happening in the body

The body is past the point where argument is possible, and shutting down is what that looks like from outside. Research on couples in conflict has tracked physiological arousal alongside behaviour, and found that heart rate and related measures climb steeply during a row well before anybody would describe themselves as overwhelmed. [gottman-dissolution-physiology]

Above a certain level of arousal, several things degrade together: taking in new information, holding what was just said in mind long enough to answer it, and generating anything constructive. What is left is an urgent need for the input to stop. Silence is the cheapest way to achieve that.

This is why the most reasonable-seeming response fails. Explaining the point again, more clearly, adds input to a system that has already stopped processing it, and the shutdown deepens. Nothing about the quality of your argument is being evaluated at that moment, because the equipment that would evaluate it is offline.

What happens to arousal when a row keeps going, and when it pauses Illustrative
0 25 50 75 100 Physiological arousal Start 5 min 10 min 15 min 25 min 40 min Kept going Twenty-minute break at 15 Overwhelm threshold

A schematic of the pattern described in this article, drawn to show shape rather than measured values.

The flat line is the thing to notice. Once a conversation crosses it, more talking cannot resolve anything, and the two paths after the fifteen-minute mark are the entire practical content of this article.

Why it reads as contempt when it usually is not

Because the behaviours look identical from outside, and the person on the receiving end has no access to the difference. Silence, a turned head and a flat “fine” are what withdrawal looks like, and they are also what dismissal looks like.

So the receiving partner reasonably concludes they are being dismissed, and does the thing that follows from that conclusion: raises the volume, follows them, demands a response. This is the point at which two people who both want the argument to end start doing precisely the things that extend it. Our piece on anxious attachment covers why the pursuing half of this feels so unbearable, avoidant attachment covers the withdrawing half, and the relationships guide covers the wider picture both sit inside.

The pattern is worth taking seriously rather than waiting out. Withdrawal during conflict has been found among the behaviours predicting relationship deterioration, and the same body of work found it markedly more common in men than in women. [gottman-dissolution-physiology] The reason usually offered is that men reach and hold high arousal more readily in conflict, so they hit the shutdown threshold first and more often. That is an explanation rather than an excuse, and it points at what to change.

What the person shutting down should know

That leaving without a return time is the version that does the damage, and that saying one sentence first costs almost nothing. Most of the harm here comes from the ambiguity rather than the break.

“I need twenty minutes and I will come back at nine” converts the single most alarming thing you can do into the single most useful thing you can do. It is the same twenty minutes. The difference is that your partner spends them waiting rather than concluding that you have written the relationship off.

What happens during those minutes decides whether they work. Rehearsing your case, replaying what they said, or drafting the reply you will deliver on return keeps arousal exactly where it was, and you will come back activated to a person who has been waiting. Something genuinely absorbing and physically settling is what brings it down. And the return has to happen, on time, without being chased, or the protocol becomes another way of avoiding the conversation and stops being trusted.

The twenty minutes, used properly

Call the break out loud, name a time to come back, then start this. The rule for the whole timer is that you are not allowed to think about the argument.

20:00

Get physically away from where the row was. Do something absorbing enough to occupy you: walk, wash up, stretch, play something. If your mind goes back to the argument, notice it and put your attention on something in the room. You are not preparing your answer and you are not deciding who was right; you are getting your body back under the line so that the conversation can work when you return. Then go back, on time.

What the person being shut out should do

Stop adding input, and agree the protocol later rather than demanding it now. Both halves are counterintuitive and both matter.

In the moment, reduce the load. No further points, no following them from room to room, no ultimatum about talking now. This feels like conceding and it is not; it is the only move that shortens the episode.

Ask for a return time rather than a conversation. “When can we pick this up?” is answerable by somebody who is flooded. “Why do you always do this?” is not.

Do the repair when you are both calm. Agree, on an ordinary evening, that either of you can call a break, that calling one is allowed and is not an escape, and that whoever calls it names a time. A protocol agreed in peacetime is the only kind that survives a row.

Notice whether it is a shutdown or a strategy. If breaks are called only when the subject is inconvenient, never returned to, and used to end conversations permanently, that is a different problem. Emotional manipulation covers where that line sits.

When to seek help

Speak to a couples therapist if the same argument keeps restarting without ever finishing, if breaks are never returned to, or if contempt has become the ordinary tone between you. Ask about emotionally focused therapy or the Gottman method by name; both work directly on this cycle and both have evidence behind them. [apa-psychotherapy-topic]

Go individually if you are the one shutting down and it happens outside arguments too, since that pattern also appears in freeze responses after trauma and is worth understanding on its own. Go sooner if the withdrawal comes with control over money, movement or who you see, which is coercive control rather than a communication problem, and needs a different kind of help.

Contact your local emergency services or a crisis helpline if you feel unsafe or have thoughts of harming yourself.

How MyFreud can help

MyFreud is a mobile app that helps you find solutions to problems that have affected your mind and productivity. Daily tracking is genuinely useful here because these episodes cluster: they land on tired weeks, on the days after bad sleep, and in the hour before dinner far more often than people believe, and seeing that pattern makes the break protocol easier to agree to. Live coaching sessions are worth it for building that protocol as an actual agreement rather than a good intention, which is the step most couples skip. The notepad is where the point you were making goes when you call the break, so that you do not have to hold it while you calm down.

Download MyFreud and start today: App Store or Google Play.

Frequently asked questions

What is stonewalling?

It is withdrawing from a conversation while still physically present: going silent, avoiding eye contact, giving minimal answers, busying yourself with something else, or leaving the room. It differs from taking a deliberate break in that nothing has been agreed and no return has been promised. Relationship researchers describe it as one of the recognisable conflict patterns that predicts a relationship deteriorating, which is why it is worth addressing rather than tolerating.

Is stonewalling a form of abuse?

Not on its own. Most stonewalling is a shutdown under overwhelm and the person doing it usually experiences it as being unable to speak rather than as choosing not to. It becomes something else when it is used deliberately and repeatedly to punish, control or extract compliance, particularly when it is one part of a wider pattern of controlling money, time or contact with other people. The distinction is intent and pattern, and if you are in the second situation, that is coercive control rather than a communication problem.

Why does my partner shut down when we argue?

Most often because their body has passed the point at which taking in new information is possible. Once physiological arousal climbs far enough, access to reasoning, memory for what was just said and the ability to construct a reply all degrade, and what remains is an urge to make the input stop. That is why continuing to explain does not get through: the problem is not that they have not understood the argument, it is that they have temporarily lost the machinery for processing it.

How do I get someone to stop stonewalling me?

Not in the moment, and not by pressing harder. During an episode the only thing that helps is lowering the load: stop the flow of new points, agree a time to come back, and physically separate. The actual repair happens outside the argument, when you agree in advance that either of you can call a break, that a break is not an escape, and that whoever calls it names a time to resume. Agreeing that when calm is what makes it usable when neither of you is.

How long should a break in an argument be?

At least twenty minutes, and often longer, because that is roughly the floor for elevated arousal to come back down. Under that, people return still activated and the row restarts from where it stopped. What you do during it matters as much as the length: rehearsing your case or replaying what they said keeps the arousal up and wastes the break entirely. Something absorbing and physically calming works better, and the return time has to be honoured or the break becomes abandonment.

Can stonewalling be fixed?

Usually, and more easily than the people in it expect, because the fix is procedural rather than personal. Most couples improve substantially just by agreeing a break protocol and using it, since the damage comes from the escalation cycle rather than from either person. It is harder where contempt has built up over years, where one person has stopped believing the other will return, or where the withdrawal is deliberate. Those are worth taking to a couples therapist rather than working out alone.

References

  1. 1.Gottman JM, Levenson RW ( 1992). Marital processes predictive of later dissolution: behavior, physiology, and health. Journal of Personality and Social Psychology 63(2). doi:10.1037/0022-3514.63.2.221
  2. 2.American Psychological Association ( 2026). Psychology topics: psychotherapy. American Psychological Association. apa.org .