Couples dynamics
What to say when your partner goes silent during an argument
Scripts to re-engage a partner who stonewalls you without triggering more withdrawal.
The couple comes in mid-fight, or they describe the same fight that happened on Tuesday. The pursuing partner asked a question. The other partner did not answer. The pursuer escalated. The other partner went deeper into silence. By the time they reach you, both of them are operating on the assumption that the other one is doing this on purpose.
Neither is. The mechanism is physiological, and the pursuing partner’s behavior is what keeps it in place.
What flooding actually looks like
Inside the silence, the withdrawer’s heart rate has crossed one hundred beats per minute. Adrenaline and cortisol are running. The part of the brain that processes complex sentences and runs empathy has gone offline. They are not deciding to ignore the question. They are physically unable to process the sentences being directed at them. They are hearing noise and sensing threat.
The pursuer’s mistake is treating this physiological state as a behavioral choice. They assume that explaining the point more clearly, or raising the volume to break through, will produce a response. From inside their nervous system, this is logical. From inside the withdrawer’s nervous system, the increased volume is increased physical threat, and the only available move is to go further inside.
This is the demand-withdraw loop in action. One nervous system manages anxiety by pursuing resolution. The other manages anxiety by reducing stimulation. The strategies are mutually exclusive. Both partners believe they are doing what survival requires.
What the pursuer wants to hear
The pursuer arrives in your office wanting a script that produces a response from their partner. They want the magic phrase. The phrase does not exist. There is no version of “the right thing to say” that gets a flooded brain back online while it is still being pursued.
This is the hardest piece of feedback to deliver, because it asks the pursuer to abandon the strategy they have been using to feel safe. Their anxiety has been telling them that if they stop pursuing, the conversation will go nowhere forever. The work is to show them that the pursuit itself is what produces the foreverness.
The shift you are asking them to make
The pursuer’s goal moves from extraction to stabilization. They are no longer trying to get information out of the partner. They are trying to make it safe for the partner to come back online.
This means dropping the rope mid-tug. It means accepting that the conversation is not going to resolve right now. The hardest piece is tolerating their own anxiety about the unresolved conversation without using their partner as the regulator for that anxiety.
Most pursuers can hold this for thirty seconds the first time they try it. Two minutes is a clinical win. Tracking the duration with them between sessions gives them a measurable target other than getting a response.
The lines that actually drop the rope
Each of these works because it removes the demand for an answer. The words themselves are not the active ingredient.
“I am pushing hard because I am anxious, and I can see I am making it difficult for you to talk. I am going to stop for a minute.” This names the dynamic without blame. The pursuer owns their part. The withdrawer hears that the pursuit is paused.
“You look like you are at your limit, and I am getting there too. Let’s take twenty minutes. I am going to read in the other room, and we can check in then.” This is the time-out with the time-in. The “I will return at this time” is the load-bearing piece. Without it, leaving reads as abandonment.
“I do not need an answer right now. I just wanted you to know how I am feeling. We can talk about the solution later when we are both less heated.” This separates the expression of feeling from the requirement to solve. It lowers the cognitive load on the silent partner to roughly zero.
“I can see you have gone quiet. I am interpreting that as you ignoring me, but I might be wrong. Can you tell me what is happening for you?” This is the hostile-attribution-bias check. Most pursuers do not know they are running one. Naming the assumption and inviting correction often surprises the withdrawer into a response that would not have been available under direct pressure.
What twenty minutes is actually for
Twenty minutes is roughly how long it takes for cortisol and adrenaline to clear once the threat signal is gone. Shorter breaks return both partners to the conversation still flooded. The exact number matters less than the principle: long enough for the physiology to reset, short enough that it does not become a different kind of avoidance.
The time-in is what keeps the break from becoming abandonment. “Twenty minutes, then we check in” gives the withdrawer a finish line they can run toward. “Let’s drop it for now” gives them an open question that produces its own flooding.
What to watch for in the next session
Did the pursuer use one of these moves? Did the withdrawer come back at the stated time?
If both happened, that is the new baseline. Reinforce the structure and look for the second-order pattern: what does the conversation look like once the flooding is no longer the dominant variable?
If the pursuer says they tried the move but the partner did not come back, the question is whether the partner was given a real out (specific time, no implied threat in the pursuer’s tone) or a performance of one. Most failures here are about the pursuer’s residual demand leaking through the structure.
When the pursuer says the move felt fake, take it seriously. The move IS unfamiliar. It will feel fake for several weeks before it feels natural. The fake-ness is the sensation of operating outside the established pattern. Trust that, and keep going.
When the silence is not flooding
Sometimes the silence is being used to extract a behavior rather than to regulate. The signal is whether it ends only when the silent partner gets what they want. If the partner can re-engage on the same topic later without conditions, you are working with flooding. If they re-engage only after the pursuer concedes or apologizes for raising the topic in the first place, you are working with a coercive silence, and the intervention shifts.
That distinction is worth holding while you work. The intervention for flooding fails on coercion, and the intervention for coercion makes flooding worse.
Continue reading with a Rapport7 membership
Get full access to 1,500+ clinical guides, directives, audiobooks, and weekly case supervision.
View Membership OptionsCreate a free account to keep reading
Sign up in 30 seconds. Free accounts get 1 full article, guide, or directive per week, the Rapport7 Assessment Map, and more. No credit card required.
Create Free AccountYou've used your free item for this week
Upgrade for unlimited access to all 1,500+ clinical guides, directives, audiobooks, and weekly case supervision.
Upgrade Now