The Silent Treatment and Stonewalling: Two Things That Look Identical

You ask a question. Your partner goes quiet. Not a thoughtful pause, a flat, unreachable silence that stretches past the point where an answer would be natural. From outside it looks like one thing: a person refusing to talk. But the difference between the silent treatment and stonewalling is not in what you see. It is in what is happening inside the person who has gone quiet. That difference decides whether the situation has a repair path or whether it is a control behaviour you need to treat as a serious problem.

John Gottman's observational work at the Gottman Laboratory at the University of Washington gave the field its most precise answer to what stonewalling is. It is one of the Four Horsemen, the cascade of criticism, contempt, defensiveness and stonewalling that predicts relationship deterioration. But the key finding is that stonewalling is not a choice to be stubborn. It is a physiological event.

During conflict, when one partner's heart rate climbs past roughly 100 beats per minute, the body enters a state Gottman calls flooding. Diffuse physiological arousal takes over: adrenaline and cortisol flood the system, blood moves away from the higher cognitive centres, and the capacity for rational speech collapses. The person is not refusing to engage. They are, temporarily, unable to. In Gottman's laboratory studies, this state made constructive interaction impossible regardless of intention.

This is the distinction that matters. Stonewalling is a stress response with a physiological signature. The 100 bpm threshold is a population average, and your personal threshold may be higher or lower. The person withdraws, the heart rate stays elevated, and recovery takes time. Conflict recovery time is measurable in minutes, not hours, and it is trainable. Partners who learn to recognise flooding early can exit the cycle before the Four Horsemen cascade gets traction.

What Happens Inside the Flooded Partner

The Gottman stonewalling definition is specific: withdrawal from interaction as a response to emotional flooding, occurring while the person remains physically present. The physiological arousal is the engine. When heart rate exceeds the flooding threshold, vagal tone drops, emotional dysregulation sets in, and the person's ability to hear a partner's words collapses. They are not giving the silent treatment. They are drowning.

This is why the standard advice to "just talk it out" fails with stonewalling. You cannot talk someone down from a flooded state with more words. The repair attempt has to come before the flood, or you have to wait it out and return when the physiology has settled. In Gottman's data, the bid response rate, how often a partner turns toward a bid for connection rather than away, is a stronger predictor of long-term outcomes than the raw frequency of conflict. The master skill is not avoiding conflict. It is exiting the negative cycle and returning to connection.

The prevalence data makes clear how common this is. In one observational study of community couples, 85% of men and 76% of women exhibited stonewalling at least once during a 15-minute conflict discussion. This is not a rare dysfunction. It is a normal human response to feeling threatened, and it becomes a problem when the pair never learns to recognise it and recover.

How Coercive Silence Differs from Flooding

The silent treatment as control is not a flooding response. It is a deliberate, instrumental behaviour pattern. The person using it is not physiologically overwhelmed. They are punishing. The silence is a weapon, deployed with specific rules that distinguish it from stonewalling.

The duration is the first tell. Stonewalling lasts minutes to hours, tied to the arc of physiological arousal. Coercive silence stretches for days or weeks, long past any plausible recovery window. The termination condition is the second tell. Stonewalling ends when the body settles and the person can talk again. Coercive silence ends only when the target apologises, capitulates, or meets a demand. The function is the third: stonewalling is self-protection; coercive silence is used to punish perceived infractions, enforce compliance, and assert dominance.

Context matters too. Coercive silence appears alongside other controlling behaviours: isolation, monitoring, financial restriction, threats. It is listed on the Duluth Power and Control Wheel under emotional abuse. The target reports feeling erased, anxious, walking on eggshells, desperate to regain the partner's approval. In England and Wales, coercive control became a criminal offence under Section 76 of the Serious Crime Act 2015; Scotland and Ireland followed in 2018, and California in 2020. The legal recognition is recent. The pattern is old.

Flooding is the mechanism that turns a solvable disagreement into an unsolvable one. The research is clear on what happens: when heart rate crosses the threshold, roughly 100 bpm as a population average, the higher cognitive functions that support empathy, perspective-taking and coherent speech go offline. You are not being difficult. Your brain has literally stopped running the software that makes constructive disagreement possible.

This is why the pursuer-withdrawer loop is so hard to break. One partner escalates to get a response, criticism, then contempt, the two most corrosive of the Four Horsemen, and the other withdraws to reduce intensity. Each behaviour triggers more of the other. The escalation raises the withdrawer's heart rate, which pushes them deeper into flooding, which makes them quieter, which makes the pursuer escalate further. The cortisol synchrony during conflict becomes negative, a marker of dysregulation in Gottman's observational work.

The practical implication: you cannot resolve a disagreement while either partner is flooded. The repair attempt has to happen before the flood, or you have to agree to call a timeout and return when the physiology has settled. Conflict recovery time is trainable. Partners who practice this get faster at returning to baseline. The Gottman Method treats this as a skill, not a personality trait.

Recognise Flooding and Call the Timeout

How to respond to stonewalling depends entirely on which of the two situations you are in. If your partner is flooded, the response is not more words. It is a repair attempt: any action that de-escalates tension and signals a return to connection. A hand on the arm. A change of subject. A stated intention to come back in twenty minutes. The research shows that the repair attempt success rate is a stronger predictor of relationship outcomes than conflict frequency. Partners who can exit the cycle survive; those who cannot do not.

The sequence matters. First, recognise the signs of flooding in yourself and your partner: rapid breathing, tight jaw, raised voice, or the opposite, the flat, unreachable quiet. Second, call the timeout explicitly. "I am flooded. I need twenty minutes to settle, and I will come back." Third, actually leave and actually return. The bid response rate in the moments after return determines whether the repair attempt lands. Fourth, when you do return, use the language of Nonviolent Communication: state what you observed, what you felt, what you need, without blame.

What the Research Says About Repair

This works because stonewalling is a physiological event, not a moral failure. The flooding threshold varies individually, so your personal number may not be exactly 100 bpm. The Gottman research that produced the divorce prediction accuracy of 83% at six years, based on the observational coding of 130 newlywed couples, is the source for the Four Horsemen cascade and the ratio that matters. A higher positive-to-negative interaction ratio during conflict is better, but the threshold effect matters more than the absolute number: partners above the 5:1 ratio are not distinguishable from each other in outcome terms.

If the silence is coercive control, the repair path does not exist. You cannot Gottman your way out of a situation where one partner is using silence to punish, dominate and enforce compliance. The practitioner assessment tools make the distinction clinical: stonewalling is measured with Gottman's Specific Affect Coding System (SPAFF) in laboratory observation; coercive control is assessed with the Coercive Control Scale (CCS), developed by Dutton, Goodman and Schmidt. The first is an involuntary stress response. The second is a voluntary, instrumental behaviour pattern.

Do not look for techniques to communicate better with someone who is using silence as a weapon. That advice would teach you to accommodate abuse. Coercive control is a pattern that escalates, and it is legally recognised as a crime in multiple jurisdictions. The lifetime prevalence of coercive control in intimate partner relationships is estimated at 5-10% for women in high-income countries, based on the WHO multi-country study. If the silence is accompanied by other controlling behaviours, the relevant help is a domestic violence advocate or a coercive control specialist, not a couples therapist.

Couples therapy is contraindicated in coercive control situations because it assumes equal footing and mutual goodwill. The therapy-too-late trap is real: partners enter treatment after contempt is entrenched and repair systems have atrophied. But the trap is worse when the problem is not a repair system failure but a power imbalance. Know the difference. Ask the question the research answers: is the silence ending because the physiology settled, or because you gave in?

Stonewalling versus taking a break is the distinction that separates a healthy repair skill from an unhealthy withdrawal pattern. A deliberate, agreed break is a repair attempt: both partners consent to a timeout, agree on a return time, and the break ends with reconnection. Stonewalling has no agreed return. It is a unilateral withdrawal that ends when the flooded partner can speak again, or never, if the pattern becomes entrenched.

The practical test is simple. Did you agree on the break together? Did you set a return time? Did you actually return? If the answers are no, you are not taking a break; you are stonewalling. The attachment theory lens helps here: a secure attachment style tolerates short separations because the return is predictable. An avoidant pattern uses distance to manage threat, and the return becomes unreliable. The Gottman Method and Emotionally Focused Therapy differ on how to treat this. Gottman is behavioural and skill-building; EFT is attachment-focused and experiential. Both agree that the return is the test.

For the separated co-parent, this distinction is daily operating procedure. You are running a child-rearing partnership with someone you are no longer romantically tied to, and the skills required are closer to a businesslike alliance than a marriage. The break has to be explicit, time-bound and honoured, because the cost of a failed return is not just your own dysregulation. It is your child watching the pattern.

The Four Horsemen versus normal conflict is a distinction the Gottman research codes with precision. Normal conflict includes complaint, anger and disagreement without contempt. The Four Horsemen are specific destructive patterns: criticism attacks the person rather than the behaviour, contempt expresses superiority through sarcasm, eye-rolling and hostile humour, defensiveness deflects responsibility, and stonewalling withdraws entirely. In the observational data, contempt is the single strongest predictor of separation.

The cascade is the point. Criticism leads to defensiveness, which leads to contempt, which leads to stonewalling. Each step raises the physiological arousal of both partners, pushing them toward the flooding threshold. The positive-to-negative interaction ratio during conflict is a dimensionless ratio, and higher is better, but the threshold effect matters more than the absolute number. Partners above 5:1 are not distinguishable from each other in outcome terms. The cultural and historical contingency of love matters here too: what counts as contempt is not universal. What reads as hostile humour in one culture is banter in another. The Gottman coding system was built on a specific sample, and the pop-culture version has drifted from the findings.

The repair attempt is the master skill, and it matters more than the frequency or intensity of conflict. The research is unambiguous: the repair attempt success rate, the percentage of repair attempts accepted, is a stronger predictor of long-term outcomes than the raw conflict count. Partners who fight often but repair well do better than those who fight rarely but repair poorly. The bid-response loop is the underlying mechanism: relationships are built or eroded in small moments of attention-seeking, and the pattern is stable enough to predict outcomes over years.

The numbers give this shape. The bid response rate is a percentage of bids acknowledged or turned toward, and the difference between happy and unhappy partners is large in the observational data. Conflict recovery time, measured in minutes to return to baseline physiological state, is trainable. Cortisol synchrony during conflict, measured as a correlation coefficient, becomes negative in dysregulated pairs, a marker that the stress response is not resolving.

The practical translation is direct. When your partner makes a bid, a question, a touch, a shared observation, turn toward it. When you are flooded, say so and take the break. When you return, make the repair attempt explicit. The Gottman Method's "magic ratio" is a shorthand for this: five positive interactions for every negative one, but the ratio is a threshold, not a target. The skill is in the repair, not in the arithmetic.

The Therapy-Too-Late Trap

There are situations where the normal route, recognise flooding, take a break, return and repair, is closed. The most common is the therapy-too-late trap: partners enter treatment after contempt is entrenched and repair systems have atrophied. The research suggests the window for effective intervention is earlier than most people believe. If you have been in the pursuer-withdrawer loop for years, the repair attempt success rate is already low, and the skills need to be rebuilt from scratch rather than refined.

When Silence Is a Weapon

The second closed route is when the silence is coercive control. No amount of repair attempts will work because the other person is not trying to repair. They are trying to dominate. The failure case here is not a failed technique but a failed assessment: mistaking control for flooding and staying in a situation that is escalating. The recognised signs are the duration (days or weeks, not minutes), the function (punishment, not self-protection), the termination condition (your capitulation, not their recovery), and the context (other controlling behaviours: isolation, monitoring, threats). If those signs are present, the route is not a couples therapist. It is a domestic violence advocate.

Clinical Crisis, Not Relational Pattern

The third closed route is when one partner is in a mental-health crisis or active addiction. Flooding is a normal stress response; a crisis is a clinical condition. The distinction matters because the repair attempts that work for flooding are useless in a crisis. The response is not a timeout. It is professional help. Nothing in the Gottman Method is designed for this situation, and nothing here is a substitute for it.

The silent treatment versus stonewalling distinction suits partners who are fighting but not abusive, people who want to stay together and need the vocabulary to recognise flooding before it becomes a pattern. It suits the separated co-parent who needs to run a child-rearing partnership with a businesslike alliance, where the timeout-and-return protocol is a professional skill. It suits the culturally curious reader who wants to understand that the Four Horsemen are observations from a specific laboratory and a specific era, not universal truths about love.

It does not suit the person living with coercive control. If the silence is a weapon, this advice is not for you. The repair path is closed, and the exit route is the one that matters. It does not suit the pair where one partner is in active addiction or untreated mental illness. The issue is clinical, not relational. And it does not suit anyone looking for a guarantee: the research predicts outcomes with 83% accuracy at six years on a sample of 130 newlywed couples, but your relationship is not a statistic. It is a joint project of repair, and the distinction between the two silences is where the work begins.

• Gottman flooding threshold: Roughly 100 beats per minute, a population average that varies individually
• Gottman divorce prediction study: 130 newlywed couples, observed longitudinally from the 1980s, 83% accuracy at six-year follow-up
• Stonewalling prevalence in community couples: 85% of men and 76% of women in one observational study, at least once during a 15-minute conflict discussion
• Coercive control prevalence: Estimated 5-10% of women in high-income countries (WHO multi-country study)
• Coercive control legislation: England and Wales 2015 (Section 76, Serious Crime Act); Scotland 2018; Ireland 2018; California 2020 (SB 1141)
• Coercive control assessment tool: Coercive Control Scale (Dutton, Goodman, Schmidt); stonewalling is coded with Gottman's SPAFF system
• Repair attempt success rate: A stronger predictor of relationship outcomes than conflict frequency
• Positive-to-negative ratio threshold: 5:1; couples above it are not distinguishable from each other in outcome terms

Is the silent treatment always stonewalling?

No. Stonewalling is a physiological flooding response with a specific signature: heart rate above threshold, cortisol elevation, impaired cognition. The silent treatment can be stonewalling, but it can also be a deliberate coercive control behaviour. The test is the termination condition: stonewalling ends when the physiology settles; coercive silence ends when you capitulate.

How long does flooding last?

Conflict recovery time varies individually, but it is measurable in minutes, often 20 to 30 minutes for the heart rate to return to baseline. If the silence lasts days or weeks, it is not flooding. It is something else.

Can stonewalling be fixed?

Yes. Recognising flooding early and taking an agreed break is a trainable skill. The Gottman Method teaches this explicitly. The repair attempt success rate improves with practice.

What if my partner refuses to take a break?

That is a different problem. A repair attempt requires two people. If one partner refuses every attempt to de-escalate, the issue may not be flooding. It may be a power imbalance. Assess the other signs: duration, function, context.

Is couples therapy safe in coercive control situations?

No. Couples therapy assumes equal footing and mutual goodwill. Coercive control involves a power imbalance, and therapy can give the controlling partner new language for old behaviour. A domestic violence advocate is the appropriate first step.

Does the 100 bpm threshold apply to everyone?

No. It is a population average from Gottman's laboratory studies. Your personal flooding threshold may be higher or lower. The point is not the exact number; it is that you have a threshold at all, and it is worth learning yours.

What is the difference between the Gottman Method and Emotionally Focused Therapy?

Gottman is behavioural and skill-building, focused on observable interactions and repair attempts. EFT is attachment-focused and experiential, working with the emotional patterns beneath the behaviour. Both are evidence-based; neither is appropriate for coercive control.