Communication & Conflict — What To Say, How To Listen, And What To Do When It Goes Wrong

In short: Forty years of research from John Gottman's Love Lab can be summarised in a single finding: how you handle conflict predicts relationship survival more reliably than any other variable. Four specific behaviours — criticism, contempt, defensiveness, stonewalling — predict divorce with over 90% accuracy in his studies. The behaviours that predict success are quieter: bids for connection accepted, repair attempts received, softened start-ups instead of harsh ones, and a 5:1 ratio of positive to negative interactions during conflict.

"Communicate more" is one of the most repeated and least useful pieces of relationship advice ever given. The relationships that work don't communicate more — they communicate differently. Most of the difference comes down to a small number of specific behaviours that research has been pointing at for decades, and that anyone can learn.

What 40 years of John Gottman's research actually shows

John Gottman, a psychologist at the University of Washington, spent four decades observing couples in his Love Lab — an apartment-style observation suite where couples lived and were filmed having conversations. His team then followed up years later to see who was still together. The predictive accuracy his methods developed reached over 90% in some studies. That's a remarkable level of certainty for psychological research.

The findings consolidated into a model that has held up across replication. Conflict frequency doesn't predict divorce. Couples who fight rarely don't necessarily stay together longer than couples who fight often. The variable that matters is how the fighting goes and what happens afterward.

Four specific conflict behaviours — Gottman calls them the Four Horsemen of the Apocalypse — predict divorce with high reliability:

Criticism is attacking the partner's character rather than complaining about a behaviour. "You never help around the house" is a complaint; "You're so lazy" is criticism. The first targets an action; the second targets the person. Criticism activates defensiveness and starts the conflict on bad ground.

Contempt is the single strongest predictor of divorce in the entire dataset. Eye-rolling, sneering, mocking, name-calling, dismissive sighs. Contempt signals that one partner has stopped seeing the other as an equal. Couples with regular contempt divorce at 10x the rate of couples without it. It's also the most corrosive variable, because it changes the partner's attachment-system state in ways the contemptuous partner usually can't perceive.

Defensiveness is deflecting responsibility — meeting any complaint with counter-attack or excuse-making. It feels self-protective in the moment, but it tells the other partner that you'd rather argue than understand what they're saying.

Stonewalling is emotional withdrawal during conflict. Physical presence but psychological exit. It usually starts as a coping strategy (when the conflict feels too overwhelming to engage with) but signals to the other partner that the relationship isn't worth fighting for. Stonewalling is more common in men than women in heterosexual relationships, partly because men's autonomic nervous systems are more easily flooded by emotional intensity.

Repair attempts — the actual engine of survival

The single most counter-intuitive finding from Gottman's work is that successful couples don't avoid the Four Horsemen entirely. They notice them faster and repair afterward. A repair attempt is any signal — verbal or non-verbal — that one partner is trying to de-escalate or reconnect during or after conflict.

A repair attempt can be a small joke. A hand on the arm. The phrase "I love you" said mid-argument. A statement like "this isn't going anywhere, can we try again in twenty minutes." A genuine apology. The specific form matters less than the underlying signal: I am here, I am not your enemy, I want to come back to you.

Gottman found that the rate at which repair attempts are accepted versus ignored is the single best behavioural predictor of relationship outcomes — better than personality match, better than reported satisfaction. Successful couples don't have fewer fights. They catch the worst moments earlier and repair faster.

The skill that complements repair is softened start-up: how you bring up a difficult topic in the first place. Conversations that begin with criticism ("you always...") end in defensiveness 96% of the time, regardless of how reasonable the underlying complaint is. Conversations that begin with "I feel X when Y happens" end in resolution far more often. The first 60 seconds of a conflict predict the next 60 minutes with disturbing accuracy.

Bids for connection — the small moves that matter most

Outside of conflict, the most important variable Gottman identified is what he calls bids for connection. A bid is any small attempt to engage your partner — a comment about something interesting, a question, a touch, a look. Most bids are tiny ("look at that bird"; "I had a weird thing happen at work"). They happen dozens of times a day.

What matters is whether the partner turns toward, turns away, or turns against the bid. Turning toward is engagement — even a brief acknowledgement counts. Turning away is ignoring or being too distracted to respond. Turning against is responding with irritation or dismissal.

In Gottman's six-year follow-up of newlyweds: couples who stayed married had turned toward each other's bids 86% of the time. Couples who divorced had turned toward 33% of the time. That gap — formed by hundreds of tiny daily interactions — predicts more than any conscious decision either partner makes about the relationship.

The implication is uncomfortable: most relationships don't end because of one big thing. They end because thousands of small bids weren't received, and the emotional bank account quietly drained.

Active listening — the actually useful version

"Active listening" is the most over-prescribed and under-practised technique in relationship advice. The simplified version ("repeat what your partner said") is widely mocked because it sounds patronising in real conversations. The actual research on active listening is more useful than the cliché version.

Three behaviours genuinely matter: not interrupting (waiting until the other person has finished a thought), checking that you understood (not by parroting, but by reflecting the underlying point — "so you're saying X, even though it's hard to admit"), and responding to what they said rather than to your prepared rebuttal. The third is the hardest. Most of us listen for openings to make our point. Real listening means staying with what the other person is actually saying long enough that you might change your own mind.

Shelly Gable's research on active-constructive responding adds an important layer: how you respond to your partner's good news matters more than how you respond to their bad news. The four response styles — active-constructive (enthusiastic engagement), passive-constructive (mild positive), active-destructive (criticising the good news), passive-destructive (ignoring it) — predict relationship satisfaction over time more reliably than conflict patterns do.

When your partner gets defensive (or you do)

Defensiveness is usually a sign that the conversation has felt like an attack on the person rather than an attempt to address a behaviour. It's also one of the easiest patterns to reverse, once both partners can name what's happening.

The reverse: in heated moments, before responding, ask "what's the most reasonable interpretation of what they just said?" Then respond to that. If you can't find a reasonable interpretation, the conversation has gone past productive — pause it. When communication breaks down, the most useful single move is a 20-minute pause with an explicit agreement to return.

The pause has to be explicit. "I need 20 minutes to calm down — I'll come back and we'll finish this" is a repair attempt. Walking off without comment is stonewalling. The two look similar but produce opposite outcomes.

The silent treatment and stonewalling — what's actually happening

The silent treatment looks like deliberate punishment. Often it isn't — it's flooding. Gottman's physiological research found that during heated conflict, men's autonomic nervous systems become overwhelmed at lower thresholds than women's. Heart rate rises past 100 bpm, the body's stress response activates, and the person becomes physiologically unable to engage productively.

The stonewalling response — going silent, withdrawing, leaving the room — is the body trying to stop the flooding. It's not a choice in the conscious sense, but it lands as deliberate withdrawal. The result is a frustrating loop: one partner pursues for engagement, the other physiologically can't engage, the pursuer escalates because the silence reads as contempt, and the stonewaller withdraws further.

The fix: name flooding when you feel it, take a 20-minute physiologically-genuine break (no rehearsing the argument during the break — read a book, walk outside), then return. Most floods de-escalate within 20 minutes if the person is genuinely doing something else.

How LoveCertain factors communication into matching

Communication style is one of the four dimensions in our matching algorithm — weighted at 15%. We don't try to match people on identical styles (some of the strongest relationships are between people with different but compatible styles). We match on compatibility: whether your conflict patterns interlock productively, whether your processing speeds match, whether one of you needs to talk things through while the other needs to think first.

See exactly how the matching works. The 15% weighting is intentional — communication matters but it's improvable. Values (40%) and life stage (25%) matter more because they're harder to change. Communication, by contrast, is genuinely teachable — which is why this whole hub exists.

Common questions about relationship communication

What's the single most important communication skill in a relationship? Repair after conflict. Successful couples aren't defined by how rarely they fight — they're defined by what happens after a fight. The ability to come back together, acknowledge what happened, and reconnect is what Gottman identified as the strongest behavioural predictor of relationship survival.

How do I get my partner to listen? Start the conversation differently. Conversations that open with criticism ("you never...") end in defensiveness 96% of the time. Conversations that open with "I feel X when Y happens" — naming your own experience rather than judging theirs — almost always go further. The first 60 seconds set the next 60 minutes.

Is the silent treatment always a red flag? Not necessarily. Often what looks like deliberate silence is "flooding" — the partner's stress response shutting down their capacity to engage. The fix isn't punishing the silence; it's naming it and agreeing on an explicit 20-minute pause with a clear plan to return.

How often should couples have hard conversations? Regularly enough that small frictions don't compound into big resentments. The research on "weekly check-ins" — a deliberate 15-30 minute conversation about how the relationship is going — consistently shows higher satisfaction in couples who do them. The frequency matters less than the existence of the ritual.

Can communication patterns actually change? Yes, and faster than most people expect. The single biggest factor isn't motivation but specific technique. Couples who learn softened start-up and structured repair often see measurable change within 4-6 weeks of consistent practice. That's faster than almost any other psychological intervention.

The articles below go deeper on each of these communication skills.