How to Do Marriage Therapy Honestly When You’re 70% Out the Door

RELATIONSHIPS • July 24, 2026

The Chair Closest to the Door

Adaugo picks the chair closest to the door. She has done this for six sessions now without noticing the pattern until her therapist names it gently in week seven. It’s a small brown armchair, slightly too soft, with a worn patch on the left arm where a hundred other clients have rested a hundred other elbows, and Adaugo has claimed it every single week without once registering the choice as a choice. Her husband sits across from her, hands open on his knees, saying the things a person says when he has been told that saying them might save something. He has brought a legal pad twice now, notes from a book on repair he found online, and there’s something about the earnestness of that legal pad that makes Adaugo’s chest hurt in a way she doesn’t have language for yet. Adaugo nods. She agrees, out loud, that yes, she would like to reconnect, yes, she remembers who they used to be, the two of them on a rented bike in a city she can’t quite place anymore, laughing at something neither of them could explain to anyone else. Inside, she is already gone. She has been gone, she realizes later, since sometime last winter, maybe the Tuesday in February when she sat in her car in the parking garage under her office for eleven minutes before going inside, not crying, just still, and what she is doing in that chair by the door is not couples therapy. It’s a performance of couples therapy, staged for an audience of one: herself, so she can tell herself later, whatever later holds, that she tried everything a reasonable person could try.

Adaugo is a composite, a blend of patterns I’ve seen across many driven women in my practice, not any single client, with details changed to protect privacy. But the chair by the door is real in the sense that matters, the sense that shows up again and again across very different marriages, very different histories, very different reasons for arriving at the same quiet standstill. A woman books the appointment, shows up on time, pays the copay, says true things in a room built for truth, and still leaves every session more certain of what she already knew walking in the first time. She drives home the same way every week, past the same gas station, and notices herself rehearsing what she’ll say at dinner about how the session went, already curating the story before she’s even parked the car.

What I see consistently is that this isn’t dishonesty exactly, and it rarely comes from malice. It’s something quieter and more tender than that: a woman caught between two loyalties, one to the marriage she built with her own hands over a decade or more, and one to the truth she can no longer unknow, no matter how many hours of therapy she logs trying to talk herself back into an earlier certainty. She isn’t lying to her husband, not exactly. She’s protecting him, and maybe herself, from a sentence she hasn’t finished forming, one she’s terrified will become permanent the moment she says it out loud in a room with witnesses. If you’re 70 percent out the door and still showing up to therapy anyway, still paying for the sessions, still driving there on a Tuesday evening when you’d rather be anywhere else, you already know the particular exhaustion of holding both things at once, of being fully present and fully gone in the same fifty minutes.

Here’s the paradox nobody names clearly enough: pretending you’re all in wastes the process, because the work being built on top of a false premise cannot hold real weight. But declaring you’re all out, flatly and finally, before you’ve actually tested that in the room with a skilled witness, forecloses it just as thoroughly, closing a door that honesty might have left usefully open. The work that helps isn’t about picking a side before you walk in the door, performing either devotion or detachment for the therapist’s benefit. It’s learning to say, out loud, in front of the person you married, exactly where you stand today, even when today’s answer is “I don’t know” or “mostly not here” or “seventy percent gone and I hate saying that number.” Saying the true percentage instead of a rounder, more comfortable number, the number that won’t alarm anyone, is often the single most useful thing that happens in the entire course of treatment.

This post is not a guide to fixing your marriage, nor a case for leaving it. It’s a clinical look at what honesty requires when you’re ambivalent, why ambivalence is information rather than failure, and how to tell genuine effort from effort performed to feel absolved later. If this sounds like the quiet math behind your own relationship patterns, this is for you.

What Is Honest Ambivalence?

Ambivalence gets treated, culturally, as a character flaw. Make up your mind. Commit or leave. Fish or cut bait, as though a marriage of fifteen years were a decision you could make the way you pick a restaurant. But in the clinical sense, ambivalence isn’t indecision, it’s the accurate perception of a genuinely mixed situation, one with real love braided through real exhaustion, real history braided through real doubt. You love parts of your marriage, the way he still remembers how you take your coffee, the family you built together, the private jokes nobody else would understand, and you are done with other parts, the loneliness inside the relationship, the years of feeling more like a co-manager than a partner. Both things are true at once, sitting right next to each other without canceling out. The discomfort you feel isn’t a sign you’re broken or indecisive or bad at commitment. It’s a sign you’re paying attention to something most people flatten into a single, easier story.

Discernment counseling, a related but distinct approach, was built for exactly this bind, and I wish more couples knew it existed before they spent eight months in standard marriage therapy wondering why nothing was landing. Unlike standard marriage therapy, which assumes both partners want to repair things and mostly need better tools and better communication, discernment work assumes one or both might be genuinely, honestly unsure, and it’s structured around a narrower question first, before any tool-building starts: is this marriage even worth the kind of full, committed effort real repair requires, and does everyone in the room actually agree that it is. That’s a fundamentally different question from “how do we communicate better,” and treating the two as the same question, starting repair work before anyone has answered the prior question honestly, is often exactly where couples therapy quietly stalls out for months without either partner understanding why.

In my work with clients, the women who do best in this process aren’t the ones who arrive with the clearest answer, the tidiest verdict, the most confident percentage. They’re the ones who can tolerate not having one yet, who can sit in the genuine not-knowing for weeks or months without collapsing it into a premature decision just to relieve the discomfort, while still being specific about what they’re feeling, session to session, moment to moment. That specificity, “today I felt some warmth when he made the coffee without being asked” or “today I felt nothing when he touched my shoulder, and that scared me more than anger would have,” is more clinically useful than a general verdict on the whole marriage. It gives the work something concrete and textured to actually respond to, rather than an abstraction everyone keeps circling without landing on.

Vague ambivalence lets you avoid the harder inventory: what you can actually tolerate long-term, what you genuinely can’t, and what you haven’t tested yet because you’ve been protecting yourself, or him, or both of you, from an answer that might be inconvenient or painful or final. Naming the specifics is uncomfortable, sometimes acutely so, in a way that makes a person want to reach for a vaguer, softer sentence instead. It’s also the only way through, part of what makes healing from earlier relational wounds possible inside a current relationship instead of around it, at a safe, theoretical distance.

None of this is a diagnosis of your marriage, nor a verdict on whether you should stay or go, and I want to be careful about that, because it would be easy to read a post like this as a subtle argument for leaving. It’s a description of a stance, a way of being honest inside uncertainty, that tends to produce clearer outcomes over time than either forced positivity or premature exit. You’re allowed to not know yet, for as long as not-knowing genuinely takes. You’re not allowed, if you actually want the process to work, to pretend you know more than you do. This differs from most writing on fearful avoidant attachment, but the two often show up in the same room together.

The Clinical Mechanics of Staying and Leaving

Long before anyone called it discernment counseling, researchers tried to understand what holds a relationship together and what predicts when it comes apart. Harold Kelley, an American social psychologist known for interdependence theory, described how partners weigh the costs and rewards of staying against their expectations and their sense of what else might be available. That weighing isn’t cold or unromantic. It’s simply how commitment gets built or eroded, whether or not a person is conscious of doing the math.

Building on that foundation, Caryl Rusbult, an American social psychologist, described how commitment grows from three ingredients: satisfaction with the relationship, the quality of perceived alternatives, and how much a person has already invested, in time, shared history, children, identity. Her framework explains why a woman can be deeply unsatisfied and still feel bound. Investment alone can hold a person in place long after satisfaction has quietly left the building.

This distinction has real stakes, not just emotional ones. A trial protocol built around a program called Healing Hearts Together, described under PMID 40766010, links relationship distress to measurable cardiovascular risk, and frames couple-level support as tied to healthier outcomes for both partners. Chronic, unresolved relational distress doesn’t just cost you emotionally. It appears to have a body, over years, and the body is yours.

There’s a cost on the other side of the ledger too. A pilot feasibility trial of a brief, single-day couples intervention for veterans, indexed at PMID 42175859, notes that separation and divorce are themselves associated with a deterioration in mental and physical health, and that access to couple treatment is often limited by logistics, cost, and geography. Staying has a cost. Leaving has a cost. Neither is free.

George Levinger, another American social psychologist, studied close relationships and mapped the stages by which marriages form, endure, and eventually dissolve. One detail from that work has stayed with me: dissolution is rarely a single event. It’s a sequence, often a long one, and a couple can sit somewhere in the middle of it for years before either partner names it out loud. If you’ve been quietly deliberating a long time, you’re not behind schedule. You’re, in Levinger’s terms, in a later stage of a process already underway.

None of this tells you what to do. It gives you language for what you’re already sensing: satisfaction, investment, and alternatives are three separate variables, not one feeling, and the ache of “should I stay or should I go” often comes from averaging them into a single verdict when they need to be weighed one at a time. For women untangling this from a history of complex trauma, the averaging problem tends to be harder, because the nervous system wants a single answer faster than the facts can produce one.

How This Shows Up in Driven Women

Afia is a composite too, drawn from patterns across clients rather than any one person, with identifying details changed. She runs an operations team of thirty people and has never once, in four years, missed a deadline. In therapy, she is fifteen minutes late to name what she actually wants to say, and when she finally says it, it comes out sideways: “I think I’ve been doing this so I can tell people I tried.” Not to save the marriage. To have evidence, later, that she gave it a real shot.

That distinction, effort as evidence versus effort as genuine attempt, is one of the clearest patterns I see in driven women specifically. You know what it looks like to perform diligence for an audience, because you’ve done it in every performance review you’ve sat through. So when you walk into couples therapy already mostly out, some part of you starts building the file: dates attended, books read, effort documented, in case anyone ever asks whether you really tried.

I don’t see this in every driven woman ambivalent about her marriage. Maybe three or four out of five. The exception tends to be women who’ve done enough of their own work already, often through relational trauma recovery, that they can tell genuine effort from performance without a therapist pointing it out.

For the rest, the pattern connects to something older. If you learned early that love had to be earned through achievement, showing up to therapy “correctly,” saying the right things, can feel like the only form of effort you know. It’s the same overfunctioning that shows up at work, except the stakes are your marriage, and nobody is handing out a score.

What research on relationship satisfaction, including the work behind PMID 41809743, suggests is that partners’ felt experience of love tracks more closely with perceived behavior than stated commitment. In plain terms: it isn’t the effort you can document that changes anything. It’s whether your husband can feel a difference in how you show up. Documentation doesn’t move that needle. Presence does, when it’s real.

Afia’s overfunctioning in therapy wasn’t a character flaw. It was the most sophisticated tool she owned, repurposed for a job it was never built to do. Recognizing that doesn’t fix her marriage. It frees her from building a case file instead of finding out what’s true, the same recognition that tends to interrupt people-pleasing as a trauma response in other parts of her life.

The Difference Between Effort and Performance

There’s a specific tell I listen for in session, and it isn’t what a person says about wanting to save the marriage. It’s what happens in their body when their partner does something small and kind. Real effort produces a flicker, a softening around the eyes, something involuntary. Performed effort produces a script: the right words, delivered on cue, nothing underneath moving.

A mixed-methods study of the Hold Me Tight relationship education program, delivered by paraprofessionals rather than licensed clinicians and described under PMID 40545740, found that structured relationship work could shift felt connection even in genuinely hard conditions. What that suggests, modestly, is that structure itself carries some of the weight. You don’t have to feel certain for the work to do something, only willing to actually show up inside it, rather than performing showing up from a distance.

The American poet Sara Teasdale wrote a line that I think about often in this exact context, from a poem that isn’t about marriage therapy at all but somehow names the whole dilemma anyway:

I read that line, in this context, as a description of the exact tension driven women bring into an ambivalent marriage: you want a partner who chooses you freely, not one who stays because leaving feels too costly. Performed effort, on either side, is possession without choosing. It’s staying because the alternative feels unbearable, dressed up as trying. Genuine effort, whatever it produces, is closer to actually choosing, with the door acknowledged as open.

This connects to a pattern I see in women recovering from trauma bonding in earlier relationships: the confusion between effort that proves something and effort that changes something. If your nervous system learned early that love required proving, discernment work is partly the process of unlearning that, the same unlearning underlying most narcissistic abuse recovery, inside a relationship that may be worth saving.

Both/And: Wanting Out and Wanting to Try

You can want out and still want to try. These aren’t contradictory positions requiring you to pick one before the work starts. They’re both true, often in the same session, sometimes the same sentence. The therapy that helps doesn’t ask you to resolve that tension before you walk in. It asks you to bring the tension into the room instead of resolving it alone, at 1 a.m., in your head, where it has no chance of being tested against reality.

What I see get missed constantly is the assumption that wanting out disqualifies you from trying honestly, as though ambivalence were bad faith. It isn’t. A person can show up fully present, genuinely curious about what might shift, and still be assessing, in parallel, whether what shifts is enough. That’s what discernment looks like from the inside, and it’s more honest than arriving with a fixed verdict and calling the sessions that follow “trying.”

The cardiovascular research referenced earlier, the Healing Hearts Together protocol under PMID 40766010, is worth returning to here, because it points at something Both/And thinking makes room for: staying in an unresolved, high-conflict holding pattern has its own cost, separate from whatever you eventually decide. “Let’s just keep going indefinitely, without naming anything” isn’t the neutral, safe option it can feel like. It’s a choice too, with its own physiological price tag.

Similarly, the Hold Me Tight findings under PMID 40545740 suggest that connection can shift even under real ambivalence, so you don’t need to resolve your uncertainty before something changes. Sometimes the shift itself generates the information you were waiting for. You don’t have to know the ending to start the chapter honestly.

Wanting to leave does not mean you failed at wanting to stay. Giving real effort does not mean you’ve forfeited the right to leave if it doesn’t land. Both/And isn’t a compromise position. It’s a more accurate description of what ambivalence is, and treating it as anything less honest is where couples work quietly breaks down. This same capacity is often what’s missing when setting a boundary feels impossible: you don’t need to stop caring to tell the truth.

The Systemic Lens: Who Gets to Be Ambivalent

Ambivalence is not distributed evenly. A driven woman weighing whether to stay in her marriage operates inside a culture with contradictory scripts for her. She’s told to communicate more, try harder, not give up on the family, while quietly asked why she’s still in a marriage that isn’t working, as though the ambivalence itself were the failure. Women with an anxious attachment style feel this pressure hardest, since staying quiet has felt safer than naming the truth.

Access matters too, and it isn’t evenly distributed. The veterans’ pilot trial referenced earlier, under PMID 42175859, notes that access to couple treatment is often limited by cost, location, and scheduling long before either partner gets near a clinical decision. A driven woman with resources has more room to explore ambivalence in a structured way, and that gap is systemic, not personal.

The idea that marriages move through recognizable stages before ending, described earlier, is worth revisiting here, because those stages don’t unfold identically across class, culture, or gender. A primary earner faces different pressure than a woman who is financially dependent. None of this determines what you should do. It means the “just decide already” pressure from friends rarely accounts for your actual position.

There’s also a specific cost to the children when a marriage dissolves, and the systemic lens requires naming it honestly rather than using it as a bargaining chip to stay. A conceptual framework for family art therapy with children going through divorce, described under PMID 41424063, uses narrative and attachment perspectives to describe how transitions are handled, not whether they happen, as the variable that most shapes outcomes for kids. In plain terms: how you and your partner handle what comes next matters more than the decision itself.

Naming these systemic pressures doesn’t remove them. But it lets you separate what’s actually your ambivalence from cultural noise dressed up as your own doubt. You’re allowed to want out of a marriage that looks fine from the outside. You’re allowed to keep trying even when people think you should have left already. Neither instinct needs their permission.

The Way Ahead

If you’re reading this from the chair closest to the door, here’s what the research and the clinical picture add up to, taken together. The work on perceived partner behavior under PMID 41809743 suggests that what moves the needle is real, felt behavior, not documented effort, which means performing diligence in sessions is largely wasted motion. The Healing Hearts Together protocol under PMID 40766010 and the veterans’ pilot trial under PMID 42175859 together suggest that unresolved limbo and separating both carry real health costs, so there’s no cost-free option hiding in the indecision. The Hold Me Tight findings under PMID 40545740 suggest that structured work can shift connection even under genuine ambivalence, so you don’t have to resolve your uncertainty before doing the work. And the family art therapy framework under PMID 41424063 suggests that if you do separate, the handling of that transition matters enormously for any children involved, reframing the goal from a clean ending to a well-handled one.

Taken together, this research points somewhere specific: honesty about where you actually stand, session by session, is not a risk to the process. It’s the only version with a real chance of producing something you can trust. Ambivalence tolerated and spoken tends to move. Ambivalence hidden and performed tends to calcify into false hope, or a decision you make for reasons you can’t name later.

Practically, this might mean naming your actual percentage out loud to your therapist, even when it feels disloyal. It might mean asking whether your process is couples therapy, which assumes mutual investment, or discernment work, which doesn’t. It might mean noticing when you’re building a case file instead of testing something real, the way attachment patterns explain an outgrown marriage long before either partner names what’s happening.

It might also mean getting curious about your own history with honesty in close relationships. If you learned young to manage other people’s feelings before naming your own, often visible now as codependency in driven women, that pattern doesn’t disappear the moment you sit down in a couples therapist’s office. It often intensifies there, because the audience is the person you married.

This is also, often, where good trauma-informed therapy for driven women earns its keep: not by telling you whether to stay or go, but by helping you tell your own signal from everyone else’s noise, including the noise inside your own head that sounds like certainty but is actually just exhaustion.

You don’t have to arrive at certainty before you’re allowed to be honest. You don’t have to know if you’re staying to stop performing that you already know you are. The most useful thing you can offer the process, and yourself, is the true number, today, even when it’s uncomfortable to say out loud.

Warmly, Annie.

Related Reading

• Rusbult, Caryl E., and Bram P. Buunk. “Commitment Processes in Close Relationships: An Interdependence Analysis.” Journal of Social and Personal Relationships, 1993.
• Kelley, Harold H., and John W. Thibaut. Interpersonal Relations: A Theory of Interdependence. Wiley, 1978.
• Levinger, George. “Marital Cohesiveness and Dissolution: An Integrative Review.” Journal of Marriage and Family, 1965.
• Teasdale, Sara. Love Songs. Macmillan, 1917.

Medical Disclaimer