7 Questions to Ask Yourself Before Deciding

The questions in this piece will not decide for you. Nothing can. What they can do is surface information you may already have and have not yet let yourself see. Each question is anchored in a specific research tradition, and each is sequenced deliberately — the easier ones first, the harder ones last. Most people, by the time they reach the final two, find that their own answers are pointing in a direction. Whether they are ready to act on it is a separate question.

How to use the questions

Read them slowly. Do not try to answer them all in one sitting. Consider writing the answers down privately — the research on decision-making consistently finds that written honesty produces more accurate self-reports than spoken ones, particularly on topics the mind has incentives to avoid.

There is no score. The questions are not designed to produce a number. They are designed to make the shape of your actual situation visible, so that whatever decision you eventually make is informed by more than the loudest emotion of whatever day you make it on.

1. When did you last feel that your partner was glad you existed?

John Gottman's research identified fondness and admiration as one of the most reliable markers of marriages that last. In his longitudinal samples, couples who retained fondness through hard periods recovered; couples who lost it generally did not. The practical test is memory: how recently, and how vividly, can you recall your partner doing or saying something that made you feel specifically seen — not useful, not tolerated, but seen.

If the answer is "recently and often," that is data. If the answer is "I have to reach back years," that is also data. It does not decide anything. It does locate your marriage on the fondness axis of the research.

2. What happens when you reach for repair?

Gottman's later work foregrounded what he called repair attempts — the small bids during or after conflict that signal, in whatever form, "I want to come back to you." A touch, a joke, a specific apology, a softening of voice. His finding was that stable couples accept repair attempts more often than they reject them; couples headed for divorce miss, rebuff, or do not offer them at all.

Notice, over a week or two, what happens when you try to repair — and what happens when your partner does. You are not looking for perfection. You are looking for whether the attempt lands.

3. Which of the Four Horsemen live in your marriage?

Named once more because they are the highest-leverage diagnostic in the field: criticism, contempt, defensiveness, stonewalling. The presence of one, occasionally, in an otherwise warm marriage, is not the signal. The signal is whether they are habitual, whether they are normal to both of you, and whether contempt is among them.

Contempt — eye-rolling, sarcasm, mockery, name-calling — is the single strongest divorce predictor in Gottman's observational data. It is also the most normalized of the four in many long-term unhappy marriages. The honest version of this question is: has contempt become boring?

4. If nothing changes in the next five years, what does your life look like?

This question borrows from decision-theory work on projection under uncertainty. The future, by definition, is unknowable. But people asking this question typically are not asking it honestly of the present. The default assumption is that the current trouble is a phase. The question pushes on that assumption.

If you genuinely believe your marriage is in a hard season that will pass — because you have specific evidence of what will shift, or a plan, or a recent turn — the answer is usually workable: tolerable, improving, recognizably your life. If you cannot produce a concrete mechanism for how things will be different, and the projection simply extends the present, notice your reaction. People often encounter, here, the grief they have been deferring.

5. What are the commitments, and what are the investments?

From Caryl Rusbult's Investment Model: separate the question of what makes you committed from the question of what you are committed to. Investments — years, children, shared finances, intertwined identity, social networks — are the sunk cost side of commitment. They are real. They are also not the marriage itself.

If you try to answer "why am I still here?" and the answer is dominated by investments ("we've been together twenty years," "the kids," "the house," "I can't imagine starting over"), the Investment Model describes what you are experiencing: high commitment driven by high investment, potentially decoupled from satisfaction. That is not a verdict. It is a diagnostic. It tells you what is currently holding the marriage together and how robust that substrate actually is.

6. If your child lived in this marriage as an adult, what would you tell them?

This is a perspective-switch question, drawing indirectly on the developmental research (Hetherington, Amato) on what children actually learn from the marriages they grow up watching. The uncomfortable finding in that literature is that children internalize their parents' relationship as a template for their own adult partnerships — not always, but reliably enough to matter.

The question is therefore: imagine, ten years from now, a grown child of yours is in this marriage. Same dynamics. Same dominant tone. Same frequency of warmth. Same frequency of contempt. What would you tell them? If the answer is "I would tell them this is what love looks like," that is informative. If the answer is something you would not let yourself say out loud to your own children, that is also informative.

7. What are you avoiding?

Decision researchers consistently find that the topics people refuse to let themselves think about are the most informative ones. Not the complaints you rehearse, but the conversations you do not have. Not the fantasies you enjoy, but the calculations you suppress. The specific moment with your child you cannot picture. The specific conversation with a therapist you cannot bring yourself to schedule. The specific truth about your own behavior that you have been careful not to examine.

Ambivalence almost always includes an avoidance pattern. The avoidance is not a moral failing; it is a protective mechanism that keeps the decision from accelerating before you are ready. But the contents of the avoidance — the specific things you cannot look at — are usually the closest thing the mind offers to a verdict. The question is whether you are willing to look.

After the questions

When you reach the end of a framework like this, two reactions are common. The first is a pull to act quickly — to resolve the dissonance by doing something. The research on decision-making consistently argues against this. The purpose of a framework is to make the situation legible, not to manufacture a deadline.

The second reaction, common in marriages that are recoverable, is a renewed willingness to try specific interventions — couples therapy, a concrete conversation that has been deferred, an explicit repair of a specific rupture. The research on Emotionally Focused Therapy (Sue Johnson) and the Gottman Method is clearest for exactly this situation: couples who recognize their own patterns and are willing to work on them have meaningful recovery rates, especially when contempt is not already entrenched.

The third reaction, common in marriages the research would describe as in decay, is the quiet settling of a decision that has been forming for a long time. People frequently describe this moment as "nothing new, just permission to see what was already there." That is, in fact, what a question framework built on decades of research is designed to provide — not new information, but permission to notice information that has been accumulating for a long time.

The bottom line

No research framework can tell you whether to stay or leave. What a framework like this one can do is slow the decision down, locate your marriage on several well-studied axes, and make sure that whichever direction you eventually move in, you move in it with your eyes open and the evidence — including your own answers — honestly in view.

Common questions