Is It Me or Them? A Framework for Honest Self-Assessment

It's 1:40 in the morning and you're typing the question into a search bar because there's no one else you can ask without it getting back to them. Is it me? Am I the problem?

Here's what you'll usually find: either a chorus telling you it's definitely them (run, they're toxic), or a self-improvement sermon implying it's definitely you (do the work, look inward). Both answers sell well, because both flatter someone. Neither is honest.

The honest answer, for most recurring conflicts between two basically decent people, is: both of you, asymmetrically. Not 50/50. Almost never 100/0. The real question isn't "which of us is the problem?" It's "what's my actual share?" That question, unlike the first one, can be worked on. This article is a set of tools for answering it fairly.

First, accept that your gut can't be trusted here

Not because you're a bad person. Because you're a person.

Psychologists have documented for decades that we don't explain our own behavior and other people's behavior with the same instrument. In their classic review of what's known as the self-serving attributional bias, Dale Miller and Michael Ross described the pattern: we tend to credit our successes to ourselves and hand our failures to circumstance. I snapped because I was exhausted and you kept pushing. You snapped because that's how you are.

A related tilt, the actor-observer asymmetry proposed by Edward Jones and Richard Nisbett, holds that actors explain their behavior through the situation while observers pin it on personality. When Bertram Malle put the hypothesis to a meta-analytic test across decades of studies in 2006, the legend turned out to be more modest than the textbooks claimed, but with a crucial wrinkle: the asymmetry showed up most clearly for negative events. The unflattering stuff. Which is to say: the bias is at its strongest precisely in the moments a fight is made of.

Now put two people in a kitchen, both running this software in opposite directions. You experience your sharp tone as what the situation forced; they experience it as who you are. They experience their withdrawal as self-protection; you experience it as proof they've stopped caring. Nobody is lying. You are both using instruments that read wrong in your own favor.

You are the unreliable narrator of your own fights. So are they. Start there.

The framework: four tests

Since instinct is rigged, use procedure. Run your recurring conflict through these four tests. Write your answers down, because the biases above are much better at bending memories than bending ink.

1. The pattern test

Does this happen with everyone, or just with them? If three exes, a sibling, and a coworker have all independently told you that you disappear when things get hard, the common denominator is standing in your shoes. If a behavior shows up only in this relationship, meaning you're not like this with anyone else and weren't before, that points at the dynamic between you, or at what you're responding to. Run the test in both directions: their pattern across relationships counts too. One person's lifelong pattern meeting one person's first occurrence is real information about asymmetry.

2. The reversal test

If they did exactly what you did, what would you call it? You "needed space." If they'd left mid-sentence, would you call it stonewalling? You were "just being honest." If that identical sentence had been aimed at you, in that tone, would it have been cruelty? The rule: you must use the same label for the same behavior, whichever direction it travels. If the label changes when the direction changes, you've found your bias, live and in person.

3. The tape-recorder test

What would a stranger hear? Imagine last night's argument as an audio file played to a neutral stranger, someone who gets no backstory, no context about how much you've endured, no access to what you meant. Just the words and the tone. What would they say each of you did? Try writing the fight out as a screenplay: only dialogue and observable action, no adverbs of intent, no "(justifiably)". Most people discover their transcript contains lines they'd wince at, and also lines they genuinely don't need to apologize for. Both discoveries are the point.

4. The outcome test

Judge your strategy by its results, not its intentions. You keep explaining, in ever-better words, why you're hurt, and it reliably produces the same terrible night. At some point, "but I'm communicating!" stops being a defense. A behavior that predictably makes things worse is part of the problem regardless of how justified it feels from inside. This test cuts both ways too: if you've genuinely changed an input (softer openings, real apologies, actual timeouts) and the output never changes, that's data about what isn't yours to fix.

The question under the question

One more reframe, and it might be the most liberating thing in this article. John Gottman's observational research found that the majority of the conflicts couples have are perpetual: recurring disagreements rooted in lasting differences of personality and need, not solvable disputes with a right answer waiting at the end. The spender and the saver. The planner and the improviser. The one who processes out loud and the one who needs silence first.

If your recurring fight is one of these, and it likely is, then "which of us is the problem?" is a category error. Neither of you is. The difference is permanent; what's optional is the pattern you've built around it: the contempt, the chasing, the shutting down, the score-keeping. You can't fix being different people. You can absolutely fix how you fight about it.

When it really is mostly them

Fairness has two edges, so let's say this part plainly. Sometimes the honest audit comes back lopsided. If you're met with contempt (mockery, sneering, being talked about like you're lesser) that's not a two-sided pattern; that's the most corrosive behavior in Gottman's research, and its presence is not your fault. If every repair you attempt gets slapped away, if you've genuinely changed your inputs for months and nothing you do alters the weather, if you're only ever offered blame and never offered curiosity, you can stop auditing yourself for the answer. Asking "is it me?" endlessly, against that kind of evidence, isn't humility anymore. It's a way of staying.

And a note in the other direction: if you read this whole article looking for ammunition, skimming each test for how it convicts them, that too is a result.

What to do with your share

Once you have an honest estimate, say it turns out you're maybe a third of this particular mess, the move is not guilt. Guilt is just self-focus wearing a sad costume. The move is ownership of your slice, out loud, without invoicing them for theirs in the same breath. "I've noticed I go silent and leave you alone in arguments. I'm working on staying." Full stop. No "and you..."

Your share is the only part of the problem you control. Which means it's also the only lever you actually hold. That's not a burden. That's the good news.

This article is for self-reflection and education, not a substitute for professional advice or couples therapy.