Why some conversations now feel impossible
Part one of a trilology
I’ve lost count of the number of people who’ve said some version of this to me lately. “I don’t know how to talk to them anymore.”
Not because the topic is complex. Not because emotions are involved. But because something more basic has collapsed, the shared assumption that conversation is about learning something together.
In therapy rooms, classrooms, and online spaces, I’m seeing the same pattern repeat. People arrive with conclusions already welded to their identity. Evidence is treated not as information but as an insult. Questions are heard as attacks. Curiosity is replaced with certainty.
When that happens, what looks like disagreement is actually something else entirely.
It’s a breakdown in shared epistemology, a failure of agreement about how we decide what is true, what counts as evidence, and whether beliefs are allowed to change.
Once that ground goes, conversation stops being conversation.
When belief becomes identity
There’s a particular kind of exhaustion that comes from trying to reason with people who aren’t actually in dialogue with you. They’re arguing at you, not with you, and that isn’t a personality quirk. It’s a psychological state.
Across countless interactions, a familiar pattern shows up. People defend a belief with no interest in evidence that contradicts it. Not because they’re unintelligent. Not because they’re malicious. But because their belief has become fused with their sense of self.
When belief becomes identity, evidence stops being informative. It becomes threatening.
Psychology describes this as cognitive fusion, the experience of thoughts and beliefs as literal truths rather than provisional interpretations. When someone is cognitively fused with a belief, challenging it feels like challenging them.
Add motivated reasoning to the mix, the well-documented tendency to selectively interpret information in ways that protect existing beliefs, and belief-updating becomes psychologically dangerous rather than intellectually routine.
This is why some conversations feel impossible.
It’s not that two minds can’t meet. It’s that one side has already decided the conclusion, and anything that doesn’t support it is processed as a threat to identity or belonging. Our nervous systems evolved to prioritise social survival over accuracy. In modern life, that wiring hasn’t changed, but the stakes have shifted from physical survival to ideological allegiance.
So we end up arguing not about ideas, but about who we are allowed to be.
That’s also why these interactions are so draining. You’re trying to engage in dialogue, which requires openness and revision. The other person is engaged in defence. Those are fundamentally different activities.
True conversation requires at least three conditions.
A willingness to consider evidence.
A separation between identity and belief.
And the humility to revise one’s position when warranted.
Remove any one of those, and what remains is not dialogue but performance.
When someone says, “I don’t know how to talk to this person anymore,” they’re often describing the loss of shared epistemic ground. Not disagreement. Not conflict. But the absence of the basic conditions that make conversation meaningful in the first place.
What to do instead
When real dialogue is no longer available, the task quietly changes.
The goal stops being persuasion, clarity, or being understood. The goal becomes self-regulation and discernment.
The first thing to notice is where the effort is going. If you’re working harder to stay calm, careful, and precise than the other person is to stay curious, you’re not in a conversation. You’re in an asymmetrical emotional labour arrangement.
That’s useful information.
Instead of asking, “How do I explain this better?” it’s often more grounding to ask, “Is this person currently capable of dialogue?” Not as a judgement. As a diagnostic.
When the answer is no, there are still psychologically healthy options available.
You can shift from debate to boundary.
You can disengage without contempt.
You can stop offering evidence to someone who has already decided what it means.
You can conserve energy rather than spend it trying to earn permission to be heard.
For neurodivergent people especially, this matters. Many autistic and ADHD adults are trained, implicitly or explicitly, to over-explain, to justify their reasoning, and to stay in conversations long past the point where they are reciprocal. Knowing when to stop is not avoidance. It’s nervous system literacy.
Sometimes the most stabilising move is to name reality internally, even if you never say it aloud.
“This isn’t a space where learning is happening.”
“This isn’t mutual.”
“This isn’t safe for my energy.”
Once you see that clearly, the pressure to perform clarity dissolves.
Knowing when conversation is no longer the task
There’s an important psychological shift that happens when you stop trying to rescue the conversation and start respecting your own limits.
Some people are listening to learn. Others are listening to defend. The difference matters more than the topic under discussion.
If the aim is understanding, growth, or shared meaning-making, conversation only works when both parties are willing to be changed by what they hear. Without that, no amount of patience, evidence, or careful phrasing will create connection. You’ll simply exhaust yourself trying to meet someone who isn’t standing on the same ground.
Sometimes the most psychologically healthy response isn’t to push harder or stay longer.
It’s to recognise that dialogue requires consent from both sides.
And when that consent isn’t there, stepping back isn’t failure.
It’s clarity.
Part two of this trilogy: When conversation disappears, grief follows
Part three of this trilogy: Rebuilding identity without consensus
References
American Psychological Association. (2023). Cognitive biases. https://www.apa.org/monitor/nov01/biases
Festinger, L. (1957). A theory of cognitive dissonance. Stanford University Press.
Haidt, J. (2012). The righteous mind: Why good people are divided by politics and religion. Pantheon Books.
Kahan, D. M. (2017). Misconceptions, misinformation, and the logic of identity-protective cognition. Yale Law School Public Law Research Paper, No. 605.
Stanovich, K. E., West, R. F., & Toplak, M. E. (2013). Myside bias, rational thinking, and intelligence. Current Directions in Psychological Science, 22(4), 259–264.
Wikipedia contributors. (2025). Motivated reasoning. https://en.wikipedia.org/wiki/Motivated_reasoning
Wikipedia contributors. (2025). Cognitive fusion. https://en.wikipedia.org/wiki/Cognitive_fusion