What political breakups reveal about how we think, and what to do about it
By James A. Petersen PhD
The text message arrives on a Tuesday afternoon, and a thirty-year friendship ends in two sentences. The cousin who taught you to fish stops returning calls. The brother-in-law who stood up at your wedding is no longer welcome at the holiday table. Multiply those small severings by tens of millions, and you have a portrait of America in 2025.
A new study published in PNAS Nexus puts a number on what most of us already feel. Researchers Mertcan Güngör and Peter Ditto at the University of California, Irvine analyzed survey data from thousands of American adults and found that 37 percent of us have ended a relationship over politics. Friends bear the heaviest cost at 62 percent of these splits, followed by family members at 40 percent, coworkers at 29 percent, and romantic partners at 10 percent. More than half of those who experienced one breakup experienced more than one.
The trend is accelerating. Ninety-six percent of respondents who recalled a political breakup placed it in 2016 or later. The 2024 election produced a higher breakup rate than 2016 did, in roughly half the time.
We are not arguing about facts. We are sorting people.
The asymmetry, briefly
The data shows a partisan gap worth naming. 47 percent of Democrats reported a political breakup compared to 29 percent of Republicans, with independents at 39 percent. Among those who experienced a split, 66 percent of Democrats said they were the ones who ended it, compared to 27 percent of Republicans. That gap is real, and it is part of the story.
But the gap is not the whole story, and dwelling on it misses the more important point. Whichever side is doing more of the cutting in any given era, the underlying mechanism is the same. Identity hardens. Disagreement starts reading as betrayal. The other side stops being people with different views and becomes, in the language of the study, opponents whose motives are presumed selfish and whose positions are presumed extreme.
That mechanism is bipartisan. It is also the subject of my book.
Why facts stopped mattering
In The Facts Don’t Matter, I argue that the modern information environment has inverted the relationship between belief and evidence. We do not assemble facts and arrive at convictions. We arrive at convictions, usually through the tribe we belong to, and then we shop for facts that confirm them. The internet has made that shopping trip frictionless. Whatever you already believe, a thousand sources are waiting to tell you that you are right.
This is not new. Confirmation bias is older than the printing press. What is new is the speed, the scale, and the social cost of dissent. When everyone in your feed agrees with you, the person who disagrees is not just wrong. They are intruders.
The Güngör and Ditto study contains a finding that, for me, lands harder than the headline number. People who experienced political breakups did not just feel cold toward the opposing party. They overestimated how extreme the other side’s actual views were. They assumed selfish motives where none existed. And critically, the breakup itself appeared to make those distortions worse, not better.
Cutting ties does not resolve the disagreement. It amputates the corrective.
This is the heart of the matter. When we stop talking to people who see the world differently, we lose the only mechanism that keeps our own picture of reality honest. The friend who pushes back, the uncle who remembers when we were wrong before, the coworker who quietly notes that the data does not actually say what we claim it says; these are not obstacles to clear thinking. They are clear thinking. Remove them, and we drift into a hall of mirrors where our most confident beliefs go unchallenged and our worst instincts about the other side calcify into certainty.
What the breakups are really about
Notice what is missing from the study’s findings. There is no evidence that political breakups happen because one party finally understood the other’s policy positions and concluded they were untenable. The breakups are not the product of careful argument. They are the product of escalating emotional charge attached to political identity, until the cost of staying in the relationship feels higher than the cost of walking away.
This tracks with what I have spent the last several years writing about. In four decades of leading people, first as a Navy officer on nuclear submarines; then as an entrepreneur and executive; and now leading executive coaching and leadership training companies, I have watched the same pattern play out in boardrooms, wardrooms, and family rooms. When identity becomes fused with position, disagreement becomes existential. Every challenge to the position feels like an attack on the self. And the self defends itself the only way it knows how, by removing the threat.
In the military, this pattern is a luxury we cannot afford. A submarine crew at four hundred feet cannot have a political breakup. A combat unit cannot dismiss the soldier whose worldview offends them. High-performing teams under real pressure develop something most civilians have lost the muscle for; the ability to disagree hard, sometimes furiously, while remaining bound to the mission and to each other. That muscle is not a personality trait. It is a skill, and like any skill, it can be trained.
Five practices for staying in the room
If the study is right, and I believe it is, then the path forward is not better arguments. The path forward is better discipline about how we hold our own positions and engage with people who hold different ones. Five practices, drawn from the book and from years of coaching leaders through high-stakes disagreement:
What is at stake
The researchers warn that political breakups carry real costs to democracy. They are right. A republic depends on the assumption that citizens who disagree can still share a table, a town, a country. When that assumption fails at scale, the institutions built on top of it begin to fail too.
But I want to close on something smaller and more personal. The study counts relationships. Behind every number is a friendship that took twenty years to build and twenty minutes to end. A grandparent who will not meet a grandchild. A wedding invitation is never sent. A funeral attended by half the people who should have been there.
These are losses that do not show up in the polls and cannot be undone by an election. They are the texture of an ordinary life, and we are forfeiting them at a rate previous generations would not have recognized.
The opposite of polarization is not agreement. It is the willingness to stay.
The facts will not save us. They never could were going to. What might save us is the older, harder discipline of treating the people who disagree with us as people, holding our own certainties more loosely than feels comfortable, and refusing the easy exit that walking away always offers.
That is the argument of The Facts Don’t Matter, and it is the argument I will keep making, because the study published last week tells me it has never been more urgent.
About the book
The Facts Don’t Matter (2025) examines why the modern information environment has weakened our capacity for honest disagreement, and what individuals, leaders, and institutions can do to recover it. It is the fourth book in a six-book leadership series by James A. Petersen PhD, retired U.S. Navy Captain, executive coach, and Adjunct Professor at The American College of Financial Services.
Study cited: Güngör, M. and Ditto, P. H. (2026). Political breakups: Interpersonal consequences of polarization. PNAS Nexus, 5(5), pgag067. https://doi.org/10.1093/pnasnexus/pgag067. Survey data collected April 2025; full datasets available via OSF and the Roper Center for Public Opinion Research.