Coalitional Psychology: A Feature, Not a Bug — And That's the Problem

James Madison didn't have the vocabulary of evolutionary psychology. But when he wrote in Federalist No. 10 that the causes of faction are "sown in the nature of man," he was making precisely the claim that evolutionary science has since confirmed: the tendency to organize into competing coalitions, and to subordinate principle to coalition membership, is not a correctable flaw in human character. It is the architecture. Washington said it differently in his Farewell Address, warning that the "spirit of party" kindles animosity, distorts perception, and ultimately serves as "a fire not to be quenched." Neither man thought you could educate it away. Which is why they spent their political lives designing structural friction against it rather than appealing to virtue to overcome it.

We have largely abandoned that project. And the consequences are visible everywhere, if you're willing to look at them clearly.

Coalitional psychology is the evolved tendency to track group membership and calibrate behavior--including moral judgment--according to what the group requires. It was adaptive for most of human history because exclusion from the group was, in the ancestral environment, often a death sentence. The psychology that survived is therefore one that monitors social standing obsessively, conforms to coalition norms under pressure, and applies moral standards with far greater rigor to outsiders than to members. As much as it appears so, this isn't weakness or hypocrisy in the ordinary sense. It is a deeply embedded survival algorithm operating in a world it was not designed for.

The political expression of this is what Madison called faction and what we now experience as the two-party system at its worst: not a contest of principles but a contest of coalitional loyalties, in which the primary question about any given issue is not "is this true?" or "is this right?" but "which side does this help?" Once that becomes the operating logic, stated principles become instruments of coalition warfare rather than genuine commitments. They are deployed when useful and suspended when inconvenient, and the people doing the deploying often don't experience this as dishonesty. From inside the coalitional frame, it feels like realism.

This explains a range of phenomena that otherwise seem like blatant hypocrisy but are actually something more systematic and more intellectually interesting (and valuable to understand). This is not to condone the behaviors, but not to misake the pathways required to avoiding them.

The Epstein network is one of the most clarifying recent examples, precisely because it should have been, on the stated principles of almost everyone involved, an unambiguous case. The documented conduct (the systematic sexual exploitation of girls as young as twelve and thirteen by wealthy and powerful men) is exactly what organized feminism, progressive institutions, mainstream media, and most conservative family-values rhetoric all claim to exist in order to oppose. The names in the flight logs include political donors, conservative and liberal icons, and figures central to the institutional infrastructure. The names in the broader social network touch figures across the political spectrum. With a few brave exceptions, the response from virtually all of those institutions has been, at best, muted management.

The near-universal tolerance of deny-and-delay is itself evidence. If the silence were confined to one ideological camp it might be explained by simple partisanship. The fact that it crosses party lines, institutional affiliations, and stated ideological commitments points to something operating at a deeper level: the network's connections ran through enough of the broader elite social world that pursuing it fully would damage almost every major coalition simultaneously. The bipartisan instinct has therefore been the same: manage, delay, minimize, and rely on the public's historical tendency to move on.

A revealing specific silence is that of the #MeToo movement, which had demonstrated real institutional reach and whose explicit mission was precisely this kind of accountability. It never seriously extended to the Epstein network, and the most parsimonious explanation is that too many figures central to the progressive and media coalition were implicated for that thread to be pulled without risk of unraveling something much larger. The stated principle was real. The coalitional constraint appears to be stronger.

The people driving accountability have been, almost without exception, coalitionally independent: journalists without institutional backing, commentators whose skepticism of elite institutions crosses party lines, political actors whose enemies happen to overlap with the network. What unifies them is not ideology but the absence of the institutional relationships that generate silence.

What I've come to believe is that Epstein is not an extreme case that reveals how bad things can get. It is a window into the normal: what happens routinely when status hierarchies are steep enough and institutional friction is weak enough. The rock music culture of the 1970s produced the a similar pattern at a smaller scale. Jimmy Page openly dated Lori Maddox when she was fourteen years old. This was not hidden. It was known, tolerated, and in certain circles celebrated, because the moral gravity field around men at the apex of a status hierarchy is strong enough to suspend ordinary moral evaluation in the people around them. (You are certainly thinking of other promient examples.) The underlying psychology is identical. What varies is the scale of the hierarchy, the degree of institutional protection, and the presence or absence of people with sufficient independence to apply friction.

The same mechanism operates with equal clarity far from the music world and Hollywood. When senior military commanders invoke Jesus in official contexts, the men in that room (with explicit professional obligations to constitutional limits on state religion) don't seem to experience themselves as violating anything. They're performing coalitional solidarity in the highest-stakes hierarchical environment in American life. The military is one of the most explicitly coalitional institutions humans have ever built, with its own identity markers, its own internal status hierarchy, and an explicit survival dependence on group cohesion. Evangelical Christianity is itself a powerful coalition with its own boundary signals and internal logic. When those two coalitions overlap heavily, as they do in significant portions of the American officer corps, the result is a reinforced identity in which invoking Jesus in a military context doesn't seem to be processed as a category error. It's processed as affirmation: we are the right kind of people, fighting for the right kind of cause. The constitutional principle and the coalitional signal run on separate tracks, and the coalitional signal wins. It nearly always does, across the political spectrum, in institutions of every ideological description.

The Founders' structural response was built on a precise understanding of this problem. You cannot fix coalitional capture with better people, because the psychology will always be there. Steep enough hierarchies will always activate the full deference and protection response. The only available intervention is structural: design institutions that prevent any single status node from becoming powerful enough to suspend independent judgment in the people around it.

Regulatory capture is the proof of concept for why this matters and why structural independence cannot be assumed. An agency created to provide friction against an industry (i.e., to hold it accountable and to apply external scrutiny) gradually gets staffed by people from that industry, funded by its political allies, and socially embedded in its professional world. The coalitional logic does the rest without any explicit conspiracy required. The SEC, the FDA, and the FAA--these are not primarily stories of individual corruption. They are stories of coalitional psychology dissolving structural independence over time. The institution doesn't fail all at once. It drifts, because the people inside it are running the same status-tracking, coalition-maintenance algorithm everyone else is running, and the industry they regulate is the most powerful status node in their environment. No one decides to stop doing their job. The coalitional gravity bends judgment incrementally until the institution serves the interest it was designed to check.

This is precisely what the Founders were designing against. Not bad actors, but the predictable operation of ordinary human psychology in proximity to power. Separation of powers, an independent judiciary, a free press, and federalism--these are all attempts to maintain sources of scrutiny structurally insulated from the coalitional capture that will otherwise inevitably occur. The degree to which those structures have eroded is the degree to which the normal reasserts itself.

Pick any significant political issue today and the coalitional fracture is immediately visible. People you would expect to criticize a position based on their stated principles do the opposite, and do so with apparent conviction. Fiscal conservatives unbothered by deficit spending under their own party. Civil libertarians comfortable with the suspension of due process when the accused belongs to the other coalition. Progressive institutions silent on the treatment of women in certain cultures because criticizing them required criticizing coalition political relationships. The stated principle didn't change. The coalitional calculus did.

What may be historically distinctive about our moment is not the fracture itself (every era of intense factional conflict has produced it) but its pervasive reach. When the primary inputs into political identity flow through media ecosystems engineered to maximize coalitional intensity, the signal saturates environments that used to provide buffer. The dinner table. The friendship group. The family reunion. Most of us have experienced this directly: relationships strained or broken not over personal conduct but over coalitional allegiance, and the strange accompanying inability to call out bad behavior in one's own coalition regardless of how visible it is. The Founders could not have anticipated a technology infrastructure specifically optimized to keep people in a state of continuous coalitional alarm, and the structural friction they designed was not built to withstand it.

I'll confess that arriving at this framework came with some personal relief. For a long time I experienced genuine confusion (and more than occasional frustration) at what appeared to be moral discontinuity in people and institutions I was trying to honestly understand. I kept searching for the unified intellectual framework that would reconcile their stated principles with their actual behavior, assuming that one must exist and that I was simply missing it. The frustration came from that assumption. If people are operating from coherent ethical positions and their behavior contradicts those positions, you have to conclude either that they're lying or that they're failing. There is an emotional appeal to either conclusion, but neither turns out to be the full story.

But let's be clear: the anthropological frame doesn't require you to abandon moral judgment. It relocates it. The question shifts from "why is this person being hypocritical" to "what are the conditions we've allowed that produce this behavior, and what do we actually need to change?" Those are questions that can actually be engaged. The military commanders invoking Jesus, the institutions silent on Epstein, the friends who cannot criticize their own coalition--they are not, in most cases, consciously choosing principle violation. They are running an ancient algorithm in a modern environment, and the algorithm is working exactly as designed. Understanding that converts frustration into something more useful: a clear-eyed assessment of the structural conditions we would have to change for the behavior to change.

That assessment is not comfortable. It suggests that the problem is not solvable, as much as we assume it is, by a better idea, or by electing better people, or by better education, or by more forceful moral argument directed at individuals whose positions are not, at their root, intellectual. It's solvable by remembering that agreed-upon cultural, political, and economic boundaries need to be set to discourage coalitional capture and abuse. Madison knew this. Washington knew this. The question they left us, and that we have done a poor job of answering, is whether we are serious enough about the problem to rebuild and maintain the structural boundaries and friction that are the only things that have ever worked against it.

One can argue that the agreement on these boundaries can only come after a crisis, when the consequences have been so severe that the importance of the boundaries, and the broad impact of not having them, has become obvious. Perhaps so.