Why Every Road Leads Back to Ideology

There is a strange symmetry in how ideology survives contact with reality. When an ideology succeeds, it takes the win as proof of its truth: see, it worked, we were right all along. When an ideology fails, it takes the loss as proof it wasn't applied properly: it didn't work because we didn't go far enough, weren't pure enough, weren't rigorous enough in our commitment. Success confirms the ideology. Failure calls for more of it. There is no outcome, in principle, that an ideology cannot metabolise into evidence for itself.

You can watch this mechanism operate cleanly in any field whose institutional survival depends on the size of the problem it exists to solve. An academic discipline organised primarily around advocacy for a cause faces a genuine structural bind: its stated mission is to make the problem smaller, but its funding, staffing, and public relevance depend on the problem staying large — or, ideally, on convincing everyone it is larger than previously understood. This creates a powerful disincentive to ever declare victory. If the problem were shown to be shrinking, so would the case for the discipline's budget, its centres, its professorships, its claim on public attention. So instead of success being reported as success, it tends to be reported as evidence of how much further there is to go, how entrenched the problem really is beneath the surface, how urgently more resourcing and more authority are needed to finish the job. Whether or not the underlying problem is real — and often it is — the incentive structure guarantees that the field will rarely, if ever, be the one to tell you it's been solved. A tribe organised around a fight has very little use for a peace it would have to admit it won.

This isn't a flaw in any particular ideology. It's the definition of one. A belief system stops being a hypothesis about the world and becomes an ideology exactly at the point where it can no longer lose. And once a belief system can no longer lose, it has quietly stopped being about the world at all. It has become about itself.

This matters more than it sounds like it should, because the alternative to ideology is not "having no strong views." The alternative is something harder and rarer: self-skeptical discernment — holding a position while remaining genuinely willing to find out you were wrong, adjusting not just your tactics but your actual beliefs when the evidence says so.

Almost nobody does this by default, and there's a good reason why, one that has nothing to do with stupidity or bad faith. Beliefs held as part of a group identity aren't processed the way ordinary facts are. When a belief is fused with belonging — with who your people are, which side you're on, who you'd be betraying by changing your mind — updating it stops feeling like learning and starts feeling like defection. Researchers call this identity-protective cognition: the finding that people don't process threatening evidence as neutral information, they process it as a threat to their standing in the group whose worldview that belief expresses. And here's the part that should worry anyone who assumes education is the fix: some of the strongest evidence of this effect shows up not in the least reflective people, but in the ones who think most carefully about their positions. More cognitive horsepower doesn't reliably make someone more accurate under identity pressure. It can just make them a more sophisticated builder of reasons why the group was right all along.

This is why ideology is so remarkably stable, and why it is the enemy of the thing I actually think matters — not correctness, but correctability. A person or an institution that has outsourced its judgement to an ideology doesn't need to be right. It only needs the ideology to remain unfalsifiable, and ideologies are, almost without exception, built to be exactly that. They come with built-in explanations for every disconfirming case before the case has even happened: the failure was sabotage, or insufficient commitment, or bad implementation, or the wrong people in charge, or not enough time. Notice that these explanations work identically regardless of what actually happened. That is not a coincidence. That is the design.

This is where the argument I've been making elsewhere — that not all values are equally valuable — actually earns its keep. If you accept that some values genuinely serve human flourishing better than others, you are immediately faced with an uncomfortable follow-up question: how would you know if you were wrong about the values you hold? An ideology answers that question by making sure you never have to find out. Self-skeptical discernment answers it by building the capacity to find out directly into how you hold your beliefs in the first place — treating your own current position as a working hypothesis rather than a flag to defend.

This is not the same as relativism, and it's worth being precise about the difference, because they get confused constantly. Relativism says no position can be evaluated against any standard, so nothing is really wrong. Self-skeptical discernment says the opposite: positions absolutely can be evaluated, which is exactly why you have to stay open to evidence that yours might not hold up. Relativism protects you from ever being wrong by making "wrong" meaningless. Discernment exposes you to being wrong on purpose, because that exposure is the only mechanism by which anyone actually gets closer to right. The scientific method, for example, has led to unprecedented good in the world (see here for a website dedicated to showcasing this, contra to the daily news, which is reliably miserable). The scientific method is precisely the systematic search for reasons why one's ideas might be wrong; disciplined humilty in action.

Outside of science, in day to day life, this looks like a small number of habits that are individually unremarkable and collectively rare. Naming, in advance, what evidence would actually change your mind — not vaguely, but specifically, before you need it, while you have no stake in the outcome yet. Noticing when an explanation for failure is suspiciously reusable — when it would apply just as well to any other outcome as to the one that actually happened. Treating the loudest, most certain voices on your own side with the same scrutiny you automatically apply to the other one. And perhaps hardest of all: being willing to say, in public, to your own people, "we were wrong about that" — not as a tactical retreat, but because it happened to be true.

None of this requires abandoning conviction, where warranted. It requires the much harder discipline of holding conviction the way a scientist holds a good hypothesis rather than the way a fan holds a team: with real commitment, but commitment that is answerable to the world rather than immune to it.

I'll double down, to conclude: not all values are equally valuable. Some genuinely serve human flourishing better than others in empirically verifiable ways. But the only honest way to act on that claim is to remain permanently willing to discover you've got your "personal values ranking" wrong — and ideology, whatever banner it's flying, exists precisely to make sure that discovery never has to happen.

If you find yourself nodding along thinking about how your political outgroup engages in this evidence-immune ideological thinking all the time, then I fear you might be especially vulnerable to missing the point here.