I have a list. You probably do too. Ideas so obviously correct that debate feels like obstacle rather than process. If I could just implement them, skip the stakeholders and the objections and the committees that meet to schedule future meetings, things would be better. I'm not imagining myself a dictator. I'm imagining myself effective.

This is the seed of something. Not evil, not ideology. Just confidence meeting friction. The certainty that I see clearly and others obstruct.

The uncomfortable truth is that every authoritarian system begins here. Not with a villain twirling his mustache but with someone who believed, genuinely, that they knew what needed to be done and that the machinery of consent was simply in the way.

China built 27,000 miles of high-speed rail while California spent fifteen years and $100 billion failing to complete a single line. Shanghai raises towers in months that would take American cities a decade to permit. When COVID arrived, China imposed order (whatever we think of its methods and whatever we now know about its origins) while democracies descended into tribal performance.

The person who finds this impressive isn't a fascist. They're someone who has watched democratic institutions calcify into vetocracies, where every stakeholder can block but none can build. They're asking a reasonable question: why can't we just get things done?

This is the honest version of the seduction, and dismissing it as authoritarianism-in-disguise doesn't help anyone. The appeal is real. The trains really do run. The buildings really do rise. The mess of democratic deliberation really is absent.

So what's the problem?

The problem is epistemic before it's moral.

Friedrich Hayek, writing in the shadow of both fascism and Soviet communism, identified something that applies to all centrally planned systems regardless of ideology. He called it the knowledge problem. The insight is simple but its implications are vast: no central authority can know what it needs to know to plan effectively.

Consider what a price actually is. When lumber costs more after a hurricane, that single number encodes information about damage assessments, supply chain disruptions, transportation bottlenecks, and urgency of need. Thousands of actors who never communicate directly can coordinate their behavior through this signal. The contractor in an undamaged county sees the price rise and delays his non-urgent project. The sawmill in another state increases production. The homeowner weighs repair against replacement. None of them needs to know why the price moved. The information is already in the number.

No planning bureau can replicate this. Not because bureaucrats are stupid or corrupt (though some are, as in any human institution) but because the knowledge doesn't exist in any one place until the system creates it. It emerges from millions of actors making billions of decisions, each drawing on local information that is, by definition, illegible to central authority.

The factory floor worker knows which safety protocols are theater and which prevent disasters. The shop owner in Chengdu knows her customers' actual needs, not the needs the planning commission imagines. The engineer knows which corner was cut to meet the quota. This is what Hayek meant by "dispersed knowledge." You can compel its disclosure, but you cannot compel the insights that only emerge when people are free to act on what they know.

This reframes the debate entirely. The argument against central planning isn't primarily about freedom as an abstract value (though it's that too). It's about information processing. How does a complex society aggregate what it needs to know? The answer, uncomfortable as it may be, is that you can't plan your way to it. You can only create conditions where it emerges.

But surely good intentions help? Surely wise planners with the right values can avoid the failures of cruder regimes?

Thomas Sowell identified two fundamentally different visions of human nature that shape how people think about this question. The constrained vision holds that human nature is fixed and flawed, so you design systems that work despite human limitations: checks, balances, distributed authority, feedback loops. The unconstrained vision holds that human nature is perfectible, so the right people with the right intentions can engineer better outcomes than messy, unplanned processes produce.

The China model assumes the unconstrained vision at scale. If we get the right technocrats, remove the friction of democratic debate, and align incentives correctly, we can plan flourishing.

The knowledge problem doesn't care. The planner who genuinely wants to help the Uyghur farmer still cannot know what that farmer knows about his land, his family's needs, his community's dynamics. The planner can only see what's legible to the state. And legibility requires simplification. And simplification erases.

This is why authoritarian efficiency and human rights abuses aren't separate phenomena that happen to coexist. They're structurally linked. The Uyghurs aren't being erased because the CCP is uniquely evil. They're being erased because their way of life is illegible to the planning apparatus. They don't fit the grid. Their religion, their language, their land use patterns, their family structures: all of it resists the standardization that central planning requires. The Tibetans face the same pressure. So do the house churches. So did the Falun Gong.

When you optimize for legibility, you eliminate what doesn't fit. Not always through violence (though often enough). Sometimes through relocation, reeducation, cultural suppression, economic strangulation. The methods vary. The logic is constant.

And then there's brittleness.

Systems that punish bad news eventually stop receiving it. This is how the Great Leap Forward killed tens of millions: local officials reported grain surpluses to meet quotas while people starved in the fields. The information the system needed most (we are failing, people are dying, the policy isn't working) was precisely the information it made dangerous to transmit.

Chernobyl followed the same pattern. So did COVID's initial emergence in Wuhan, where doctors who tried to warn of a novel pathogen were silenced as rumor-mongers. By the time the information reached decision-makers who could act, weeks had been lost.

Authoritarian efficiency trades resilience for speed. It works until it doesn't. Then it shatters. The feedback loops that democracies build in (a free press, opposition parties, independent courts, the capacity to vote out failure) are irritants to efficiency. They're also how complex systems learn. Remove them and you get a system that can execute but cannot adapt, cannot hear what it needs to hear, cannot correct course until correction is catastrophic.

But there's a softer version of this temptation. Not China's hard authoritarianism but Scandinavia's gentle socialism. High taxes, generous benefits, social trust, trains that run on time. It works. Danes report happiness. Swedes live well. No surveillance state required.

So why not import the model?

The usual answer involves homogeneity, and there's something to it, though not quite what people assume. The relevant factor isn't ethnic composition. It's preference variance: how much do different communities within the society diverge in their visions of what a good life looks like?

In Denmark, the answer is: not very. Centuries of shared history, Lutheran heritage, geographic isolation, and small population produced a rough consensus about work, family, leisure, and the state's proper role. When the Danish government invests heavily in parental leave and elder care, it's encoding priorities most Danes actually hold. The fit between what policy provides and what citizens want is reasonably tight.

Now consider what genuine pluralism means. Not diversity as aesthetic (different cuisines, different festivals, a pleasing variety of faces in the stock photos) but actual divergence in conceptions of flourishing. What counts as a life well-lived for an Amish farmer in Pennsylvania? A tech founder in San Francisco? A rancher in West Texas? A Hasidic family in Brooklyn? A commune in rural Oregon? A Pentecostal congregation in Alabama?

These aren't different priorities within a shared framework. They're different frameworks entirely. And this changes what kinds of systems can work.

Socialism, even the gentle kind, requires choosing whose preferences to encode. The state must decide what to fund, what to subsidize, what to prioritize. In a society with narrow preference variance, this feels like consensus. Everyone more or less agrees about what flourishing requires, so collective provision of it generates little friction.

In a genuinely pluralistic society, that same provision becomes imposition. Someone's vision of the good gets encoded. Others accommodate or resist. The more pluralistic the society, the more acute the problem becomes.

The American founders understood this, even if they lacked our vocabulary for it. The Constitution doesn't prescribe a vision of the good life. It creates space for divergent visions to coexist. Federalism, limited central authority, rights as constraints on government: these aren't inefficiencies to be overcome. They're load-bearing walls for a society that contains multitudes.

There's a distinction here worth holding onto: collective provision of means versus collective imposition of ends.

A society can pool resources for infrastructure, security, basic education, a safety net beneath which no one falls. These are means. They expand what individuals and communities can do without dictating what they should do. Social Security doesn't tell you how to live in retirement. It gives you resources to live as you choose.

But when collective provision starts encoding particular visions of flourishing (what family structure to incentivize, what values to instill, what ways of life to privilege) it crosses into imposition of ends. And in a pluralistic society, this is where the trouble begins. Not because such visions are wrong, but because they're contested. One community's flourishing is another's oppression.

This maps onto something The Prometheus Dispatch has explored before: equality of opportunity versus equality of outcome. Providing means is about opportunity. Everyone gets access to the tools. What they build with those tools is their own. Imposing ends is about outcome. Everyone arrives at the same place, which requires constraining the journeys that would lead elsewhere.

The former is compatible with pluralism. The latter isn't.

So where does this leave us?

I still have my list. The ideas I'd implement if I could, the certainty that I see clearly where others fumble. I don't think that impulse is going away. It's human. It might even be useful in small doses, the engine that drives people to build and advocate and try to make things better.

The question is whether we can hold that impulse alongside a structural humility. The recognition that my certainty (your certainty, anyone's certainty) is not a sufficient basis for overriding the distributed intelligence of a free society. Not because I'm wrong (though I might be). Because even if I'm right, the system that lets me impose my rightness is the same system that lets others impose theirs. And I have no guarantee that the others who gain that power will share my values.

Tocqueville saw where this leads. Not jackboots and gulags, necessarily. Something softer. A state that "does not tyrannize, but it compresses, enervates, extinguishes, and stupefies a people, till each nation is reduced to nothing better than a flock of timid and industrious animals, of which the government is the shepherd." We would be cared for. We would be managed. We would be safe. We would have traded the burden of self-determination for the comfort of being told.

The authoritarian and the soft paternalist offer the same bargain, differing only in degree. Relief from the friction of freedom. An end to the exhausting work of deliberation, disagreement, and compromise. Someone who sees clearly, finally in charge.

The seduction is real. So is the cost.

I don't know how to make democratic institutions less sclerotic, less captured, less maddening in their inability to act. That's a genuine problem and pretending otherwise is dishonest. But I've come to think the messiness we hate is doing something we can't see until it's gone. The delays and vetoes and competing interests are how distributed knowledge enters the system. How preferences that don't fit the grid get represented. How bad news reaches decision-makers before it's too late.

The alternative isn't efficient democracy. It's the slow or sudden loss of something we won't know how to rebuild.

The question we're left with isn't whether the Chinese model is wrong. It's whether we're still the kind of people who can bear the weight of the alternative.