I. The Word Stopped Fitting
For most of my adult life, when someone asked about my politics, I said I was a conservative. It was the shortest true answer available, and for a long time it was close enough.
It was not always the answer. I started out a Democrat, or thought of myself as one. My first presidential election was 1992, and the candidate I remember liking was Ross Perot, the Texan with the charts who warned that American jobs would go south if NAFTA passed. Nine years later the towers came down, and something in how I thought about my country and its enemies moved and did not move back. In 2004 I voted for George W. Bush. After that the drift ran one direction, and it did not slow down.
Somewhere along the way the word stopped fitting. It told people I would probably vote Republican, which was true, and nothing about why, which was the part that mattered. It put me alongside men whose foreign wars I thought ruinous, whose deference to corporations I thought had helped hollow out the towns they claimed to speak for, and whose politics was mostly a career. I was hostile to communism and to liberalism alike, but “conservative” could not tell you what I was for, and it had nothing to say about the things I cared about most.
So I tried something. Not another online quiz, though I will come to those, but an interview: eighty-seven questions, asked one at a time by an AI told to behave less like a test and more like a political philosopher who keeps noticing what your answers imply. What came back was not a score. It was a map, and the map had a center I had never quite put into words.
II. What the Quizzes Measure
Most people who have wondered about their own politics have taken one of the online quizzes. The best known is the Political Compass, which plots you on two axes (economic left and right, authoritarian and libertarian) and drops a dot on a square. Others promise to match you with a party or a candidate. I have taken several. They were not useless. They were answering a smaller question than the one I was asking.
The first problem is that the categories come first. Before you answer anything, the quiz has already decided which distinctions matter, and all it can measure is how far you stand from boxes someone else drew. A man can be culturally traditional, suspicious of free trade, hostile to socialism, attached to private property, doubtful about democracy, opposed to foreign wars, and more wary of large corporations than of his county sheriff. Put him on two axes and every one of those facts is averaged into a single dot, and the dot tells you almost nothing about him.
The second problem is that everyone gets the same questions, which wastes most of them. Once a man has said plainly that unrestrained markets are dangerous, there is no point asking him three more times whether he wants more government or less. The useful question is why he rejects them. The answer might be socialism, or distributism, or tariffs, or old-fashioned anti-monopoly sentiment, and those are very different men. A fixed quiz cannot ask that question, because it does not know it needs to.
The third problem goes deepest. A quiz records positions, not the principles underneath them. Two people can both favor restricting immigration, one because he wants higher wages and the other because he believes a nation is a historical people and not a stack of paperwork. They check the same box. They do not hold the same politics.
What comes out the other end is a verdict like “you are 82 percent conservative.” It is precise, and it is empty. It tells you where you stand relative to a word. It does not tell you what you believe.
III. An Interview, Not a Test
The alternative was simple to state. Instead of a fixed questionnaire, I asked the AI to conduct an interview: one question at a time, each chosen partly in light of everything I had already said.
Most questions came with lettered options, but the options were prompts rather than boxes. I could pick one, pick several, reject all of them, qualify an answer, or say something else entirely. “No opinion” was recorded as undecided, not forced into a guess. If I corrected myself later, the correction replaced the earlier answer. After each reply the AI said briefly what the answer seemed to show, noted any tension it created with what I had said before, and then chose the next question.
There was no quota. The ground was wide: first principles such as authority, liberty, and equality; constitutional structure; nation and citizenship; economics; family; religion and public morality; foreign policy; justice and civil liberties; the institutions and elites that run things. But the order and the depth were set by my answers. A reply that closed a question moved the interview on. A reply that opened one kept it there. The interview was to stop when further questions stopped changing the picture, and for me that took eighty-seven.
Two instructions mattered most. The AI was not to assign a label at the start, and it was not to collapse the result into left and right, a party, or a number on a compass. It was also not to tell me what I ought to believe. Its job was to be interviewer, classifier, and pattern detector, and to answer one question at the end: given everything this man has said, what political philosophy does he actually hold?
That second instruction was not a formality. I have written before about how readily these systems present one contested moral framework as the neutral starting point and judge everything else against it. An interviewer who has already decided which answers are respectable will steer toward them, however politely. The point of the exercise was to be mapped, not managed.
IV. Where the Branches Turned
The method is easiest to see in the moments where one answer changed the direction of the whole interview. Three of them did most of the work. The questions below are as the AI put them. My answers were spoken aloud, and I have tidied them for print without changing what they say.
Liberty and order
When liberty and social order collide, which should usually win? Imagine an activity that is peaceful and voluntary, but that you believe is clearly corrosive to family life, public morality, or the long-term health of society.
A. If it doesn’t violate someone else’s rights, government should leave it alone.
B. Government may legitimately restrict it if the social harm is serious enough.
C. It depends heavily on whether the issue should be handled by family, church, local government, state government, or the federal government.
D. Some combination of B and C.
My answer:
A is wrong. Just because something doesn’t violate someone else’s rights doesn’t mean the government should leave it alone. It harms the individual, and who knows what he might have contributed to civilization if he weren’t pursuing something destructive. So yes, government may legitimately restrict it if the harm is serious enough.
Up to that point I had talked a good deal about rights and limited government, and a fixed quiz would have filed me as a libertarian. The interview noticed instead that I had just rejected the libertarian test of legitimacy: that anything which harms no one else’s rights is beyond the law’s reach. The questions that followed turned to public morality, inherited moral tradition, and who gets to decide what counts as social harm.
The franchise
Should every adult citizen have an equal vote?
A. Yes. Universal adult suffrage is the fairest and most legitimate system.
B. Generally yes, even though voters vary enormously in judgment and civic virtue.
C. No. Voting should probably be conditioned on some demonstration of stake, responsibility, competence, citizenship, or civic qualification.
D. I’m skeptical of mass democracy itself. Political representation should be filtered through institutions, local bodies, property, family, profession, or some other mediating structure rather than simple one-person-one-vote majoritarianism.
My answer:
D. I’m very skeptical of mass democracy itself. Universal suffrage is not the fairest or most legitimate system. That people who abuse the system and live off it should have an equal vote with the producers, the net contributors to the society, is outrageous. So, D.
A conventional survey would have gone on to ask about voter access and turnout. The interview dropped all of that and asked what representation should rest on instead: property or taxation, the household, civic qualification, the local community, or some mixture. I said all of them mattered, and I reached for Heinlein’s Starship Troopers, where the vote belongs to those who have served. That opened a line of questions about national service, political privilege, and the difference between the rights every citizen holds and the authority to govern.
Markets and the common good
What do you think about capitalism?
A. Free markets are generally the best system, and government should interfere as little as possible.
B. Markets are useful, but they are subordinate to the national interest, family stability, and social order.
C. Large corporations and finance can become as dangerous as government and should be restrained when they damage communities, workers, or national independence.
D. Protectionism, tariffs, industrial policy, and limits on offshoring are legitimate if they preserve domestic production and national strength.
E. Some combination of B, C, and D.And underneath it: if a policy makes the country richer overall but destroys local industries, weakens families, and hollows out towns, is it still a good policy?
My answer:
No, free markets are not generally the best system. Left unchecked, runaway capitalism can be as destructive to a nation as communism. Markets are useful, but they should be subordinate to the national interest. Large corporations and finance can absolutely become as dangerous as government and should be restrained when they damage communities, workers, and national independence. So B and C, and D as well. If a policy makes the country richer but destroys local industries, weakens families, and hollows out towns, it is a bad policy. Capitalism should exist to let people live a good life. A line on a chart going up forever is not the goal. A people well cared for is.
That one answer ruled out both laissez-faire and socialism, and the interview stopped asking whether I wanted more government or less. It asked instead about concentrated corporate power, antitrust, foreign ownership, banking and usury, automation, and whether ordinary families ought to own property. What emerged was not “right-wing economics” but something older and less familiar, which the next section describes.
V. The Map
After eighty-seven questions, the AI offered a short label: Christian paleonationalist. The longer version ran like this. A Christian traditionalist and paleonationalist: strongly federalist and localist, anti-majoritarian, family-centered, producerist and protectionist in economics, noninterventionist abroad, with Hoppean sympathies toward constitutional monarchy and a restricted franchise.
That is a mouthful, and the label matters less than what sits under it. When the AI was asked for the one principle that kept recurring beneath the answers, it did not say “right wing.” It said inheritance and stewardship. Each generation receives goods it did not make, among them its faith, its families, its nation, its laws, its land, and its memory, and it is not free to spend them down. Politics exists to secure the common good of a people across generations: to keep a Christian moral order, and to protect the families, property, local communities, and institutions through which that inheritance passes. Nearly every answer I gave turned out to be an application of that.
Authority. Political authority comes from God. Consent matters, but a majority vote does not make a thing legitimate. Christianity should hold a recognized place in public life and law. Non-Christians are owed civil toleration, and no one can be made a Christian by force, but the law need not pretend every faith is equally authoritative. That includes laws against blasphemy, and restraint of cultural industries that work steadily against the moral order the law exists to keep.
Representation. Voting is a trust, not a natural right. It should rest on things that give a man a stake in the outcome and the judgment to use it: property, independence, contribution, civic knowledge, service, the care of a household, roots in a place. The household, not the isolated individual, is the natural unit of representation. A citizen permanently dependent on the government has a plain conflict of interest in holding an equal share of power over it. Senators should again be chosen by state legislatures, as they were before 1913, and the House should represent actual communities rather than counted heads.
Constitution and crown. The Constitution is sound and badly interpreted. It should be read strictly, with real federalism, a far smaller administrative state, a restrained Supreme Court, and states that keep the right of peaceful separation. I am also open to constitutional monarchy. Hans-Hermann Hoppe’s argument persuades me here: an elected official holds power for a few years and has every incentive to spend the country’s capital while he can, while a hereditary ruler has reason to hand on a realm in better condition than he found it. The king I have in mind is bound by Christian law, the constitution, custom, the courts, and the lesser authorities beneath him. A king who turns tyrant may be resisted and removed.
Nation. A nation is a historical people, bound by language, religion, custom, memory, ancestry, and inherited institutions. It is not a set of papers. Newcomers can join it, but by real assimilation, not by filing a form. Immigration policy exists to preserve the nation, not to raise GDP. Citizenship should be hard to earn and open to those who become English-speaking Christians, contribute, and commit for the long term. Even then, holding citizenship and holding power over the country are different things, and the second should wait on deeper roots. America itself is less one people than a federation of related regional peoples, Texas, the South, New England, Appalachia, the Midwest, and those differences should be kept rather than ground down.
Economy. I hold private property and inheritance in high regard. That means low and flat taxes, no property tax, balanced budgets, decentralized banking, and as many families as possible owning something of their own. But the economy exists for the household, not the other way round. That justifies tariffs, domestic industry, antitrust, limits on foreign ownership of farmland and strategic assets, usury laws, and care for the land. Automation is good when it serves the men whose work it changes. Looking back, the candidate I liked in 1992 was the first sign of this.
Family and morals. The family is the basic unit of society. Marriage is a covenant, not a private contract, and divorce should require cause. Parents hold primary authority over education. I oppose the welfare state and would return charity to families, churches, and local institutions. But I favor policy that makes marriage and children easier to afford, because it builds independence rather than dependence. Pornography, prostitution, hard drugs, and commercialized vice should be prohibited, with the heaviest penalties for those who profit from them. Culture is upstream of politics, and the law has a duty to the moral climate a people lives in.
The sword. Punishment should aim at deterrence, incapacitation, and restitution, and the death penalty is legitimate for the gravest crimes. Citizens should be able to bear the same kind of small arms a soldier carries, for defense, for the militia, and as a final check on tyranny. Policing belongs to sheriffs and localities. Abroad, the nation should be overwhelmingly strong and slow to act. It should defend its borders and its people, and leave off remaking other countries. The American empire built after 1945 has done real harm at home.
Beneath all of it is a view of hierarchy. Men are equal before God and before the law, and unequal in almost everything else. A healthy society cultivates a governing class out of family continuity, property, education, service, and Christian character. It also strips standing from members of that class who prove corrupt or incapable, and it holds all of them to the same civil law as everyone else.
VI. What I Am Not
The AI was also asked which labels would mislead, and that list turned out to be nearly as useful as the map.
I am not a libertarian. Individual autonomy is not the highest political good, and I favor moral legislation no libertarian could accept. Hoppe shaped how I think about democracy and monarchy, but he is an influence on one part of the map, not a description of the whole. I am not a classical liberal either, because the rights I care about exist inside a moral, religious, and national order rather than standing above it. I am not a free-market capitalist in the modern sense, because I hold markets subordinate to families and the nation, and I am not a socialist, because property, inheritance, and the independent household are the ground I stand on. I am not a democrat in the philosophical sense, since a majority vote does not make anything right. Nor am I an absolutist: the king I would accept answers to God, to the law, and to the lesser authorities, and may be resisted. I am not a centralizer of any kind, since I would keep power with states, towns, families, and armed citizens, down to the right of a state to leave. And I am not simply a Republican. The party’s wars, its corporate loyalties, and its habits as a career cut against nearly everything above.
No single historical order matches the map, but several come close in parts. The nearest may be the Dutch Anti-Revolutionary tradition of Guillaume Groen van Prinsterer and Abraham Kuyper, Reformed Christians who held that authority comes from God rather than from the people, and who built a politics around the family, the church, and the other institutions that stand between the individual and the state. Its living heir, the Reformed Political Party (SGP) in the Netherlands, keeps explicitly Christian principles in a modern parliament, though it is more democratic and more pluralist than I am. The early American republic supplies the federalism, the state-chosen Senate, and the property qualifications for voting that many states kept well into the nineteenth century. The older constitutional monarchies of Europe supply the idea of a hereditary crown held within law. And the American paleoconservatives supply the rest: suspicion of empire, attachment to borders and inherited culture, and distrust of the global and the managerial. Several of the thinkers behind these traditions have places in the library I am building.
VII. Where the Map Pulls Against Itself
A map that fit together perfectly would be suspect. Mine has places where two principles I hold pull in different directions, and the interview found them. Three matter most.
Equal law, unequal franchise. When the interview asked whether different classes of people should live under different legal codes, I answered without hesitation: the same civil law for every citizen, full stop. Yet I do not think every citizen should hold the same share of power. The line runs between civil equality and political equality. Everyone stands under the same law; not everyone governs. The distinction is coherent, and it is an old one, but it leaves real work to do. Someone has to decide which rights belong to every citizen and which are privileges of governing.
A national moral floor in a decentralized union. I want most of life governed locally, by states and counties and towns, with the right to leave the union intact. I also want certain Christian moral standards to bind the whole nation, a floor the states may build above but not dig beneath. The reconciliation is that local government handles administration and ordinary life, while the first principles of moral order may be national. That settles the matter in principle. In practice every hard case will turn on whether the issue at hand is local policy or part of the floor.
Liberty within an order. I value the First Amendment, and the liberties of person, papers, property, and arms, as highly as anyone. I also favor blasphemy laws and limits on movements, communist or openly anti-Christian, whose purpose is to destroy the constitutional order itself. The distinction is between ordered liberty, which exists within a prior Christian order, and liberal neutrality, which treats every view as equally entitled to govern. I hold the first. But it raises a question I have not answered: who decides when dissent has become subversion? Whoever holds that judgment holds a great deal of power, and the map does not yet say who it should be.
The other tensions the interview found resolve the same way. Property is strong but not absolute against the common good. Public money that builds families is different from public money that breeds dependence. A crown can give a federation continuity without ruling it. In each case the apparent contradiction turns out to be a distinction between levels of political life.
Which leaves the question underneath all of them. Where exactly do the boundaries run between family, church, town, state, nation, crown, and the law of God, when those authorities disagree? It is not a new question. Abraham Kuyper spent his life on it and called his answer sphere sovereignty: each institution holds its authority from God within its own domain, and none may swallow the others. I suspect it is the question I will be working on for some time.
VIII. Try It Yourself
The case study here is mine, and readers will agree or disagree with the map as they please. The method is not mine to keep. It would work as well for a socialist as for me, because it does not care what you believe. It cares whether it has found out. What makes it work is an interviewer that follows your answers instead of its own categories, and keeps asking until it finds the principle you have been applying without naming it.
Most people’s politics are more particular than their party, and most people have never been asked the second question that follows their first answer. It is worth being asked.
Below is a reusable version of the prompt, written afterward so that anyone can run the same interview. Paste it into whichever AI you prefer and set aside an evening. If it starts telling you which of your answers are respectable, remind it of the rules.
Conduct a deep, adaptive interview designed to map my political beliefs as precisely as possible.
Do not begin by assigning a political label.
Ask one question at a time.
After each answer:
- Briefly summarize what the answer appears to indicate.
- Identify important distinctions, tensions, or implications.
- Ask the next question based partly on previous answers.
- Ask follow-ups when an answer is unclear or important.
- Allow mixed answers.
- Record “no opinion” as undetermined.
- Replace earlier answers when I correct myself.
Do not use a fixed question quota.
Explore political philosophy broadly, including government, democracy, constitutionalism, religion, nationhood, immigration, economics, family, foreign policy, justice, civil liberties, elites, education, and social order.
Do not reduce the result to left/right, authoritarian/libertarian, Democrat/Republican, or a numerical political-compass score.
Stop when further questions no longer materially change the profile.
At the end, identify: my core governing philosophy, primary political tradition, secondary influences, preferred political institutions, economic philosophy, theory of nationhood, religious and moral framework, family philosophy, foreign policy, civil-liberties philosophy, political temperament, closest intellectual relatives, labels that would be misleading, internal tensions, the best concise description, and the deepest recurring principle underlying my answers.
Your goal is not to tell me what to believe. Your goal is to determine: given everything I have told you, what political philosophy do I actually hold?
The label you get at the end matters less than the question the interview keeps putting to you. Given everything you have said, what do you actually believe? Most of us have never been asked it that way. It is a better question than the one on the ballot.
