The Myth of Libertarianism

Tribe first, theory later

[My writing here sort of fell off a cliff after coming back from sabbatical last Fall. But look for more content soon. In the meantime you should also check out The Institute for Humane Studies’ fantastic new project, Liberalism.org. I’ll be doing a good deal of writing there this year. My first essay, “The Enemy is Power, Wherever You Find It,” is up, along with a companion podcast interview with Aaron Ross Powell.]

If you had asked me in 2015 to describe the core commitments of American libertarianism, I could have done it in about a minute. Free markets, limited government, individual rights, skepticism of state power, free trade, open or liberal immigration, some version of non-interventionism abroad, a strong preference for constitutional constraints on executive authority. There would have been edge cases and internal disputes, sure, but the center of gravity was clear enough that you could gesture at it.

Try to do the same today, ten years later, and you run into trouble almost immediately. In the public-facing, movement-adjacent side of libertarianism — the one that reaches audiences through podcasts, YouTube, X, and the tech-intellectual networks of the last few years — the center of gravity has shifted in ways that would have seemed inconceivable a decade ago. Peter Thiel, perhaps the most influential libertarian-adjacent figure in American finance, funded J.D. Vance’s Senate campaign and helped bring him into the Trump orbit, where Vance now openly describes himself as post-liberal. Dave Smith and Tom Woods, two of the most-listened-to libertarian podcasters in America, have made common cause with Trumpism on immigration. A significant fraction of their audiences has gone along for the ride.

The picture is different in the academic wing of the movement, which is where many of the readers of this Substack will have encountered libertarianism. Classical liberals and libertarians in universities, at Reason, and at institutions like Catoand the Institute for Humane Studies have largely stayed where they were — defending open immigration, free trade, limited executive power, and constitutional constraints, and most of them have been among the administration’s sharper critics. Ilya Somin has spent the last year writing essentially without pause about the administration’s assaults on constitutional limits. Justin Amash, still the most prominent libertarian politician in recent memory, has been unsparing. These sticky ideologues now find themselves watching former fellow travelers in the popular movement make arguments that would have gotten them laughed out of libertarian circles in 2015, and some of them feel like strangers in a movement they thought they belonged to.

What’s going on?

* * *

One answer — the one most libertarians instinctively reach for — is that whichever side has drifted isn’t really libertarian anymore. Depending on who’s telling the story, either the Thiel-Vance wing has abandoned libertarianism for reactionary nationalism, or the Reason wing has sold out to the very liberalism it was supposed to oppose. Each camp accuses the other of being fake libertarians who’ve misplaced the true principles of the movement.

I want to suggest a different diagnosis: that both camps are real libertarians, and that the reason libertarianism can’t settle the question of who counts is that libertarianism was never as philosophically coherent as libertarians have tended to assume. The current fracture isn’t an aberration. It’s what libertarianism has always looked like. We just didn’t notice, because for a brief window in the late twentieth century the sociology of the movement happened to produce the appearance of a consensus.

The most useful framework I’ve found for describing this phenomenon comes from a recent book by Hyrum and Verlan Lewis called The Myth of Left and Right. The Lewis brothers’ central claim is that “left” and “right” are not coherent ideologies. They’re tribes. The political positions associated with left and right don’t hang together because of any underlying philosophical essence. They hang together because they happen to be the positions currently held by the left-wing and right-wing tribes. Conservatives went from being free traders to being protectionists, from being hawks on Russia to being doves, from championing federal power to hating it, and the shifts aren’t puzzling once you see that the tribes came first and the ideologies came second. People anchor into a tribe, usually for sociological reasons — family, friends, a formative book, a single resonant issue — and then pick up the tribe’s positions as a matter of social conformity. The positions change as the tribe changes. The “essence” that’s supposed to link them all is a story told after the fact.

The Lewis brothers aren’t writing about libertarianism — their target is the essentialist reading of the political spectrum as a whole. But their framework gave me a vocabulary for something I had already come to suspect, both from years inside the libertarian movement and from tracing its intellectual history. Libertarianism isn’t an exception to the Lewis thesis. It’s a smaller version of the same phenomenon, and if anything its philosophical pretensions make the tribal dynamics easier to see once you know to look.

* * *

Let me start with a brief version of the history (the longer version of which is laid out in The Individualists). The term “libertarian” began in the 1850s as a self-description for a French anarcho-communist named Joseph Déjacque, who believed that private property and the state were two sides of the same coin and that a real commitment to liberty required the abolition of both. Through most of the nineteenth century, “libertarian” was used by anarchists of various stripes — some who favored private property, some who rejected it — to distinguish themselves from authoritarian socialists. The term was consciously positioned on what we’d now call the left.

By the early twentieth century, the label had broadened into something close to a generic anti-authoritarianism. Charles Sprading’s Liberty and the Great Libertarians (1913) used “libertarian” to cover “Republicans, Democrats, Socialists, Single-Taxers, Anarchists, and Women’s Rights advocates” — a genuinely big tent united only by the commitment not to impose one’s views on others by force. A few decades later, under the influence of figures like Leonard Read and the Foundation for Economic Education, “libertarian” had been narrowed to mean support for free markets and limited government. By the 1970s, the Nozick-Rand-Rothbard synthesis had narrowed it further, to the point where many libertarians treated the label as synonymous with a particular form of rationalist, rights-based free-market absolutism.

That’s not where things stayed. In the 1990s and early 2000s, the label fragmented again. Bleeding-heart libertariansstarted talking about social justice. Left-libertarians drew on Tucker and Proudhon to critique actually-existing capitalism. Paleolibertarians went in the opposite direction, making alliances with the cultural right. And now, in the 2020s, the rise of the tech-right, the neo-reactionary movement, and the Thiel-Musk orbit has produced another set of self-identified libertarians whose views on immigration, trade, executive power, and free speech would have been nearly unrecognizable to the libertarians of twenty years ago.

The Lewis framework predicts this kind of drift. An essentialist framework doesn’t. If libertarianism were really defined by a fixed philosophical essence, you wouldn’t see these kinds of swings. You’d see the label getting more precise over time, not less. You’d see cumulative convergence on what libertarian principles actually require, not recurring bouts of fragmentation and re-fragmentation. The pattern we actually see — a label whose meaning shifts as the sociological composition of the movement shifts — is exactly the pattern the Lewis brothers describe for left and right.

* * *

The second piece of evidence is theoretical, and for me it’s the more decisive one. It’s that libertarianism’s foundational principles can’t actually do the work libertarians sometimes claim they do. They’re too open-ended to yield determinate answers to concrete political questions.

Take property, which is the closest thing libertarianism has to a settled core. Libertarians who all claim to be defending property rights disagree radically about what those rights require. Herbert Spencer and Lysander Spooner thought intellectual property was a straightforward extension of property rights; Murray Rothbard thought it was an affront to them. Henry George thought land was categorically different from other property, justly owned by the community; Rothbard called that view “intellectually and morally beneath contempt.” Benjamin Tucker thought legitimate property in land required ongoing occupancy and use; Robert Nozick thought the original appropriation of unowned land, once performed, generated permanent rights. These are libertarians all claiming to defend the same foundational commitment while disagreeing about what that commitment actually is.

The same story plays out for every other core libertarian principle. Does the non-aggression principle permit pollution? That depends on how you define “aggression” and whose property rights you recognize, and libertarians disagree sharply about both. Does individualism require open borders or permit the kind of restrictive “covenant communities” that Hans-Hermann Hoppe defends? Libertarians who share the same foundational premises reach opposite conclusions. Does skepticism of authority lead to anarchism or just to minimal government? Libertarians have been arguing about that since the 1970s and will still be arguing about it in 2075.

The deeper point isn’t that libertarians happen to disagree about these questions. It’s that the foundational principles underdetermine the conclusions. A commitment to property rights doesn’t tell you what property is, how it gets acquired, whether it’s inheritable across generations, whether it applies to ideas, whether land is a special case, or what happens when property-based claims conflict. A commitment to non-aggression doesn’t tell you what counts as aggression, what counts as a defense against aggression, or how to weigh the competing claims of parties who each believe themselves aggrieved. A commitment to individualism doesn’t tell you how to think about collective action problems, inherited obligations, or any of the other ways individual lives turn out to be embedded in structures.

The principles do some work. They rule out, for example, the more extreme forms of state socialism or totalitarianism. But they can’t do anything close to the work of generating determinate libertarian positions on the contested questions of our politics. Which means that when libertarians arrive at positions on those contested questions, the real work is being done somewhere else — by a particular interpretation of the principles, combined with a particular reading of history, a particular set of empirical assumptions, and a particular constellation of intuitions about which tradeoffs matter. And those interpretations, readings, assumptions, and intuitions are, for most of us, not the product of independent philosophical reasoning. They come from the particular sociological path we took into libertarianism — the books we read, the teachers we had, the institutions that trained us, the peers we argued with at a formative moment.

* * *

This is why the current fracture in libertarianism is so illuminating. The tech-right wing and the academic classical-liberal wing aren’t disagreeing about libertarian principles. Both sides are drawing on the same vocabulary of property, liberty, individual rights, and skepticism of power. The disagreement is about how to interpret that vocabulary and apply it to the contested questions of the current moment — and the interpretations track the sociology of the sub-tribes, not the logic of libertarian philosophy. The tech-right lineage came into libertarianism through a particular set of institutions (Founders Fund, certain podcasts, a distinctive set of internet intellectuals), and it’s going in one direction. The classical-liberal academic lineage came in through a different set of institutions (IHS, Cato, the Liberty Fund seminar circuit, Mont Pelerin), and it’s going in another. Both sides are doing what libertarians have always done, which is to reason from shared premises to conclusions consonant with the sensibilities of their particular intellectual community.

The essentialist view can’t account for this without calling one side fake. The sociological view doesn’t have to. It just says that libertarianism, like every other ideological identity, is held together primarily by a set of sociological anchors — institutions, readings, friendships, conventions — and that when those anchors shift or diversify, the views do too. When one of my friends finds himself newly sympathetic to Trump-style immigration restriction while still describing himself as a libertarian, he isn’t being incoherent. He’s responding to a shift in the sociological composition of the tribe he’s part of, and reverse-engineering a philosophy to fit. So, for that matter, are the rest of us — the ones who feel we’ve stayed put. Our “staying put” is itself the product of a different but equally sociological anchoring. None of us is reasoning from pure libertarian axioms to specific political positions. The axioms aren’t powerful enough to do that kind of work, and we aren’t disembodied philosophical agents to begin with.

* * *

There is a version of this argument that ends in nihilism — if libertarianism is mostly sociology, why bother? That isn’t the conclusion I draw, and it isn’t the one the Lewis brothers draw either. Their own recommendation, at the end of The Myth of Left and Right, is a kind of particularism. If the political spectrum isn’t tracking any real essence, we shouldn’t try to think our way through politics by locating positions on a single axis. We should disaggregate — take issues one at a time, think them through on their own terms, and let our conclusions go where the specific case takes us, even when the resulting pattern doesn’t fit any existing ideological mold.

I find that broadly congenial, and not only because it’s where the present argument pushes. It aligns with a pluralistic, non-ideological sensibility about moral and political philosophy that I’d want to defend on other grounds anyway. Isaiah Berlin‘s value pluralism, Bernard Williams‘s skepticism of systematic ethical theory, the long tradition that insists there are many goods, that they don’t reduce to one another, and that practical wisdom about particular cases can’t be replaced by the application of a master principle — all of that sits much more comfortably with me than the monism of an Ayn Rand or a Murray Rothbard, for whom political philosophy is the deductive application of a single axiom to every case that comes along.

So the right response to seeing libertarianism as partly sociological isn’t to stop holding libertarian views. It’s to hold them in a different spirit — less as conclusions derived from an axiom, more as provisional positions arrived at by thinking as carefully as I can about particular issues, with libertarian principles as one important input rather than the only one. On some questions — mass incarceration, occupational licensing cartels, the drug war, closed borders — I’m confident. On others — climate policy, labor organizing, antitrust, the regulation of AI — I’m much less so, because the tradition I’ve inherited offers multiple defensible answers and I’m not sure I’ve landed on the right one. That kind of issue-by-issue tentativeness can feel like weakness next to the certainty of the ideologue. I’ve come to think it’s the stance that actually follows from taking libertarian principles seriously, once you see they were never powerful enough to do the work that ideological certainty requires of them.