Martin Elson
The non-aggression principle is the rule most libertarians would name as their core, usually alongside self-ownership, from which Rothbard derived it: no one may start the use of force against another person or their property. It is a good rule. It is not a starting point. On its own it can’t tell an attack from a defense, it needs a theory of property before it can say what counts as theft, and the attempts to prove it from logic alone prove much less than they claim. As the conclusion of something deeper, it holds up well.
The case for non-aggression
Rothbard and Rand both made non-aggression the core of political ethics, and the Libertarian Party once asked its members to certify that they didn’t advocate the initiation of force. The appeal is plain. One test settles a great deal: taxation is taking, conscription is forced labor, war is killing at scale, and none of them is excused by being done by a state. Stephan Kinsella prizes the principle for its precision. Aggression, on his account, is crossing the borders of someone’s body or property without consent, and libertarianism is the theory of where those borders lie. A clear line makes conflict over scarce things settleable.
Gary Chartier gives the strongest defense, because he derives the principle instead of assuming it. He works from the natural law tradition of John Finnis and Germain Grisez. Human welfare has several basic aspects, among them life, knowledge, friendship and play, and they can’t be weighed against each other on a single scale. So no one can justify deliberately damaging one person’s good for the sake of another good, however large. Add the Golden Rule, and you get a robust set of property rights and very little reason to accept a body with authority to attack people’s persons or take their things. Chartier is honest about the result: it is something like the NAP, not a simple version of it. Jan Narveson reaches a similar rule from a social contract instead of natural law. Two roads to the same place is itself evidence that non-aggression is a conclusion.
Hans-Hermann Hoppe aims higher: a proof that no one can deny. To argue at all, he says, is to seek agreement without force and to use your own body as you choose. Anyone who argues against libertarian property rights therefore contradicts himself in the act of arguing. If it worked, non-aggression would be as undeniable as logic. Frank van Dun has defended and extended the argument at length. Stephan Kinsella applied the same move to punishment: an aggressor can’t consistently object to force used against him, so he is estopped from complaining.
Where it goes wrong
Attack or defense? Stated alone, the principle fits everything from pacifism to shooting trespassers. Alex Strekal laid out the spectrum. Pacifists allow no force at all. Thick libertarians allow defense against a real threat to life. Thin libertarians count force to protect property as self-defense. Objectivists add retaliation and even preemptive war. All four call their own violence defense, so the rule can’t referee between them: the dispute is over what its words mean. Chartier asked whether “initiation of force” settles it. Strekal’s answer was that Rothbard himself needed a separate theory of proportionality to rule out the harshest readings, and that theory is not in the principle. Thick, Not Thin takes up the political side of this; here the point is logical. The principle needs a definition before it can do anything, and the definition does the work.
Theft needs a theory of property. What counts as stealing depends on what counts as owning. A Lockean, a Georgist and a mutualist will each read non-aggression differently, because each draws the borders of property differently. Kinsella says so openly, which is to his credit, but it concedes the point: the principle is a theory of property dressed as a principle of peace. Read strictly, it also forbids too much. Matt Zwolinski, himself a libertarian, notes that every exhaust pipe sends particles across someone’s border; a strict reading bans pollution and nearly every small risk, and no one lives that way. David Friedman raised the problem decades earlier, asking whether shining a flashlight onto a neighbor’s land is trespass, and Robert Nozick gave a chapter of Anarchy, State, and Utopia to risk and compensation without a clean answer. Where the borders of property belong is the subject of Property Without Propertarianism.
Punishment is force after the fact. Defense meets a present threat. Revenge comes when the threat is over. Strekal held that violent punishment is itself aggression, and Chartier agreed: he rejects punishment and keeps restraint and restitution. Rothbard allowed a criminal to lose rights in proportion to what he took, and called the result frankly retributive: “a tooth (or two teeth) for a tooth.” Whatever one makes of that, it comes from outside the principle, which says nothing about what may be done after the harm. Kinsella’s estoppel argument tries to supply it from inside, but it inherits Hoppe’s gap: what a person can consistently say doesn’t settle what others may do to him.
The only absolute. In a note on Mike Gogulski’s rejection of voluntaryism, Strekal put his finger on an odd line. Non-aggression is treated as the one categorical rule, while every other value is left to taste. Once that line is drawn, everything objectionable about the state can come back in private form, as long as it is reached by contract. Strekal never saw anyone defend the line, and neither have I.
The proof proves too little. Hoppe’s argument comes from the discourse ethics of Karl-Otto Apel and Jürgen Habermas: the claim that arguing commits you to norms you can’t consistently deny. Hans Albert, the critical rationalist, had already attacked Apel’s version. Every attempt at a final justification, Albert showed, ends in a regress, a circle, or an arbitrary stop, and a performative contradiction shows only what a speaker presupposes while speaking, not what anyone owes anyone afterward. Habermas himself drew democratic deliberation from the same starting point, which shows how little the starting point settles.
Robert Murphy and Gene Callahan, both sympathetic to Hoppe’s conclusions, made the objection concrete. At most the argument shows that, during a debate, you can’t consistently deny your opponent control of the body parts he is using to argue. Slaves and prisoners have argued well without owning themselves. Standing on someone’s land while you debate him doesn’t make the land yours, and nothing in the argument reaches the homesteading of external things, which is where libertarian property theory does its real work.
The strongest case against
Chartier pressed the best objection to this page in a reply to Strekal. Why must the principle be grounded at all? Respectable moral philosophers, W. D. Ross and Henry Sidgwick among them, held that some principles are known directly. And a movement can hold Aristotelians, Kantians and consequentialists together on non-aggression without settling whose foundations are right.
The best version of the intuitionist reply is Michael Huemer’s. He reaches anarchist conclusions from ordinary moral judgments: we would condemn a private person who did what the state does, so the state needs a special authority it can’t show. But look at how Huemer works. He starts from many judgments about cases, weighs them against each other, and lets them be revised. That isn’t an axiom; it is a presumption built from cases, which is the view of this page.
On the coalition, Strekal’s answer still stands. Foundations matter once practical disagreement arrives. An anti-racist and an ethnic nationalist can share the principle and nothing else, and a principle with no grounds can’t say which world it is meant for.
Where this lands
Non-aggression is a strong presumption, not an axiom. Anthony de Jasay’s presumption of liberty puts it right: liberty is the default, and whoever would use force carries the burden of proof. That burden is heavy, and the state almost never meets it. The presumption rests on something prior, as Strekal said: we accept non-aggression because we respect people, not the other way round. On this site that respect is the floor from What Rights Are: no one may be used as an input. Defense against a present threat is not aggression; restraint and restitution replace punishment; and property is a conclusion of liberty, not its master. Non-aggression is where you arrive, not where you begin.
Sources
- Murray N. Rothbard, For a New Liberty (Macmillan, 1973), ch. 2, and The Ethics of Liberty (Humanities Press, 1982), ch. 13 (punishment and proportionality).
- Ayn Rand, “The Nature of Government,” in The Virtue of Selfishness (Signet, 1964). Force barred from social relations.
- Stephan Kinsella, “What Libertarianism Is” (Mises Institute, 2009). Aggression as crossing property borders.
- Gary Chartier, “Natural Law and the Non-Aggression Principle” (Liberale Law, February 2010). original · copy here. See also Anarchy and Legal Order (Cambridge University Press, 2013).
- Alex Strekal (as brainpolice), “Non-aggression Is Not An Axiom” (Polycentric Order, July 2009), with Chartier’s replies in the comments. archived · copy here
- Alex Strekal (as brainpolice), “Putting the NAP in Its Proper Context” (Polycentric Order, January 2009). The spectrum from pacifism to Objectivism. archived · copy here
- Alex Strekal (as brainpolice), note on Mike Gogulski, “Why I am not specifically a voluntaryist.” Non-aggression as the only absolute.
- Hans-Hermann Hoppe, A Theory of Socialism and Capitalism (Kluwer, 1989), ch. 7, and The Economics and Ethics of Private Property (Kluwer, 1993). Argumentation ethics.
- Hans Albert, Treatise on Critical Reason (1968; Princeton University Press, 1985), and Kritik des transzendentalen Denkens (Mohr Siebeck, 2003), his collected essays against Apel and Habermas.
- Jürgen Habermas, Moral Consciousness and Communicative Action (MIT Press, 1990). Discourse ethics.
- Robert P. Murphy and Gene Callahan, “Hans-Hermann Hoppe’s Argumentation Ethic: A Critique” (Journal of Libertarian Studies 20, no. 2, 2006). PDF
- Matt Zwolinski, “Six Reasons Libertarians Should Reject the Non-Aggression Principle” (Libertarianism.org, April 2013).
- Michael Huemer, The Problem of Political Authority (Palgrave Macmillan, 2013).
- Anthony de Jasay, Justice and Its Surroundings (Liberty Fund, 2002). The presumption of liberty.
- Jan Narveson, The Libertarian Idea (Temple University Press, 1988). Non-aggression from a social contract.
- Frank van Dun, “Argumentation Ethics and the Philosophy of Freedom” (Libertarian Papers 1, 2009). The strongest defense of Hoppe.
- Stephan Kinsella, “Punishment and Proportionality: The Estoppel Approach” (Journal of Libertarian Studies 12, no. 1, 1996): 51–73.
- David Friedman, The Machinery of Freedom (2nd ed., Open Court, 1989). The flashlight and the problem of small risks.
- Robert Nozick, Anarchy, State, and Utopia (Basic Books, 1974), ch. 4. Prohibition, compensation and risk.