The closing chapter, against libertarians who say “some crimes ought to be committed”: the lifeboat (owner’s rules; else first come, first served), Rothbard read charitably as not endorsing two spheres of ought, “ought is ought” restated, then the chapter’s centrepiece — a syllogism that claims to derive the NAP as a true ought-statement from is-statements and thereby “dismantle Hume’s guillotine” — and finally the categorical ought promised in ch. 6, which turns out to be Rand’s life-base by retortion: to argue is to have chosen life, so life is undeniably good, so every is implies an ought, so error is immoral and “the true is the good.” Findings. (1) The syllogism does not cross the gap: its first premise embeds the ought it concludes, and it slides from not-obligatory-to-aggress to obligatory-not-to-aggress (¬O(A) → O(¬A)) without noticing. (2) The life-base argument is Chris’s repair to Rand from the Objectivism Boundary page, minus the scope discipline that page insists on — it shows you cannot coherently argue you have not chosen life; it does not show life is “undeniably good.” (3) His own examples (crops in the sun, sunscreen) are hypothetical oughts conditional on life — so “ought is ought” collapses into the vault’s Trunk 1 by his own hand. (4) The chapter quietly concedes the vault’s scope structure: guardians ought save their drowning children is asserted as a universal ethical principle, while ch. 5 made neglect legally harmless — ethics has positive duties, law does not. That is the enforceability gradient, and it settles the “neglect is not harm” test the way the vault would.
Source: liquidzulu-wiki → src/content/libertarian-ethics/law-subset.md (n: 7, 2023-04-25, ~8.6k words, of which ~3k is a closing quote bank) · live at liquidzulu.github.io/law-subset. CC0; quoted, not copied.
Links: LiquidZulu hub, Ch. 6 — Defensive Force — where “ought is ought” was promised its proof, Is-Ought and Performative Grounding — the vault’s own dissolution of the gap, and the two tests, The Objectivism Boundary — Chris’s repair to Rand, Scope Confusion — the enforceability gradient, Legal Theory — core principles — “mercy lives in the people, not the code”, The Constitutive/Elective Distinction, Ch. 5 — Children — “neglect is not harm”, now placed
LZ: many Austrians treat “people ought not violate rights” as controversial, holding that there can be a virtuous crime — the man who grabs a would-be suicide at the cliff’s edge. Rothbard’s lifeboat: eight seats, more than eight people. LZ: unsatisfactory conclusions are not refutations, and warping ethics to intuitions is “the mystic power of whim.” The boat has an owner; his rules govern (“whatever arbitrary rules he sets ought be abided”); if he is dead and set none, the boat is abandoned and first come, first served — “tough luck! Life isn’t fair and neither is law — the prime goal of law is justice not fairness.” Rothbard’s “normal vs abnormal situations” would be two spheres of ethics, and no principled normal/abnormal line exists, so that is particularism. His drowning-children case is answered by a universal principle: “guardians ought save their children from drowning insofar as they have the ability to” (“not proven — this is a law course”). Rothbard’s “moral course of action vs the rights of the case” is read as hedging — narrowing the legal theory so it “may be slotted into any general ethical theory” — not as endorsing two oughts; Rothbard’s “even if Smith should throw someone out, he is still liable, and the owner may pardon or not” shows the theory holds regardless. Then: “Ought is ought, there is not ought-but-personal and a separate ought-but-political.” The prudential “you ought wear a coat” is an is in disguise; “I may want to rape a baby but it does not follow that I ought obey [the whim].”
(vault) — Rothbard’s answer is the vault’s, and LZ’s reading of it concedes the structure he then denies. Rothbard’s lifeboat verdict — Smith is liable; the owner or heir may forgive or prosecute — is the legal-theory README’s principle 4, “mercy lives in the people, not the code,” exactly. And Rothbard’s distinction between what Smith should do and the rights of the case is the scope confusion page’s enforceability gradient: one ought for the ethics, a narrower enforceable subset for the law. LZ reads this as Rothbard hedging rather than distinguishing, then supplies the very distinction himself: the drowning-children principle is a positive ethical duty (“guardians ought save”), while ch. 5 made a guardian’s refusal to feed legally harmless. Ethics has positive duties; law does not. That is two tiers — not two oughts, but one ought with two enforcement regimes — and it settles the “neglect is not harm” test the vault’s way: the asymmetry is the enforceable tier, not the whole ethics. He has the structure and refuses the name. The suicide-grab then falls out cleanly on the vault’s reading (a crime; restitution owed; forgiveness near-certain) and awkwardly on his: the jumper’s will is known (ch. 5 — overriding a stated will is criminal), so the grab is a crime, and since there is one ought and crimes ought not be committed, one ought not save the jumper. He does not say this; it is what “ought is ought” plus ch. 5 entails.
LZ (vs Konrad Graf and the “value-free” Hoppeans): Graf says legal theory is a categorical, praxeological account of what infringes the NAP, advisory to ethics, not a sub-field of it. LZ: “law is pointless if it is not a subset of ethics — A is a crime, but so what?” Hoppe himself says his proof “remains entirely in the realm of is-statements.” LZ’s reply is a new syllogism: (1) “I ought engage in aggression X” → p ∧ ¬p (a priori true is-statement); (2) ¬(p ∧ ¬p); (3) ∴ “I ought not engage in aggression X” (a priori true ought-statement). “¬NAP → p∧¬p, ∴ ¬(¬NAP) = NAP — but how could this be if facts and values ‘belong to different logical realms’?”
(vault) — two errors, one of them the standard one. (a) The ought is in the premises. Premise 1 is a statement about an ought-proposition; it is not ought-free — it says a certain ought-claim entails a contradiction. Modus tollens then returns the negation of that ought-claim. Nothing crossed from is to ought; an ought went in and an ought came out. Hume’s point was never that you cannot reason among oughts. (b) The deontic slip. What modus tollens delivers is ¬(“I ought aggress”) — it is not the case that I ought to aggress. LZ writes down “I ought not aggress.” Those are different: ¬O(A) is no obligation to aggress (aggression is at most permissible or optional); O(¬A) is an obligation not to aggress. The gap between them is the entire content of the NAP, and the syllogism steps over it in the notation itself (“¬(¬NAP) = NAP” treats the NAP as a proposition, not a norm). Add the ch. 2 point that “I ought aggress” only contradicts anything when it is argued — as a proposition it is consistent — and premise 1 is false as an is-statement about propositions and true only as a claim about performances. The vault’s own is-ought page dissolves the gap differently and more modestly: the questioner is already standing on the ought side; nothing is derived, and the page’s 2026-09-10 correction is precisely that the move shows pragmatic indispensability, not a theorem.
LZ: “a preference for X over ¬X is itself an ethical claim, so choice itself has value implications.” Anyone who evaluates, argues, or acts “is necessarily making the choice of existence over non-existence” — to negate the value of life one must be alive and choosing to remain so, “thus life is undeniably good”; the good = pro-life = in correspondence with reality; evasion = anti-life = incoherent (“is it true that there is no truth?”). Rand: “every is implies an ought”; sunshine → grow crops in the sun, put in windows, wear sunscreen — “all these evaluations are demanded by the cognitions involved”; these oughts “cannot be coherently negated.” Error is bad; the honest mistake is forgivable, the “crusading irrationalist” (Peikoff on Marxists and inherently dishonest movements, at length) is wicked. Hume’s guillotine “implies the complete demolition of ethics to the chattering applause of the most evil philosophies known to man.” A world of total unreason would be Mao, Hitler, Stalin; a primitive rational society would flourish. Peikoff: truth is good, falsehood evil; the boss who says “your idea is correct — good day” like a dead fish vs the one who evaluates. “Every crime implies an evasion of objective law, evasions are per se errors, errors are per se immoral. The true is the good, and the good is the true.”
(vault) — this is Chris’s repair to Rand, without the discipline; and his own examples are hypothetical oughts. The Objectivism Boundary records exactly this move as Chris’s: “I always crossed the ‘if you care to live’ with a performative contradiction — if you don’t care to live, that is pretty easy to fix, and yet you are not doing it.” The antecedent is revealed by continued action rather than asserted. And that page fixes the scope in the same paragraph: pragmatic indispensability, not semantic self-defeat — the denial can be coherently stated, and the suicide actually lives it, so it does not prove anyone ought to live; what it shows is that you cannot coherently argue you have not chosen life while sustaining it. LZ takes the same argument to “life is undeniably good” — the overextension the vault corrected on 2026-09-10 in its own pages. Then his examples do the vault’s work for it: “other things being equal we ought grow our crops in the shine” is conditional on the end of life and on ceteris paribus — a prudential, hypothetical ought of the very kind ch. 6 dismissed as “an is in disguise.” Every substantive ought in the chapter has the form given the choice to live, … — and Peikoff’s own sentence says so (“if one chooses to live and to be objective”). So “ought is ought” resolves, by his own examples, into one hypothetical ought conditional on agency — which is Trunk 1, the vault’s constitutive/elective position (mind-dependent and constrained). He has been arguing against the vault’s framework from inside it since ch. 1; this is the chapter where the frame is visible in his own sentences.
(vault) — “error is per se immoral” is a doctrine, not a derivation, and it moralises disagreement. Evasion is immoral follows from the life-base for the evader; honest error he exempts; the difference between the two is an intent inquiry into the philosopher’s mind (mens rea for ideas, one chapter after mens rea for acts). Applied as Peikoff applies it — Marxist academics are “crusading irrationalists,” followers “willfully self-deluded” — it makes every persistent philosophical opponent wicked, which is the closed-system reflex LZ himself criticised in the video (“Randroids … screaming about rationalism”). It is also the mechanism behind Chris’s earlier observation that “pragmatism” is a slur in his world: on this doctrine, an error of method is a moral fault, so an opponent’s method is denounced rather than tested. The vault’s symmetric-grounding rule (ground-truth both sides to equal depth) exists for the opposite reason. And “the true is the good” as a biconditional proves too much: a true fact about a bacterium’s replication is not “good” for the man it kills; the Peikoff passage he quotes only claims that given the choice to live, every truth bears on it — which is the conditional again.
(Chris) — closing read. This is fair, and the last chapter makes it solid. When libertarian ethics is challenged — save the drowning child, the ball on the lawn — the answer tends to bottom out in “well, you just should.” That has drawn enough pushback that many libertarians now treat the theory as a legal theory, not an ethical system — which is fair. (vault) Note what that retreat is: it is Konrad Graf’s position (§2 above) — law as a categorical account of what infringes the NAP, advisory to ethics — the position this chapter exists to refute. The community has moved to the view LZ argues against, and it moved there because the categorical ought has no answer at the cases where it is needed. “You just should” is the empirical face of a hypothetical ought being asserted as a categorical one.
Rothbard, Hoppe, Kinsella, Block and others saying “law is a subset of ethics.” (vault) Nothing here changes the ledger: the vault agrees law ⊂ ethics; the dispute was never the subset claim but the one-ought claim, and on that the quotes are silent.
What holds. Law ⊂ ethics (the vault’s position too); Rothbard read charitably, and correctly, as narrowing the legal theory rather than splitting the ought; the lifeboat verdict (liable, pardonable — “mercy lives in the people”); the rejection of a normal/abnormal line as particularism; the reply to Graf’s “value-free” praxeology — a right no one ought to respect is not a right.
What does not. The syllogism embeds the ought it derives and slides from ¬O(A) to O(¬A) (§2). “Life is undeniably good” is Chris’s repair to Rand overextended past the scope the vault fixed for it (§3). Every substantive ought in the chapter is hypothetical on the choice to live — “ought is ought” collapses into Trunk 1 by his own examples (§3). “Error is immoral” moralises disagreement and installs mens rea for ideas (§3). And the chapter asserts a positive ethical duty (save your drowning child) alongside ch. 5’s legal harmlessness of neglect — the enforceability gradient, used and unnamed (§1).
One-line read for the hub: the categorical ought never arrives — what arrives is Chris’s repair to Rand without its scope discipline, and every ought in the chapter is conditional on life; the course ends inside the vault’s framework, calling it something else.
philosophy, objectivism, libertarian-law, natural-law, law, ethics