Libertarian Legal Theory

Applied legal philosophy from a big-L Libertarian framework. Applied Trunk 1.

Links: The Four Trunks, Morality, Scope Confusion, Civilizational Cycles

Trunk: Applied Trunk 1 — Normative Ontology. Legal norms are constitutive of cooperative rule-systems.

Framework

Starting position: big-L Libertarian — Non-Aggression Principle (NAP), natural law, private property as axiomatic. Hoppe, Rothbard, and the argumentation ethics tradition.

Core principles:

Aggression is binary — you violated someone’s property/person or you didn’t. Intent is irrelevant to classification.
Restitution makes the victim whole — automatic, objective, no intent analysis needed.
Retribution deters future violations — set by contract, common law, or victim’s discretion.
Mercy lives in the people, not the code — voluntary human judgment handles nuance the law deliberately excludes.

Framing principle

The current system is closer to libertarian law than libertarians often admit. Marriage, corporations, common law retribution — centuries of practical solutions to the same property-rights problems. The libertarian contribution isn’t “start over” — it’s principled evaluation of what the existing system got right and where it drifted. See Words of Wisdom.

Method

Topics are explored through adversarial debate — Chris takes one position, Claude takes the opposite, and we stress-test from first principles.

Debates

Mens Rea — does criminal intent have a place in libertarian law?
Government Formation and Structure — why government forms (division of labor); the ancap challenge; the Constitution as a distributed system protocol
Aggregation vs. Sorting — Why Collective Choice Keeps Collapsing to Two — the two ways to reconcile many preferences: aggregate into one outcome (voting, provably defective and worse as dimensionality rises — median voter in 1-D, McKelvey chaos in ≥2) or sort people among many outcomes (markets/exit, which removes the theorems’ target rather than beating them). Parties are a dimensionality-reduction device, so unrepresentation is quantization error; increasing returns to coordination (law as a network good) is the one force collapsing both voters onto two parties and legal orders onto one system — Nozick’s dominant-agency argument in economic dress. Closes with: Duverger fixes the cardinality of parties, not their identity, so realignment is a change of basis

Analysis

Registration and Scope Creep — the structural pattern of regulatory expansion beyond original justification; now also innocuousness as a selection filter (the entry stage) and the revocable (may) vs. constitutionalized (must) distinction that decides whether a delegation is a check or a transfer
Lazy Law — (thesis, Chris’s term) banning the tool because the wrong is hard to reach; over-reach plus substitution; a bright line on the act vs. on the tool; the impulsive-act boundary
The Slack in the Law — (legal-theory thesis) enforcement has always been rationed by cost, and the gap between what the law forbids and what is punished is where liberty has been living. A legal system has two independent parameters — how much the code forbids, and how expensive it is to catch you — and every society has been safe because only one was ever turned up at a time. Thick code + expensive enforcement is where we live, tolerable because the slack absorbs the excess; thin code + cheap enforcement is the Misesian ambition and, in Chris’s words, “next door to complete totalitarianism”; thick code + cheap enforcement is the empty cell no society has occupied, and a surveillance layer now makes it reachable — nothing about the law changes, everything about your exposure to it does. Carries the deterrence vs. prevention distinction behind “we could build cars that don’t speed, but is that really reasonable?” (prevention removes the offence and the virtue together, and has no exception handling); ⭐ the observation that automation does not enforce the law but the machine-legible subset of it — detection cost is uncorrelated with severity, so plates and speeds become real offences while fraud and coercion stay as under-enforced as ever; ⭐ the point that many provisions are adjudication defaults, not conduct commands, and an automated enforcer cannot tell the two apart; and four disanalogies to “a cop could have watched you” — cost, discretion, confrontation, aggregation — with red-light cameras as the twenty-year rehearsal and the right to face your accuser as the live objection when the accuser is a machine. ⚠ Registers its own strongest counter as an open question: slack and discriminatory enforcement may be the same phenomenon described by someone with different priors, which puts it in direct tension with Registration and Scope Creep — two-way ledger wired. Specimen: the flock-camera semifinal. From Chris ⭐⭐ Three additions from the drug-legalization thread (2026-09-09), and each upgrades the thesis rather than illustrating it. (1) The criterion the page was missing — who holds the slack, and by what procedure. Chris on decriminalization (“something is either legal or not legal“) is legally right — Gonzales v. Raich means the CSA applies in “legal” states, and only executive discretion keeps the feds out (Cole Memo 2013, rescinded 2018; and since April 2026 federal legality turns on whether your state issued you a medical licence). His objection looks like it contradicts his objection to flock cameras, and doesn’t: jury slack has a designated holder and a public procedure; prosecutorial non-enforcement has neither; automation deletes the gap for everyone. Yields a three-way split the page now runs on — accidental slack is a symptom, designed slack is a feature, undesignated slack is what policy fights are actually about. (2) Proximate cause is the law’s own answer to “everything affects everything.” But-for causation proves too much (“the universe is a closed system… does this mean anything for a society? No”); Palsgraf exists to cut the chain; and it is a standard by necessity — so an enforcement layer that can only apply rules is structurally incapable of the doctrine that stops every prohibition swallowing everything. With three tests that do sharpen it (severability, chain of free choices, ⭐ Fletcher’s reciprocity of risk — the one that bites the driving analogy), and the design space as one inequality: expected social cost ≤ tax ≤ cost of evasion, whose right-hand bound California measured. (3) ⭐⭐⭐ Rules and standards are the endpoints of a dial, and the number of classes is the dial — from Chris’s “the whole class system was an attempt at this… maybe subclass to allow for finer control.” A rule is a schedule with one class; a standard is a schedule with a class per case. Finer classes cut error but spend legibility and add capture surface. And the CSA schedules show the failure subclassing can’t fix: they are a scalar over a four-dimensional risk space, which is why marijuana sat in I with heroin while fentanyl sat in II. Driving already has the fix — class + endorsements, i.e. more ladders, not more rungs. Plus a test for “who controls the licensing?”: is the licence a filter or a quota? Uncapped objective filters are competence checks; capped ones are cartels with a government stamp.
Common Law as a Discovery Procedure — (legal-theory thesis) a legal code is a central plan for conduct and meets the Mises–Hayek calculation objection unchanged — substitute decide every case for price every good and the argument goes through. Adjudication is the institutional answer with a market’s shape: it does not demand the information ex ante but generates it at the dispute, from parties who were there, elicited under adversarial pressure, decided by people embedded in the context — and precedent is the aggregation step, a compressed public summary of many private judgements, what price is to production knowledge. Hayek named the two things: nomos (grown law, spontaneous order) vs. thesis (made law, the plan). ⚠ Four disanalogies stated up front rather than buried — sparse slow feedback (a market with one price update per decade would not clear either), lock-in (stare decisis preserves error as well as knowledge), judges are not price-takers, and the strongest: law needs ex-ante knowability that markets do not — you want the price when you trade, the rule before you act — which is a genuine competing value the mechanism cannot supply, and why every mature system runs statute over common law rather than choosing. ⭐ Sharpens the automation worry: automated enforcement is a mechanism for converting nomos into thesis — it does not repeal the discovery procedure, it starves it, because disputes stop reaching adjudication once the machine has disposed of them, so no precedent forms and no local knowledge is elicited. ⚠ Leads with its own deflater: modern systems are overwhelmingly statutory, so the mechanism may describe a system that no longer exists. From Chris
Corruption — The Floor Under the Mud-Sling — corruption = diversion of entrusted power from its constitutive purpose to a private end; a non-indexical floor orthogonal to legality (the legal×corrupt grid; you can be legal and corrupt, illegal and clean); requires a trust to betray; adversarial ≠ corrupt; the honest-tyrant / charter-vs-institution-type fork; the test that separates a real charge from a partisan mud-sling. Feeds the administrative-agency question
Administrative Agencies and the Power to Execute — the “fourth branch” vs. Article II; a unitary-executive reading grounded in Humphrey’s Executor → Seila → Loper Bright → Trump v. Slaughter (2026, Humphrey’s overruled) + the Fed carve-out; the DACA/removal asymmetry; and the resolution that gap-filling is lawmaking, so the design question is who holds the discretion and can they be fired (accountable executor vs. insulated agency; the nondelegation fork)
The Represented Unit — what a bicameral upper chamber is for (representing the pre-existing power centers whose consent formed the union — Lords/states/Bundesrat), and what breaks when you swap the represented unit: the U.S. 17th Amendment (1913) held Senate allocation but changed what the seats represent, voiding the founding bargain, removing the states’ last voice (after 1865 removed exit), and manufacturing the modern malapportionment complaint. Federalism’s two purposes — proximity (sorting) and surface area (capture-ROI). Sparked by the 17th-Amendment specimen
Defensive Force and Collateral Damage — causationism; defense/retribution distinction; gentleness principle
Self-Ownership and Property Rights — self-ownership→property-rights is the weak link; conflict avoidance grounds both independently
Restitution for Death — the framework’s hardest case; containment vs execution vs indentured restitution

Upcoming

Marriage under natural law
Shared ownership — foil on file: LiquidZulu Course 3 §4, the strongest ancap statement of the no answer (co-owners cannot both win a dispute), whose own company example is co-ownership under one name
Abandonment & residency-franchise — Chris’s rule (title lost only by declaration or death without transfer) and the open renter/hotel/$1-invitee voting lines: Mentis Wave borders review
Fraud as aggression without mens rea
1A scope — “speech” vs “expression”; can 1789 text cover 2026 landscape
2A as preventive check — amendment’s purpose is deterrence, not last-resort revolution
Constitutional analysis as its own page — the document as check against civilizational cycles