Navigate
HomeStart here
MusingsResearch & long-form
BuildingProjects & learnings
WorkProfessional practice
RunningTraining & races
AboutValues & identity
Life & PlacesCulture, food, travel, cities
Notes & ArchiveJournals, essays, portfolio
Research essayKnowledge & ResearchUpdated March 2026Law-and-game-theory synthesis
Musings

Knowledge & Research · March 2026

Law for States: How Public Law Works Without a Sovereign

Breaking down Goldsmith & Levinson's Harvard Law Review paper (2009)[1] that argued international law and constitutional law face the same structural problems — and solve them the same way.

Why This Paper

Goldsmith & Levinson do something rare — they take two fields that barely talk to each other (international law and constitutional law) and show they're solving the same puzzle. This page is my attempt to map the argument, connect it to trade practice, and make the theory tangible. WTO disputes, Section 232 tariffs, IEEPA rulings — the exact enforcement gap G&L diagnose, playing out in real time.

The Question

Why does law constrain states — domestically and internationally — when states hold the monopoly on force?

HLR 2009
Published in
Harvard Law Review
3
Structural problems
Uncertainty, Enforcement, Sovereignty
12+
Theorists engaged
Hobbes to Ackerman
70+
Years of doctrine
1945 UN Charter to present
01

The Argument in 60 Seconds

Five-bullet executive summary

Expand
02

The Three Problems

G&L identify three structural problems shared by international and constitutional law. Each tab shows the problem, its domestic parallel, and a concrete case.

International Law

What does a treaty actually require? The ICJ interpreted the UN Charter's prohibition on “the use of force” in Nicaragua v. US (1986)[2], but the US rejected the court's jurisdiction. Who decides what the law means when there is no supreme court?

Constitutional Parallel

What does the Constitution actually require? The Second Amendment’s "well regulated Militia" clause has been interpreted as both an individual right (Heller, 2008) and a collective right. The Supreme Court decides, but 5-4 splits show how thin the margin of certainty is.

Case in Point

WTO Appellate Body: The US blocked appointments in 2019, effectively killing the WTO's “supreme court.”[3] Trade disputes are now unresolvable through the institution designed to resolve them — proving that uncertainty is not just theoretical.

03

The Theorists

G&L engage 12 thinkers across political science, legal philosophy, and constitutional theory. Click any card to see who they are, what they claim, and how G&L respond.

Thomas HobbesSocial Contract
John AustinLegal Positivism
H.L.A. HartLegal Positivism
Hans MorgenthauClassical Realism
Kenneth WaltzStructural Realism
Alexander WendtConstructivism
Robert KeohaneNeoliberal Institutionalism
Harold Hongju KohTransnational Legal Process
Thomas FranckLegitimacy Theory
James MadisonConstitutional Design
Bruce AckermanConstitutional Moments
Louis HenkinInternational Legal Process
04

The IR Theory Landscape

International relations scholars disagree on two fundamental questions: are state preferences fixed or socially constructed, and is cooperation rare or frequent? G&L move the conversation from the realist pessimism corner toward the institutionalist-constructivist zone — and then argue the same move is needed in constitutional theory.

Figure 1

IR Theory Landscape: Where Scholars Stand

Preferences exogenousPreferences endogenousHow are state preferences formed?RareFrequentCooperationMorgenthauClassical RealismWaltzStructural RealismKeohaneNeoliberal Inst.WendtConstructivismKohTransnationalAustinPositivismFranckLegitimacyThe G&L move

G&L move the conversation from the realist corner, where cooperation is rare and preferences are fixed, toward the constructivist-institutionalist zone, where cooperation is frequent and institutions shape preferences. Their key insight is that this move applies equally to domestic constitutional law.

05

Game Theory Meets Trade

The classic prisoner's dilemma explains why trade wars happen even when both sides lose.[5] Without institutional mechanisms, each country has an incentive to protect its own industries while hoping the other stays open. The WTO solves this by creating an iterated game with reputational stakes.

Figure 2

Trade as Prisoner's Dilemma: The WTO Payoff Matrix

Country BLower TariffsRaise TariffsCountry ALower TariffsRaise TariffsMutual GainBoth lower tariffsA: +3 / B: +3(Pareto optimal)Sucker's PayoffA opens, B protectsA: -1 / B: +5(A loses industries)Free RideA protects, B opensA: +5 / B: -1(B loses industries)Trade WarBoth raise tariffsA: +1 / B: +1(Nash equilibrium)WTO makesthis movepossible

Without institutions, both countries defect and raise tariffs: the Nash equilibrium. The WTO creates an iterated game with transparency, dispute resolution, and reputational costs, enabling the move to mutual cooperation. G&L's point is that constitutional structures do the same thing domestically through credible commitment mechanisms.

G&L's Insight

The WTO doesn't enforce trade law through force — it creates an institutional structure where cooperation becomes self-enforcing through repeated interaction, dispute resolution transparency, and reputational costs. This is exactly how constitutional checks and balances work: no branch has an army to enforce its judgments, but the institutional architecture makes defection costly.

In Practice

This is the Section 232 / WTO dispute risk in abstract form. When the US imposed steel tariffs under Section 232 “national security” authority[6], it was a defection move in the iterated game. The WTO dispute panels ruled against it — but the US ignored the rulings. G&L would say: the enforcement gap trade lawyers navigate every day is the same gap constitutional lawyers see when the executive ignores judicial rulings. Same problem, different courtroom.

06

The Precommitment Problem

The theory of precommitment begins with Ulysses.[7] In the Odyssey, Ulysses knows that when his ship passes the Sirens, their song will overwhelm his judgment. He will want to steer toward them. He will believe, in that moment, that it is the right thing to do. So he orders his crew to bind him to the mast and plug their own ears with wax. He constrains his future self — not because he distrusts himself in general, but because he knows that in a specific, predictable situation, his preferences will betray his interests.

This is the logic of constitutional precommitment. A society writes down its highest commitments — free speech, due process, equal protection — precisely because it knows that future majorities, under pressure, will want to abandon them. The Bill of Rights is the mast. Article V, which makes amendment nearly impossible, is the rope. The point is not that future generations are stupid. The point is that the situations in which rights are most needed are exactly the situations in which popular will is most likely to override them.

G&L's insight is that this logic extends identically to international law. A treaty is a precommitment device: a state binds its future administrations to trade rules, human rights norms, or arms limitations because it knows that future leaders, facing domestic pressure or geopolitical temptation, will want to defect. The dead-hand problem — why should the past bind the future? — is the same in both domains. And so are the answers: Madison says make the binding nearly impossible to undo; Ackerman says allow democratic override in extraordinary moments.

Figure 3

The Precommitment Problem: Binding Future Sovereigns

UnconstrainedSovereign(full discretion)RatificationConstitution / Treaty(self-binding act)ConstrainedFuture Sovereign(bound by prior act)Dead Hand ProblemWhy should the past bind the future?Madison: Article V(near-impossible amendment)Ackerman: Moments(democratic override)Treaty analog

G&L's key insight is that the dead-hand problem is identical in both domains. A constitution adopted in 1787 binds citizens in 2026; a treaty ratified in 1995 binds future administrations. Madison's near-impossible amendment rule and Ackerman's democratic override theory both have international analogs.

07

The Fiat Money Analogy

This is the emotional center of the paper. G&L argue that law works like fiat currency — it has value because everyone acts as if it has value.

The Dollar

A $100 bill is a piece of paper. It has no intrinsic value. It works because the Federal Reserve, commercial banks, merchants, and citizens all act as if it's worth $100. The shared belief is self-reinforcing: because everyone accepts it, everyone has reason to accept it.

The Law

A constitutional provision or treaty has no physical force. It constrains states because executives, legislators, judges, and citizens all act as if it constrains them. Compliance is the equilibrium — not because of enforcement, but because everyone expects everyone else to comply.

When Belief Breaks Down

Hyperinflation (Weimar, Zimbabwe, Iran) shows what happens when shared belief in currency collapses. The legal analog: when states stop acting as if law binds them — Russia annexing Crimea, the US ignoring ICJ rulings — the system doesn't crash overnight but erodes incrementally.

“The truly unsettling implication is not that international law might fail — it's that constitutional law operates on the same fragile foundation. Both are fiat systems. Both depend on collective belief. And belief, once lost, is extraordinarily difficult to rebuild.”

The fiat money parallel is mine — G&L lay the groundwork, but this is where I think the argument lands.

08

Key Cases & Moments

1945

UN Charter signed. Prohibition on use of force (Art. 2(4)) and self-defense exception (Art. 51) established.

1945

ICJ Statute establishes the International Court of Justice. No compulsory jurisdiction — states must consent.

1950

European Convention on Human Rights. First binding international human rights instrument with an enforcement court.

1986

Nicaragua v. United States. ICJ rules against US mining of Nicaraguan harbors. US withdraws from ICJ compulsory jurisdiction.

1995

WTO established. Dispute Settlement Understanding creates binding trade adjudication — the closest thing to international enforcement.

1998

Rome Statute creates the International Criminal Court. US, Russia, China refuse to ratify.

1999

Kosovo intervention. NATO acts without UN Security Council authorization. "Illegal but legitimate" — the enforcement gap in action.

2003

Iraq War. US/UK bypass UNSC for invasion. International law's enforcement failure at its most visible.

2008

Medellín v. Texas. SCOTUS rules ICJ judgments not directly enforceable in US courts. The domestic-international gap, crystallized.

2009

Goldsmith & Levinson publish 'Law for States' in the Harvard Law Review. The paper that unifies these threads.

2019

US blocks WTO Appellate Body appointments. The trade system's "supreme court" effectively dies.

Feb 2026

SCOTUS strikes down IEEPA tariffs 6-3. Statutory precommitment constrains executive power — the dead hand working as designed.

09

Competing Explanations

Three frameworks for why states comply with law — each with different implications.

Austin Is Right

International law is not real law. States comply when it serves their interests and defect when it doesn't. The WTO, the ICJ, and the UN are forums for power politics dressed in legal clothing. Compliance without enforcement is just coincidence.

Implication: invest in power, not institutions.

Koh Is Right

Law works through internalization. Transnational legal process embeds norms in domestic institutions, bureaucratic routines, and professional culture. States comply because their own lawyers, agencies, and courts enforce international norms from the inside.

Implication: invest in legal infrastructure and institutions.

Morgenthau Is Right

Law reflects the balance of power, period. The US follows WTO rulings when convenient and ignores them when not. The system “works” only to the extent that powerful states find it useful. When they don't, it doesn't.

Implication: law is a tool of the powerful, not a constraint on them.

10

Theory Meets Practice

Three connections between G&L's framework and what trade lawyers deal with every day.

1

The Enforcement Gap Is Your Business Model

G&L show that international law lacks centralized enforcement — and that’s exactly why trade lawyers exist. The job is to navigate the gap between what the WTO rules say and what states actually do. The Section 232 steel tariffs, the IEEPA executive orders — each one is a state testing whether the system will enforce its own rules. Trade law practice doesn’t enforce international law; it makes the non-enforcement manageable.

2

Precommitment Devices Are What You Draft

Every treaty provision, every MFN clause, every dispute settlement mechanism you negotiate is a precommitment device — an attempt to bind future governments to current agreements. G&L would say you’re doing exactly what Madison did with Article V: creating structures that constrain future sovereigns. The dead hand problem isn’t academic for you — it’s the core of treaty drafting.

3

The Fiat Money Problem Is Getting Worse

When the US blocks WTO Appellate Body appointments, ignores ICJ rulings, and imposes tariffs under emergency powers that courts later strike down — it’s eroding the shared belief that makes the system work. G&L’s fiat money analogy says the system can absorb individual defections, but cumulative erosion is catastrophic. We’re watching this happen in real time. The question is whether the belief can be rebuilt, or whether we’re in the early stages of a run on the bank.

11

Sources

Goldsmith & Levinson, “Law for States: International Law, Constitutional Law, Public Law” — Harvard Law Review, Vol. 122, 2009

The paper itself. The unified theory of public law that bridges international and constitutional fields.

Thomas Hobbes, Leviathan (1651)

The foundational text on sovereignty and the state of nature. G&L’s starting point for why law without a sovereign seems impossible.

John Austin, The Province of Jurisprudence Determined (1832)

The command theory of law. G&L’s primary foil — they argue Austin’s definition fails at the constitutional level.

H.L.A. Hart, The Concept of Law (1961)

Primary and secondary rules framework. G&L accept the framework but argue international law has more secondary rules than Hart acknowledged.

Hans Morgenthau, Politics Among Nations (1948)

Classical realism. The strongest challenge to international law’s relevance — and the one G&L take most seriously.

Kenneth Waltz, Theory of International Politics (1979)

Structural realism. Anarchy as the defining condition. G&L use this as baseline, not conclusion.

Alexander Wendt, “Anarchy Is What States Make of It” — International Organization, Vol. 46, 1992

Constructivism’s founding text. G&L draw on constructivist epistemology throughout.

Robert Keohane, After Hegemony (1984)

Neoliberal institutionalism. How cooperation persists without a hegemon. G&L’s closest intellectual ally.

Harold Koh, “How Is International Law Enforced?” — Indiana Law Journal, Vol. 74, 1998

Transnational legal process. The internalization mechanism that makes law work from the inside out.

Louis Henkin, How Nations Behave (1979)

The empirical observation that launched modern IL scholarship: “almost all nations observe almost all principles of international law almost all of the time.”

Footnotes

[1]Goldsmith & Levinson. “Law for States: International Law, Constitutional Law, Public Law.” Harvard Law Review, Vol. 122, 2009.The core paper. Argues that international law and constitutional law face identical structural problems — uncertainty, enforcement, and sovereignty — and should be studied as a unified field of public law.
[2]ICJ. “Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of America).” 1986.The ICJ ruled the US violated international law by supporting Contra rebels and mining Nicaraguan harbors. The US withdrew from compulsory ICJ jurisdiction in response — a defining moment for the enforcement gap.
[3]Pauwelyn. “WTO Dispute Settlement Post 2019: What to Expect?” Journal of International Economic Law, Vol. 22, 2019.Documents the WTO Appellate Body crisis after the US blocked new appointments starting in 2017, rendering the body non-functional by December 2019. The institutional collapse G&L's framework predicted.
[4]Medellín v. Texas, 552 U.S. 491 (2008).The Supreme Court held that ICJ judgments under the Optional Protocol are not directly enforceable as domestic law. The case crystallizes the gap between international legal obligation and domestic enforcement authority.
[5]Keohane, Robert O. After Hegemony: Cooperation and Discord in the World Political Economy. Princeton University Press, 1984.The foundational text on how international institutions enable cooperation through iterated games, reduced transaction costs, and reputation effects — even without a hegemon to enforce agreements.
[6]WTO Dispute DS564: “United States — Certain Measures on Steel and Aluminium Products.”Multiple WTO members challenged US Section 232 tariffs on steel and aluminum imposed in 2018 under national security authority. WTO panels ruled the tariffs violated GATT obligations, but the US has not complied.
[7]Elster, Jon. Ulysses and the Sirens: Studies in Rationality and Irrationality. Cambridge University Press, 1979.The origin of the Ulysses metaphor for constitutional precommitment. Elster argues that binding future selves is rational when agents can predict their own future irrationality — the logic G&L extend to treaties.

Related Reading

lawinternational-lawconstitutional-lawgame-theoryIR-theory