“What Is Law? A Coordination Model of the Characteristics of Legal Order” by Gillian K. Hadfield and Barry R. Weingast asks whether legal order requires a centralized enforcement agency. The paper argues it does not. It presents a model in which a system of decentralized enforcement, carried out collectively by individual agents, sustains legal order. The article is part of the authors’ broader project to decouple law from the state and to show that the former does not conceptually depend on the latter (see, e.g., Hadfield & Weingast, 2013; Hadfield & Weingast, 2018).
Hadfield and Weingast’s central thesis is that the defining characteristics of legal order do not conceptually depend on centralized, coercive state enforcement. Legal order instead emerges as a coordination equilibrium sustained by decentralized collective punishment, so long as a shared institution provides authoritative, publicly knowable classifications of wrongful conduct. On this view, law functions as a coordination technology. Its core function is not to command through force, but to align expectations.
Methodologically, the paper offers a piece of foundational law & economics modeling. Rather than evaluate the efficiency of particular legal rules, Hadfield and Weingast adopt a rational-choice framework to explain the emergence of legal order itself. They rely on standard tools—repeated games, belief updating, and equilibrium analysis—but deploy them at a highly abstract institutional level. The aim is not empirical prediction, but an existence proof: to show that legal order with rule-of-law characteristics can, in principle, arise without centralized enforcement.
The paper proceeds in three steps. First, the authors construct a formal repeated-game model in which enforcement is entirely decentralized and individual agents hold heterogeneous, private views about what counts as wrongful behavior. Second, they show that an equilibrium with effective deterrence can still be sustained if a third-party institution supplies a common logic for classifying conduct, even though that institution lacks enforcement power. Finally, they demonstrate that the institutional features required for this equilibrium—generality, impersonality, stability, openness, and publicity—closely mirror the attributes traditionally associated with the rule of law.
Why Anyone Enforces the Rules
Importantly, the model is not anarchic. It posits a single institution responsible for promulgating rules and determining compliance, while enforcement remains decentralized. That structure raises a central question: why would individuals participate in collective punishment whenever the “judiciary” declares that someone violated its rules? The answer contains the paper’s central insight.
The authors argue that the “judiciary” applies its own criteria of right and wrong—criteria that differ from, but partially overlap with, those used by individual agents. Individuals have an incentive to join collective punishment only when that overlap is sufficiently high. Only then will punishment deter the behavior they themselves want deterred.
Participation also depends on expectations about others. Punishment is effective only if exercised collectively, so an individual must believe others will join. That requires the judiciary’s criteria to overlap not just with one participant’s standards, but with those of other participants across the system.
This dynamic creates an additional incentive to participate in enforcement. By joining collective punishment, individuals signal that the overlap remains high enough to justify continued participation in the system.
Why the Rule of Law Emerges
After presenting the model, the authors ask what properties the “judiciary’s” rules must have to sustain the coordination they describe. They find a striking convergence between those properties and the rule-of-law characteristics long identified by legal philosophers: generality, impersonality, stability, open process, etc. (Lon L. Fuller, 1969; Joseph Raz, 1977). The paper thus offers an explanation for why legal systems display these familiar features.
Much of applied law & economics treats these attributes as background conditions, and jurisprudential theories from H.L.A. Hart to Lon Fuller treat them as conceptual or normative requirements. Here, by contrast, the features arise endogenously from the coordination and incentive constraints facing decentralized enforcers.
Having drawn this connection between decentralized enforcement and the core properties of legal order, the authors pose a final, intriguing question: can these characteristics persist in a system of fully centralized punishment? They conclude that a system relying exclusively on centralized punishment would struggle to maintain them.
What the Model Doesn’t Claim
It is important not to misread Hadfield and Weingast’s argument. They do not claim decentralized enforcement of law is practically desirable or feasible. Their model abstracts away many essential features of social life.
The framework assumes participants have roughly equal power and lack outside incentives that would conflict with those generated within the model. It also assumes no possibility of exit for buyers or sellers, and a substantial commonality of interests and values among participants.
In other words, the model might reasonably describe medieval Iceland, but modern settings rarely satisfy these conditions.
Law Without the State
Still, the paper has real value. Its main contribution is conceptual. A common view holds that law and the state—understood as a system of centralized coercion—are conceptually inseparable, such that one cannot exist without the other (see, e.g., John Austin, 1995).
Hadfield and Weingast’s model challenges that claim. The essential attribute of law, on their account, is its capacity to coordinate. Once coordination is in place, coercion becomes secondary. If law fulfills its coordination function, enforcement can, at least in theory, be coordinated among individuals acting independently of one another.
Social, historical, psychological, and other contingent factors may explain why this rarely occurs in practice. But the paper’s point is narrower: legal order does not conceptually require state enforcement.
Implications, Extensions, and Objections
Hadfield and Weingast’s framework also sheds light on contemporary governance settings where centralized enforcement is weak or absent. Examples include international trade regimes, transnational arbitration, blockchain-based dispute-resolution systems, and informal commercial communities operating across jurisdictions. In these environments, compliance often depends less on state coercion than on shared classificatory institutions—arbitral bodies, industry standards, or reputational intermediaries—that coordinate expectations about acceptable conduct and trigger decentralized sanctions when violations occur.
The paper also opens several avenues for future research. One concerns distributional conflict. The model assumes sufficient overlap in agents’ interests, leaving open how legal order survives when classifications systematically favor some groups over others. Another concerns institutional change: the authors bracket how a common logic emerges, adapts, or degenerates over time. The analysis also invites empirical investigation into which real-world institutions function primarily as classificatory coordinators rather than enforcers, and how they maintain legitimacy without coercive authority.
One criticism is normative. Carmen Pavel (2021) argues a decentralized legal system may lack equality before the law and equal access to remedies. See, however, a recent response by Robert Reamer (2025).
A different, though related, line of inquiry asks whether the mechanism Hadfield and Weingast describe counts as “enforcement of law” at all. The enforcement they model is carried out by private citizens, not public officials, who are not legally required to participate and whose methods and scope of participation are not legally regulated. One might therefore question whether a legal system that lacks rules governing punishment is complete. If not, some form of “executive” authority may remain necessary after all. On this point, see Joshua Kleinfeld’s “Enforcement and the Concept of Law” (2011), which offers helpful insights into what defines enforcement and can be read in dialogue with Hadfield and Weingast’s account.
A Foundational Law & Economics Account of Legal Order
Although the paper engages jurisprudential debates, its analytic core is unmistakably law & economics. It explains legal order in terms of incentives, expectations, and strategic interaction, rather than moral obligation or sovereign authority. Compliance arises not because rules are normatively binding, but because following them is a rational response to others’ anticipated behavior. The paper thus treats law as an endogenous institution whose stability depends on incentive compatibility—the explanatory strategy characteristic of law & economics analysis.
The article does not resemble standard law & economics analyses of tort, contract, or criminal doctrine. It instead fits within the canon at a more foundational, institutional level. In this respect, it aligns with the institutional and political-economy strand of law & economics associated with Douglass North’s work on institutions (1990), Avner Greif’s analyses of self-enforcing governance (2006), and Avinash Dixit’s theory of lawlessness and governance (2011). Like these contributions, the paper treats law not as an exogenous constraint imposed by the state, but as an institution that must be incentive-compatible and self-sustaining. It also builds on the law & economics literature on collective punishment and reputation-based enforcement, most notably Paul Milgrom, Douglass North, and Barry Weingast’s analysis of medieval trade institutions (1990), while generalizing that framework beyond commercial settings to a theory of legal order.
Within the law & economics canon, Hadfield and Weingast’s contribution is distinctive because it shifts attention from the efficiency of legal rules to the preconditions of legality itself. It belongs to a small but growing body of work that treats legal order as an equilibrium phenomenon, rather than a policy instrument. In doing so, it helps bridge institutional economics, political economy, and jurisprudence, and offers a framework likely to remain influential for scholars interested in law beyond the state.
Further Reading:
- Gillian K. Hadfield and Barry R. Weingast, “Law without the State: Legal Attributes and the Coordination of Decentralized Collective Punishment,” Journal of Law and Courts, Vol. 1, No. 1 (2013).
- Joshua Kleinfeld, “Enforcement and the Concept of Law,” Yale Law Journal, Vol. 121, No. 2 (2011).
- Paul R. Milgrom, Douglass C. North, and Barry R. Weingast, “The Role of Institutions in the Revival of Trade: The Law Merchant, Private Judges, and the Champagne Fairs,” Economics and Politics, Vol. 2, No. 1 (1990).
- Carmen Pavel, “The Rule of Law and the Limits of Anarchy,” Legal Theory, Vol. 27, No. 1 (2021).
- Robert Reamer, “Covenants without the Sword: Consent and Enforcement in Market Anarchism,” Legal Theory (2025).
