What this covers
Barry Weingast joins Russ Roberts to challenge the standard definition of law as rules enforced by centralized state power. Instead, Weingast argues that law is fundamentally defined by a set of legal attributes—generality, universality, and prospectivity—plus the presence of a legal steward function. This framework allows genuine law to emerge and persist in decentralized communities without government coercion. The conversation moves through historical cases where law arose without sovereign enforcement: medieval Iceland, ancient Greece, and the California Gold Rush. In each, communities developed rules to govern disputes and allocate resources, then enforced them through social pressure and graduated sanctions rather than state machinery.
The discussion traces how law coordinates expectations, drawing on Hayek's insight that rules matter not merely as commands but as coordination devices—what others expect to happen shapes what you must do to avoid danger or breach. Weingast contrasts this with the natural state, the historical norm in which restricted access to power and dispersed violence potential produce chronic disorder. Two practical threads run through: decentralized collective punishment systems, illustrated in Robert Ellickson's study of Shasta County ranchers who enforced norms through gossip and livestock trespass before threatening dire consequences; and the contrast between legal orders and the aid-industry approach that transplants legislatures and constitutions into societies lacking the conditions for them to function. The speakers also examine why trading societies and small, stable groups more readily develop law, and why separating law's nature from its enforcement mechanism matters for understanding how stable rules actually take root.
Weingast argues that law is defined not by centralized state enforcement but by a set of legal attributes (generality, universality, prospectivity) plus a 'legal steward' function, so genuine law can and does emerge in decentralized communities without government coercion.
- Law's defining features are legal attributes plus a steward, which are conceptually separate from the enforcement mechanism
- Historical cases (medieval Iceland, ancient Greece, California Gold Rush, Shasta County) show law emerging without centralized coercion
- Hayek's insight that law coordinates expectations explains why decentralized rule systems function
Law coordinates expectations through stewards who publicly interpret rules for unforeseen cases, enabling decentralized communities to enforce order without central authority.
- What distinguishes law from informal social norms is the presence of an official 'steward'—judges, a council of elders, or even the men who meet at the diner—who makes common-knowledge announcements about how to extend existing rules to unforeseen circumstances where self-serving parties each propose interpretations favoring themselves.
“there's some kind of official steward. That is, there's someone or some group or some body--maybe they are judges, maybe they are a council of elders, maybe the men in the community that meet at the diner every morning for breakfast.”
- In stable communities, the key feature distinguishing legitimate enforcement from feud is that no punishment occurs without the community agreeing on what to do; an individual who simply retaliates because they were hurt is engaging in a feud, and the whole purpose of law is to substitute legal reasoning for such vengeance.
“the key is, is that no punishment occurs without the community agreeing on what to do. So it's not an individual that's saying, 'Oh they hurt me; I'm going to retaliate.' That's a feud.”
- Law above mere social norms requires serious punishment that can be administered through decentralized collective punishment—a graduated hierarchy where, in Ellickson's Shasta County study, recalcitrant ranchers first face gossip, then have cattle driven into the hinterland, and ultimately are warned that the volunteer fire company might not arrive in time to save their house.
“there we do see a hierarchy of punishments... The first is the community starts to gossip about you. If that doesn't work then you can't find some of your cows... the volunteer fire company couldn't guarantee that if there was, if his ranch caught on fire, that they couldn't get there on time to save the house.”
- In Hayek's vision, judges (the stewards) do not invent new law but discover or find existing law by understanding what the litigating parties and others expected; because no judge can know every expectation, this is an ongoing evolutionary process that never reaches maturity as the changing world makes previously appropriate rules obsolete, as seen in the explosion of intellectual property rights over the last 25 years.
“Those who first attempted to express the rules did not invent new rules but were endeavoring to express what they were already acquainted with; and this endeavor is that of discovering something which exists not as one creating something that is new.”
- Following Hayek, law is not only rules of conduct but rules that coordinate people's expectations: the relevant question is not just 'if the speed limit is 65, how fast should I drive?' but 'how fast are other people going to be driving?'—because I need to know others' expected behavior before deciding my own, and this coordination function is the real key to the rule of law.
“In some situations, keeping that so-called law, abiding by that legislation, is a violation of that law. Because everybody expects to go 72... If you are going 58, you are actually making it dangerous for people.”
- Even modern legal systems have no unique steward; there are state courts supervised by federal courts and the Supreme Court, legislation by Congress, presidential signatures and vetoes, executive orders, and regulatory agencies all making rules, yet they fit together as a functioning whole.
“There's not a unique steward here, either. We've got many different levels of courts... There's legislation that's passed by Congress... And then of course there are executive orders. So there are a large number--and then there are all these regulatory agencies out there making rules. So our society has no unique steward. And yet they all fit together”
Law requires three defining attributes: rules must be general, universal, and prospective, distinguishing legal order from personal dictatorship.
- A dictator's commands are not law because cronies receive different deals based on how powerful they are, which violates generality (a name is attached to particular privileges) and universality, making the system personal rather than legal.
“if different cronies get different deals by virtue of how powerful they are, then that's certainly not general because that means there's a name attached to those particular kind of privilege.”
- Legislation written in general language can still violate the principle of generality when it is clearly designed to cover a single specific circumstance, as with the Munn v. Illinois grain elevator law applying to cities over 100,000 (only Chicago) or the TARP electric-car subsidy that effectively applied only to General Motors.
“the law was written as, every city that had more than 100,000 people in the state of Illinois were subject to this regulation. Every single one of them. Chicago.”
- Law is differentiated from other forms of order (especially dictatorship) by attributes: rules are general (apply to a wide variety of circumstances), universal (apply to everybody not just specific named people), and prospective (known in advance, not made up after the fact), along with consistency, publicity, and feasibility.
“law has certain characteristics that differentiate it from other forms of order, dictators in particular [?]. That is, the rules are general; they are prospective; they are universal.”
- Historically the rule of law has meant rules that are known in advance, stable, general, and apply universally—the legal attributes; but in the last 25 years there is a tendency to lump all desirable things (good governance, equity) into the rule of law, which conflates good things with what the rule of law actually is.
“historically, the rule of law has meant that the rules are known in advance; they are stable; they are general... In the last 25 years there's a tendency to lump all good things that people want--such as good governance, equity, and the like--to be considered as part of the rule of law... they are not the rule of law.”
Most societies remain natural states with restricted access, dispersed violence, chronic instability, and low accumulation; developed societies supporting property rights are rare exceptions.
- In natural states, the opportunity to grow through savings or innovation is low because accumulation cannot be relied upon to persist; if you lack the violence potential to protect yourself, accumulating wealth makes you a target, so the natural state tends to be poor.
“If you don't have violence potential to protect yourself, you become a target if you accumulate.”
- Societies with dispersed violence potential struggle to sustain stable rules because adaptive efficiency requires adapting to changing circumstances, and when circumstances change the relative capacities of groups shift; for cooperation to persist each group with violence potential must consider itself better off under peace than violence, which credible-commitment problems and asymmetric information disrupt.
“in order to cooperate and have peace they need to consider themselves better off under peace than violence. And one of the problems with changing circumstances is that the relative capacities change all the time.”
- Developed societies that support property rights, contracts, stable political systems, and economic growth are relatively rare—no more than about two dozen on the planet, roughly 10-15% of all countries—while the rest remain natural states.
“We now talk about 'developed societies,' but there's not more than--many more than--2 dozen on the planet. So, approximately 10-15% of all countries. All the rest are a very different form of natural states.”
- Most societies throughout history have been 'natural states' that limit access to organizations, restricting competition in politics and economics, distributing a set of privileges, and dispersing violence potential, producing chronic disputes and violence; these are by far the most prevalent and long-lived form of government in human civilization.
“most countries, most societies in the world have been what we call 'natural states.' These are states that limit the access, limit access to organizations, so there is limited competition in both politics and economics. There's a set of privileges and the violence potential is widely dispersed.”
- Decentralized emergent legal orders are rare and observable precisely because there aren't hundreds of thousands of them; in most cases a local strongman or dictator arises producing a hierarchical, non-legal system, so the named historical examples are exceptional rather than typical.
“the places where you see this--there's a reason why you can name these examples. And that is because there aren't hundreds of thousands of them. This is not the typical case.”
Legal orders emerge from social transformation of violent disputes into nonviolent bargaining, as demonstrated by Gold Rush mining camps that created rules for claims and conflict resolution.
- Law can be characterized as the slow progression of transforming violent disputes into nonviolent ways of bargaining conducted in some public fashion, as in Peter Stein's work, and this is what emerged in the Gold Rush camps that needed to defend against both internal disputes and external raiders.
“Peter Stein, whose 1984 book on Legal Institutions characterizes law as the slow progression of transforming violent disputes into nonviolent ways of bargaining that involve some form of public fashion.”
- Mining camps formed during the Gold Rush primarily for protection against outsiders and to allocate claims given high population density, generating general rules like a maximum 15-square-foot claim per miner and the 'Shovel Rule' (Jackass Gulch Rule) allowing five days' absence before a claim is deemed abandoned, with exceptions for circumstances like a broken leg.
“each miner is allowed at most 15 square feet to dig in... the Shovel Rule, which I think has this name 'Jackass Gulch Rule' or something, which has to do with: you are allowed to be away for 5 days. If you wait more than 5 days then it is considered that you have abandoned your claim.”
- Law with the three defining features can emerge in communities lacking centralized enforcement, as in the California Gold Rush of 1848, where there was no territorial government to create or settle property rights yet a set of rules emerged to govern mining claims.
“the Gold Rush occurred in 1848 and California was not organized even as a territory at the time... there was no territorial government, and so no place to create property rights much less settle disputes over property rights... a set of rules emerged in this case that have the 3 features we define as law.”
- The efficiency of coordinated traffic rules becomes visible when they fail: when a power outage turned a complex Stanford intersection's lights to flashing red, lines stretched for miles because every car had to stop and there was confusion over whose turn was next, demonstrating that established expectations, not the physical infrastructure, do the coordinating work.
“a private plane one day crashed into the lines leaving the power plant in Palo Alto, and the electricity in Palo went out. And so the lights were flashing red. And the lines were miles long. Because nobody could get--every single person had to stop, wait their turn; there was confusion as to whose turn was next.”
Trading societies escape natural states by creating law to facilitate exchange; ancient Greece emerged through commerce and private adjudication rather than centralized courts.
- Societies like ancient Greece emerged from the natural state because they were trading states; facilitating trade and getting rich beyond a self-sufficient landlord's level required a different set of rules, leading them to create law with legal attributes and stewards and a much wider body of citizens.
“I think it had to do with that they are trading states. And that in order to facilitate trade and get rich beyond what you could do by being a self-sufficient landlord, requires a different set of rules.”
- Ancient Greece had law but no judges; it used a system of private adjudication where private individuals were responsible for apprehending wrongdoers, prosecuting the case, bringing evidence and witnesses, and sometimes even dishing out punishment, mixing private and public ordering.
“what they have is a system of private adjudication, whereby individuals, private individuals are responsible for apprehending wrongdoers, prosecuting the case, making the case, and sometimes even dishing out the punishment.”
- Athenian juries used hundreds of jurors (often 201, drawn from a pool of 6000 male citizens) rather than a handful because large numbers reveal what the average citizen expects in a given situation, serving the law's function of coordinating expectations rather than risking peculiar outcomes from a small sample.
“part of the reason for having large numbers of jurors is that that way you are getting a sense of what the average Athenian citizen expects in this situation.”
Aid organizations transplant developed-world legal institutions into natural states without understanding that law requires dispersed violence potential, stewards, and legal attributes to function.
- Aid organizations (World Bank, IMF, USAID) design legal systems based on a flawed conception of law as sovereign commands, leading them to transplant developed-world institutions (legislatures, constitutions, separation of powers) into natural states where dispersed violence potential makes such institutions fail, because they lack the necessary legal attributes and steward function and ignore that law is separate from enforcement.
“they have a very different conception of what the law is. And as a consequence I think the nature of the systems they design are bound to fail because they don't have all the features that are necessary.”
- The way rules are enforced is secondary to the nature of law itself; what matters is people obeying rules so as to create order, and the enforcement mechanism can be separated conceptually from the legal attributes, a point Hayek makes in Law, Legislation, and Liberty.
“this is something that Hayek mentions in Law, Legislation, and Liberty book--that the way the rules are enforced are secondary in comparison to the nature of people obeying rules so as to create order.”
Successful constitutional settlements persist when populations accept them as legitimate and defend them; stability requires limiting stakes of power to prevent coups.
- A rule for creating stability is that the political bargain should limit the stakes of power; if a society can limit how much is at stake in controlling government, it lowers the chance of disruptive coups, as illustrated by Chile in 1973 where a left government threatening property rights led large landowners to support a bloody military coup.
“one of the rules, I think, of creating stability is the nature that the bargain should limit the stakes of power... if societies have the ability to limit the stakes then they lower the chances that this kind of coup will happen.”
- Chile's democratic transition around 1989-1990 succeeded because the outgoing dictatorship still controlled the army, forcing the left to accept the Pinochet constitution with veto enclaves (life-appointed senators giving the right a senate majority); over time as appointed senators died the left gained control through constitutional means and came to accept the constitution as legitimate, illustrating that a successful constitution is one people consider appropriate and defend.
“they accepted the Pinochet constitution, which has a bunch of peculiar features... it creates a bunch of veto enclaves... there were a number of senators that were appointed for life... over time what happens to these senators for life? They're not immortal. And so the left begins to take command over these veto enclaves”
- The two biggest social science literatures on law—Law and Economics, and Positive Political Theory and the Law—both assume a legal order exists and rules are enforced, then ask what the rules should be or how they are chosen, thereby begging the prior question of what law actually is.
“the two biggest literatures in social science on law are Law and Economics; and then Positive Political Theory and the Law in political science and law. And neither of them ask this question. They both assume that there's a legal order and that the rules are enforced.”
Contracts in complex relational deals operate on reputation and expectation-setting rather than court enforcement; norms accompanying new market platforms evolve endogenously.
- Courts are good at enforcing publicly verifiable terms (such as whether a bond paid the prescribed interest on time) but have no clue what parties intended in complex relational deals common among firms, so many such disputes never go to court and contracts function largely to set expectations and operate on a reputational basis.
“Courts are really good at saying... was the bond paid back. Courts are really great at that. When you talk about these complex relational deals that are so common among firms, of course have no clue what the parties intended.”
- New apps like Uber and AirBnB create markets and exchanges between people who wanted to exchange but previously could not, and the norms that accompany these activities are endogenous and will change, with the young adopting them easily while older generations struggle.
“What these are doing is creating markets and exchanges between people who wanted to have exchange but no longer could. And I think that we are just beginning to see where that's going to go.”