Haute Lumière
Commerce · VI.04 · MMXXVI · daylight
The Commons That Held
Volume VI — Governance and the Commons
THE PLATE
THE LETTER
You have been told that the commons is a tragedy, and then — if you read a little further — that Elinor Ostrom disproved it. Both of those are summaries, and a summary is a poor instrument to build with.
This chapter is not about whether commons can work. That question is settled and it is settled in VI.02, which sets out Ostrom's eight design principles and the meta-analytic record behind them; it is not re-derived here. This chapter asks a narrower and more useful question, and it asks it the way a lawyer reads case law rather than the way an advocate reads an example.
Take the commons that actually held — for five hundred years, for eight hundred — and read them as a body of decided cases. What does the rule say, word for word? What does it cost to enforce, in money or in hours? Who specifically enforces it, and what do they get for doing so? And what happened the last time somebody tested it in earnest?
Those four columns produce something the inspirational literature does not: a number. The enforcement ratio — the annual cost of keeping a rule, divided by the annual value of the resource it keeps. Call it E/V. It is the figure a modern designer needs before anything else, because it decides whether a governance system can be paid for out of the thing it governs, or whether it must be subsidised by somebody outside — at which point the outsider governs.
We will compute E/V for three commons in three centuries on three continents. We will also do the arithmetic that spoils the mood, which is survivorship: these are the ones still standing, and we do not know the size of the graveyard. That is the honest negative, it is unavoidable, and it has a research design attached so that it does not have to stay true.
— The Editors
DISCOVERY
Five that held, read as case law
Valencia: the Tribunal de les Aigües de la Vega de València.
Every Thursday at noon, in front of the Apostles' Door of Valencia Cathedral, eight elected síndics sit down in a half-circle and hear water cases. The court is entirely oral. It is conducted in Valencian. There is no written pleading, no lawyer, no transcript of argument and no appeal. A case is called, the accused answers, the síndic of the canal concerned withdraws from the vote, the others decide, and the sentence is spoken aloud. The whole thing frequently takes less time than reading this paragraph would take aloud.
The rule. Each of the eight canals — Quart, Benàger-Faitanar, Tormos, Mislata, Mestalla, Favara, Rascanya, Rovella — draws from the Turia in a fixed order, and each irrigator takes water in a fixed turn. The offence is almost always the same: taking water out of turn, or taking more than the turn allows.
The document. The irrigators' right to administer their own water dates in written form to the settlement after 1238, confirmed in the Furs de València from 1239 — 787 years of continuous documented jurisdiction. The tradition claims a caliphal origin around 960; that claim is argued rather than established, and Glick's Irrigation and Society in Medieval Valencia treats it with the care it deserves. We do not use it in any arithmetic here.
The test. Twice in living memory. The Spanish state, codifying its courts in the Ley Orgánica del Poder Judicial of 1985, could have absorbed the tribunal; the statute instead names it as a customary and traditional court and leaves it alone. Then the city itself: Valencia's expansion has eaten the huerta the canals irrigate, and the answer, after decades of pressure, was the Generalitat's Llei de l'Horta de València of 2018, which protects the irrigated land as a landscape with a law rather than protecting the tribunal with a preservation order. The rule survived by having its resource defended, not itself.
Törbel, in the Valais.
Robert Netting spent years in a Swiss alpine village of a few hundred people and produced in Balancing on an Alp the most carefully documented commons in the literature. Ostrom built on it.
The document. On 1 February 1483 the villagers of Törbel drew up articles of association governing the alp, the forest and the waste lands, restricting rights in them to citizens of the village — 543 years ago, and the village's land documents run back to 1224. The rule that matters most entered the record in 1517 and is one sentence long.
The rule. No citizen may send more cows to the alp than he can feed through the winter.
Read it again, because it is a masterpiece of instrument design and almost nobody notices why. It does not cap the herd. It does not meter the grass. It ties summer extraction to a winter asset that is visible from the road. You cannot hide hay. The audit happens in February, on foot, by people already walking past the barns.
The test. Swiss federal agricultural policy. When Switzerland moved to direct payments, summering support came to be paid per normalised stocking unit, with the stocking of each alp inventoried and both over- and under-grazing penalised — the regime consolidated in the Direktzahlungsverordnung of 23 October 2013. The federal government, looking for a workable rule five centuries later, adopted the commons' own: count the animals against the capacity, and pay against the count.
The iriai forests of Japan.
Iriai (入会) is the customary right of a village to enter and harvest a commons — fuel, fodder, thatch, timber. At the Meiji cadastre the area under such tenure was on the order of 12,000,000 hectares. Something like 3,000,000 hectares of iriai-derived tenure survives.
The rule. The mountain is closed. It opens on stated days, announced in advance, and on an opening day the whole village goes up at once, at a signal, with a stated number of people and tools per household, for a stated number of hours. Margaret McKean's fieldwork on the village codes shows this pattern across regions that had no contact with one another.
The document. The Civil Code of 1896 carried iriai forward into modern Japanese law, defining it by local custom rather than by a national rule — a legislative decision to let eight hundred village codes each remain the law of their own mountain.
The test. The Act on Promotion of Modernization of Rights Relating to Iriai Forests, passed in 1966, offered villages a route out: convert the common into co-ownership or a production forest partnership, with state assistance. Many took it. Many did not. That is the cleanest natural experiment in this entire literature — one offer, one date, two outcomes — and we return to it in the Arithmetic, because it is also the best available answer to survivorship.
The English commons that survived enclosure.
Roughly 6,800,000 acres of England — 2,751,862 hectares — were enclosed by act of Parliament. About 370,000 hectares of registered common land remain in England today. The interesting question is not why so much went. It is why any stayed.
The rule. From the Statute of Merton of 1235 — 791 years — English law allowed a lord to approve, that is to enclose, part of the waste provided sufficient pasture remained for the commoners. Common rights attach to land, not to persons, and they do not lapse merely because nobody exercised them last season.
The test, and it is the finest piece of case law in this chapter. In 1874 the Corporation of London, wanting to save Epping Forest, did not petition. It bought the rights of commoners in the forest and sued as a commoner — Commissioners of Sewers of the City of London v. Glasse. Sir George Jessel, Master of the Rolls, held the enclosures unlawful and ordered thousands of acres thrown open. The Epping Forest Act 1878 then vested the forest in the Corporation as conservators for the public, in perpetuity.
One commoner was enough. The right was not majoritarian, not proportional and not extinguishable by disuse; a single holder of a right of lopping was standing to void twenty years of enclosure. Every commons designer should carry that sentence.
The later documents matter too, and not all of them were kind: the Commons Registration Act 1965 required rights to be registered by a deadline and extinguished those that were not, and the Commons Act 2006 reopened and tidied the register. Most recently, the Supreme Court in May 2025 upheld the public right to camp on the Dartmoor commons under the Dartmoor Commons Act 1985, against the owners of the soil.
The copyleft commons.
The modern one. The GNU General Public Licence, first published in 1989, governs a body of software on which most of the world's infrastructure runs. Thirty-seven years is not seven hundred, but it is four times the working life of most corporate governance structures and it has been tested in court.
The rule. You may use, copy and modify without asking. If you distribute, you distribute the source under the same terms. That is it — one conditional obligation attached to one event.
The test. Welte v. Sitecom Deutschland, Landgericht München I, 19 May 2004: the first judicial enforcement of the GPL anywhere. Later, in Artifex Software v. Hancom (N.D. Cal., 2017), a United States court accepted that breaching the licence could also be breach of contract. The commons won both times, and it won by being a boundary rule that a judge could read in an afternoon.
THE ARITHMETIC
What it costs to hold a commons, and what we cannot know
The instrument: E over V.
E / V = annual cost of enforcement
--------------------------
annual value of the resourceA commons pays for its own government out of the thing it governs. When E/V is small, the users can carry it without anybody's permission. When E/V is large, somebody outside must pay — and whoever pays, governs. That is the whole reason to compute it.
Valencia, worked. The eight canals irrigate about 11,000 hectares. Put intensive huerta gross output at €8,000 per hectare per year — an assumption, labelled as one — and the resource is worth €88,000,000 a year.
Now the cost side. Eight síndics, fifty-two Thursdays, about two hours each with travel: 832 hours a year, valued at €25 an hour, is €20,800. Sixteen canal guards at €32,000 fully loaded is €512,000. Administration, the alguacil, the records and the premises: €120,000. Total €652,800.
E / V = 652,800 / 88,000,000 = 0.74 %Move the output assumption and the answer moves with it: at €12,000 a hectare, 0.49 percent; at €5,000, 1.19 percent. Under one percent to keep seven centuries of water law running.
A New Mexico acequia, by a different route. The same rule family crossed the Atlantic and is now enforced by labour rather than by salary. Take a ditch with sixty parciantes, each irrigating about $9,000 of output a year — $540,000 in all. Enforcement is the annual limpia, one compulsory day of ditch-clearing per parciante, sixty days valued at $150 each, plus a mayordomo's stipend of $2,500. E is $11,500, and E/V is 2.13 percent — of which the labour is 1.67 and the stipend 0.46.
The copyleft commons, by a third route. Four thousand kernel contributors in a year, at an average quarter of a working year each and $250,000 fully loaded, is $250,000,000 of annual effort. The entire enforcement apparatus — the licensing organisations, the compliance staff, the litigation — is on the order of $4,000,000 a year. E/V is 1.60 percent.
Three cases, three continents, three methods of payment, and a band of 0.49 to 2.13 percent.
And here is the discipline, because this is exactly where an instrument starts lying. One person chose every assumption in those three calculations in one sitting, with the band already in mind. The convergence is therefore a hypothesis, not a finding, and it will stay a hypothesis until somebody reads documents that share none of its assumptions. Four exist and are public: the annual budget of a Valencian comunidad de regantes; the accounts of a Swiss alp corporation under the direct-payment regime; the Software Freedom Conservancy's Form 990; and Maine's Marine Patrol appropriation set against the Department of Marine Resources' landings value. Any one of those turns the band into evidence or kills it. Until then it is a well-made guess, and it is labelled as one in lib/verify/VI_04.py as well as here.
Why the ratio is low: monitoring is a by-product of use.
The surviving commons do not buy surveillance. They arrange the work so that watching happens anyway.
Törbel. Suppose you wanted to police the alp directly: a hundred days of summering at two hours of deliberate stock-counting a day is 200 hours. The wintering rule replaces that with sixty households' hay counted at a quarter of an hour each — 15 hours, a 13.33× reduction — and those hours are spent on a path the counters walk anyway.
Iriai. The mountain is open ten days and closed 355. On a closed day, moving a cartload requires passing sixty households making two trips each — 120 observations. At a modest five percent chance that any one passer-by notices, the chance of moving unseen is 0.95 to the power of 120, which is 0.0021. Detection 99.79 percent, with nobody employed to watch. The path is the monitor.
Valencia. The rotation itself is the audit. The irrigator whose turn is next is standing there, because it is his water.
The Balenciaga cut: the survivors are the ones that punish least.
Every designer's instinct on being shown a rule that holds is to ask how hard it bites. Look at the actual penalties. Valencia's fines are small. Törbel's were small. The iriai codes fined in sake and confiscated the day's cut. The commons that lasted seven hundred years are the ones with the gentlest first sanction in the room.
That is not softness. It is arithmetic. Deterrence is probability times severity, and the two are not interchangeable in the way the product suggests.
commons p = 0.95 S = 5.00 E[penalty] = 4.75 sd = 1.09
distant regulator p = 0.02 S = 250.00 E[penalty] = 5.00 sd = 35.00The same expected cost. A severity ratio of 50×, and a variance ratio of 32.12×. A small certain fine is a cost a household budgets for, like fuel. A large improbable one is a gamble, and a household near its margin will take that gamble, because the distribution of outcomes now contains a branch where nothing happens at all.
And there is a second, structural reason the survivors kept the fine small, which is the one a designer can use. A penalty large enough to ruin a neighbour cannot be levied by a neighbour. It requires a court, a court requires evidence, evidence requires a record, a record requires delay — and delay destroys the certainty that was doing all the work. Valencia keeps no transcript and allows no appeal precisely because its sentences are too small to be worth appealing. The gentleness of the penalty is what keeps enforcement inside the village, and enforcement inside the village is what makes E/V less than one percent. Raise the fine and you export the enforcement, and the moment you export it you are paying for it.
The honest negative: survivorship, sized.
Everything above is drawn from commons that are still here. We have no register of the ones that are not, and the ones that are not may have looked identical from the inside.
We can at least size the bias. In England, 2,751,862 hectares were enclosed by act against 370,000 hectares of registered common land remaining: 11.85 percent of that area still common, over the 276 years since the wave began — an annual loss of 0.77 percent. In Japan, 3,000,000 hectares of 12,000,000 survive — 25 percent over the 153 years since the Meiji cadastre, an annual loss of 0.90 percent.
Two national series, measured by different people for different purposes, agreeing on the order of magnitude. The implied half-life of commons area is 89.7 years in England, 76.5 in Japan, 83.1 as a mean.
Törbel's charter is 543 years old. That is 6.53 halvings, which puts it — crudely, as an order of magnitude and nothing better — at about 1 in 93.
Read that correctly or it becomes a different kind of lie. Both series measure area, and area can be lost while an institution holds; conversely a commons can retain every hectare and lose its rule. The two national estimates were not measured here. What the number is good for is one thing only: the cases in this chapter are drawn from the extreme tail of a distribution whose body nobody has observed. Anything this chapter says about what the survivors share is a description of the tail. It may be the cause of survival. It may be a correlate. It may be what tails look like.
What would settle it, and each of these is a real, executable programme:
- The English register as a cohort. The Commons Registration Act 1965 gave a hard deadline and therefore a dated census. Take every unit registered by 1970, and count how many still have active commoners in 2026. That is a denominator with a birth date, and it exists.
- Japan's 1966 Act as a natural experiment. One offer, one date, villages that converted and villages that did not, with ministry records of both. Compare survival by which arm they fell into.
- The Swiss summering inventory. Switzerland counts alp operations annually for payment purposes. That is a whole population with entries and exits, not a sample of successes.
- The IFRI and Nepal irrigation databases built by Ostrom's group, which were designed from the start to include systems that failed.
Until one of those is run, the correct posture is the one this chapter takes: describe the survivors precisely enough to be contradicted, and say plainly that the base rate is unknown.
DREAM
What becomes ordinary
In the version of this that has already happened, a governance system arrives at its first review with an enforcement ratio on the front page, and everyone in the room knows what a good one looks like.
The fisheries council, the housing co-operative, the data trust, the shared compute pool and the catchment partnership all report the same figure in the same units: what it cost us to keep our own rules this year, and what the thing we were keeping was worth. Nobody finds this unusual. It is simply what a constitution's accounts look like.
Rules are written so that compliance is visible as a by-product of ordinary use. The question who will watch this has stopped being answered with a job title and started being answered with a mechanism: the rotation, the shared opening day, the winter count, the published log. When somebody proposes a rule that would need a new inspector, the room asks what the Törbel version of it would be — the rule that binds the thing you cannot hide — and usually finds one.
First sanctions are small, immediate and public, and nobody mistakes that for weakness. Escalation exists and is written down, but the first rung is deliberately light enough to be applied by a peer on the same afternoon, because everyone has read the arithmetic and knows that certainty is doing the work.
Commons carry their founding document with a date on it, and the date is on the wall. New members are handed the document rather than the culture, because a culture cannot be handed over and a document can. When the rule is changed, the change is dated too, and the old text is kept.
And the state, when it arrives, arrives the way the Spanish state arrived in 1985 and the Swiss state arrived with its stocking units: it recognises the rule the users already keep, and it pays against their count. That turns out to be cheaper for everybody, including the state, and the reason is on the front page of the review.
DESIGN
The four columns, and how to fill them for something you are building
Read as case law, the five survivors give a specification. It has four parts and each part is a document, not an intention.
1. A boundary with a date. Not the community. A written statement of who holds the right, made on a day, kept, and amendable only by a stated procedure. Törbel has 1 February 1483. Valencia has the Furs. Iriai has the village code and then the Civil Code's articles on common rights. The copyleft commons has a licence text with a version number. If you cannot name the date of your boundary document, you do not have a boundary; you have a mood.
Write it so that a single member's right is sufficient standing to challenge a breach. That is the Epping Forest lesson and it is the cheapest enforcement mechanism ever designed: one person, unanimity not required, disuse no defence.
2. A rule that binds a visible proxy. Do not regulate the thing you cannot see. Find the quantity that (a) moves with extraction and (b) is impossible to conceal, and bind that. Winter hay for summer grazing. Distribution for copying. The turn in the rotation for the water taken. Ask of every draft rule: what would somebody have to hide, and could they?
3. A forum that meets on a schedule, hears orally, and decides the same day. Thursday at noon. The opening-day assembly. The annual ditch meeting before the limpia. A forum that meets when there is trouble is a forum that meets during trouble; a forum that meets on a fixed day mostly meets and finds nothing, which is how it stays cheap and how everyone stays practised.
4. A first sanction small enough for a peer to apply today. With an escalation ladder behind it, published, and used rarely. The first rung is not a deterrent — the certainty is the deterrent. The first rung is a ritual of noticing, and its job is to make the fact that somebody noticed common knowledge.
The sequence, for something being built now.
Days 1–30. Write the boundary document and date it. List every candidate rule. For each, name the visible proxy, or discard it.
Days 31–60. Compute V — the annual value of the resource, honestly, in the units the users already use. Then cost E for each candidate rule: who watches, for how many hours, paid how. Any rule whose E/V exceeds a few percent is rewritten or dropped at this stage, not after it has failed.
Days 61–90. Set the forum's day and hold the first one with no business on the agenda. Publish the sanction ladder. Register the boundary document somewhere outside your own control — a public register, a notary, a repository with immutable history — because a boundary document that only you hold is a boundary document only you can change.
DESTINY
How it holds when nobody is pushing
Three things keep these systems running after the founders, and all five cases have all three.
The rule pays for itself out of the resource. E/V under a few percent means no external funder is needed, and no external funder means no external veto. This is the load-bearing one.
The forum meets whether or not there is business. Valencia sits on Thursdays in weeks with no cases at all. An institution that only convenes for trouble loses the practice of convening, and then trouble finds it without a venue.
The document outlives every member. Nobody in Törbel signed 1483. That is the point: the rule is inherited rather than renegotiated, and inheritance is much cheaper than consent.
Now the failure modes, named.
It fails when V collapses. This is the most common modern death and it is not a governance failure at all. When the harvest stops being worth more to the users than their next-best use of the same hours, E/V goes to infinity however well the rule is drafted, and the commons converts into an amenity that must be paid for from outside. Much of England's registered common land is in exactly this state: the rights exist, the graziers do not. The condition this whole chapter needs, and which is usually absent in modern settings, is a resource whose annual harvest is still economically live for the people governing it.
It fails when the sanction is exported. Raise the first penalty past what a peer can apply and enforcement moves to a court, cost rises by an order of magnitude, certainty falls, and the rule quietly stops being kept.
It fails when the visible proxy stops being visible. Mechanise the harvest, move the work indoors, or let the rotation be recorded privately, and the free monitoring evaporates. The rule has not changed; its enforcement has just acquired a budget.
It fails when the boundary document is lost, or was never dated. Then every dispute becomes a dispute about the rule rather than under it.
And it fails when an outside appeal becomes cheaper than compliance. If a member can get a better answer from a distant authority than from the forum, the forum is decorative within a generation. Valencia has no appeal. That is not an archaism; it is the load-bearing beam.
DELIGHT
What it feels like
There is a particular pleasure in a court that takes four minutes.
Somebody is accused of taking water on Tuesday that was not theirs. They say what happened. Two other men say what they saw. The síndic of their own canal turns his chair slightly away and does not vote. The others say a number. It is over, and everyone walks back to the square in the heat, and the man who was fined stops for coffee with the man who reported him because they are neighbours and will be irrigating next to each other on Thursday.
The delight is in the proportion of it. Nothing is inflated. No one has been made an example of, no one has hired anybody, no one has written anything down except a figure in a fine book. A small wrong was noticed publicly and corrected immediately, and then the day went on — which is, when you have watched the alternative, an almost luxurious way to live.
That is what a governance system at 0.74 percent feels like from the inside. It feels like almost nothing at all, which is exactly why it is still there.
OPERATIONALIZE THIS
At the level of finance
The instrument: a commons charter with a funded enforcement line and a sanctions escrow.
You are constituting a shared resource — a co-operatively held asset, a shared data or compute pool, a catchment, a fishery, a building, a jointly maintained codebase. The structure below is what turns a memorandum of understanding into something a treasurer can underwrite.
The structure. A charter (in England and Wales, most naturally a company limited by guarantee or a co-operative society with the charter as its rules; in the United States, an LLC with the charter as the operating agreement; for a software commons, a licence plus a fiscal host). Three things are in the rules rather than in a policy document, because policies are amendable by whoever is in the room: the boundary and its amendment procedure; the forum's fixed meeting day; and the sanction ladder, with the first rung applied by a peer.
The balance-sheet treatment. The resource is held by the entity, not by the members. Members hold a right of use appurtenant to their membership, which is not a security and is not transferable separately from it — that non-severability is the Commons Act 2006 lesson and it prevents the slow leakage of rights to people with no stake in the resource. Enforcement is an operating expense, budgeted annually, disclosed as a single line. Disclose E next to V. A commons whose accounts do not show what its own government cost is not governing itself; it is being governed by whoever absorbs the cost.
The sanctions escrow. Fines are not revenue and must not be. Pay them into a restricted fund applied to maintenance of the resource, with the schedule published. The moment fines fund the enforcers, the enforcers acquire an interest in infractions, and every user can see that they have.
The counterparty. For the first facility, the members themselves — a levy per unit of use, set to cover E with a modest reserve. Only once you have two years of a stable E/V should you take the charter to an external lender or a public funder, and then the ratio is the credit case: a borrower whose governance costs under two percent of the resource it governs is a borrower whose covenants can be monitored cheaply.
The first ninety days.
| Day | Action | Artifact |
|---|---|---|
| 1–15 | Draft and date the boundary document; name every right-holder | The dated charter |
| 16–30 | List candidate rules; name the visible proxy for each | Rule table with proxies |
| 31–45 | Compute V in the users' own units | The valuation note |
| 46–60 | Cost E per rule; cut any rule that cannot be watched cheaply | The E/V sheet |
| 61–75 | Publish the sanction ladder; open the escrow account | Ladder and escrow |
| 76–90 | Hold the first forum on its fixed day, with nothing on the agenda | Minutes of an empty session |
The number that decides it. One figure, on the front page:
annual cost of enforcement
E / V = --------------------------- < 0.02
annual value of the resourceTwo percent is the top of the band this chapter computed, and the band is a hypothesis rather than a finding — so treat the threshold as a design target and a prompt, not a law. If your draft governance cannot be kept for two percent of what it keeps, the rule is wrong before the people are. Go back and find the proxy you cannot hide, because that is the move that took Törbel's monitoring from 200 hours to 15.
And the second number, which decides whether the first one will survive: can a peer apply the first sanction today, without anyone's approval? If no, your enforcement is already exported and your ratio will not hold.
APPRECIATIVE QUESTIONS
Twelve, for a room
Discovery — what is already working
- Where in this organisation does somebody already notice a breach without being paid to look? What is it about how the work is arranged that makes them see it?
- Think of a rule here that everybody keeps without anyone checking. What is the visible thing it is attached to?
- What is the oldest written agreement we still actually follow, and who wrote it? What has kept it readable?
Dream — what becomes possible
- If our governance cost under two percent of what it governs, what could we afford to hold in common that we currently cannot?
- Imagine the forum that meets on a fixed day and usually finds nothing. What would we start bringing to it once we trusted it to be there?
- If every member's single right were sufficient standing to challenge a breach, what would change about how carefully we draft?
Design — what we build
- For the rule we most want people to keep: what would somebody have to hide in order to break it, and could they? If they could, what is the version that binds something they cannot hide?
- What is the smallest first sanction a peer could apply this afternoon without anyone's approval — and what is it about it that makes the noticing public?
- What is the date on our boundary document, and where is it held that is outside our own control?
Destiny — how it holds
- What would have to be true for this to still be running when nobody here remembers writing it?
- If the value of the resource fell by half, which of our rules would still pay for themselves — and what would we do about the rest, on purpose, in advance?
- Where could a member get a faster answer from outside than from us, and what would it take to make our forum the quicker route?
WORKS CITED
Ostrom, E. (1990). Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge University Press.
Cox, M., Arnold, G. and Villamayor-Tomás, S. (2010). "A Review of Design Principles for Community-based Natural Resource Management." Ecology and Society, 15(4): 38. [The design-principle record is treated in VI.02 and is not re-derived here.]
Glick, T. F. (1970). Irrigation and Society in Medieval Valencia. Harvard University Press.
Maass, A. and Anderson, R. L. (1978). …and the Desert Shall Rejoice: Conflict, Growth, and Justice in Arid Environments. MIT Press.
Ley Orgánica 6/1985, de 1 de julio, del Poder Judicial (Spain), art. 19.3.
Llei 5/2018, de 6 de març, de la Generalitat, de l'Horta de València.
UNESCO (2009). Representative List of the Intangible Cultural Heritage of Humanity: Irrigators' tribunals of the Spanish Mediterranean coast.
Netting, R. McC. (1981). Balancing on an Alp: Ecological Change and Continuity in a Swiss Mountain Community. Cambridge University Press.
Swiss Confederation. Verordnung über die Direktzahlungen an die Landwirtschaft (Direktzahlungsverordnung, DZV), SR 910.13, of 23 October 2013.
McKean, M. A. (1992). "Management of Traditional Common Lands (Iriaichi) in Japan," in Bromley, D. W. (ed.), Making the Commons Work: Theory, Practice and Policy. ICS Press.
Civil Code of Japan (Act No. 89 of 1896), arts. 263 and 294.
Act on Promotion of Modernization of Rights Relating to Iriai Forests (Act No. 126 of 1966), Japan.
Statute of Merton 1235 (20 Hen. 3).
Commissioners of Sewers of the City of London v. Glasse (1874–75) LR 19 Eq 134.
Epping Forest Act 1878 (41 & 42 Vict. c. ccxiii).
Shaw-Lefevre, G. (Lord Eversley) (1910). Commons, Forests and Footpaths. Cassell.
Neeson, J. M. (1993). Commoners: Common Right, Enclosure and Social Change in England, 1700–1820. Cambridge University Press.
Thompson, E. P. (1991). Customs in Common. Merlin Press.
Turner, M. (1980). English Parliamentary Enclosure. Dawson.
Commons Registration Act 1965; Commons Act 2006; Dartmoor Commons Act 1985.
Darwall v. Dartmoor National Park Authority [2023] EWCA Civ 927; [2025] UKSC 20.
Rodgers, C. P., Straughton, E. A., Winchester, A. J. L. and Pieraccini, M. (2011). Contested Common Land: Environmental Governance Past and Present. Earthscan.
Rivera, J. A. (1998). Acequia Culture: Water, Land, and Community in the Southwest. University of New Mexico Press.
Welte v. Sitecom Deutschland GmbH, Landgericht München I, 21 O 6123/04, judgment of 19 May 2004.
Artifex Software, Inc. v. Hancom, Inc., No. 16-cv-06982 (N.D. Cal. 2017).
Kuhn, B. M., Sebro, A. K. and Gingerich, D. Copyleft and the GNU General Public License: A Comprehensive Tutorial and Guide. Software Freedom Conservancy, successive editions.
Becker, G. S. (1968). "Crime and Punishment: An Economic Approach." Journal of Political Economy, 76(2), 169–217.
Baland, J.-M. and Platteau, J.-P. (1996). Halting Degradation of Natural Resources: Is There a Role for Rural Communities? Clarendon Press.
Agrawal, A. (2001). "Common Property Institutions and Sustainable Governance of Resources." World Development, 29(10), 1649–1672.
Poteete, A. R., Janssen, M. A. and Ostrom, E. (2010). Working Together: Collective Action, the Commons, and Multiple Methods in Practice. Princeton University Press. [On case-selection and the limits of what a case literature can establish.]
Note on figures. Every figure in this chapter is computed in lib/verify/VI_04.py and reprinted there in the form the page shows it. The enforcement ratios rest on declared assumptions, each labelled ASSUMED, APPROX or ORDER OF MAGNITUDE in that module; the band they produce is a hypothesis and the four published documents that would test it are named in the module and in the Arithmetic. The survivorship arithmetic uses area, not institutions, and says so where it is printed.