communicoupling

The tragedy of the
commons is optional.

In 1968 Garrett Hardin proved, on the back of a napkin, that every shared pasture must die: each herder pockets the whole gain of one more animal and pays only a fraction of the damage, so ruin arrives by arithmetic. Yet Swiss alpine meadows, Spanish irrigation courts, and Japanese village forests ran for centuries — no privatization, no state. Elinor Ostrom went to look, and found what the napkin left out: Hardin's tragedy is the equilibrium of herders who cannot talk. Real appropriators talk, and out of talk they build institutions — boundaries, monitoring, graduated sanctions — that make the commons govern itself.

Her lab confirmed the field: mere face-to-face communication, with nothing enforceable behind it, reliably lifts cooperation in commons games; add low-cost monitoring and sanctions that start gentle and escalate, and cooperation holds. Below is a working pasture: eight harvesters, a stock that regrows by its own law, and Ostrom's machinery as switches. Run the napkin. Then turn the herders' voices on.

monitoring · mchance a rule-breaker is seen
0.50
self-interested · nof 8 harvesters, the rest conditional
3
regrowth · rhow forgiving the resource is
0.30
Institutions
the stock
100
of a carrying capacity of 100
yield efficiency
net harvest vs the sustainable maximum
norm strength
conditional cooperators still complying
The pasture, over time
green: the stock · gold: net yield per round · red ticks: sanctions levied
0 rounds
stock S net yield / MSY sanction
The harvesters
number = current effort (‰ of stock taken) · warm: self-interested · cool: conditional · violet: outsider · red ring: fined this round

quota under communication: per head

Watch what happens
The Ostrom package: communication sets a quota, monitoring at 0.5 sees half of all raids, and sanctions start gentle and escalate. The stock holds near half capacity indefinitely — with these dials it takes roughly six sanctions in four hundred rounds to keep the peace.
dS/dt = rS(1−S/K) − S·Σeᵢ The pasture regrows by its own logistic law and shrinks by the sum of everyone's effort. Whether Σeᵢ stays inside the regrowth is a question no single herder can answer — only the institutions among them can.

the napkin, formalized

The tragedy is real — for herders who cannot speak.

Hardin's argument is a prisoner's dilemma with grass. Each harvester chooses an effort; the whole benefit of extra effort is private, while the depletion it causes is shared eight ways. For a mute stranger, restraint is strictly dominated: whatever the others do, taking more pays more, until the stock crosses the point where regrowth cannot keep up and the equilibrium harvests the pasture to its floor. In the instrument, that is the Hardin's napkin preset: eight self-interested harvesters, no institutions, and with these dials the stock collapses by round six — after which the group scrapes along at a few percent of what the pasture could have yielded forever.

Ostrom's observation is that the napkin smuggles in its conclusion through its assumptions: no communication, no shared rules, no way to see or sanction each other, no boundary around who may harvest at all. Those are not facts about commons; they are facts about strangers. Every long-lived commons she documented had built the missing pieces — clear boundaries around the appropriators, rules congruent with local conditions and made by the people bound by them, monitors accountable to the community, sanctions that begin as a raised eyebrow and escalate only with repetition, and cheap ways to resolve disputes. The tragedy is not the fate of shared things. It is the fate of shared things among people who have been denied every instrument this page lets you switch on.

what to try

Three experiments on the pasture.

01

Run the napkin, then the fix

Press Hardin's napkin and watch six rounds of arithmetic do their work. Then press the Ostrom package and run four hundred: the stock holds near S = 50, yield efficiency sits above 80%, and sanctions are levied only a handful of times. Same grass, same greed — different couplings.

02

Find the deterrence threshold

With graduated sanctions on, sweep monitoring m slowly upward from zero. Raids are deterred when the expected fine beats the raid's gain — with these dials, at roughly m ≈ 0.35. Below it, sanctions fire constantly and fail; above it, they almost never fire and succeed. A working rule is one that rarely needs using.

03

Erase the boundary

From the working Ostrom regime, switch boundaries off. Every 25 rounds a newcomer arrives who harvests outside the rules. The insiders keep paying for monitoring; the pasture collapses anyway — with these dials, around round 77. Unowned is survivable; unbounded is not. Ostrom put boundaries first for a reason.

neither market nor state

Communication is the substrate the institutions are built from.

The textbook offered two exits from the tragedy: privatize the pasture, or put a Leviathan over it. Ostrom's cases are a third thing — self-governance built out of communication. Notice what each switch in the instrument actually is. The quota exists because the harvesters met and made it common knowledge. Monitoring is neighbors seeing neighbors, reported where everyone hears. A graduated sanction is a message before it is a punishment: we saw, we remember, and we are still willing to treat you as one of us. Its gentleness is information, not weakness. The harsh regime preset deters from the first offense but nets far less. Its surveillance apparatus eats a third of the sustainable yield, and fear teaches compliance without teaching membership.

The cheap talk preset shows the machinery's weakest honest setting — and its most instructive one. Communication alone, with nothing enforceable behind it, lifts the group to roughly two-thirds of the sustainable yield at the start: the conditional cooperators comply, and even the self-interested moderate while they think they are watched. Then erosion: unpunished raids accumulate, the cooperators' norm decays round by round, and the pasture eventually goes the napkin's way — just slower. Talk starts the institution; only monitoring and sanction let it persist. For commons too large for a meeting — oceans, the atmosphere, an open-source ecosystem — Ostrom's later answer was polycentricity: many overlapping small governments of the resource, each built on the same conversational machinery, none sovereign over the whole.

the mapping

Mechanism ↔ social life.

In the modelIn the world
the stock SThe fishery, the aquifer, the team's slack, the maintainers' attention.
effort eᵢHow hard each party pulls on the shared thing.
communicationThe meeting where the quota becomes common knowledge rather than one person's wish.
monitoring mWho can see whom defect — neighbors on adjacent plots, code review, the irrigation ditch walked daily.
graduated sanctionsThe raised eyebrow before the fine before the exile; punishment that still carries an invitation back.
boundariesWho counts as an appropriator at all — why "unowned" and "ungoverned" are different words.

where it tears

Limits.

The principles were drawn from small, stable worlds.

Ostrom's cases are villages: repeat play, long shadows of the future, people who cannot avoid each other at church. Scaling the design principles to oceans, the atmosphere, and anonymous digital commons is the open problem — her later polycentric program is a research direction, not a solved theorem, and this simulation's eight named harvesters flatter the easy case.

"The community manages it" is not automatically just.

A boundary that keeps the commons alive also keeps people out, and the insiders' institution can be parochial, hereditary, or cruel while remaining perfectly sustainable. The model scores yield and compliance; it does not score who was never allowed to be a harvester in the first place. Sustainability and equity are separate audits.

Real appropriators strategize about the rules themselves.

These agents play within institutions handed to them by preset buttons. Ostrom's people make the rules, bend them, renegotiate them — rule-making is itself a commons with its own free-riders. That recursion, the heart of her design principle about collective choice, is exactly what a fixed payoff engine cannot contain.