Who belongs, and who decides that

The edge of a community is where its structure is tested, and almost nothing about it is written down in advance.

That is not carelessness. Writing down what would get somebody removed feels, in year one, like planning for a betrayal that has not happened among people you have just chosen. So the question is deferred, and it stays deferred right up until the group is holding a live situation involving a real person with friends in the room.

The four

• The purpose that quietly changed — nobody voted to become something else, and enough people arrived wanting something else that it happened
• The norm nobody could enforce — a commitment central to the community’s identity, with no process behind it
• Rules that spread into private life — the opposite failure, and the one that makes members leave without complaining
• Removal with no process — grounds invented during the case they are used on

Notice that two of these point in opposite directions. A community can fail by enforcing nothing and fail by enforcing too much, and the same missing document causes both. A group with no stated scope has no basis for acting and no basis for restraint, so it oscillates: years of tolerating everything, then one decisive over-correction that costs it three members.

The one that is missing from this list

There is a fifth pattern here that RCOS’s stress tests do not catalogue, and it is worth naming because it decides how much every other rule on this page matters: exit that is not really available.

Nobody writes a rule forbidding departure. What blocks it is structural — an individual stake in the property that only converts to money when a buyer appears, an open-ended notice period, a queue with no position in it. Each part was put there by somebody being reasonable, and together they mean that leaving costs more than staying and disagreeing.

This is documented outside our field. Ran Abramitzky’s study of the Israeli kibbutzim found that a community’s wealth functions as a “lock-in device” that increases the value of staying, and that higher wealth reduces exit while sustaining a high degree of income equality. A stake you forfeit by leaving is a lock-in device whether or not anyone designed it as one.

It matters here because every membership rule on this page is only as strong as the exit behind it. A community whose members cannot leave has not achieved commitment. It has achieved captivity with good intentions, and its agreements are no longer consented to in any meaningful sense — they are merely endured.

Is that the community’s business?

A community has written a scope statement. It governs: the land and buildings, the shared budget, the common meals, the work rota, and who becomes a member. It states as out of scope: members’ incomes, their religious practice, and their relationships. Nothing else is written down anywhere. Five things now happen — decide each one against that document, not against what you would want.

1. A member takes a well-paid job with an oil company. Several members find this incompatible with why they are all here.

Out of scope — they may dislike it, and disliking it is not authority correct answer Right, and this is the case that proves the document is real. Income is explicitly out of scope, so there is nothing to decide and nobody has to win an argument. If the community wants this to be its business, the route is to declare an identity constraint that is testable and to do it in the calm — not to discover one during a case about a named person. In scope — the agreements cover this Nothing in the statement reaches it. Reading a general commitment into a scope that lists incomes as out of scope is exactly the move the document exists to prevent, and the next person it is used on will have watched how it was invented. Partly — separate the part that is governed from the part that is not There is no governed part here. No land, no budget, no rota and no meal is affected — only how a number of members feel about a colleague’s employer.

Where this comes from

2. A member starts a small furniture business, using the shared workshop on weekday evenings and three car-park spaces for deliveries.

In scope — the agreements cover this correct answer Right, and note it is not in scope because it is a business. Their income is nobody’s business; the workshop and the car park are listed, and sustained private use of a shared asset is a decision the community is entitled to make. Framed that way it is an ordinary allocation question rather than a referendum on somebody’s livelihood. Out of scope — they may dislike it, and disliking it is not authority Income is out of scope. The workshop and the car park are explicitly in it, and this uses both, regularly, in a way that affects everyone else’s access. Partly — separate the part that is governed from the part that is not A reasonable instinct, and here the ungoverned part is not really in dispute — nobody is proposing to rule on what they earn. The question in front of the group is entirely about the shared assets.

3. A member decides to home-school their two children. Several others think this is a mistake and that the children will be isolated.

Out of scope — they may dislike it, and disliking it is not authority correct answer Right, and for a different reason from the oil job: this is not on either list. The default rule is what decides it — anything not explicitly declared as in scope is out of scope, so silence resolves in the member’s favour. That is the opposite of how an unwritten arrangement drifts, where silence resolves in favour of whoever feels most strongly. In scope — the agreements cover this It appears nowhere in the statement. A community that treats an unlisted matter as governed has no scope statement in practice, only a list of things it happens to have thought of first. Partly — separate the part that is governed from the part that is not There is no governed part. If the children’s presence at common meals or on the land raised a real question, that would be in scope — but concern about a parenting decision is not that question wearing a disguise.

Where this comes from

4. A member has not done a rota shift in four months and has not said why.

In scope — the agreements cover this correct answer Right, and the easy half is that the rota is listed. The harder half is what the community may do about it, which the scope statement does not answer — this is where a ladder of cheap, early, reversible steps matters, because a group with only one available response tends to wait until it is furious and then use it. Partly — separate the part that is governed from the part that is not Nearly, and worth holding onto: the rota is governed, and why they have stopped may not be. A community can require the shifts without acquiring a right to the reason. Out of scope — they may dislike it, and disliking it is not authority The work rota is on the list. A scope statement that cannot reach four months of unexplained non-participation is not protecting anybody.

Where this comes from

5. A member’s new partner moves into their home. The partner does not come to common meals, does not do rota shifts, and has not applied for anything.

Partly — separate the part that is governed from the part that is not correct answer Right, and doing the separation out loud is what keeps this from becoming a row about somebody’s relationship. The relationship is explicitly out of scope. Who becomes a member is explicitly in it, and so is the rota — so the live questions are what status a resident non-member has and what the household owes, neither of which requires anyone to have a view about the couple. In scope — the agreements cover this Much of it is, and saying so without qualification is how the conversation ends up including opinions about the partner. The statement lists relationships as out of scope, and that clause earns its keep on exactly this kind of evening. Out of scope — they may dislike it, and disliking it is not authority The relationship is out of scope; a person living on the land without any defined status is not. A community with no answer here has an undefined membership state, which is where expulsion-without-process cases usually begin.

Where this comes from

If it is already happening

Write the scope before you write the rules. A community arguing about whether it may have an opinion on somebody’s diet, or their guests, or their income, is arguing about scope with only a specific case in front of it — which is the worst possible moment. What resolves it is a statement of what the community governs and, crucially, what it does not.

Handle the live case with what you have, and say so. If somebody has done something the group cannot tolerate and there is no process, the honest move is a process invented transparently for this case, stated as such, and reviewed afterwards. That is far better than pretending a rule existed, and far better than doing nothing.

Then write the general version afterwards, not during. Rules written in the middle of a case are written about a person, and everybody knows it — including the next person they get applied to.

What prevents it

Three documents most communities have none of: a scope statement saying what is governed and what is not, identity constraints stated so they could actually be tested, and an exit and separation protocol covering both directions — voluntary and forced.

A fourth thing is worth separating while you are in there. Most communities hold one founding statement that is simultaneously why they exist, what they hope to become, and what they are doing about it this decade — which means the plan cannot be revised without reopening the purpose, or the purpose moves whenever the plan is updated. Vision, purpose and mission are three different documents with three different change rules, and pulling them apart is a morning’s work.

Sources & further reading

1. 1 RCOS stress tests — Membership & Boundaries — the four documented patterns this lesson covers, in specification form
2. 2 RCOS Core — Layer 0: Identity & Scope — purpose, scope, and identity constraints that have to be testable
3. 3 RCOS Core — Layer 1: Membership System — membership states, rights and obligations, and the rules on exit
4. 4 Abramitzky, R. — The Limits of Equality: Insights from the Israeli Kibbutz (QJE, 2008) — community wealth as a lock-in device that raises the value of staying
5. 5 Membership: getting in, and getting out — the joiner-facing version — the four stages and what a fair agreement contains

Written by EcoHubs members, with AI assistance for drafting and editing, and reviewed by a person before publication. Facts are checked against the sources listed; anything we could not verify is marked. How this is written