From Five People to the First Local Core

Five is a working starting size, not a formula

A small circle has one major advantage: coordination costs are still manageable. People can meet, learn what each person can do and quickly discover whether their agreement works. But research on group performance does not produce one universal optimal team size. The effect of size depends on the task, the coordination burden, the knowledge required and the degree of interdependence.

The Community Tool Box therefore does not prescribe a single number for an initial steering group; it emphasises enough people to do the work, but not so many that coordination becomes the work. This article uses five as a practical beginning, not as a political or organisational dogma.

If a task needs three people, we do not recruit two more to satisfy a formula. If ten distinct skills are genuinely needed, the circle expands. The first test is always: do we have enough capability and a simple enough process to carry out the concrete task?

One of the weakest beginnings for a local core is an overly broad promise: “we will change society”, “we will build a new community”, or “we will unite everyone who thinks like us”. Such goals make a group dependent on identity, rhetoric and internal belonging before it has accomplished anything.

A stronger beginning is one need revealed by “Local Resource Map: What Do We Already Have Around Us?”: organise tool sharing, connect five food producers with nearby buyers, create a phone network for older neighbours, coordinate shared transport or test a backup communication path. The task should be small enough that after thirty days we can show whether anything actually improved.

OECD guidance on participation similarly starts by asking whether there is a genuine problem on which participation can have meaningful influence. This article applies the same sobriety locally: first the need, then the people and process—not the other way around.

Choose reliability and complementarity, not sameness of opinion

The first five people do not have to be five best friends. It is more useful if the group combines different capabilities: someone who understands the problem, someone who can organise, someone with technical knowledge, someone connected to a wider network and someone good at noticing risks others miss. One very ordinary quality also matters: does the person show up when agreed and do what they accepted?

Difference does not mean manufacturing conflict. It means the core should not become a circle of people who merely confirm the same worldview to one another. “Shared Values Without Ideological Uniformity” already established that cooperation needs enough shared values, not complete ideological uniformity.

The Community Tool Box recommends recruiting people with energy, time, relevant connections and a real willingness to participate, while also benefiting from diverse representation. A first core is stronger when its members complement one another than when they simply resemble one another.

Five people can do something useful. Five people do not thereby represent a neighbourhood, village, municipality or “the people”. This boundary matters. Many new hierarchies begin when a small active group gradually starts presenting itself as the voice of a passive majority.

The first agreement should therefore contain a sentence such as: “We act in our own name and within this task; we represent the wider community only where we have received an explicit mandate to do so.” A core running a tool-sharing pilot can manage that pilot. It cannot infer from this that it has authority over unrelated local questions.

This is a practical safeguard against the dynamics analysed in “When Does a Delegate Become a Ruler?” and “Institutional Capture: When the System Starts Serving Itself”. A delegate does not become a ruler, and a useful initiative does not become a captured institution, when the scope and source of mandate remain visible.

The first meeting ends with a one-page founding agreement

Good intentions are not enough. The first or second meeting should end with one short document. We do not need a fifty-article legal statute, but we do need shared memory that cannot later be rewritten by whoever remembers the meeting most confidently.

On one page, record the problem, purpose, what will be done in the next thirty days, what is outside scope, who currently participates, how decisions are made, how tasks are assigned, how money and data are handled, how someone leaves and when the agreement will be reviewed. Research on team charters suggests that investing early effort in explicit teamwork agreements and work strategy can improve later performance.

The document is not a sacred constitution. Its value is that when disagreement appears, the group can point to a common agreement and change it together. Ostrom’s research offers a useful principle here: people are more likely to respect rules when they have genuinely participated in making them.

Every group quickly needs functions: someone convenes the meeting, someone records decisions, someone contacts partners, someone checks costs. The problem begins when a temporary function turns into personal status: the founder is always the leader, the person with the password is always the gatekeeper, or the person holding funds becomes an untouchable treasurer.

This article therefore favours short, described and where practical rotating roles. In a small pilot, the coordinator may chair only the next meeting, the recorder serve for one month, and the person holding funds should never be the only person able to verify the balance. Where expertise requires continuity, we do not rotate mechanically, but we still define the mandate and review it.

The International Cooperative Alliance emphasises democratic member control and the accountability of elected representatives to members. A local core is not yet a cooperative, but the principle transfers well: the function is accountable to the group; the group is not subordinate to the function.

Choose the decision rule before the first serious dispute

“How Do We Make Decisions Without a Permanent Ruler?” already showed that no decision method fits every situation. For a first local circle, a simple division is enough. Routine implementation decisions belong to the person who accepted the task within an agreed boundary. A change of purpose, use of common money or acceptance of a new obligation should require a decision by the whole core under a pre-agreed procedure.

Some groups may use consensus, others consent with the possibility of a reasoned objection, and others a majority rule. More important than the label is clarity about who may decide what, when wider agreement is required, and what happens if no decision is reached.

OECD guidance on meaningful participation stresses clear purpose, accountability and a genuine possibility of influence. A small group also loses trust when people are invited to participate but the real decision has already been made elsewhere.

A small circle benefits from closeness, but closeness also creates social pressure. If everyone knows the founder, if members are friends, or if one person controls an important resource, saying “this is a bad idea” can become surprisingly difficult. Safe dissent is therefore an operational property, not a decorative value.

Research on psychological safety shows why teams learn and correct errors more effectively when members can raise concerns without fear of humiliation or retaliation. This article translates this into simple rules: criticise the proposal, not the person; doubt is not disloyalty; a safety or legal concern can pause implementation long enough to be checked; nobody loses membership merely for voting against the majority.

The Community Tool Box similarly recommends that groups create their own ground rules early. When rules are jointly made, it is easier to invoke them even when the person breaking them is the loudest or holds the most informal power.

Membership is voluntary, boundaries are clear and the doors are not sectarian

The core needs to know who is currently responsible for the pilot. That requires a membership boundary. But a clear boundary is not the same as a closed identity. Someone can help with one task without adopting the group’s entire worldview. New people should know how to join, and existing members should know how to leave peacefully.

The general principle of freedom of association is fundamental here: nobody should be compelled to belong. The ICA similarly links membership with voluntariness, openness and acceptance of member responsibilities. For this article, this means the core does not use guilt, family pressure, religious affiliation, political loyalty or economic dependence as the price of participation.

If the circle begins dividing people into the “right” and “wrong” kinds, or demands loyalty far beyond the concrete task, we have left practical local cooperation and begun building an identity hierarchy. That is a signal to review the rules immediately.

Communication, data and money should be minimal but traceable

A small initiative does not need six channels, a complex CRM system and a database of personal information. It needs one agreed communication channel, a short decision log, a task list and a clear place for documents. Less system often means less friction and fewer new control points.

If money appears, the rule becomes stricter: purpose, amount, who holds it, who may spend it and how everyone can see the balance. During a first pilot, avoid large pooled funds and obligations until the group has shown that it can manage small ones. The same principle applies to personal data: collect only what the task genuinely requires.

The Community Tool Box includes clear procedures for decision making, information sharing, conflict resolution and finance among basic organisational mechanisms. This article uses the lightest possible version: enough traceability for accountability, not so much bureaucracy that the process becomes its own purpose.

The first project should be small, measurable and preferably reversible

The first local core does not yet need a legal entity, office, logo or large budget. It needs evidence that a few people can make a promise together and keep it. Choose a task with a visible end: produce ten working contacts, run one tool-sharing session, coordinate one transport need, establish one call tree or test one backup communication channel.

The pilot should be small enough that failure does not cause major harm, but concrete enough to teach something. If it requires more money, liability or legal commitment than the group understands, break it into smaller steps. Reliability is not proved by the size of the promise but by the relationship between what was promised and what was actually delivered.

After thirty days, review the outcome rather than the mood. What did we promise? What happened? What failed? Who actually carried the work? Is the group still needed? If the answer is no, dissolving the core is a legitimate success condition rather than an organisational failure.

Twelve questions for the first thirty days of a local core

Before the circle becomes an organisation, it should be able to answer twelve simple questions:

1. What one concrete need currently connects us? 2. What will we do in the next 30 days? 3. What are we explicitly not taking on? 4. Who currently participates and on what voluntary basis? 5. What capability or connection does each member bring? 6. Who may decide routine tasks and who may change the purpose? 7. What roles exist, how long do they last and who can replace them? 8. How can someone safely say no, flag an error or request a pause? 9. How does someone join or leave? 10. Where are decisions, money and necessary data recorded? 11. By what measure will we know the pilot succeeded or failed? 12. Exactly when will we review the entire agreement?

Without these answers we do not yet have a core—we mostly have goodwill. If we have them and complete one small project, we have done something more important than creating another committee: we have demonstrated the capacity for voluntary collective action without a permanent ruler. “Transition Without Collapse: Building Parallel Institutions” will continue from here: how small, tested capabilities can grow into parallel institutions and gradually take on functions without first destroying existing systems and creating a vacuum.