Affiliation Enables Representation, Not Participation

1. Who the system can see

The list of people RSM names is deliberately broad: a farmer, scientist, landowner, chef, agronomist, entrepreneur, investor, citizen, patient, researcher or practitioner. Any of them may have an intent, a capability, capacity, knowledge, relationships, evidence, an offer or a seek in their own right. They do not acquire these by joining an organization, and they do not lose them by leaving one.

Organizations and communities still matter a great deal. The Concept Paper describes them as associative structures: they let people coordinate collectively, hold an institutional identity, govern shared resources and make commitments under explicit authority. A cooperative can do things no individual farmer can, and a research institution can stand behind evidence in ways a single scientist cannot. But these structures, in RSM's account, enrich and extend the person; they do not create the person's right to participate. Association is voluntary, and nothing a person is entitled to do for themselves depends on it.

This is not only a statement of values. It changes what a coordination system can find. If the unit of participation is the organization, much of the capability relevant to a regenerative outcome is invisible: the knowledge of an independent advisor, the latent capacity of a landowner who has never sold anything, the willingness of a person to take part in a formation without representing anyone but themselves. Self-first participation is what lets the persistent network described in Why Regenerative Systems Need a Shared Model include them.

2. Self, Person, Participant

RSM makes the premise precise through three related concepts, which its canonical model calls the foundation of agency.

The Self is the sovereign human actor from which the right to participate originates. It is conceptual: RSM does not store a Self as a record. The Person is the identity that represents a Self in the network, a stable opaque identity together with a minimal, governed, progressively enriched profile. The Concept Paper is careful about the relationship between the two. The Person record is a representation of the Self, not the source of its agency. Keeping the Self out of storage is a way of saying that no record, and therefore no one who administers a record, is where a person's right to take part comes from.

The Participant is the concept that most systems get wrong. In RSM it is a contextual role of agency, played by a Self acting as a Person, by an organization, or by a governed community. It is never a synonym for an organization. The same person may be a producer in one formation, an advisor in another and a buyer in a third, and the same organization may be a processor in one context and a storage provider in the next. Identity persists while roles change. Facilities, living systems, places and resources may be subjects of a formation or contribute to it without being participants at all.

The distinction is easy to lose in engineering. When the first plan for the RSM reference kernel was reconciled against the canon, one of the shorthands it corrected treated Participant as an entity with person and organization variants, which would have quietly made a role into a kind of thing. The correction is described in the Fieldnote The First Plan Had Semantic Debt.

3. Two lanes: acting for yourself and acting for others

Once a person can participate without an organization, the model has to say what changes when they act for one. RSM draws two lanes.

A Person acting for Self derives authority from being the subject and owner of the expression or action. A farmer offering her own capability, a researcher describing his own expertise, a landowner expressing an intent about land he governs: none of these requires an organization, and none requires anyone's permission beyond the person's own.

A Person acting for an organization or a community is doing something different. This is representation, and it requires explicit authority: in the canonical model's terms, a verified affiliation, a stewardship grant or a delegation. The Concept Paper spells out the boundary from the other side. Self-sovereignty does not confer authority over other entities. A person cannot commit an organization's capacity, disclose its private evidence, join it to a community or purchase on its behalf without explicit authority. Organizations and communities have their own governance and act through authorized representatives, stewards or delegations.

The two lanes protect each other. The first protects the person: they never have to surrender their own agency to an organization merely to become discoverable or useful. The second protects the organization and everyone who relies on its commitments: no one can speak for it simply by being near it. Hence the sentence that gives this essay its title: affiliation enables representation; it does not enable participation. And hence the invariant behind it: participation is not representation.

4. Why affiliation and membership confer nothing

The hardest part of the model to accept, because it runs against how organizations often work in practice, is that the ordinary ways people are connected to institutions carry no authority at all. An Affiliation relates a person to an organization: owner, executive, employee, advisor and so on. A Membership relates a person or an organization to a community, with its own lifecycle from invitation to activity to removal. In RSM neither, by itself, confers the right to act for the organization or the community. Authority, in the canonical model, is a separate concept: the capacity to take a consequential action, composed of a verified basis together with the capability to act. The concepts that explicitly grant it are Stewardship and Delegation.

The reason is that consequential acts in a formation, such as committing capacity, disclosing evidence or accepting terms, are exactly the acts other participants will rely on. If a lender extends transition financing on the strength of a buyer's indicative commitment, it needs to know that the person who made the commitment could make it. Being an employee of the buyer is not the same as being authorized to commit its demand, and an advisor's title says nothing about whether they may disclose the records they have seen.

This is one of the places where RSM is already enforced in software rather than only described. The reference kernel resolves, for each consequential operation it supports, whom the actor is acting for and on what basis. A person acting for themselves is allowed without any organization. A person acting for an organization without valid authority is refused, because authentication alone is not authority. A person acting for an organization under a valid stewardship or delegation is allowed. And a person who is merely an employee, advisor or member, with no delegation, is refused. The kernel's acceptance rules add a detail that matters: an implementation that passed the refusal cases simply by rejecting every organization-scoped operation would fail, because legitimate representation must succeed.

One question here is genuinely open. The canon names a verified affiliation as a possible basis of authority, but it does not yet define how a verified, authority-bearing affiliation differs from an ordinary one in a way a system can check. The kernel's response is to refuse such a request rather than guess, and to rely on stewardship or delegation for representation in the meantime; closing the gap is recorded as governed work on the canon itself, not something to be patched in software. The Fieldnote Fail Closed When the Canon Is Silent describes that decision.

The same logic extends beyond people. The runtime architecture requires every operation to record both who is acting and under whose authority, and it applies this directly to software agents: an agent may have the technical capability to issue a commitment, but the operation is valid only if the participant has delegated that authority. It applies to collectives as well. When a formation such as a cooperative participates in a larger one, its commitment does not automatically mean that each member has committed, unless the cooperative holds that authority. When a collective is one actor and when it is many is the question of the Seednote Is a Cooperative's Commitment Its Members'?.

5. Agents, subjects and contexts

The final part of RSM's account concerns entities that matter to a regenerative outcome but do not act at all. The canonical model gives every actor or context one of three agency statuses. An Agent can hold intent, make claims, consent, commit and act under its own or delegated authority: a person acting for themselves, an organization, a governed community. A Subject or contributor can be the subject of activity, observation, evidence and outcome, or contribute to a configuration, but performs no institutional act: a living system, a facility, a place considered as a reference, a resource. A Context is a durable environment that organizes agents and subjects, such as a community considered as an environment or a place considered as a spatial setting. An entity may carry more than one status by role.

The Concept Paper's example is brief and exact: a watershed is observed; it does not commit. A field, a soil, a herd or a landscape can be observed, made the subject of claims and evaluations, managed and affected, but it does not consent, sign or promise. Modelling it as an actor would be a category error. Modelling it as a mere attribute of its owner would be a different error, because living systems have their own trajectories that a regenerative network exists to observe. The same care applies to facilities: a mill can be constrained in a given period even when the organization that operates it is generally capable, so a facility can carry its own capability, capacity and constraints rather than simply inheriting its operator's.

The purpose of the three statuses, in the Concept Paper's phrasing, is that reasoning always asks which entity is acting, under whose authority, and in what capacity. They also leave a question the sources do not answer: if a living system cannot be a party to a formation, how does its interest enter one? The Seednote Living Systems Do Not Sign Contracts holds that question open.

6. What a Self-first model changes

Put together, these distinctions describe a network in which participation is broad and authority is narrow. Anyone can take part in their own right. Very few acts can be performed for someone else, and each of those requires a grant that can be checked, scoped and, where its semantics allow, revoked. Discovery does not create participation either: a person or an organization can be relevant to an outcome without having agreed to take part in it.

For coordination, that combination is what makes commitments worth relying on. A formation is assembled out of undertakings by many parties, and each party has to be able to trust that the others' undertakings were made by someone entitled to make them. For people, it is what makes the network worth joining: they can be seen, contribute and build a history without first surrendering their agency to an institution. The Concept Paper compresses the whole position into three short sentences that open its governing statement: Self is foundational. Association is voluntary. Authority is explicit.

Sources

• RSM Concept Paper v3.0, §3 Self is Foundational; §4 The Foundational Triad; §9 Sovereignty, Authority, and Trust.
• RSM Canonical Model v3.0, §3.2 The three agency statuses; §4 Foundation of Agency; §5 Actors, Contexts, Associations.
• RSM Runtime and Federation Architecture Specification v1.0, §3.2 Actor Identity; §3.3 Delegation; §9.3 Boundary Preservation.
• RSM Normative Conceptual Reference v1.1, §4.3 Living System; §4.4 Facility.
• RSM Reference Implementation and Acceptance Specification v1.0, §7.2 Delegated Authority.