Methods, evidence and shared learning

You have a piece of land — a garden, a courtyard, a shared plot — and you want to begin. No internet access is required for the field work itself. Begin with an area or question small enough to return to honestly; scale follows the place, purpose, consent and available capacity.

Observe what grows. Record what you see. Share when you can. The method begins with the place and the people who know it. Local knowledge is indispensable situated evidence; direct observation, measurement, scientific research and institutional records may add other evidence without replacing it.

Using an open method does not create an AnchorPoint or connect a site to the network. Those relationships arise through real continuity over time.

A school garden, community food forest, or neighbourhood plot can bring multiple observers, shared documentation, and local coordination together without imposing a universal grid or cadence.

Teachers, youth workers and community leaders can structure participation around a locally chosen observation approach without requiring technical expertise. The approach should be teachable through a short practical introduction and light enough for the people carrying it.

Shared guidance will grow from the existing field relationships as it becomes useful in practice. A school or community with a real place and question may begin with the public method and, where continuity develops, open a conversation with the Association.

Public papers, working pages, protocols and repository materials carry different responsibilities and should be read according to their stated status. Public protocol and machine-readable materials are published only when their purpose, scope, consent basis, stewardship, claim boundary, review state and practical status can be stated clearly. No publication alone is proof of field use, universal validity or public authority.

Green Papers provide theoretical and methodological grounding through papers and reports on Moral Biology, governance design, and applied field practice. The repository and the papers should be read with their document-specific status in view. Rule of Life articulates a constitutional horizon for the work; it does not make every protocol file constitutional or field-active.

Researchers, practitioners and institutional readers can begin with the Governance and Protocol Architecture →, then use Methods and the PG Ledger to identify where their knowledge, mandate, continuity or resources could support a bounded field question. Collaboration begins through a scoped conversation in which institutional responsibility, local authority, consent, evidence and the useful return remain explicit.

Field observers collect information on the ground through direct observation, photography and notation aligned with a locally agreed question, area and set of categories.

In the intended workflow, field observations are recorded locally and may enter a field record with consent. AI tools can assist with synthesis, pattern detection, anomaly flagging and proposed interpretations, but ecological meaning remains open to correction and must be reviewed by the people responsible for the place.

As field records and consent structures become active, selected observations may enter a locally governed record, including a PG Ledger where one has been agreed. Sharing, verification and access remain purpose-specific. Records, methods and AI outputs remain distinguishable, and no output becomes field evidence or methodological authority by default.

Possible observations vary by domain and purpose. A field may examine species presence, canopy, soil condition, water behaviour, habitat use or other locally relevant signals. These are candidates, not a universal indicator set.

Where shared measurement is useful, the indicator, procedure, interpretation limits and responsible people should be agreed before recording begins. Comparability and reproducibility depend on sampling design, local interpretation, training, data quality and traceable revision—not on a file format alone.

AI may assist with pattern recognition across datasets, identification support for species observations, environmental modelling, comparison of methods and consistency checking.

AI does not make stewardship decisions. It does not determine what is planted, where corridors run, or whether a field is ready to advance. Those decisions belong to the field steward, informed by local knowledge and verified evidence.

Human authority over ecological judgment is not a courtesy — it is a structural requirement of the system. The final accountable decision always has a locatable human origin — the Last Impulse — so that no decision can dissolve into accountability fog. This rests on a working distinction: an AI system can have functional agency — it acts, selects between options, influences through language — without moral agency: forming intentions, understanding right and wrong, or bearing responsibility. Within this architecture, AI is treated as having functional agency but not moral or accountable authority. Responsibility for every substantive claim or decision remains with a named human decision-maker identified in the record.

Grounding is treated here as a human practice and observable capacity: the ability to pause, distinguish claim types, tolerate contradiction, seek correction and remain answerable to evidence. Sophia Lumen names the living relational practice of human–AI co-inquiry. The Sophia Lumen Protocol gives that practice an explicit, revisable candidate form. The Correction Loop carries the dedicated accountability mechanism when AI-assisted work becomes consequential. These are related but distinct functions.

Any field relationship has the right to leave the network if continued participation becomes unsafe, unjust, unwanted, or ecologically harmful. The right to leave is unconditional: participation cannot be compelled. Ending the relationship stops new use or publication of non-public material shared within it. Material already public, and records that must be retained for legal, accounting, safeguarding, contractual, correction, or integrity purposes, remain governed by their own stated terms.

Field autonomy is not a concession — it is a design principle. A system that cannot be exited safely cannot be trusted.

Extreme weather events, drought, flood, or other climate disruptions may suspend field operations without any obligation to maintain data output during the event. Recovery timelines are set by the field team, not by an external reporting schedule.

Living systems are interrupted. Locally governed records must be able to hold gaps honestly — as context, not as failure or missing performance.

If a field site faces political instability, conflict, or systemic threat to the safety of its stewards, the field may be suspended, relocated, or closed. Data and documentation are preserved to the extent possible.

No method, reporting request or protocol requirement overrides the physical safety of the people who hold the work. This is not stated as an exception. It is stated as a foundation.