Adapter Principles

Adapters define how the Governance Substrate Model is translated into specific domains without distorting its invariants. They are not implementations. They are translation layers that allow the same structural logic to operate across different contexts, scales, and regimes.

This layer exists to prevent both rigidity and dilution.


What an Adapter Is#

An adapter is a context‑sensitive translation mechanism that:

  • Preserves core invariants.
  • Respects local constraints.
  • Avoids importing unnecessary structure.
  • Enables interoperability without uniformity.

Adapters allow the model to fit without being forced.


What Adapters Are Not#

Adapters are not:

  • Domain‑specific governance systems.
  • Policy templates.
  • Best‑practice checklists.
  • Authority grants.

They do not decide outcomes.
They preserve structural coherence during translation.


Core Adapter Principles#

Invariant Preservation#

Every adapter must explicitly identify:

  • Which invariants apply.
  • How they manifest locally.
  • Where they are at risk.

If an invariant cannot be preserved, the adapter is invalid.


Minimal Sufficiency#

Adapters should introduce:

  • The least structure required to maintain alignment.
  • No additional enforcement mechanisms.
  • No narrative framing.

Excess structure increases fragility.


Phase Sensitivity#

Adapters must:

  • Declare the assumed system phase.
  • Adjust posture accordingly.
  • Avoid importing scaling logic into emergence.

Phase mismatch is the most common adapter failure.


Reversibility First#

All adapter‑introduced structures must:

  • Be reversible.
  • Include rollback paths.
  • Treat irreversibility as a failure signal.

Adapters that lock systems in place destroy learning.


Legibility Preservation#

Adapters must:

  • Keep decision logic visible.
  • Avoid opaque automation.
  • Preserve human interpretability.

Translation that reduces legibility is misalignment.


Authority Neutrality#

Adapters must not:

  • Expand authority.
  • Centralize control.
  • Normalize enforcement.

If authority increases, the adapter has failed.


Adapter Design Process#

Effective adapters follow a disciplined sequence:

  1. Identify applicable invariants.
  2. Map local constraints and affordances.
  3. Detect phase and coupling level.
  4. Introduce minimal translation structure.
  5. Stress‑test for reversibility and legibility.
  6. Document assumptions and failure modes.

Adapters are designed, not improvised.


AI and Adapters#

AI may assist by:

  • Mapping invariant‑to‑context translations.
  • Detecting regime mismatch.
  • Stress‑testing adapter assumptions.
  • Highlighting legibility loss.

AI must not:

  • Generate adapters autonomously.
  • Optimize for domain metrics.
  • Declare adapter validity.

Adapter judgment remains human.


Failure Mode#

Adapters fail when:

  • Domain norms override invariants.
  • Convenience replaces coherence.
  • Authority is smuggled in as “efficiency.”
  • Translation becomes implementation.

At that point, the model fragments.


Adapters are how the model travels without breaking.

When designed with restraint and clarity,
they allow governance to remain coherent
across domains, cultures, and scales —
without becoming rigid or imperial.