Trust is the confidence that other agents will act in good faith within the shared method — that they will follow conventions, respect governance, and not exploit the endeavor’s openness for purposes incompatible with its closure conditions.
Secondary intension
Something functions as trust when it has all of:
- Relational basis: trust exists between agents, not as a property of individuals. It is produced and maintained through interaction, not declared.
- Defeasibility: trust can be withdrawn. Unlike method (which is documented and persistent), trust must be continuously sustained through behavior consistent with shared commitments.
- Non-formalizability: trust cannot be fully captured in specifications. Method can document what agents should do; trust is the confidence that they will. The gap between method and trust is irreducible.
- Preconditionality: trust is a precondition for method-based coordination. Without trust, shared method becomes surveillance — a system of verification rather than coordination.
Distinguished from
-
Method: method is the codified system of conventions. Method provides the shared language; trust provides the confidence that the language will be used in good faith. Method without trust produces compliance regimes. Trust without method produces informal arrangements vulnerable to drift.
-
Consensus: consensus is a governance mechanism that produces decisions. Trust is what makes consensus possible — the confidence that participants will engage honestly and accept outcomes they disagree with. Consensus without trust degenerates into veto games.
-
Verification: verification checks whether method was followed. Trust is what makes verification unnecessary for most interactions. A zero-trust architecture verifies everything; a trust-based architecture verifies at boundaries.
In the semiotic framework
Trust is the interpersonal precondition for the non-interference theorem’s practical application. The theorem proves that agents on disjoint regions produce order-independent results. But the theorem assumes agents stay within their declared regions. Trust is the confidence that they will — without trust, region declarations become unenforceable boundaries rather than coordination structures.
Trust stands in the same relation to multi-agent coordination that habit stands in to method: it is the implicit, relational substrate on which the explicit, formal structure depends. Method can encode trust-building practices (transparent governance, append-only decision records, shared repository access), but cannot substitute for trust itself.
Vulnerability
When trust is the coordination medium, destroying trust is the most efficient form of disruption. Counterinsurgency doctrine identifies this explicitly: infiltrators, informants, and snitch-jacketing attack the trust network rather than the organizational structure. An endeavor’s governance must account for this vulnerability — not by replacing trust with verification (which eliminates the coordination advantage) but by maintaining practices that sustain and repair trust (transparent decision-making, accountable processes for handling concerns, evidentiary standards for accusations).