Is one term having different meanings in different departments a contradiction?

No. It is ambiguity, which is a feature of natural language. Semantic Governance represents it as multiple Canonical Concepts inside the same Meaning Space, distinguished by context labels. A contradiction is when two definitions claim to mean the same thing in the same context, but disagree.

When does reuse count as evidence of stability?

Each time a new ingestion resolves to an existing Canonical Concept without modification or contradiction, confidence strengthens. Stability accumulates from reuses over time relative to drift events resolved.

When is a divergence flagged as drift?

When a new definition resolves to an existing Canonical Concept in the same context but contradicts the existing canonical definition — measured by low overlap between the prior and new wording. The system routes it for governance review rather than overwriting silently.

Semantic Governance Examples

Coverage — Insurance Meaning Space

The word "Coverage" has at least five governed meanings inside the Insurance Meaning Space. Each is a distinct Canonical Concept with its own definition, owner, and provenance. None of these meanings contradict each other — they coexist in different contexts.

Department	Context Label	Canonical Concept	Definition	Confidence
Underwriting	Policy Scope	Coverage (Policy Scope)	The scope of risks insured under a policy, including perils, locations, and exclusions.	0.90
Claims	Claims Limit	Coverage (Claims Limit)	The maximum financial obligation of the insurer for a covered loss under a specific policy.	0.92
Reinsurance	Reinsurance Cession	Coverage (Reinsurance Cession)	The share of risk transferred from the primary insurer to a reinsurer under a treaty or facultative agreement.	0.95
Product	Product Offering	Coverage (Product Offering)	A marketable insurance product line offered to a defined customer segment.	0.90
Compliance	Regulatory Control Coverage	Coverage (Regulatory Control Coverage)	The extent to which a business activity falls within a regulator's jurisdiction or required control framework.	0.92

Why this is not drift

These five meanings are not drift, contradictions, or errors. They are different governed meanings in different contexts within the same Meaning Space. Glossary tools that force one definition per term destroy this distinction; semantic governance preserves it as ambiguity, resolved by context.

What would be drift

Drift would occur if Claims submitted a new definition of "Coverage (Claims Limit)" that contradicted the existing one — for example redefining it as a deductible rather than a maximum obligation. The system would flag this as a same-context contradiction and route it for governance review.

Evidence

Reproduced by scripts/test-upload-workflow-coverage.ts: 5 distinct canonical concepts produced from 5 department submissions, 0 false cross-context drift, second upload of a Reinsurance definition reused the existing concept and raised confidence from 0.95 to 0.97 with no duplicate created.

Frequently asked

Is one term having different meanings in different departments a contradiction?: No. It is ambiguity, which is a feature of natural language. Semantic Governance represents it as multiple Canonical Concepts inside the same Meaning Space, distinguished by context labels. A contradiction is when two definitions claim to mean the same thing in the same context, but disagree.
When does reuse count as evidence of stability?: Each time a new ingestion resolves to an existing Canonical Concept without modification or contradiction, confidence strengthens. Stability accumulates from reuses over time relative to drift events resolved.
When is a divergence flagged as drift?: When a new definition resolves to an existing Canonical Concept in the same context but contradicts the existing canonical definition — measured by low overlap between the prior and new wording. The system routes it for governance review rather than overwriting silently.