When the set learns but the organization does not: premature learning closure and an evaluative framework for action learning conversion

Action Learning can produce genuine learning without equivalent organizational conversion. This paper calls the resulting evaluative risk premature learning closure: treating evidence of learning as evidence of a stronger conversion claim. Drawing on Action Learning literature and anonymized practice material used illustratively, it develops the Action Learning Conversion Framework (ALCF), an evaluative framework with four analytically distinct dimensions: diagnosis as epistemic adequacy, jurisdiction as actionable capacity, consequence as reality contact, and retention as temporal continuation. Questioning operates across them as a cross-cutting constitutive Action Learning mechanism rather than as a fifth criterion. A criterion need not become a visible blockage; each specifies a different evidentiary question. Premature learning closure occurs when evidence sufficient for one claim is used to support a stronger claim it cannot establish. The ALCF is recursive and empirically contestable, offering a way to locate where learning stops without reducing conversion to a single transfer or implementation gap.