Boundary Selection and Dependency Exclusion: How the Edges of an Account Decide Which Dependencies It Can Show

An account can be accurate in every sentence and still omit a condition needed to explain what happened. A report can correctly declare a task complete while leaving unresolved the wider problem it is taken to settle. This paper defines Boundary Selection and Dependency Exclusion as the operation through which a rule delimiting an account also excludes a dependency relevant to its explanatory or completion claim. It distinguishes ordinary choices of scope from consequential exclusion and explains why placing a condition outside the focal event need not prevent the account from identifying it as a cause. Drawing on framing, contrastive causal explanation, and boundary critique, the paper uses Diane Vaughan’s research on the Challenger launch decision to illustrate how broader historical inquiry can change an explanation. It provides practical questions for examining an account’s boundaries, the dependencies they exclude, and who can reopen them. Part of the On Human Modeling series on Generative Operations.