Counterfactual Contribution of Prevention: Why the Work That Prevents a Harm Can Be Seen Only Against What Would Otherwise Have Happened

An uneventful outcome does not by itself show whether preventive work made a difference. The same absence of harm can be compatible with an effective intervention or with a threat that would never have materialized. This paper sets out Counterfactual Contribution of Prevention, a structural constraint on what observed success can establish about the work that preceded it. The paper distinguishes the intervention, the observed outcome, and the specified course of events that would have followed without the intervention. It relates this account to counterfactual theories of causation and examines how models or comparisons across settings can support estimates of prevented harm. Each comparison requires evidence and assumptions; the absence of harm alone does not justify either crediting or dismissing the intervention. Published simulations of ozone depletion provide the documented case. The paper examines comparisons between modeled reference and unregulated futures at corresponding dates. Those comparisons concern the broader pathway of controls on ozone-depleting substances, including responses preceding the Montreal Protocol, and remain dependent on the model and its assumptions. Part of the Relational Conditions and Dynamics series within On Human Modeling, the analysis identifies questions for evaluating prevention: what would otherwise have happened, what supports that comparison, what uncertainty remains, and who supplies the comparison through which preventive work receives credit.