Why it is a question
THEORY-133 states, and the record has re-derived, the framework result: given that the higher-level property F makes a difference to the effect G, its realizer P fails to be a cause exactly when some nearest situation without P has F realized otherwise and G still present. Whether that holds is a fact about each system’s counterfactual structure. List and Menzies’s claim that special-science relations are typically realization-insensitive is an observation about practice, and Pernu (LIT-147) disputes their neuroscience example. Until 2026-10-20 this was THEORY-133’s promote_when.
What would answer it
For a particular system, interventions that change the realizer while holding the higher-level property fixed leave the effect in place, and interventions that change the higher-level property change it. The answer for that system goes the other way if changing the realizer alone changes the effect. Arguments about the metaphysics of causation in general cannot settle it, since the point is that the answer is system by system.