--- title: "PHI403 Lecture 03 — Evidence Based on What?" type: source tags: [philosophy-of-science, ontology, epistemology, evidence, operationalisation] authors: [Anjum, Rani Lill; Rocca, Elena] year: 2023 venue: "PHI403 Causation in Science, NMBU" kind: handout raw_path: "raw/Philosophy of Science/PHI302 03 Evidence Based on What.pdf" created: 2026-04-20 updated: 2026-04-20 key_claims: - The ontological question "what is causation?" and the epistemological question "how do we discover causation?" must be kept distinct. - Epistemology cannot dictate ontology — identifying causation with its evidence (Hume's constant conjunction, operationalisation) is a representational fallacy. - Methods restrict the type of evidence available; scientific methods thus shape but do not exhaust causation. - Each method picks out some feature of causation (regularity, difference-making, manipulability, tendency, mechanism, probability) without being identical to causation. --- # PHI403 Lecture 03 — Evidence Based on What? Articulates a distinction that runs through the whole course: **[[concepts/causation|causation]]** as an ontological matter (what causation *is*) versus causation as an epistemological matter (how we *discover* it). Confusing the two is the **operationalisation fallacy** — Berkeley identifying material objects with perceptions, Hume identifying causation with constant conjunction. Scientific methods each pick out a different *symptom* of causation: regularities, difference-makers, probability raisers, manipulability, tendencies, mechanisms. None is identical to causation, and none is immune to counterexamples. Therefore causal evidence is not the same as causation, and choice of method implicitly commits one to a particular ontology of causation. The lecture's ontology-first prescription: what causation *is* should inform our choice of method, not the reverse. This sets up the methodological-pluralism argument of lecture 20. ## Connections Back-link: [[sources/2023-anjum-rocca-phi403-causation-in-science]]. Concepts: [[concepts/causation]] · [[concepts/philosophical-bias]] · [[concepts/methodological-pluralism]].