---
title: Causal Pluralism (vs Causal Monism)
type: concept
tags: [philosophy-of-science, causation, pluralism, monism, methodology]
sources: ["[[sources/2023-anjum-rocca-phi403-causation-in-science]]", "[[sources/2023-anjum-rocca-phi403-lecture-20-plural-methods-one-causation]]"]
created: 2026-04-20
updated: 2026-04-20
---

# Causal Pluralism (vs Causal Monism)

Two axes on which one can be monist or pluralist about [[concepts/causation]]: the **ontological** axis (how many kinds of causation exist) and the **methodological** axis (how many methods are needed to detect causation).

## Positions

- **Causal monism** — causation is one single thing. Often paired with **causal primitivism**: causation cannot be analysed into anything more fundamental (correlation, counterfactual dependence, probability raising, manipulation).
- **Causal pluralism** — causation is not one thing but many; perhaps a family-resemblance concept or a disjunction (e.g. difference-making *or* influence). Different cases of causation may be different *kinds* of causation.
- **[[concepts/methodological-pluralism]]** — even if causation is ontologically one, multiple methods are needed epistemically because causation has multiple *symptoms* (difference-making, regularity, manipulability, probability-raising, energy transference) — and none of them mark causation perfectly.

## The course's position

Anjum & Rocca defend **ontological monism + methodological pluralism** ([[sources/2023-anjum-rocca-phi403-lecture-20-plural-methods-one-causation]]): causation is one thing (a real property of the world, often grounded in dispositions), but because no single symptom is reliable, science needs plural methods — statistical, experimental, mechanistic, singular-case — to triangulate it. This generalises Feyerabend's and Longino's [[concepts/methodological-pluralism]].

## Implications
- Rejects the strict [[concepts/evidence-hierarchy|evidence hierarchy]] of EBM that privileges RCTs over all else.
- When methods disagree, an **evidence-ranking scheme** is needed rather than automatic dismissal of the lower-ranked method.
- Qualitative and quantitative methods answer *different causal questions* and neither is inherently better.

## Related
[[concepts/causation]] · [[concepts/methodological-pluralism]] · [[concepts/evidence-hierarchy]] · [[concepts/dispositionalism]]