Who gains from this particular simplification? 2. **What is absent?** — What causes and contexts have been omitted? 3. **Is the cause proportionate to the effect?** — Can the designated cause actually produce the attributed effect at scale? 4. **What is the attribution structure?** — Does this treat