Discussion about this post

User's avatar
Philip's avatar

Once upon a time, there was an old man who used to go to the ocean to do his writing. He had a habit of walking on the beach every morning before he began his work. Early one morning, he was walking along the shore after a big storm had passed and found the vast beach littered with starfish as far as the eye could see, stretching in both directions.

Off in the distance, the old man noticed a small boy approaching. As the boy walked, he paused every so often and as he grew closer, the man could see that he was occasionally bending down to pick up an object and throw it into the sea. The boy came closer still and the man called out, “Good morning! May I ask what it is that you are doing?”

The young boy paused, looked up, and replied “Throwing starfish into the ocean. The tide has washed them up onto the beach and they can’t return to the sea by themselves,” the youth replied. “When the sun gets high, they will die, unless I throw them back into the water.”

The old man replied, “But there must be tens of thousands of starfish on this beach. I’m afraid you won’t really be able to make much of a difference.”

The boy bent down, picked up yet another starfish and threw it as far as he could into the ocean. Then he turned, smiled and said, “It made a difference to that one!”

Expand full comment
Nathan Barnard's avatar

I think the counterargument to this is that plausibly addressing root causes has extremely large effects across a wide range of problems. I obviously agree that there's no normative case for preferenceing root causes, but I think you've underrated the empircal case.

For instance, the root cause of the continuation of many communicable diseases is poverty caused by bad and weak government. It's plausible that we should target interventions at this instead, for instance by trying the replicate the success of the world and bank and IMF improving the economic policy of China and India respectively.

I think the analogy to the dying gun crime victim is a poor one. In that case, it's extremely clear that the marginal effect of helping the gun crime victim directly is much greater than what would have been done counterfactually with ones time for gun policy - but this is often not the case for considering interventions (although plausibly is for Pepfar.) I also this that it activates deontic intuitions, are it's fine to make non-consequentalist arguments, but I don't think that that's what was being argued for in the rest of the peice.

Expand full comment
40 more comments...

No posts