Some philosophers want a general theory of causation that agrees with the results of various thought experiments about causation. And some philosophers want a theory of causation that vindicates our intuition that causation is ultimately an objective non-perspectival feature of the world. In contrast, I want to know what happens if we abandon these aims and instead look for a theory of causation that starts from the question: why is causal knowledge especially useful for serving our subjective and somewhat parochial aims?
Case Study: Group Selection in Evolutionary Biology
I consider the issue of group selection explanations in evolutionary biology, and I show how treating causation and causal explanation as partially subjective solves a number of paradoxes about group selection explanations.
Negative Point: Against Thought Experiments
To support my approach to studying causation, I give some reasons to think that thought experiments about causation are unreliable.
Negative Point: Against Objective Explanatory Levels
Consider the fact that the demand for petrol is inelastic (the amount of petrol that car owners will buy doesn't very decrease much as the price of petrol increases). It turns out that this fact about inelastic demand is good at explaining many things about the economy, for example the price of petrol when there is a shock to the supply of petrol. This leads many philosophers to reason in something like the following way:
(1) Without the concept of inelastic demand, our explanations of economic phenomena would not be as good. (2) This is true even for a super-intelligent being who knows all the lower-level facts about biology and psychology and physics, for example. (3) So the higher-level concept of inelastic demand is a privileged concept. It is privileged not merely in the sense that it serves our subjective aims, but in a much more objective sense(4) So many of the concepts that higher-level sciences like economics use for causal explanations are objectively privileged.  (5) This also suggests that any theory of causation should ensure that there is a tight connection between causation and objectively privileged categories. (6) Indeed one might go as far as to claim that the only features of the world that have causal power are those that are objectively privileged.
Although this is somewhat of a caricature, I think this basic way of thinking about causation, and explanatory levels, and privileged concepts is mistaken.
I challenge premise two from this argument, the idea that higher-level categories do explanatory work that is in-principle indispensable:
Negative Point: Against Objectively Privileged Categories
I also challenge the move between premise two and premise three, the idea that explanatory indispensability is a good indicator of what categories are objectively privileged.