Interesting post, but I remain skeptical of electoral politics as charity because I am skeptical that even an extremely strong rationalist can identify the best candidate.
But here is an experiment that would reduce my skepticism.
Find a pair of experimental subjects I would consider strong rationalists who have both informed themselves about some election past or present. Since it would be convenient for our pool of potential subjects to contain people who want the experiment to succeed or want it to fail, ask the subjects to swear that they have never communicated with each other about the election. Even better are pairs that have had little communication at all. If they cannot so swear, drop that pair from the experiment. Ask each subject which candidate would maximize the expected global utility of the election. Record their answers.
Actually, we are picking triples (a, b, e) where a and b are rationalists (and subjects) and e is an election past or present. Find as many triples as practical subject to the above conditions. Of course, take pains not to limit the experiment to subjects who share the same political outlook or the same political and moral intuitions. If the number of triples in which a and b agree on a candidate is high enough, it would greatly increase the expected utility I assign to this form of charity.
Interesting post, but I remain skeptical of electoral politics as charity because I am skeptical that even an extremely strong rationalist can identify the best candidate.
But here is an experiment that would reduce my skepticism.
Find a pair of experimental subjects I would consider strong rationalists who have both informed themselves about some election past or present. Since it would be convenient for our pool of potential subjects to contain people who want the experiment to succeed or want it to fail, ask the subjects to swear that they have never communicated with each other about the election. Even better are pairs that have had little communication at all. If they cannot so swear, drop that pair from the experiment. Ask each subject which candidate would maximize the expected global utility of the election. Record their answers.
Actually, we are picking triples (a, b, e) where a and b are rationalists (and subjects) and e is an election past or present. Find as many triples as practical subject to the above conditions. Of course, take pains not to limit the experiment to subjects who share the same political outlook or the same political and moral intuitions. If the number of triples in which a and b agree on a candidate is high enough, it would greatly increase the expected utility I assign to this form of charity.