Amia Srinavasan's critique of "Doing Good Better: Effective Altruism and a Radical New Way to Make a Difference" by William MacAskill
is excellent, pointing out how much Effective Altruism (EA) simply takes for granted (e.g., capitalism itself, and the status quo).
But the worst is that EA is psychopathic -- as psychopathic as Darwinian evolution itself: Evolution's sole criterion is maximizing (“satisficing,” really) net survival and reproduction, and EA utilitarianism’s sole criterion is maximizing net utility. Both turn a blind, "rational" eye on collateral damage, including proximal collateral damage.
That’s not morality, it’s mathematics. And treating emotion as if it were just a vice or a distraction is not a virtue. In fact, it was (ironically) Darwinian evolution itself (the origin of sentience, hence suffering, hence all moral problems) that implanted empathy and compassion in mammals and birds (at least), probably in the adaptive service of reproductive success (in altricial K-selected species, at least, of which we are one). Without those traits we’d all be psychopaths (as r-selected, precocial species may be).
In the trolley problem, any mother who would not flip the switch to save her own child rather than another’s would be a psychopath. If it was for the sake of saving two children of another instead of her own child that she failed to flip the switch then she’d be an EA utilitarian — and a psychopath.
Altruism needs to be compassionate, not just “effective.” And charity begins at home (or it never begins at all). Nor would an uncharitable world be a hospitable one to live in: It would be rather like a zombie world. Surely an (emotionally!) weighted combination of EA and proximal compassion would be better than EA alone.