Re: (1), I can’t do too much better than what I already wrote under “How should we assign weight to different groups of people?” I’d say you can go about as elite as you want if you are good at telling how the relevant people think and you aren’t cherry-picking or using the “No True Scotsman” fallacy. I picked this number as something I felt a lot of people reading this blog would be in touch with and wouldn’t be too narrow.
Re: (2), this is something I hope to discuss at greater length later on. I won’t try to justify these claims now, but other things being equal, I think it favors
more skepticism about most philosophical arguments (I think a lot of these can’t command the assent of a broad coalition of people), including arguments which depend on idiosyncratic moral perspectives (I think this given either realist or anti-realist meta-ethics)
more skepticism about strategies for doing good that “seem weird” to most people
more respect for the causes that top foundations focus on today
more effort to be transparent about our thinking and stress-test unusual views we have
But this framework is a prior not a fixed point, so in case this doesn’t settle issues, it just points in a specific direction. I’d prefer not to get into details defending these claims now, since I hope to get into it at a later date.
I’d say you can go about as elite as you want if you are good at telling how the relevant people think and you aren’t cherry-picking or using the “No True Scotsman” fallacy.
If your advice is “Go ahead and be picky about who you consider elites, but make a good-faith effort not to cherry-pick them and watch out for the No True Scotsman fallacy” then I may merely agree with you here! It’s when people say, “You’re not allowed to do that because outside view!” that I start to worry, and I may have committed a fallacy of believing that you were saying that because I’ve heard other people argue it so often, for which I apologize.
I really appreciate your thoughts on this thread :-)
I think that on any given specific question, it’s generally possible for a sufficiently intelligent and determined person to beat out elite conventional wisdom. To the extent that you and I disagree, the point of disagreement is how much investigation one has to do in order to overturn elite conventional wisdom. This requires a subtle judgment call.
To give an example where not investigating in sufficient depth leads to problems, consider the error that GiveWell found in the DCP-2 cost-effectiveness estimate for deworming. A priori one could look at the DCP-2 and say “Conventional wisdom among global health organizations is to do lots of health interventions, but they should be trying to optimize cost-effectiveness, deworming is the most cost-effective intervention, there are funding gaps for deworming, so the conventional wisdom is wrong.” Such a view would have given too little weight to alternative hypotheses (e.g. it being common knowledge among experts that the DCP-2 report is sloppy, experts having knowledge of the low robustness of cost-effectiveness estimates in philanthropy, etc.)
Great. It sounds like may reasonably be on the same page at this point.
To reiterate and clarify, you can pretty much make the standards as high as you like as long as: (1) you have a good enough grip on how the elite class thinks,(2) you are using clear indicators of trustworthiness that many people would accept, and (3) you make a good-faith effort not to cherry pick and watch out for the No True Scotsman fallacy. The only major limitation on this I can think of is that there is some trade-off to be made between certain levels of diversity and independent judgment. Like, if you could somehow pick the 10 best people in the world by some totally broad standards that everyone would accept (I think this is deeply impossible), that probably wouldn’t be as good as picking the best 100-10,000 people by such standards. And I’d substitute some less trustworthy people for more trustworthy people in some cases where it would increase diversity of perspectives.
Great, thanks. The unjustified examples for (2) help.
I’m interested in how the appropriate elite would vary in size and in nature for particular subjects. For instance, I imagine I might place a little more weight on certain groups of experts in, say, economics and medical effectiveness than the top 10% of Ivy League grads would. I haven’t thought about this a great deal, so might well be being irrational in this.
Re: (1), I can’t do too much better than what I already wrote under “How should we assign weight to different groups of people?” I’d say you can go about as elite as you want if you are good at telling how the relevant people think and you aren’t cherry-picking or using the “No True Scotsman” fallacy. I picked this number as something I felt a lot of people reading this blog would be in touch with and wouldn’t be too narrow.
Re: (2), this is something I hope to discuss at greater length later on. I won’t try to justify these claims now, but other things being equal, I think it favors
more skepticism about most philosophical arguments (I think a lot of these can’t command the assent of a broad coalition of people), including arguments which depend on idiosyncratic moral perspectives (I think this given either realist or anti-realist meta-ethics)
more adjustment toward common sense in cost-effectiveness estimates
more skepticism about strategies for doing good that “seem weird” to most people
more respect for the causes that top foundations focus on today
more effort to be transparent about our thinking and stress-test unusual views we have
But this framework is a prior not a fixed point, so in case this doesn’t settle issues, it just points in a specific direction. I’d prefer not to get into details defending these claims now, since I hope to get into it at a later date.
If your advice is “Go ahead and be picky about who you consider elites, but make a good-faith effort not to cherry-pick them and watch out for the No True Scotsman fallacy” then I may merely agree with you here! It’s when people say, “You’re not allowed to do that because outside view!” that I start to worry, and I may have committed a fallacy of believing that you were saying that because I’ve heard other people argue it so often, for which I apologize.
I really appreciate your thoughts on this thread :-)
I think that on any given specific question, it’s generally possible for a sufficiently intelligent and determined person to beat out elite conventional wisdom. To the extent that you and I disagree, the point of disagreement is how much investigation one has to do in order to overturn elite conventional wisdom. This requires a subtle judgment call.
To give an example where not investigating in sufficient depth leads to problems, consider the error that GiveWell found in the DCP-2 cost-effectiveness estimate for deworming. A priori one could look at the DCP-2 and say “Conventional wisdom among global health organizations is to do lots of health interventions, but they should be trying to optimize cost-effectiveness, deworming is the most cost-effective intervention, there are funding gaps for deworming, so the conventional wisdom is wrong.” Such a view would have given too little weight to alternative hypotheses (e.g. it being common knowledge among experts that the DCP-2 report is sloppy, experts having knowledge of the low robustness of cost-effectiveness estimates in philanthropy, etc.)
Great. It sounds like may reasonably be on the same page at this point.
To reiterate and clarify, you can pretty much make the standards as high as you like as long as: (1) you have a good enough grip on how the elite class thinks,(2) you are using clear indicators of trustworthiness that many people would accept, and (3) you make a good-faith effort not to cherry pick and watch out for the No True Scotsman fallacy. The only major limitation on this I can think of is that there is some trade-off to be made between certain levels of diversity and independent judgment. Like, if you could somehow pick the 10 best people in the world by some totally broad standards that everyone would accept (I think this is deeply impossible), that probably wouldn’t be as good as picking the best 100-10,000 people by such standards. And I’d substitute some less trustworthy people for more trustworthy people in some cases where it would increase diversity of perspectives.
Great, thanks. The unjustified examples for (2) help.
I’m interested in how the appropriate elite would vary in size and in nature for particular subjects. For instance, I imagine I might place a little more weight on certain groups of experts in, say, economics and medical effectiveness than the top 10% of Ivy League grads would. I haven’t thought about this a great deal, so might well be being irrational in this.