When I ran for City Council, one of the issues I wrote about on my website was finding ways to reduce staff hours spent running open comment sessions on things -- which are basically worthless if what you wanted was some representative sense of "public opinion" -- and channel the money into running a smaller number of more intensive Deliberative Polls. The goal in the long run would be for the Council to be able to really take the temperature of the public on perhaps 2-4 major issues each year.
As it happens we have a major research center on how to do this well, just down the road. ( https://deliberation.stanford.edu/ ) I called them up to see if any of their grad students might be interested in helping to start a program, for free or cheap, as part of their research, and they were very into the idea. Alas, I lost, 45-55, to the incumbent.
(This wasn't an issue I talked about a ton, because it just doesn't have strong traction with most voters. But it was, TBH, one of the ideas I was more excited about trying to implement if I got the chance.)
I find sortition to be an intriguing idea and definitely one worth trying. But (probably due to my being a myopic American), I had a different idea about how to try it. Almost every state in the U.S. has two chambers of a legislature, largely for historical reasons. Why not try and convince one to make *one* of the two houses chosen by sortition (with each representative chosen for, say, two years)? It would mitigate the risk factor, since anything they would do would also have to clear a normal (elected) legislative body. And it could give a sense of the strengths and flaws in a *reasonably* low-risk way.
> There are some permanent political oversight bodies governed by sortition, mostly in Europe. It is hard to get information about them in English and they have all been, as best I can tell, attached to government and very local. Municipal government experiments in German Belgium and in Paris with permanent sortition councils are welcome, but unlike our proposal, they are organs of the state.
Have you read "Against Elections" by David Van Reybrouck? He's the Belgian author that promotes "Deliberative democracy" and "Citizens' assemblies" which are the closest thing to your proposal. He was the one that pushed for it in Belgium, and now we have a permanent advisory citizens' assembly here in Brussels. If you're having trouble accessing information on them maybe I can help/translate. I might also be able to get you in contact with him (I happen to know his nephew, and am currently working with an academic social choice theorist).
"Government Watch" is perhaps a good name for branding, though in academic circles you might consider calling it a "permanent citizen's assembly", so people can immediately grasp the concept and slot it into its school of thought.
Are you familiar at all with the Californian institution of the civil grand jury? If not, I think it holds lessons for you both about the potential uses and downsides of your proposal. DM me if you want a braindump on it from someone who has done multiple terms of civil grand jury service.
I think about sortition a lot. The experience argument against it set forth above is worth thinking about, but current results of elections make it moot. Almost no one acts as if they have experience. It’s the nature of a view of politics as entertainment and the delusion that one can either communicate or learn in sound bites.
Don’t think your proposed experiment is quite big enough.
I'm broadly sympathetic to sortition, with maybe a few caveats, and I generally agree with everything you've said in this post, so I only have one little quibble.
I think you're significantly underestimating the logistical and cost issues behind Government Watch. For one thing, you'd have to pay much more than the median wage because otherwise half your sample would have to take a giant pay cut to participate. For a national Government Watch, you'd have to persuade people to move. Then there's the long-run wage penalty of career interruptions. All of these will create sampling errors and significantly drive up costs.
I have to say - I'm pretty sceptical of these things. If they are institutionally spun up for dealing for individual matters I suspect without doing anything obviously unfair you could probably get whatever outcome you like (within reason) - a combination of:
- Selecting which demographics should be proportionally represented (and which ignored)
- How random selection works (eg does it go to adresses or persons on some governement ID list - doubly so if you want minority representation - you need to confirm who is a member of that class)
- What are the rules on opting out (and relatedly what is the salary), both at entry and leaving part-way.
- What process selects the evidence presented.
- How long do they get to learn/decide, what is cut off in the case of overruns.
- What determines the result - can it be fillibustered/vetoed, what level agreement.
Now a more permanent body does help - although it does become a lot more onerous for the participants. But, does permancy in this postition change peoples attitudes (power corrupts idea).
More importantly I doubt it would have much effect - consider them finding support for the death penalty and hard labour in prisons - I suspect they wouldn't persuade you. You have come to your conclusions based on some serious thought on the matter. But for the most part everyone's political opinion largely comes downstream of people who have considered things far more than they have. Without changing the minds of existing experts I doubt it could play that big of a role.
The existing system relies in part on information downstream of lifelong experts - the media will generally get one or more experts in the field for comment (and indeed shape their presentation from those experts opinion) - this system might simply add in a layer of less informed people giving their takes.
That said - no reason not to try it and see I reckon. It is certainly imaginable that it could be worthwhile.
So from what I can tell you're trying to pay a random set of 120 people to comment on stuff after much deliberation. Thoughts:
"They largely start from the premise that there is a range of conventional, acceptable opinions in the middle represented by old media gatekeepers whose positions are— by hypothesis “reasonable” and hence candidates for being right. The better approach to polarisation is, I think, much more radical. Perhaps polarisation happened because something had already gone deeply wrong."
There's two separate claims here, both of which I more or less agree with, but which point in separate directions. One is the enthymeme that in fact the old media gatekeepers are wrong or at least unrepresentative. Well, they got us in the current mess, so that's credible. The other is that something has gone deeply wrong that caused polarization. Yeah, I would agree with that too, though I'm not sure if it's the old media gatekeepers being stupid, as it seems to have gotten worse with their decline--could be social media bubbles or the structure of the two-party system or something else. Maybe all of the above!
From what we can tell, the median voter is socially conservative and economically progressive relative to the current parties:
So empowering the average citizen may not give you what you wanted. I think you'd see a much more populist body than most people on Substack (or reddit) would like. They'd probably take positions you don't like on trans issues and immigration, for instance.
To your specific issues:
I'm not sure involving the general public in foreign aid issues is necessarily going to be all that productive. PEPFAR happens to be popular (we had the good luck to have a GOP president who actually was a Christian after all), but in general people are skeptical. Although as you say people underestimate the amount of foreign aid, a lot of them don't think we should be giving any foreign aid at all, and in bad times in general people are less giving. I'm not sure giving them numbers is going to make that much of a difference.
I've taken calc, linear algebra, basic statistics, and ODE, and a lot of the AI stuff still confuses me. And I still don't have an opinion apart from noticing everyone with experience in the field who doesn't work for one of the AI companies seems to think this is dangerous. Imagine what the average person is going to think when confronted with one set of Very Smart People saying "AI will bring us infinite wealth!" and another set of Very Smart People saying "AI will destroy the world!" They'll probably just shut down the whole thing. Maybe that's what you want, I guess.
With the tax policy thing you might get somewhere, everyone is sick of doing their taxes and having the government send your their estimate like in Europe might save us a lot of time! So, kudos!
Is this a third reason, or a disguised version of one of the others?
Politics is inherently about shared resources. But not all outcomes we care about are appropriately addressed using universally shared resources. EAs want to take action for the common good. But people do not agree unanimously about what counts as a problem or how to solve them. Politics is sometimes necessary, as only one relevant shared resource is feasible, and we have to somehow come to a decision about how it will be used. But politics can be over-applied. Because of the mania for centralization and standardization experienced in the industrial age, we still suffer from a centralization hangover, and seem determined to continue applying the one-size-fits-all hair of the dog. The cost is the difficulty created for people who care who have good ideas by people who either don’t care or have kooky ideas. The alternative is to take action without waiting for politics to catch up.
In such a system, I would also be interested in the "minority opinion" from a sortition body, ie. what do the dissenting 33% agree on? This is a detail, though, the core idea is great. I suppose patrons of the Government Watch would be disqualified from being selected? Or am I thinking of it too much as a jury rather than a newspaper?
I've had similar, less sophisticated ideas in the past.
Many years ago I proposed a once-in-a-century constitutional revision for Canada that would be made up of randomly selected Canadian citizens advised by a body of former politicians, current academics and other experts.
More recently I had the idea of mandatory citizens' town halls. I find the media often do not ask the questions I want them to. The idea is that, once a month, a federal minister (different each month) would be forced to do an in-person or virtual town hall with randomly selected citizens which would be televised on CBC.
I think your idea has more promise than either of these ideas.
Would you mind posting this on the EA Forum? (There's a longtermist institutional reform tag already.) This is an EXCELLENT twist on something I have strongly supported ever since the Citizens' Climate Commission in France conclusively showed that it robustly produce people with higher information and lower time preference than any bourgeois politician.
Con: this is also a con, simply because this idea is alien to our system and conveys a kind of class hostility. Without wishing to argue that our democracy really does produce the greatest good for the greatest number, explicitly introducing adversarialism into a political system, may not work out well for those with little power.
Con: the name also lacks ambition beyond becoming a tribunate. My own idea is that we build another house of the legislature and then try to negotiate its way into the constitution as a check and balance. This is what was done in Magna Carta and the American Constitution in the judicial branch — which is a partnership between the Pyramid of Political Power and the people as constituted in an allotted body.
With experience, it can conceivably completely flip the existing foundation of the legislature with the lower (elected) house being the seat of government (in the Westminster system). But it would only do that gradually and with a lot of accumulated political experience that was widely seen to have been successful. Whether it should or not is also an open question which might make a good debating point today, but none of us know without more experience.
"There are some permanent political oversight bodies governed by sortition, mostly in Europe" My impression of the East Belgian one is that, though it's a standing council, not one convened at the behest of the Government or the legislature, it seems to be an 'agenda setting' rather than an oversight body. The standing council appoints other special purpose councils which go out and investigate an issue and report back to the standing council — this has the nice property that it creates a "blind break" between the two councils. In addition to seeking internal agreement, the special council is seeking to craft something that will be persuasive to another body of their peers. (See Alex Kovner's discussion of the blind break outlined here: https://chatgpt.com/share/67dfa3af-f874-8003-8263-e168bc54ca17. But at least as far as I know it doesn't have much of a tribunate flavour. (It was supported by all six political parties represented in the legislature, so that helps explain its collaborative, rather than antagonistic intent. Of course once constituted it can always argue for a change of purpose and the associated powers.)
I totally agree with you on the role of supermajorities. I'd rather reduce the majority to 60%, but would need to look at the stats closely (It might be a good argument for a somewhat larger body — which gives one better odds of avoiding a false positive (>60% in the body but <50% in the population). Non-supermajorities should also count — obviously in procedural matters within the body, but perhaps in other ways — but there should be a threshold in which the body can speak with authority for the considered opinion of the people.
You envisage a full-time body. I'm not so sure. As a 'start-up' I think paying people to meet — say for two days a month on weekends and 2-3 teleconferences wouldn't just keep costs down. It would improve takeup. Unless and until you have some state support, or at the very least much more public recognition of the role, I think you'll run into big hurdles in recruiting people if you're asking them to leave their job. In time to become a mature part of government, I expect it needs to be full-time. Then again you can have several of these bodies covering different areas.
I used to argue with Nick Coccoma about sortition when he was posting on Substack, and while I think you make a better case I'm going to argue with you too.
The argument against sortition is analogous to the argument against legislative term limits- term limits are a bad idea because they cap the institutional knowledge of legislators but not of lobbyists. Term limits result in a body of legislators who are less savvy, less independent, and more reliant on the institutional knowledge of others (who have their own ulterior motives in the process) because they have little time to develop that knowledge themselves. Likewise, the primary risk of citizens assemblies is over reliance on the experts (both subject matter experts and *procedural* experts) who are chosen to present the issues to a mostly uninformed body of decision makers. It's very easy to appear neutral and even-handed while presenting a biased case- by lying mostly by omission, for example, or by ignoring relevant complexities under the excuse of presenting a simple and comprehensible summary. This kind of expert capture would be mitigated somewhat when the citizen assembly was dealing with issues that are some combination of high salience and easy to understand (e.g. same sex marriage), and most pronounced when it was dealing with issues that are more obscure and difficult to wrap one's mind around.
Your Risk Management section acknowledges this failure mode, but I don't think it could be overcome by any commitment to scrupulous neutrality on the part of the administrators. It's a structural problem, not a problem of individual bad actors.
I love this idea. A great first use for initial funding seems like it would be getting good translations for the European experiments and aggregating them.
When I ran for City Council, one of the issues I wrote about on my website was finding ways to reduce staff hours spent running open comment sessions on things -- which are basically worthless if what you wanted was some representative sense of "public opinion" -- and channel the money into running a smaller number of more intensive Deliberative Polls. The goal in the long run would be for the Council to be able to really take the temperature of the public on perhaps 2-4 major issues each year.
As it happens we have a major research center on how to do this well, just down the road. ( https://deliberation.stanford.edu/ ) I called them up to see if any of their grad students might be interested in helping to start a program, for free or cheap, as part of their research, and they were very into the idea. Alas, I lost, 45-55, to the incumbent.
(This wasn't an issue I talked about a ton, because it just doesn't have strong traction with most voters. But it was, TBH, one of the ideas I was more excited about trying to implement if I got the chance.)
I find sortition to be an intriguing idea and definitely one worth trying. But (probably due to my being a myopic American), I had a different idea about how to try it. Almost every state in the U.S. has two chambers of a legislature, largely for historical reasons. Why not try and convince one to make *one* of the two houses chosen by sortition (with each representative chosen for, say, two years)? It would mitigate the risk factor, since anything they would do would also have to clear a normal (elected) legislative body. And it could give a sense of the strengths and flaws in a *reasonably* low-risk way.
> There are some permanent political oversight bodies governed by sortition, mostly in Europe. It is hard to get information about them in English and they have all been, as best I can tell, attached to government and very local. Municipal government experiments in German Belgium and in Paris with permanent sortition councils are welcome, but unlike our proposal, they are organs of the state.
Have you read "Against Elections" by David Van Reybrouck? He's the Belgian author that promotes "Deliberative democracy" and "Citizens' assemblies" which are the closest thing to your proposal. He was the one that pushed for it in Belgium, and now we have a permanent advisory citizens' assembly here in Brussels. If you're having trouble accessing information on them maybe I can help/translate. I might also be able to get you in contact with him (I happen to know his nephew, and am currently working with an academic social choice theorist).
"Government Watch" is perhaps a good name for branding, though in academic circles you might consider calling it a "permanent citizen's assembly", so people can immediately grasp the concept and slot it into its school of thought.
You might also be interested in the idea of introducing sortition elements into the voting system itself. If that sounds intriguing to you, I wrote a blogpost about it: https://bobjacobs.substack.com/p/reasons-to-vote-in-non-deterministic
Are you familiar at all with the Californian institution of the civil grand jury? If not, I think it holds lessons for you both about the potential uses and downsides of your proposal. DM me if you want a braindump on it from someone who has done multiple terms of civil grand jury service.
I think about sortition a lot. The experience argument against it set forth above is worth thinking about, but current results of elections make it moot. Almost no one acts as if they have experience. It’s the nature of a view of politics as entertainment and the delusion that one can either communicate or learn in sound bites.
Don’t think your proposed experiment is quite big enough.
I'm broadly sympathetic to sortition, with maybe a few caveats, and I generally agree with everything you've said in this post, so I only have one little quibble.
I think you're significantly underestimating the logistical and cost issues behind Government Watch. For one thing, you'd have to pay much more than the median wage because otherwise half your sample would have to take a giant pay cut to participate. For a national Government Watch, you'd have to persuade people to move. Then there's the long-run wage penalty of career interruptions. All of these will create sampling errors and significantly drive up costs.
I have to say - I'm pretty sceptical of these things. If they are institutionally spun up for dealing for individual matters I suspect without doing anything obviously unfair you could probably get whatever outcome you like (within reason) - a combination of:
- Selecting which demographics should be proportionally represented (and which ignored)
- How random selection works (eg does it go to adresses or persons on some governement ID list - doubly so if you want minority representation - you need to confirm who is a member of that class)
- What are the rules on opting out (and relatedly what is the salary), both at entry and leaving part-way.
- What process selects the evidence presented.
- How long do they get to learn/decide, what is cut off in the case of overruns.
- What determines the result - can it be fillibustered/vetoed, what level agreement.
Now a more permanent body does help - although it does become a lot more onerous for the participants. But, does permancy in this postition change peoples attitudes (power corrupts idea).
More importantly I doubt it would have much effect - consider them finding support for the death penalty and hard labour in prisons - I suspect they wouldn't persuade you. You have come to your conclusions based on some serious thought on the matter. But for the most part everyone's political opinion largely comes downstream of people who have considered things far more than they have. Without changing the minds of existing experts I doubt it could play that big of a role.
The existing system relies in part on information downstream of lifelong experts - the media will generally get one or more experts in the field for comment (and indeed shape their presentation from those experts opinion) - this system might simply add in a layer of less informed people giving their takes.
That said - no reason not to try it and see I reckon. It is certainly imaginable that it could be worthwhile.
Adam Grant beat you to it: https://www.nytimes.com/2023/08/21/opinion/elections-democracy.html . I also remember reading an argument for this in National Review back in the 90s.
So from what I can tell you're trying to pay a random set of 120 people to comment on stuff after much deliberation. Thoughts:
"They largely start from the premise that there is a range of conventional, acceptable opinions in the middle represented by old media gatekeepers whose positions are— by hypothesis “reasonable” and hence candidates for being right. The better approach to polarisation is, I think, much more radical. Perhaps polarisation happened because something had already gone deeply wrong."
There's two separate claims here, both of which I more or less agree with, but which point in separate directions. One is the enthymeme that in fact the old media gatekeepers are wrong or at least unrepresentative. Well, they got us in the current mess, so that's credible. The other is that something has gone deeply wrong that caused polarization. Yeah, I would agree with that too, though I'm not sure if it's the old media gatekeepers being stupid, as it seems to have gotten worse with their decline--could be social media bubbles or the structure of the two-party system or something else. Maybe all of the above!
From what we can tell, the median voter is socially conservative and economically progressive relative to the current parties:
https://medium.com/@xenocryptsite/revisiting-the-famous-2016-economic-views-vs-social-views-scatterplot-55016c1b8888
So empowering the average citizen may not give you what you wanted. I think you'd see a much more populist body than most people on Substack (or reddit) would like. They'd probably take positions you don't like on trans issues and immigration, for instance.
To your specific issues:
I'm not sure involving the general public in foreign aid issues is necessarily going to be all that productive. PEPFAR happens to be popular (we had the good luck to have a GOP president who actually was a Christian after all), but in general people are skeptical. Although as you say people underestimate the amount of foreign aid, a lot of them don't think we should be giving any foreign aid at all, and in bad times in general people are less giving. I'm not sure giving them numbers is going to make that much of a difference.
I've taken calc, linear algebra, basic statistics, and ODE, and a lot of the AI stuff still confuses me. And I still don't have an opinion apart from noticing everyone with experience in the field who doesn't work for one of the AI companies seems to think this is dangerous. Imagine what the average person is going to think when confronted with one set of Very Smart People saying "AI will bring us infinite wealth!" and another set of Very Smart People saying "AI will destroy the world!" They'll probably just shut down the whole thing. Maybe that's what you want, I guess.
With the tax policy thing you might get somewhere, everyone is sick of doing their taxes and having the government send your their estimate like in Europe might save us a lot of time! So, kudos!
Is this a third reason, or a disguised version of one of the others?
Politics is inherently about shared resources. But not all outcomes we care about are appropriately addressed using universally shared resources. EAs want to take action for the common good. But people do not agree unanimously about what counts as a problem or how to solve them. Politics is sometimes necessary, as only one relevant shared resource is feasible, and we have to somehow come to a decision about how it will be used. But politics can be over-applied. Because of the mania for centralization and standardization experienced in the industrial age, we still suffer from a centralization hangover, and seem determined to continue applying the one-size-fits-all hair of the dog. The cost is the difficulty created for people who care who have good ideas by people who either don’t care or have kooky ideas. The alternative is to take action without waiting for politics to catch up.
In such a system, I would also be interested in the "minority opinion" from a sortition body, ie. what do the dissenting 33% agree on? This is a detail, though, the core idea is great. I suppose patrons of the Government Watch would be disqualified from being selected? Or am I thinking of it too much as a jury rather than a newspaper?
This is a great idea.
I've had similar, less sophisticated ideas in the past.
Many years ago I proposed a once-in-a-century constitutional revision for Canada that would be made up of randomly selected Canadian citizens advised by a body of former politicians, current academics and other experts.
More recently I had the idea of mandatory citizens' town halls. I find the media often do not ask the questions I want them to. The idea is that, once a month, a federal minister (different each month) would be forced to do an in-person or virtual town hall with randomly selected citizens which would be televised on CBC.
I think your idea has more promise than either of these ideas.
Would you mind posting this on the EA Forum? (There's a longtermist institutional reform tag already.) This is an EXCELLENT twist on something I have strongly supported ever since the Citizens' Climate Commission in France conclusively showed that it robustly produce people with higher information and lower time preference than any bourgeois politician.
It's up! I'd appreciate your upvote :-) https://forum.effectivealtruism.org/posts/YgSsHurKoH3RuDe2Q/how-to-change-the-world-a-lot-with-a-little-introducing
The name "Government watch"
Pro: it captures the tribunate nature of the body — an idea from Ancient Rome's political system. It's an institution for the plebs to challenge the oligarchs who run the show. It is thus in the tradition of John McCormick's Machiavellian Democracy. https://www.cambridge.org/core/books/machiavellian-democracy/B059E5BE6F7BEF0012099517E2494CBA. Intriguingly Camila Vergara gives a much more left-wing spin to this idea and doesn't seem to reference McCormick. https://www.plebeianphilosopher.com/
Australia's John Braithwaite has also argued for tribunate powers, though his inspiration is Sun Yat Sen and other early Asian democrats which is intriguing. https://press-files.anu.edu.au/downloads/press/n9134/html/cover.xhtml?referer=&page=0#
Con: this is also a con, simply because this idea is alien to our system and conveys a kind of class hostility. Without wishing to argue that our democracy really does produce the greatest good for the greatest number, explicitly introducing adversarialism into a political system, may not work out well for those with little power.
Con: the name also lacks ambition beyond becoming a tribunate. My own idea is that we build another house of the legislature and then try to negotiate its way into the constitution as a check and balance. This is what was done in Magna Carta and the American Constitution in the judicial branch — which is a partnership between the Pyramid of Political Power and the people as constituted in an allotted body.
With experience, it can conceivably completely flip the existing foundation of the legislature with the lower (elected) house being the seat of government (in the Westminster system). But it would only do that gradually and with a lot of accumulated political experience that was widely seen to have been successful. Whether it should or not is also an open question which might make a good debating point today, but none of us know without more experience.
"There are some permanent political oversight bodies governed by sortition, mostly in Europe" My impression of the East Belgian one is that, though it's a standing council, not one convened at the behest of the Government or the legislature, it seems to be an 'agenda setting' rather than an oversight body. The standing council appoints other special purpose councils which go out and investigate an issue and report back to the standing council — this has the nice property that it creates a "blind break" between the two councils. In addition to seeking internal agreement, the special council is seeking to craft something that will be persuasive to another body of their peers. (See Alex Kovner's discussion of the blind break outlined here: https://chatgpt.com/share/67dfa3af-f874-8003-8263-e168bc54ca17. But at least as far as I know it doesn't have much of a tribunate flavour. (It was supported by all six political parties represented in the legislature, so that helps explain its collaborative, rather than antagonistic intent. Of course once constituted it can always argue for a change of purpose and the associated powers.)
I totally agree with you on the role of supermajorities. I'd rather reduce the majority to 60%, but would need to look at the stats closely (It might be a good argument for a somewhat larger body — which gives one better odds of avoiding a false positive (>60% in the body but <50% in the population). Non-supermajorities should also count — obviously in procedural matters within the body, but perhaps in other ways — but there should be a threshold in which the body can speak with authority for the considered opinion of the people.
You envisage a full-time body. I'm not so sure. As a 'start-up' I think paying people to meet — say for two days a month on weekends and 2-3 teleconferences wouldn't just keep costs down. It would improve takeup. Unless and until you have some state support, or at the very least much more public recognition of the role, I think you'll run into big hurdles in recruiting people if you're asking them to leave their job. In time to become a mature part of government, I expect it needs to be full-time. Then again you can have several of these bodies covering different areas.
I used to argue with Nick Coccoma about sortition when he was posting on Substack, and while I think you make a better case I'm going to argue with you too.
The argument against sortition is analogous to the argument against legislative term limits- term limits are a bad idea because they cap the institutional knowledge of legislators but not of lobbyists. Term limits result in a body of legislators who are less savvy, less independent, and more reliant on the institutional knowledge of others (who have their own ulterior motives in the process) because they have little time to develop that knowledge themselves. Likewise, the primary risk of citizens assemblies is over reliance on the experts (both subject matter experts and *procedural* experts) who are chosen to present the issues to a mostly uninformed body of decision makers. It's very easy to appear neutral and even-handed while presenting a biased case- by lying mostly by omission, for example, or by ignoring relevant complexities under the excuse of presenting a simple and comprehensible summary. This kind of expert capture would be mitigated somewhat when the citizen assembly was dealing with issues that are some combination of high salience and easy to understand (e.g. same sex marriage), and most pronounced when it was dealing with issues that are more obscure and difficult to wrap one's mind around.
Your Risk Management section acknowledges this failure mode, but I don't think it could be overcome by any commitment to scrupulous neutrality on the part of the administrators. It's a structural problem, not a problem of individual bad actors.
Thanks Nathaniel,
I tried to respond to that problem by proposing a 'council of elders' from the alumni of the body chosen by its members.
https://app.box.com/s/hlwwv64txsww4n3tkuz03c6ybswd5yvw
I love this idea. A great first use for initial funding seems like it would be getting good translations for the European experiments and aggregating them.
French EA-adjacent YouTubers have made very good videos on the Citizens' Climate Commission, might be worth translating:
https://www.youtube.com/watch?v=yuopYkpmvCo
https://www.youtube.com/watch?v=hqFynZQWYB8