Hmm. Well, certainly the full follow-up would be a tremendously valuable thing to have, so whenever you have the time to write it, I definitely think that you should!
But for now:
… any rationalist organization deserving of the name would carefully choose its norms, structure and bylaws to reflect those of the most successful existing organizations (empiricism!), with care taken to exclude the aspects of those organizations that are inimical to group or individual rationality.
Hm, indeed. Obvious follow-up questions:
By “rationality”, here, do you mean epistemic or instrumental rationality? Or both? (And if “both”, which is prioritized if they conflict? Or, must aspects of successful organizations that are inimical to either epistemic or instrumental rationality, and either group or individual versions of each, all be excluded?)
What should an aspiring rationalist organization do if, upon empirical investigation, it turns out that the norms, structure, and bylaws shared by all the most successful existing organizations turn out to all be inimical to group or individual rationality?
Regarding both follow-up questions, I have two answers:
Answer 1: I don’t intend for this to be a dodge, but I don’t think it really matters what I think. I don’t think it’s practical to construct “the perfect organization” in our imagination and then anticipate that its perfection will be realized.
I think what a rationality organization looks like in practice is a small group of essentially like-minded people creating a Schelling point by forming the initial structure, and then the organization evolves from there in ways that are not necessarily predictable, in ways that reflect the will of the people who have the energy to actually put into the thing.
What’s interesting is that when I say it that way, I realize that it sounds like a recipe for disaster. But also note that essentially no other organization on Earth has been formed in any other way.
Answer 2: I personally would create separate organizational branches for epistemic and instrumental focus, such that both could use the resources of the other, but neither would be constrained by the rules of the other. Either branch could use whatever policies are most suited to themselves. Think two houses of a congress. Either of the branches could propose policies to govern the whole organization, which could be accepted or vetoed by the other branch. There’s probably also a role for something like an elected executive branch, but at this point I am grasping well beyond my domain of expertise.
What’s interesting is that when I say it that way, I realize that it sounds like a recipe for disaster. But also note that essentially no other organization on Earth has been formed in any other way.
I’m not so sure about that. Perhaps you meant “no other [freestanding] organization on Earth has been formed [by people without access to massive amounts of resources] any other way”.
To elaborate on the distinction, I can point to organizations like NASA, or the Manhattan Project. They were created, from the beginning, as large groups, with pre-planned bureaucratic structures and formal lines of authority. While there was a certain level of organic growth, it’s not like these things started in a garage and grew outward from there. Similarly, in private industry, when IBM embarked on its OS/360 project, or Microsoft embarked on Windows Vista, these were not small efforts started by a group of insurgents. Rather, they were responses to concrete opportunities/threats (a new mainframe, Netscape) that were identified by the leadership of the parent organization, who then mobilized the appropriate resources.
I think this matters beyond mere pedantry. You’ve identified one way to spread rationalism—bottom up, by establishing a small group of rationalists, who then spread their doctrine outwards. I’m saying there’s another way: identify leaders, convince them that rationality is a good thing to focus on, and then have them mobilize the appropriate resources to spread rationality. If you could convince Bill Gates, Jeff Bezos, Warren Buffett and Carlos Slim that they should fund rationality with the full force of their collective fortune, then that would potentially do more in a year to spread rationalism than a small organization organically growing for a decade.
Of course, “go big or go home” has its own failure modes, but it’s not actually self-evident that it’s more risky than starting small and spreading outwards. Moreover, the really successful small groups employed a hybrid strategy. They started out as a small group, until the could amass enough resources and prestige to convince influential decision-makers that their cause was worth supporting. The canonical (pun fully intended) example is the Catholic Church, of course. It started as a small, often persecuted group of followers of a particular religious prophet, indistinguishable from the other Jewish spin-offs. However, through steady proselytizing, it grew and converted the aristocracy of the Roman empire. At that point, from the the conversion of Constantine onwards, it became the state religion, and spread rapidly wherever the Roman empire held sway.
I think MIRI also employed a hybrid strategy. I will say, it seems much easier to deploy a “go big or go home” approach after you’ve already created a minimum viable organization, rather than attempting to poach thinkfluencers without even having that groundwork in place.
Hmm. Well, certainly the full follow-up would be a tremendously valuable thing to have, so whenever you have the time to write it, I definitely think that you should!
But for now:
Hm, indeed. Obvious follow-up questions:
By “rationality”, here, do you mean epistemic or instrumental rationality? Or both? (And if “both”, which is prioritized if they conflict? Or, must aspects of successful organizations that are inimical to either epistemic or instrumental rationality, and either group or individual versions of each, all be excluded?)
What should an aspiring rationalist organization do if, upon empirical investigation, it turns out that the norms, structure, and bylaws shared by all the most successful existing organizations turn out to all be inimical to group or individual rationality?
Regarding both follow-up questions, I have two answers:
Answer 1: I don’t intend for this to be a dodge, but I don’t think it really matters what I think. I don’t think it’s practical to construct “the perfect organization” in our imagination and then anticipate that its perfection will be realized.
I think what a rationality organization looks like in practice is a small group of essentially like-minded people creating a Schelling point by forming the initial structure, and then the organization evolves from there in ways that are not necessarily predictable, in ways that reflect the will of the people who have the energy to actually put into the thing.
What’s interesting is that when I say it that way, I realize that it sounds like a recipe for disaster. But also note that essentially no other organization on Earth has been formed in any other way.
Answer 2: I personally would create separate organizational branches for epistemic and instrumental focus, such that both could use the resources of the other, but neither would be constrained by the rules of the other. Either branch could use whatever policies are most suited to themselves. Think two houses of a congress. Either of the branches could propose policies to govern the whole organization, which could be accepted or vetoed by the other branch. There’s probably also a role for something like an elected executive branch, but at this point I am grasping well beyond my domain of expertise.
I’m not so sure about that. Perhaps you meant “no other [freestanding] organization on Earth has been formed [by people without access to massive amounts of resources] any other way”.
To elaborate on the distinction, I can point to organizations like NASA, or the Manhattan Project. They were created, from the beginning, as large groups, with pre-planned bureaucratic structures and formal lines of authority. While there was a certain level of organic growth, it’s not like these things started in a garage and grew outward from there. Similarly, in private industry, when IBM embarked on its OS/360 project, or Microsoft embarked on Windows Vista, these were not small efforts started by a group of insurgents. Rather, they were responses to concrete opportunities/threats (a new mainframe, Netscape) that were identified by the leadership of the parent organization, who then mobilized the appropriate resources.
I think this matters beyond mere pedantry. You’ve identified one way to spread rationalism—bottom up, by establishing a small group of rationalists, who then spread their doctrine outwards. I’m saying there’s another way: identify leaders, convince them that rationality is a good thing to focus on, and then have them mobilize the appropriate resources to spread rationality. If you could convince Bill Gates, Jeff Bezos, Warren Buffett and Carlos Slim that they should fund rationality with the full force of their collective fortune, then that would potentially do more in a year to spread rationalism than a small organization organically growing for a decade.
Of course, “go big or go home” has its own failure modes, but it’s not actually self-evident that it’s more risky than starting small and spreading outwards. Moreover, the really successful small groups employed a hybrid strategy. They started out as a small group, until the could amass enough resources and prestige to convince influential decision-makers that their cause was worth supporting. The canonical (pun fully intended) example is the Catholic Church, of course. It started as a small, often persecuted group of followers of a particular religious prophet, indistinguishable from the other Jewish spin-offs. However, through steady proselytizing, it grew and converted the aristocracy of the Roman empire. At that point, from the the conversion of Constantine onwards, it became the state religion, and spread rapidly wherever the Roman empire held sway.
I think MIRI also employed a hybrid strategy. I will say, it seems much easier to deploy a “go big or go home” approach after you’ve already created a minimum viable organization, rather than attempting to poach thinkfluencers without even having that groundwork in place.