Game B

Jim Rutt
Entrepreneur and Founder of Game B
Jim Rutt
Entrepreneur and Founder of Game B
Jordan Hall
Co-Founder, Game B
Jordan Hall
Co-Founder, Game B

What is Game B? We sit down with the founders of this alternative civilization movement to understand its genesis, central tenets, and current state.

Show Notes

In this episode Jenny sits down with Jim Rutt and Jordan Hall, founders of the Game B movement, to understand it's history and main tenets. Jim and Jordan have thought deeply about many facets of the Denizen inquiry, this conversation dives into many core concepts related to social and institutional redesign.

  • The genesis of Game B [3:05]
  • Defining complexity science [4:04]
  • Epistemology and epistemic modesty [5:26]
  • Jim's original paper the Root Doc [5:49]
  • The Stanton meetings and establishing a political party [7:01]
  • Evolving from The Emancipation Party to Game B [9:06]
  • Defining Game B [12:23]
  • The dissolution of the original Game B cohort / shifting to "spore mode" [12:38]
  • Main tenets of Game B [16:02]
  • Design and design principles [23:04]
  • Collective intelligence and technology in Game B [24:23]
  • Governance, scale, and Dunbar's number [28:55]
  • The importance of protocols [36:56]
  • Game A's scaling advantage and self terminating nature [38:40]
  • Upgrading humans in Game B with psycho technologies [42:42]
  • Where Game B is now [52:20]
  • How a Game B transition might take place [1:00:27]
  • The relationship between theory and practice [1:03:39]
  • Permaculture vs. Game B [1:11:13]
  • The Circular Economy vs. Game B [1:12:04]
  • Game B and Economic Growth [1:13:32]


"Jordan Hall (JH): It is simultaneously the aim of Game B to be able to allow us to continue to be a technological species without guaranteeing our extinction. But then, also, to – I mean, this is sort of like two sides of exactly the same coin, to therefore be able to use technology in the process of developing a more effective collective intelligence and do so in a way which does not generate, by the design implications of how you're using the technology, some set of failure conditions in what it is you're trying to do." 

[00:00:39] Jenny Stefanotti (JS): That's Jordan Hall, one of the founders of Game B. Also, an entrepreneur. That quote felt particularly timely. And this is a Denizen podcast. I'm your host and curator, Jenny Stefanotti. 

In this episode, we're talking about Game B. When I started on this path with the Denizen inquiry, I kept hearing about Game B and I didn't know exactly what it was. I'm sure a lot of listeners are familiar with Daniel Schmachtenberger and his work. And so, I felt like it was a really important concept to dive into and understand. 

And so, for this conversation, I invited Jim Rutt and Jordan Hall to join us. Both founders of Game B. Jim is a former technology executive, former chairman of the Santa Fe Institute. Host of the podcast, The Jim Rutt Show. I'm sure many of you listen to that. And if you don't, you should definitely check it out. And also, one of the founders of Game B. 

Jordan is an entrepreneur. He's a deep thinker on so many different aspects of the Denizen conversation. Our episode with him on parenting is one of the most listened to episodes of the podcast. He is a close colleague of Daniel Schmachtenberger's. And he was part of the Game B conversations from the inception. So he's considered a founder. 

But then after the initial group disbanded, and we'll talk about that history in this conversation, Jordan, Daniel and Forrest Landry picked up and evolved the Game B thinking. 

In this conversation, we discussed the main ideas in thinking of what Game B actually is. The genesis of it. The early meetings that Jim organized and Jordan attended, and why that original group dispersed, how they think about theory versus experimentation, and their thinking on technology's role in supporting a transition to Game B. We touch on important concepts like Dunbar's number, governing at various skills, and the role of economic growth. We also touched on how Jim thinks about transitioning between Game A and Game B and the current state of the Game B conversation. 

As always, you can find show notes for this episode on our website, There you can also sign up for our newsletter. I send information on each episode to your inbox alongside announcements about other events and offerings from Denizen and our partners. 

Again, Game B just felt like something that we should all understand. And, of course, Jim and Jordan are always such a delight to listen to because they thought so deeply for so many years about so many of the topics we cover here on the podcast. So many great nuggets of wisdom in this one. 


[00:03:05] JS: Jim, the Genesis of Game B was with you. And it started with a paper that you wrote in 2012. 

[00:03:11] Jim Rutt (JR): Yep. 

[00:03:12] JS: Tell us about it. 

[00:03:14] JR: You know, I've been thinking about the problems of our society for a really long time, since about 1994. Even though I was in the belly of Game A at that time, I was a CTO of a multi-billion-dollar, multinational publishing company. Didn't get much more Game A-y than that. But the curtains were starting to rise and I was starting to realize that this system that I was in the middle of, and at least minor player in, was fundamentally rotten. That it didn't deliver a high-quality of life for most people. And this was right at the time when the wealth gap was ever-growing. The cultural capital, social capital in our societies were declining. Crime was at a very high rate at that time. That unfortunately has turned around, et cetera. 

But for a number of reasons I just started realizing that we were on the wrong road. And I didn't have any answers yet. I just had a diagnosis that this is not the right road. Then over the next several years, I kept my eyes open. Thought about it a little bit. Started learning more about complexity science, which turned out to be a hugely powerful lens in thinking about how a society unfolds over time. The fact that all the agents are strategic. And so, you can't predict what they'll do very well. And if nothing else, it gave me a good dose of epistemic modesty. The thing you gain most when you study complexity is to realize how little capacity, we really have to predict the unfolding of a complex system. 

[00:04:40] JS: For those in the audience who aren't familiar with complexity science, can you just give us the headline? 

[00:04:45] JR: Yeah, it's basically how complex things arise from simplicity. In fact, the full name of our science is the science of complexity from simplicity. As an example, you, each human, is a complex adaptive system. Then at the bottom is made up of atoms jiggering around, which then constitutes molecules, which then constitutes cells, which then constitutes tissues, which then constitutes organs and then systems within the body. And then an entity, a person. And then that person is part of an eco-system, which is then part of a planet. We have this emergent hierarchy from simplicity to complexity that's highly dynamic and highly non-linear. 

[00:05:26] JS: And you also mentioned epistemic, which is a term that some people may not understand either. So do you mind just explaining that quickly too? 

[00:05:35] JR: It basically means how do we know things and the nature of knowing. And epistemic modesty is to realize that we don't know as much as we think we might know, particularly the unfolding a complex adaptive system.

[00:05:49] JS: So tell us what was in this paper, which I think had a lot of momentum and then led to a gathering. 

[00:05:53] JR: Yeah. It was, as I said, still in the belly of the beast for many years, for several more years. And then 2000, I quit. Sold the company. Left. I went to the Santa Fe Institute. Worked there. Continued to think. Ran into Jordan Hall there actually, which turned out to be an important part of the journey. 

He was a new, young trustee at the Santa Fe Institute. We'd said howdy a few times at events but probably not much more than that. But after a trustee meeting, it must have been around 2008, we sat down and had – I don't know – a three, four-hour conversation. And it was a “holy shit, we are seeing things in a mighty similar way.”  And we then carried on the conversation sporadically from time to time for the next three or four years. And that, as well as my own thinking during that period, led to the publishing of this paper, kind of known in the history of the Game B movement as the Rutt Doc. 

And I think it was 64 pages of a brain dump of ideas. Somewhat organized. And the first person I sent it to was Jordan. Then I sent it out to a few other people. And people seemed to resonate with it. At least that it was calling out some of the issues and possibly some of the solutions. And we said, "Shit. Maybe we should do something about it." 

So a circular conversation, an email, eventually led to us establishing an online community on a system called Basecamp. And then in September of 2012, we said we need to get together in-person. In-person is critical to work these things. 

I think the first Stanton meeting was about eight people or nine people, something like that. We all flew to Virginia and we all met in the small city of Stanton. Very charming place. And then head to the Shenandoah Valley. And we said, "Hmm. What should we do here? I guess we should do something." We decided we'd start a political party. 

And so, we didn't know shit about political parties. That's all right. Didn't stop us before. And so, we spun up this effort. Started recruiting more people in. And every six weeks, the whole team from around the country and even a few International people flew into Stanton, Virginia. 

And we had another meeting. We did more work. The work continued both online – especially online between meetings. And we amazingly, rapidly, I look back at this, created a platform, a whole series of written statements, a website, a pretty complicated, too complicated as it turned out, membership model, et cetera. And launched it all in late 2012. 

By that point I think we had 25 or 30 core members. And we went out to sign up members for this new political party called The Emancipation Party. You can actually see the effort that we created in that time at And our reforms that we wrote are still up there and they're actually fairly prescient.

Being 2012, one of our reforms was called The Citizenship Wage, which is essentially the same thing as UBI. We also talked about a new monetary system. We talked about radical transparency. We talked about ending the drug wars. We talked about a whole new radically better financial system. We're really pretty far ahead of the curve. In ideas, I'll give us an A+. Quality of these ideas, they still impress me when I go back and read these reforms. 

[00:09:06] JS: So would you say that this scope of the ideas in The Emancipation Party was the same as where Game B has evolved? There was just a different theory of change in terms of how to get there where one made an assumption that you could create a new party and enact reform with the existing political system? Whereas, as we'll get into it, the other is talking about more of a kind of decentralized, distributed, experimental approach that starts at a small scale and evolves up? 

[00:09:34] JR: I think that the emancipation party was on a narrower basis. And that was actually its problem. We had naively thought that we could take the existing structure of the nation-state and late-phase capitalist industry, et cetera. Slap on ten reforms and the world would be well. Now we realized these reforms would make some changes in the capitalist structure, et cetera. But I think we underestimated the social change that had to happen along with institution building to make real change happen. After – go ahead. 

[00:10:07] JS: Well, when you say you underestimated the social change, so was it the course of these convenings or five of them, right? Over the course of I think nine months, you met every six weeks or so? 

[00:10:16] JR: That's about right. 

[00:10:16] JS: So was it in the course of these convenings and conversations it sounds like the group grew? And there was also a robust online community that happened in parallel with these convenings. The thinking evolved. It sounds like the scope of social change evolved as well as the theory of change. 

[00:10:32] JR: Yeah. Yeah. I think we realized that, at least at this time, straightforward partisan politics was not going to succeed and was insufficient. That, actually, our issues as a civilization are much broader than the details of what could be enacted in law by a political party. And they had to do with the fact that our systems are not able to adapt anywhere near the rate of change that our technologies and our rapidly closing in on the limits of the carrying capacity of the earth need us to be able to change. To be able to adapt to these new situations. To make sense of the situation we're in. To act with wisdom and courage in time to save civilization from the multiple heads of the meta crisis, which could do us in if we don't. And so, we much, much, much broaden the scope of our thinking. 

And it's funny, originally we thought of this thing that became Game B as an on-ramp to the political party. And I would say that was certainly my interest in it. And that was literally drawn that way on the whiteboard when Jordan first went to the whiteboard. And we did all of our work at the – a lot of our work at this gigantic 24-foot white board in our offices in Stanton. 

And then Jordan showed, "Here's our party. And then here's a ramp to get on to the party. We'll call that this and yadi-yada." And then over time we realized that we had kind of missed the forest for the trees. And the bigger questions were much more important than partisan politics. 

And so, the Game B movement in the broader sense became the real focus of attention of the group and the interest in the party. Eventually it atrophied away and there were only a few of us that were still interested in that. 

[00:12:10] JS: Mm-hmm. And so, right. It evolved from thinking about a political party and more partial and marginal reform to a more comprehensive thinking about a movement and thinking about culture. 

[00:12:23] JR: Correct. And in fact, the definition essentially of Game B is to create a new civilization-level social operating system. And that's way beyond the charter of an elected political party under a constitutional government. 

[00:12:38] JS: Yeah. And we'll get to that in just a moment. Because I do find this is a fascinating case study that I want to come back to on the back end of the conversation once we get more into the meat of Game B itself. And then tell everyone what happened with those convenings and with that initial cohort that was involving these ideas. 

[00:12:54] JR: Yeah. We were moving along. We were moving on different projects. We tried to build our own online system. Basically, we took a fork of the Reddit search source code. That was one approach. We tried to raise some philanthropic money. We had a growing number of people in the online community that were moving in different directions. And this was quite interesting. Here we are trying to build a social movement. So that's a civilization-level social operating system. 

And we ended up fractioning ourselves, and exploding and falling into rancor and argument. And it was quite ironic that we couldn't even make it work with maybe a hundred people at the most in our community.

[00:13:33] JS: Which I find fascinating, right? Because we've had this conversation with Forest about this meta challenge of governance, particularly governance at different scales, which I will certainly get into in the conversation. And here you have an example where governance broke down. But maybe it didn't, right? Because it went into what you referred to as spore mode. 

[00:13:54] JR: Yep. Yeah. Once it finally became clear to me that this was degenerate. There was just no reconciling the factions at this point in time. I said, "However, there's lots of good work." And every single person in the movement is good-hearted and is coming from the right place. But somehow, we just haven't figured out how to operate together. So I said, "Spore mode, people." 

And for those of you who know biology, spores are a way that ferns, in particular, and fungi reproduce. And they're kind of like teeny little seeds. And the interesting thing about spores is they can last a long time. They can last 50 years in the ground. And then when the conditions are correct, the plant will sprout or the fungi will sprout and grow up and reproduce. 

So that's why at the end of the day, rather than saying this was a waste of time. Thank you for your efforts. Blah-blah. I said, "Let us take what we have learned here – and we have learned a lot. Both good things to do and things that didn't work. And let's turn us all into spores and let us go out into the world. And perhaps at the right time, and the right place, and the right soil, and the right condition this thing might regenerate. And we'll probably add to it along the way. But game over for now. But we're not dead because we're spores that are now spread to the world." 

[00:15:09] JS: Right. Which debatably may have been the actual governance mode that makes more sense. Doing something as complex is what you were looking to do. The notion that everyone would come together and agree on something as complex as a new human civilization. And the mechanism to get there, it seems to be a fatal assumption of governance. But the ideas that you brought out and the evolution of those ideas was very important and meaningful.

[00:15:37] JR: Yeah, I think you're right. It's a very wise way to frame it. We did not invest at all in governance. We never even had a conversation on governance. And I suppose we thought that somehow governance would emerge. And it didn't. And attempts to do so failed. And it was a very strong lesson learned. And much of the time we've spent since then has been thinking about governance. So at least that lesson we did learn. 

[00:16:02] JS: Okay. So here's some things from the Game B Wiki, which I think collectively represents, to some extent, the current aggregate thinking. What is Game B? There is no set definition of Game B. Game B is an approach to thinking up a sustainable win-win alternative style of civilization instead of the unsustainable win-lose model that we have now. 

I think this is very important, this notion of win-win versus win-lose. And can you design an incentive structure where it's not a rivalrous but it's a everyone wins set of incentives? It's a memetic tag that aggregates myriad projects and visions that are modeling this potential future civilization form. It's not an ideology. It's not well-defined. But it has four fundamental characteristics. It's self-organizing. It's network-oriented. It's decentralized. And it is meta stable for extended periods of time. 

And then you talk a little bit about some goals and aims. It aims to maximize human potential in flourishing, Maslow's notion of self-actualization. Life in Game B is about self-expression and leaving no barriers to friendship, fellowship and love. 

But then you also – the Wiki also talks about some of the big goals. Solving the meaning crisis. People looking for a purpose in life. Something bigger than themselves. Solving the sense-making crisis. This is something that we've talked about with Daniel with the epistemic crisis and the work that he's doing with the Consilience Project. Solving the ecological crisis, which we have talked about as we have looked at these alternate models of capitalism or even economics that you would no longer define as capitalism. I'm interested to probe how that fits into Game B. Or if that's just too marginal of a reform on Game A. And also, I want to talk about what's the path that gets us there and how these other ideas fit into that. 

It solves the community crisis, which I think is a really important piece of the puzzle we haven't explored too much, which is that life in cities is isolated and people have oriented towards each other as transactional service providers. And there's an assumption that's baked in that humans are better off living in communities in ranges of what's called the Dunbar Number. A really important concept. We'll get into in just a second. 

And building more integrated and social communities is at the heart of Game B. I want to get back to that in just a second. It solves the surveillance crisis, it solves the economic crisis and it solves the education crisis. That's the kind of delineation of big goals of Game B, which is no small feat. 

But Game B also is civilization-level in scope. Encompasses these kinds of meta questions of governance and sense-making that we've explored. Parenting, education, healthcare, economy, food and housing are all part of it. 

Okay. What do you think of what I just said? Does that do justice to how you conceptualize or define Game B when someone asks you that question? Jim, why don't we hear from you first and let's hear from Jordan? 

[00:18:54] JR: Well, yeah, I think it's actually – it presents it in a realistic way, which is that it is still large, amorphous and not sharply-defined. And that's good. Because we're at the point where we're still exploring really what the full meaning of Game B is. And so, I think that this very broad perspective that you gave from the wiki is realistic. 

I will say each of us in Game B focuses our attention on different parts of it. Because obviously, that very wide front is too large for any single person to master or even to frankly understand what's going on. 

And so, for instance, I am very, very interested in the community aspects that you brought up. And how do we operate as communities? How do communities operate with each other at a larger scale? How do we develop essentially a fractal civilization from the bottom-up but based on solid communities at a scale around the Dunbar number? Maybe a little bit bigger than that as a way forward. Overall, I think it's as good a broad-based description as any. 

[00:19:57] JS: Jordan, how do you answer that question when someone says what is Game B? 

[00:20:01] JH: Well, it depends on who I'm interacting with. 

[00:20:03] JS: Of course. 

[00:20:05] JH: I mean, the reality – for example, the problem of something like a Wiki or, frankly, the problem of something that uses like definitions, is that if you're a kind of mind that is looking to be done with thinking, then, in this case, it will actually end up leading you astray. Because you'll think that you've actually gotten a handle on this thing if you've thought about the words that are on the page. 

But the point is, if you're somebody who's using this as a way to think more deeply, like to orient your thinking in a more effective way, then it's not bad, right? It's sort of like a cluster of stars and your sailing and they give you directionality. Very few of the points that are on there, ones that I would kind of go, "Ah. No. There's a way off." But not even the whole is actually really the thing. 

Because really the thing is bizarre. I mean, this gets very esoteric. But it involves a changing of the way that we actually do stuff, like use definitions, to make sense of the world and then to make choices together. If you're looking at it through a particular lens, it's not going to work. If you are looking at it through a different lens, that Wiki is very useful. It's full of all kinds of good stuff. 

Part of it, by the way, is the fact that Jim and I had nothing to do with it. In fact, nobody I would consider to be important, or senior, or official, or formal had anything to do with it.  It was done entirely organically by people who were, in some sense, doing it for their own purposes. Like they were gathering notes and trying to explain it to themselves and to each other. That's a performative example of the thing that it's talking about. The performativity is closer to the real thing than the wiki itself. 

[00:21:44] JS: Yep. Okay. You make an important point, Jordan, that we have a tendency to want to have a handle on something, which by nature has to be evolutionary and amorphous. 

At the same time, we want to operate in a way that is emergent and not chaotic. So we need some degree of scaffolding, right? You had mentioned the stars to point in a direction. Instead of everyone kind of going in circles, right?

[00:22:14] JH: Right. 

[00:22:14] JS: So the stake in the ground is this notion of designing something. It sounds like these sort of incentive structures that lead to these what's called omni-win decisions. Is that fair? 

[00:22:24] JH: Well, that's one piece for sure. 

[00:22:25] JS: Right. 

[00:22:26] JH: The word epistemology came up. Epistemological frameworks that shape the way that we conceptualize what possible choices might look like have the same – they also bind our choice flow. That's another one. Epistemological frameworks are sort of at the same level of importance as incentive structures. 

I mean, we can kind of operate in that way. We can say, "Okay, what sorts of categories of phenomena or notions are really, really close to the center of the important stuff to be able to design something like this?" And by the way, if we do that, all we're really doing is we're talking about what is the – what's sort of the fundamental toolkit of civilization design? 

[00:23:04] JS: Yeah. I mean, I think design is a really important concept here, which is like, what are the design principles? And how might you think about design mindsets or processes versus necessarily having answers, right? 

I had a conversation with someone just the other day. We're talking about like will we just solve all these problems and then just live happy lives into perpetuity? Versus kind of constantly having to solve the new problem and the frontier evolving, right? And I think that tendency for the former is what's sometimes flawed.

[00:23:41] JH: It's part of the problem. 

[00:23:41] JR: Yeah, it's very important to keep in mind that we have to be careful about the word design sometimes. There're long and tedious discussions on the Facebook Game B group, which for people who are interested in seeing what's going on, is probably still the best single place.  Should we use the word design? Or should we use the word garden? Or should we use the word grow? 

And if we are really practicing epistemic modesty about the unfolding of a complex adaptive system, which a civilization is, I'm going to be a little bit careful not to be social engineers and think too strictly in design terms. At least not in the design terms we're used to thinking about, which is either engineering or architecture. 

[00:24:23] JS: Well, I want to probe some of the core concepts of Game B with the two of you. One of them being collective intelligence, which is related to this complex question of governments. The wiki said something about Game B's core advantage to Game A being collective intelligence. 

Brett Weinstein has talked about Game B needing to be competitively advantageous to Game A. But I'm curious how the two of you think about collective intelligence in the context of Game B. Particularly, what role technology might play in facilitating that? 

[00:25:00] JR: I think technology is a piece but it's far from the whole thing. The social parts of the operating system are actually more important. A category, a topic, that's related to sense-making is one called coherence, which I keep coming back to as near the essence of Game B-ness. At least in the early stages. 

And coherence is the idea that people can operate together at a much higher level of effectiveness than you would think. You can think of coherence as the capability of groups of people to operate at truly high-effectiveness through shared models, knowledge, experience, vocabulary and values wrapped in trust and emotional transparency and engaging with each other in good faith. 

If we can operate like that, it's hard to do. I mean, it's even hard for two people to do. I mean, I've been able to get into coherence with two people, me and another person at a time fairly often. Three is a little harder. Four is a little harder. 25-person Zoom meeting, forget about it. But to the degree you can build coherence, you can make sense. Now that's a special small group style of sense-making.  Where technology may be able to help us out is at a larger scale. The attempts so far for large community technologically-enabled sense making has been pretty disappointing. 

[00:26:19] JS: Jordan, what's your take on collective intelligence in the role that tech-could play? 

[00:26:24] JH: Yeah. I mean, obviously, the first point is something like that's a super profound question. And so, I would just caution that whatever I'm about to say is going to be like a super tiny slice through it. 

Maybe one point is to say, yes, the notion that Brett hits on and that I sit on in a number of different languages, like the languaging of OODA Loops, the notion that one of the primary characteristics, almost even one of the primary strategic assets of Game B. is a vastly, call it stronger or a more powerful collective intelligence is true. At least in the sense that that's a fundamental proposition. 

By the way, even just saying that requires a deep question of the nature of, what are the characteristics? How do you describe – what would it mean to be a stronger, more powerful or more effective collective intelligence? 

The other side of it is the question of technology. And the first thing that came up for me, which feels a little bit humorous, is actually very fucking carefully. One of the characteristics, Game A, is a sort of a combination of factors that leads to a very unfortunate degree of recklessness with regard to the use of technology in supporting and aiding life. But more specifically actually supporting an aiding strategy. 

And it is simultaneously the aim of Game B to be able to allow us to continue to be a technological species without guaranteeing our extinction. But then, also, to – I mean, this is sort of like two sides of exactly the same coin, to therefore be able to use technology in the process of developing a more effective collective intelligence and do so in a way which does not generate, by the design implications of how you're using the technology, some set of failure conditions in what it is you're trying to do. 

[00:28:17] JS: I want to integrate a couple of different really critical components into this conversation. I've already brought up the question of technology and collective intelligence. And part of the reason I'm asking that is I think I'm curious about your thoughts on how technology relates to this really important question about scale. And then, also, with respect to collective intelligence. I want to get into what was the fissure in the original community around this question of working on oneself versus the institutions. But I think related to working on oneself, this question of the advantages potentially in Game B of the human because of the way that we engage in what you described as psycho technologies, right? 

There're a couple of different pieces of this puzzle and I want to tease them out. The first I think is just a really profound and important question that I don't think we have a pretty good handle on that I keep coming back to, which is this question of scale and governance at scale. And we've kind of touched on that but we haven't said it explicitly, Dunbar's number. Jim, why don't you explain it to us? 

[00:29:14] JR: Yeah, sure. Robin Dunbar, a British anthropologist. At least he lives in Britain now. Did a lot of research on the scale at which forager people, so-called hunter-gatherers, actually lived. And what he found was very, very seldom did a forager group get to be more than 150 people. As it approached that scale, it almost inevitably fissioned. 

He also then talked to cognitive scientists and discovered that irrespective of what society one lives in, one's not actually able to know and keep track of more than something like 150 humans. And so, that became known as Dunbar's number. There are actually other smaller Dunbar numbers. Groups of five, groups of 20, groups of 50, et cetera. But the most well-known one is 150. 

And we have at least a rough sense that that might be a good target size for the level on the ground of organizing Game B, Proto B Civium type communities going forward. So that's just the first part. But as you alluded to, if we're going to fall back to our forager ancestors, we're going to be hungry all the time and our best technology is going to be stone axes or something like that. 

How do we capture the benefits of scale and building a civilization that can span the globe and eventually span the galaxy? And yet at the same time, have the psychological and bodily grounding of living in a civilization on the ground community I should say that's of the right scale, Dunbar plus maybe as much as 10x Dunbar. But probably not larger than that. 

[00:30:48] JS: Well, let me just add a couple of quick points on the Dunbar concept, which I think is really important. One is that there's a question about how many relationships one can maintain or one feels like they actually know someone. And from a game theory perspective, I think it's quite interesting to think about what happens beyond a certain scale? Like in a scale of sub-150, ostensibly, if somebody comes and steals something from you, somebody in the community is going to know about it. And then beyond a certain scale, maybe they can get away with that and no one will know who they are. 

There are interesting informal institutional dynamics that take place or take hold at a community sub a certain number that break down once we get to scale. And once we got to larger scales, that's when formal institutions like judicial systems and laws and legal infrastructure were needed. 

Part of what I'm wondering is how technology might enable some governance and collective intelligence beyond the scales that were stable or that we've been good at historically. Jordan, what are your thoughts? 

[00:31:49] JH: When you're talking about the question of scale, there are three fundamental characteristics that you have to be considering. One, which is sort of the one that gets talked about the most, has to do with the number of – let's just call it individuals that are part of some other higher-order ontological objects. We talked about 20 and 150. And fill in the blank, dot-dot-dot. So that's the capacity of the “we,” of the thing that is scaling to include some number of subsidiary components. 

The second characteristic has to do with the implications for the subsidiary components of being part of the thing that is scaling. And this has to do with things like what the local choice-making and decision-making will look like from the inside. 

Tyson Yunkaporta I think makes a very nice distinction here between – I think he calls it growth and increase. I could have a giant pile of completely uniform grains of, say, monoatomic gold, right? So I've scaled it in the sense that I can get a big, big, big pile. It's a heap. 

But for example, linking back to the notion of complexity, organic processes will not spontaneously emerge in a heap. And so, to be able to not have the thing that is scaling – by the way, I'm about to have to throw a bunch of concepts out there that I don't think are going to be either common currency or easy to define quickly. But to avoid the thing that is scaling spontaneously transforming into a closed system, a complicated system, is the second characteristic. 

For example, ants or hive insects have developed a capacity to scale in the sense that a very large number of discrete organisms can participate in the hive but at the cost of radically reducing the actual local capacity of each distinct agent, each distinct member. And this creates limits to the capacity of the larger phenomenon in generating collective intelligence, right? There's sort of an asymptote in what it can and can't do. You can add another million ants to the ant hill, it's not going to be able to do a whole bunch of other kinds of things because of the fundamental characteristic of how it's transformed into a complicated system with regard to the individual organisms. 

And then the third has to do with some of the things you're talking about, like the carrying cost of the mechanisms that are enabling the scaling to happen. You mentioned the game theoretic transition. The tit-for-tat model. If you don't know that, you could just Google it. What it tells you is that if you're playing a prisoner's dilemma once, then the game theory looks like, always defect. But if you iterate it, then the game theory shifts to double tit-for-tat depending on exactly how you design it. 

And that's very mapped very closely to one of the boundary conditions of the Dunbar 150 number, which is, at Dunbar 150, you're effectively always playing a deeply iterative prisoners' dilemma game with everybody. And so, it spontaneously puts everybody in a game theoretic construct of double tit-for-tat as opposed to always defect. 

As soon as the number of people gets outside of that number, based on being humans – if you'd like, I can go to detail about what I mean by that, people begin to move into always defect zones because the bandwidth of the quality of memory and interaction with the other person starts to mirror a non-iterative prisoners' dilemma game, right? 

That spontaneous phase transition in the underlying dynamics of how people's choice making operates from tit-for-tat to always defect is an example of something that can actually be at the limit of a scaling vector. There're lots of others, like the carrying cost of models. Having to keep everybody on the same model, like the paradigmatic approach is sort of where we are right now. Sort of the enlightenment notion. 

Okay. We're all going to have a giant model and it’s going to describe all of reality. And the way it's going to work is, simultaneously, we're going to school everybody and squeeze this model into everybody's head so we can all communicate using the same model. And we're going to continually have to update the periphery of the model to include the novelty that we're encountering. That has a whole bunch of limiting factors on the capacity of that particular kind of system, in this case, the one we're living in, to be able to "scale" without shattering, right? You can shatter it a bunch of different ways. But those are the three fundamental characteristics to look at that are implicated in the notion of scaling.

[00:36:10] JR: Yeah, I just want to just add a slight bit of color on that phase transition. At 150, what Jordan didn't quite say is that it's an artifact of human cognitive capability that we can only keep score more or less simultaneously on 150 objects. It also turns out that chimpanzees can only do it with 20. And dogs can do it with about seven or eight because of the nature of their mental capacity. And as it turns out, that defines the size of a wolf pack, and the size of a chimpanzee band and the size of a human forager group. It turns out that there's an actual capability limit. At least it's not augmented with technology. 

And as you talk about technology, this is something where Jordan and I are very much on the same page on, which is before you actually start thinking about building technology, you've got to go down a level lower. You have to think about protocols. 

Protocols are hugely important. And if you get protocols right, a lot of good things flow from them. Get protocols wrong, you're fighting them forever. I'll give an example. What I mean by a protocol, the Internet is built on a protocol called the internet protocol, IP, and a light wrapper around it called TCP. We have the TCP/IP protocol, essentially the whole Internet is built upon that one simple protocol. 

And because they got it approximately right, the Internet out competed other pre-existing network protocols like DECnet, and SNA, and OSI and several others and has created a really huge network effect,  which we're all getting a lot of value on. 

And so, before one jumps into, "Okay, we need a better Facebook," we feel it's very important to think through at a protocol level, “What is it that we're trying to accomplish as we communicate between entities?”  And entities can be humans or they can be membranes. For instance, we talk about a Proto B or a Civium community on the ground, that has a membrane around it of some sort. Has some rules for things that enter and things that leave. 

If we think about the protocols for sending messages and even atoms across that membrane. And if we get that straight first, then we can build the technology. But to think that, "Oh, yeah. A better Facebook or a better Bitcoin is going to solve the problem of coordination at global scale from units of 150 people," I believe that's naive thinking. 

[00:38:36] JS: Jordan, did you want to build on that? Or else I'll take us in a slightly different direction. 

[00:38:40] JH: Yeah, well, there was something that occurred to me that seems like it might actually be interesting and useful. And I apologize if it's not. Jim mentioned the evolutionary biological cognitive capacity limits that are the underlying constraint on the thing that we call the Dunbar number or the various Dunbar numbers with regard to Homo sapiens. 

And the insider, the point, to say, "Look, evolution –" particularly in the context of discovering the notion of social, right? Social animals of various sorts, "– is a search algorithm." Right? It's looking for various locations in the possibility space of phenotype, in the context of genotype and the context of niche, right? It's always looking to figure out how to generate the most adaptive species. 

Now what we can say is that all the different failure conditions that show up when endeavoring to scale are the kinds of things that evolution would run into. It would be a failure condition. It would just continue to search around it, right? We've got some very large amounts of time that biological evolution has been searching the phase space of organism, right? Evolving of all ability. 

And in the context of that, Homo sapiens shows up as being a limit case of collective intelligence. A limit case of social mammals. If you look at the specific details that have to do with things like neoteny, and brain size and relationship to physiological size. And by the way, some of the limits around the pre-linguistic, early linguistic stages of communication. 

And so, we achieved a certain stable point in actual – I just call it sort of game I, or game indigenous, or game human, which is the form of Homo sapiens that came out of the evolutionary challenge and was commonplace pre what I call Game A or pre the emergence of agricultural or complicated civilization. 

And that had the Dunbar style limits to certain forms of scaling. And we should mention by the way, the Dunbar elements don't stop at 150, right? There are others that are higher and they have very complex natures of how they can operate and what sorts of things can and can be done at larger numbers of people involved with these sort of larger group sizes. 

Then if you say, "Okay. Well, that's a very neat way of describing Game A." Because Game A then looks like a search algorithm seeking on a different path of a different kind of scaling, right? What happened is Game A discovered that by basically converting Homo sapiens into social insects, creating complicated hierarchical formal systems, you could actually scale the number of people a whole lot larger than the Dunbar limit. And by the way, and therefore, always locally I'll outcompete indigenous groups that are around you. In some sense, Game A is that thing which is designed to consume indigenous humans and make them its interior. 

But as it turns out, at the cost of failure conditions of – I've diagnosed three specifics that will always cause a particular Game A instantiation to collapse in some finite time. And so, it's an unstable short-term local optimization of a particular kind of scaling that has the local advantage of being hyper-competitive against the indigenous form of scaling. It actually gets locked into an arms race, vis-a-vis, other versions of itself. But it's also always inexorably self-terminating. It has a long-term terminus in its underlying scaling design. 

This thing gives you a kind of a neat compact way if you track that move of pointing to Game B as being something that doesn't do that sort of thing, right? How do we find a way to invent, i.e. discover and/or invent a new approach to the question of scaling that doesn't have the inexorably self-terminating characteristics of any instantiation of Game A? And also, isn't subject to the set of characteristics that cause game indigenous to be fatally vulnerable to a spontaneous transformation to Game A. 

[00:42:42] JS: I think a lot about this, and I raise this a lot in our conversations – what are the fundamental cognitive limitations of the human species that we need to acknowledge when we think about what an institutional redesign looks like? Or in your case, a civilization redesign? 

We touched on this briefly when I talked to you yesterday, Jordan. We've surfaced one which is limitations in how many relationships we can hold, which then leads to different institutional outcomes. But this split that happened – and I actually laughed out loud when I was listening to Jordan, your podcast on Jim's show, when you mentioned that there was this split with the camp that believed that the focus should be on institutional redesign and the camp that believed that the focus should be on working on oneself. 

And particularly, this interesting debate that happens within the Game B community around what you call psycho technologies? Which includes everything from meditation and yoga, to nootropics, to psychedelics, to more advanced technologies that we imagine in science fiction down the line. 

What made me laugh out loud was how eventually the two of you realized it's a yes-and thing and not an either/or thing. But I want to talk about, because I think this is such an interesting debate that comes up in the context of Game B, but it is important broadly, this question about the belief within the Game B community, if I understand it correctly, was, "Well, hey we can upgrade human capacity through these technologies." Which gives Game B an advantage over Game A. I'm wondering what your thoughts are on this piece of the puzzle. It sounds like you two are sort of in different camps. Am I correct? That you were leaning in different directions on this question, which is now not a question because you agree that you need both? 

[00:44:22] JR: I think where neither of us were in the polar extremes. Though I think we had our biases, I will certainly confess to being somewhat myopic on the benefit of upgrading the units from which you build institutions and more concerned with the institutions. I think Jordan actually was a bit more balanced actually. And there were some other folks that were further in the direction of individual change was the only thing that mattered. 

But as you point out, the most important thing is I think all of us have learned over the last several years and the differences are much smaller than they used to be, which is if you're building institutions, what do you build them from? You build them from humans. Just like when you build Legos, you use Legos to the degree that the humans have upgraded their capacity through psychotechnologies of whatever sort. And that's broadly construed as you laid out. We're not just talking about brain implants, whatever. It could be anything from yoga, to traditional medical psychotherapy, to psychedelics, et cetera. To the degree we have more capable humans, we have more degrees of freedom in the institutions we design. We have more ability to build institutions that can out-compete Game A. 

And so, it's actually both together. And they essentially iterate together. Because, also, this is the interesting thing. The institutions themselves, let's think of education as an institution, will also be helpful in building, up-regulating the capacity of the individual. This literal cycling game between human capacity and institutions that upgrade human capacity, which allow us to build more functional institutions, which then out compete Game A. That's essentially the inner loop of the tactics essentially. 

[00:45:55] JS: Yeah, I have a perspective on this. But I want to hear from Jordan first and then I'll chime in. 

[00:45:58] JH: Well, I guess maybe first, just to provide a more robust definition of psychotechnology. A psychotechnology is anything that is produced by humans that has the consequence of changing the niche of the developing psyche. For example, writing and accounting, or psychotechnologies, as also our prayer and meditation. Because if your interior develops in a culture that is an oral culture, there's a different developmental landscape than if your interior develops in a literate culture. 

And so, you can see, this changes the shape of the psyche. And sometimes these psychotechnologies are broad, flat and largely unconscious like the ones I just mentioned. And some, of course, they're very pointed. Oh, I can think about breathing right now. Changing the shape of my psyche in a very specific way, right? 

But the skillfulness of the awareness. Actually, John Vervaeke is the one who gave me that term. And we very rapidly iterated on the notion of what we called a meta psychotechnology, which is – well, we can go down that rabbit hole if you'd like. 

[00:47:02] JS: Well, at least give me a hint of it. Come on. You piqued my curiosity. 

[00:47:02] JH: And the next piece is – yes, my relationship to sort of the schism or the tension inside Game B was something like frustration, which is to say, because of my developmental history as a reasonably autistic and reasonably successful tech entrepreneur, or that was kind of the identity that I was bringing into the conversation, I came with an easy skillfulness at the level of the objective and at the level of institutions. 

And by the way, at that point, I began to develop a meaningful mastery of being able to talk and play in those sorts of spaces. Whereas things like trying to think about how to have a – what does it take for me to actually change in myself to increase my EQ, that felt really scary, right? So I didn't want to go there. 

But I was very intrigued by the conversations that were coming from that end. So I was coming from one place and noticed that the other place was definitely not full of shit and there's a lot of interesting stuff there and I wanted to explore it. And then the schism happened, where there was this really interesting tension. And from my point of view at the time, at least, a very frustrating tension, where people seem to have very strong and somewhat militant positions and could no longer play nicely with each other. 

[00:48:11] JS: As the former CEO and now chairman of the Neurohacker Collective, I just assumed that you have fallen in the – and I had heard Daniel talk about – a theory of change about just upgrading humans as a first step towards a shift to a new system. That's why I assume that you were in that camp. But I mean, the way that you articulated it, everything is a psycho technology, right? Just given – 

[00:48:31] JH: Everything produced by humans. 

[00:48:33] JS: Given that we're neuroplastic, everything that we do – well, yeah. You wouldn't call it technology unless it was produced by humans. But everything that exists in the world has some sort of effect on the psyche, given neuroplasticity. 

[00:48:43] JH: Yes. And therefore, needs to be considered very deep. This is like the question of ontological design. Every aspect of our built environment feeds back on our psyche, which is then feeding back on our built environment. If you don't think about that whole feedback loop, then you're going to get caught unaware, right? Something is going to show up in your blind spot. And if you're doing big, heavy stuff, complex stuff like trying to manage entire global civilizations, you don't want a whole lot of obvious blind spots. 

Back to that notion. My current perspective is it's not so much a need to create, sort of like cybernetic superhumans, or even vaguely in that direction, as much as it is more in the direction of healing back towards baseline humans. It's more like we actually just need to find a way to bring more current contemporary people up to the level of humaning that was commonplace before the Agricultural Revolution. And then bring that kind of person into the contemporary context of the kinds of psycho technologies we could put to work. And by the way, with the guidance of metaphysico technologies, and I apologize for the esotericism there, is actually substantially more the directionality than say – for example, something like Neuralink. Like, "Okay, can I create cybernetically-enhanced humans that could engage in fiber optic telepathy?" Which could happen. But I'm not saying, by the way, that's off the table. What I'm saying is that's very much in the bucket of being real fucking careful. 

And the other stuff. Like I know for sure that if I can just get people back up to being just ordinarily the kind of thing that humans can normally be, that's got a complex system stability built into it. That sort of thing doesn't tend to go completely sideways. Particularly not asymmetrically. Like you don't have the problem of technological augmentation where you can have gigantic problems emerge from a small design characteristic. Does that make sense the direction I'm coming at it from? 

[00:50:43] JS: Yeah. I mean, I've thought a lot about how the institutional and cultural context essentially up-regulates and down-regulates certain tendencies of human interaction. For me, Burning Man was this extraordinarily eye-opening experience where you put somebody in a radically different context and people behave totally differently than anything else I've ever seen. Because all I've seen is this very Game-A-oriented, capitalism-oriented way of living and set of values that are injected into me through that institutional design. To Jim's point, I think about the feedback loop between them. 

[00:51:20] JR: It's also important to note that the forager-type people and indigenous people had a large suite of psychotechnologies in the more narrow sense. They sang and danced around the campfire until they were in a trance. They often had shamanistic practices that gave at least certain individuals in the tribe or in the band I should say. I like to use the word band as a smaller thing than a tribe and corresponds more to our indigenous period. 

And the shamans developed grand and strange stories, which somehow illuminated the nature of life and the meaning of life in some sense for the members of the band. And all those were early psycho technologies, which helped the forager bands survive and allowed those forger bands that were better at using these psycho technologies to out-compete because they had higher group coherence and higher group effectiveness than those groups that didn't have those capabilities. This is not a new thing. It goes all the way back. 

[00:52:20] JS: Game B sort of went into spore mode. And this was in 2013, right? 2014? As I understand what has happened. Jordan, and Daniel and Forest evolved the thinking. And there's been a bit of a resurgence. Where is the game B movement now? Jordan, I feel like you've got a good – as good of a handle on it as anyone. 

[00:52:42] JH: Well, let's see. Maybe first thing is – and I'm not sure how uniform this was across the entire group. But for me, what happened is in the process of the movement from – literally, we just call it Stanton 3 to Stanton 5. A series of gatherings. And Jim asked me to take on a project that he called the Deep Code project. It was sort of like the R&D project of Game B. 

And so, I did. I dove very deeply into it. You might think about it as something like how do we get clarity on the characteristics of like, "Okay, where are we? And what is this thing that is Game A? And what does it look like? And can we build a sort of a meta-model that can help us make more sense of this sort of thing? 

The result of that actually put me in a disposition where the activity, the participation in the thing that we were calling Game B became existential. Not just sort of important or interesting. That's why I came to the conclusion that the ship that we were on was heading towards an iceberg at very rapid pace and that it was not a turnable thing. We very rapidly began thinking about how to build enough lifeboats and get people on those lifeboats in what might in fact be a remarkably short period of time? 

As the last sort of five years – from 2013 to 2018 moved on, every time I sort of got more information or interacted with different groups that had different perspectives, unfortunately, the time horizon actually continued to shrink. That's the context where I show up in San Diego with people like Daniel and Forrest who just, by sure happenstance, reached effectively the same set of conclusions that came with their own background. Daniel was coming out of the integral universe and wasn't even using complexity language at the time when I met him. Forrest was coming out of the completely bespoke Forrest universe and – 

[00:54:23] JS: That's amazing. 

[00:54:25] JH: Yeah. And communicating with him was vastly harder than it is now. Because we've spent a lot of time – 

[00:54:32] JS: The three of you developed your common language. And then those of us, like myself, helped to translate that for everybody else. 

[00:54:38] JH: Right. Right. And is developing kind of the emotional capacity to undergo the transformation in yourself necessary to be able to communicate more effectively? All kinds of stuff. We built a common language. We also built a – supported each other in developing and increasing capacity to participate in the collaboration that we all knew needed to happen at large scale. 

That I know very closely. Like I was very intimate with that particular set of activities and conversations. For reasons, Brett and his wife, Heather, we kept on a relationship. But it was at a further distance. And several times we almost came into the same orbit but not quite. And then Brett and Eric, of course, have got their thing going on. And then when Olympia or Evergreen went down, Brett got pulled into the eddy around that set of dynamics and the whole IDW thing. So that kind of was a different vibe. 

But at the same time, I think still he very deeply felt and I think also still very deeply feels the meaningfulness of the conclusions in the Game B conversation and dropped that into the Zeitgeist. I think on Joe Rogan, he mentioned something like that. 

[00:55:47] JS: I really appreciated Brett's orientation around prototyping and experimentation in the talks that I've heard. 

[00:55:54] JH: There's a very large number of things to appreciate about Brett.

[00:55:57] JS: I just appreciate his sensibility about really understanding what it meant for something to be a prototype and the kind of mindsets associated with that and how that would evolve over time. 

[00:56:07] JH: Yeah. Yeah. And I think – by the way, just kind of – I think largely as a consequence of perfectly frankly many obvious virtues that Brett happens to embody and the increasing awareness or the increasing consciousness of shit's fucked up. And I don't quite know how to deal with this. So let me look outside of the box that I've been living in, right? Those are two that's been going on. 

The Game B notion, the semantics, just the term started to float around and some people started picking it up. Now, of course, we live in a social media milieu where, unfortunately, was the term that right now that I see on Twitter grifters or virtue signaling. And this I mean in a very specific way. Like counterfeiting is commonplace. And so, some people, a non-trivial number of people, who had skillfulness in marketing saw this as something that had some particular value and began to up-regulate it in the larger social media environment, which had the advantage or the benefit of increasing awareness of the notion. And it had the disadvantage of decreasing its fidelity with its underlying basis. It turned it into pop music to some extent. 

But you did see a growth. And at that point, was this probably – gosh. 2017 or so. Maybe 2018. I can't jump into this more precisely. Jim saw this happening and then chose to endeavor to catalyze more out of it by doing things like creating a number of different Facebook groups that were more formal and creating some incubator style things and things like that. 

Jim certainly knows more about the stuff that's going on in his universe, which is the Proto B movement, and the incubators and the embodied structures. And I'm sure there are frankly just like people out in the Twitter sphere, the Facebook sphere or wherever who know vastly more about the kind of the organic thing. Like you might call it the Game B movement, with which I have essentially no connection. 

[00:58:00] JR: And that's I think a good place to start for the high-level view of where Game B is today, which is it's got a basis in theory, which came out of the Stanton meetings and was then refined and expanded by Jordan, and Daniel and Forrest in particular. 

And then as Jordan alluded to, it somehow came back to life in 2018 and started to get this buzz on the Internet, on Twitter and on Facebook. And by chance, as it turned out, I had established a Facebook group in 2014 that was essentially a rendezvous point for the OGs of the old original Game B movement. And there was like one post a month somebody put up on what they were up to. And that was all there was. 

Well, when the buzz got out in the air, the Game B Facebook group was where people went. They would type in Game B and that's what they'd find. And so, it fairly quickly grew to being several thousand members. And another thing I love about it and why I continue to invest in developing that community is that a lot of new emergent leaders have appeared. 

Not only do we have the core group. But – I don't know. 18 or 20 subgroups on things like parenting, and education and increasing personal sovereignty, building Game B communities, on and on and on. Even there's a Game B group in Spanish for instance that is very clever. 

While these are contributing at the margins, the theory – and I do wish Jordan would follow it a little bit more. He might learn a few things. Truthfully, it's more about finding the people who can lead the movement forward in the next phase. Because theory alone isn't going to get the job done. You have to move from theory to practice at some point. 

And you alluded to the fact that Jordan and I have certain tensions between us. And we always have from the very beginning even though we respect each other a tremendous amount. I'm always going to be more inclined towards practice and Jordan's always going to be more inclined towards theory. 

And so, I continue to invest in things that lead eventually to practice and to experiments at scale. And to me, the most exciting thing going on in the Game B world right now on the practice side is this strong sense from many people, particularly the younger folks, the Millennials, that we need to start bringing into being Game-B-like or Game B transition communities, which we are calling Proto-Bs. 

[01:00:27] JS: So what you've outlined in your Medium post was how you imagined a transition would take place, which was essentially people who are kind of pre-Game B, I think more kind of theoretical conceptualizing Game B might think about. Well, what would education look like in Game B? What does housing look like in Game B? What are all these different pieces of the puzzle? 

And then these communities would come together at smaller scales to actually experiment putting as much of that as they could together thinking about what their relationship looked like to Game A, right? 

One of the things you mentioned, for example, is, "Okay, we could do basic health care in a relatively small community. And then we might go out to the Game A world for some things. But if we collaborated across communities, then we might be able to aggregate to ensure and do more novel institutional structures." 

And those would be experimental. Some of them would be successful. Some of them would die out. Some of them would join forces. And eventually, you would see some dominant structures becoming larger and larger and larger until they were at a scale where you could say we had transitioned to Game B. That's a very brief characterization of what that transition might look like. 

Jordan, how are you thinking – or are you thinking about – Or actually, Jim, did you want to add anything to that before… I'm just curious how Jordan is thinking about theory of change to move towards a Game B. 

[01:01:44] JR: Jordan actually added a lot to this area in theory with his Civium concept, which I'm sure he'll talk about. But yeah, I think you outlined it approximately, which is that there's a sense – let's take these theories we have, which we know are wrong in part. So maybe in total. And let's put them into action. Let's actually live together in communities and try to practice Game B, or at least Game B transition, and see what we learned. 

And further, this is very, very, very important from my perspective, is that the Proto-B movement explicitly says, “We do not have a formula for how to do this. Let us have multiple Proto-Bs with different settings” right? For instance, economic egalitarianism. Should we have total egalitarianism? Should we have partial? Should we have no economic egalitarianism? 

One could make arguments I suppose for any of those within a Game B context. Let's explore the design space of how people live together in a Game B way. And then as you alluded to, and this is also hugely important, what differentiates the Proto-B movement from, say, eco villages is we don't see these settlements for long being one-offs. They have to be designed from the beginning to interoperate with each other to have these protocols we talked about earlier. 

It might even be a monetary system for doing trade between Proto-B. Certainly, there ought to be immigration policies. It's easy to move from one Proto-B to another and to bring your person and your learnings with you. And over time, that scales up to larger and larger systems where there might even be a quite significant Proto-B economy where we trade with each other locally and over extended periods of time. And then, over time, as winning configurations in the Proto-B space start to emerge through practice, not just theory, that's the time when we're ready to transition to the early days of real Game B. I'm sure Jordan has some interesting thoughts. I'd love for him to talk about his Civium idea a little bit. 

[01:03:39] JH: Okay. Well, let's see. So let me, just maybe for a moment, just talk about the relationship between theory and practice and then the context that we're in. Because it matters. 

The point of theory is to orient the always expensive efforts of practice towards the highest likelihood, at the very least, of failing interestingly. And ideally, to fail maximally insightfully. And therefore, to create the most useful insights and learnings to then feedback the highest leverage back into theory, right? If you think all the way down to the very basic invention of the notion of thought at all, that's the point, right? 

The point of thought is, "Well, it's going to be very expensive for me to chase every single money. Is there a way for me to sort of take a look at the environment and have some set of heuristics that quickly prunes all those paths until that one's the one that I'm going to chase after?" That kind of an idea. 

When I look at the landscape, I notice that we have actually a very large number of low-hanging fruits in practice land that are oftentimes completely ignored, which is to say that they're easy, ready-at-hand, necessary, very close to the path of things we have to figure out early. And that we don't spend time on those. So that's one piece. And I'll get to that in a second. 

[01:04:56] JS: Can you give us some examples? Or do you want to come back to that? Because I'd love to hear some examples – 

[01:05:02] JH: Learning how to be a better parent. We've talked about that as a thing that we can talk about. The development of humans is simultaneously the most complete experimental harness for any kind of civilization and the most necessary condition for any viable civilization. 

If you happen to be a parent, then you are sitting smack dab in the center of the most – sort of the highest point of focus. And I assure you, as you and I both know with sort of, I'm assuming, agony. In my case, Agony. I don't know if everybody else's case. The challenge of better parenting is up front and center. And I know that I'm very, very far away from being a good enough parent in the context of where we are. 

And by the way, this is not just a me thing, right? It's my wife, right? Parenting involves relationships, and then community, and context and infrastructures. You can look at the whole problem space from that axis mundi and that gives you – sorry. Please. 

[01:05:59] JS: No. No. It's interesting. Because part of what Jim talks about in his path to it is the need to have a prototype that integrates them. And parenting I think is a really interesting example of this, where part of what makes parenting so hard is the isolation of the nuclear family. Particularly when you often don't even live in the same city as the grandparents, right? 

And so the opportunity to really rethink and redesign parenting, the opportunity space is much broader when we start to think about living in a community versus living in a nuclear family. 

[01:06:34] JH: Exactly. And of course, if you think about the notion of it's going to be a whole bunch of 20-year-olds who don't have kids are going to get together and live in a community, the stakes are actually quite low. But if you're actually uprooting your family, your kids from the context that they're currently in and having them be co-raised inevitably by other humans, the stakes are maximal. And the degree to which your skin is in the game is very high. And so, A, you're going to pay close attention. And B, you're not going to take a whole lot of shit. It's actually a very powerful experimental harness. 

And then, by the way, we can pivot to a completely different mode, which is this mode. The one we are in right now, which is entirely virtual conversations. And this would be the ethics of the virtual, right? Another thing that is a sine qua non for us to make it through this transition is we have to actually build a level of individual and collective mastery of ethics in the virtual domain rather rapidly. 

And one of the big challenges that we found ourselves in is that we willy-nilly through everybody on the planet more or less into McLuhan's Global Village without, in some sense, the least bit of consideration about how to do that in a way that is sustainable, and healthy, and developmentally relevant and doesn't create chaos in every direction, right? 

That's the thing that we can practice as well like simply. Like how do I show up in Twitter not as an asshole? Just start there. And then work backwards. Like how do I create a sub-community of people who are actually practicing the art of simultaneously having meaningful conversations but also meaningful both in the sort of the larger sense and in a local sense? Like a conversation that improves our capacity as individuals to participate in the conversation, improves the quality of the conversation, is able to bring other people in. Hint, hint scale. And is more and more capable of actually orienting itself towards the most meaningful things for it to be paying attention to. And that's micro experimentation in what we call the virtual version of Proto-B. The collective intelligence shards. All that's like a precursor. Like those are things that can and should be done. 

The other side is the problem of what I call black holes or traps and triggers that are built into our incentive structures and into our deformed psyches. And this is another thing to worry about, right? It’s unfortunately, very easy to bullshit ourselves by simulating something. Literally telling ourselves that we're doing X. And of course, really powerful by saying that to other people, we look shiny, and interesting, and maybe profound and good. But in fact, not even coming close to doing X, right? It's much, much easier to simulate something than it is to actually do it. 

And so, for example, this is why so many intentional communities blow up is that they sort of are oriented towards the surface. Can we really look like the kind of thing that looks good on the brochure? Can I look like that? But actually don't even take the time to contemplate the set – the queries that need to be thought about at the lowest, lowest levels so that the thing on the brochure is actually sitting on top of a very solid foundation, which is enormously non-trivial. 

[01:09:30] JS: And you feel like that happens out of just ignorance? 

[01:09:33] JH: No. Actually, it happens out of a very strong high-quality game theoretic evolutionary strategies in the contemporary feral environment. 

[01:09:41] JS: The black holes. 

[01:09:42] JH: Yeah. Well, the black holes were things that just show up in humaning. And then the particular black holes are the ones that are shown up in our contemporary environment. We live in a context where virtue signaling has so radically selected out virtue. That if you haven't mastered the virtue signaling game, you don't have a whole lot of currency in the social milieu. 

But unfortunately, if you've mastered the virtual signaling game, you're also caught up in an enormous amount of bullshit. And so, you're going to have a very hard time getting out of it. Which is why, by the way, the turn towards the interior showed up as being so important, as I kept noticing so many people who listened to the narrative and said, "Okay. I get it. That makes sense." But then in ways that seem very inexplicable to me will then immediately dodge doing the thing that mattered until I began to understand how important learning and how you bullshit yourself is in making strong committed choices with integrity. 

I get very concerned about circumstances where people get very excited about performative Game B. And I know why that's going to happen and how it's going to happen. But it ends up being sort of depressing. Because, again, these are people who are very close to being able to actually spend their time, and energy, and their lives, and their efforts to fail interestingly. But in fact, because they may be stuck in an eddy of simulation being too much part of what they're thinking about that the effort's not going to get as much out of it as it could. 

And by the way, I'm not saying any of these things are useless. I'm just saying that my focus is on trying to get us to a point where we have a chance. And I think you have to put that very clearly. 

[01:11:13] JS: I want to just ask one more question. We spend all this time talking about some of the things that are happening within the landscape. And we've had a center of gravity talking about capitalism and economics. If you look at something like – like is permaculture adjacent to Game B? Because – yeah. 

[01:11:30] JR: Definitely. And I've had numerous permaculture people on my podcast. I think it's a very important piece-part as I call that in thinking about what comes next. 

[01:11:40] JS: It's Game B-ish. It's just not comprehensive. 

[01:11:43] JR: Yeah, it's a piece. It's a piece-part. It's not even -ish. Because it's actually part of Game B. Permaculture will certainly be a component of building out Game B civilization. It's a very important piece. And as is, it's closely related neighbor, regenerative ecology. But again, think of these as pieces from which we build the civilization-level operating system. 

[01:12:04] JS: And what about the circular economy? Is this marginal Game A still self-terminating? 

[01:12:11] JR: It's an important topic. And it's certainly part of what it's going to take for any civilization operating system to operate within the constraints of what Mother Nature can provide over the long-term. It again would, by definition, be part of any Game B economy. Because when you say circular in the sense that we use, and reuse and use again, rather than having to go out and tear the earth up and mine. And frankly, Game A will adopt that strategy as well because it's a good strategy. 

[01:12:37] JH: I think there's a heuristic that's coming up for me as you're mentioning this that I think might be useful. If I think about something like permaculture and compare it to something like the circular economy. If I were to imagine adopting what you might call the underlying basis or like the aesthetic and ethical characteristic of permaculture, and then I imagine just sort of taking that really fucking seriously and just running it to its logical conclusion in every direction, it ends up being isomorphic with Game B. 

The circular economy isn't like that. This makes the circular economy as kind of a – it's not essentially the same as Game B. Although it is a – as Jim has mentioned, some portion of any Game B will have to have a circular economy characteristic as part of it. Whereas permaculture is actually like a cousin. Like you can get there. There are ways to get all the way there if you just take it really seriously. 

[01:13:26] JS: I totally see that. I mean, the circle – 

[01:13:28] JR: Yeah, I agree too. Like that's a – very well said, Jordan. 

[01:13:32] JS: I mean, this is a sort of top-of-mind question for me. How do you two think about the future of economic growth? 

[01:13:39] JR: Growth is a topic which is talked about in a way that's superficial way too often. And we hear words like de-growthing. And that turns people off. Even though at some level it has to be true. 

And what I would like to talk about is – let's talk about what we're really talking about. Talk about the problem of growth. And the problem of growth is the impinging upon the limits of what Mother Nature can actually provide to us sustainably over a long period of time. The amount of topsoil that we can allow to wash into the ocean, which turns out to be almost zero if you're going to have a civilization that lasts a long time. 

Growth that impinges upon Mother Nature we must back away from. In fact, the best evidence I've seen from people, I will say this carefully, is we have to reduce our capacity on the earth – taking from the Earth by 80% if we're Westerners, say, Americans, Canadians or Australians, and maybe 60% if we're Europeans, if we're going to survive long-term in a metastable fashion. So that means we have to degrowth at the level that impinges upon Mother Nature. But that does not mean we have to stop doing interesting things. We can grow into the microcosm. 

Think for instance of an Apple phone, right? There's not a whole lot of material on an Apple phone. There's some sand. There's a little aluminum. A little bit of copper. And the amount per capita is pretty de minimis. There's already plenty enough in circulation. They continue to make Apple phones forever. The chips get smaller and smaller. The applications get smarter and smarter. The networks are faster and faster. And yet, they don't impinge upon Mother Nature at all. We can grow into the microcosm and while we reduce our impact on the macro causal. And I think making that distinction is really important. 

[01:15:24] JH: I'm going to end up saying more or less the same thing with a very different set of words. I think actually the difference might be quite useful. 

[01:15:30] JS: Okay. 

[01:15:31] JH: The concepts that I'd like to present would be metric, mode and essence. Metric would be something like US GDP as indexed in dollars. And the metaphor here is – again, I'm actually stealing this straight from Tyson Yunkaporta, is I'm thinking about intelligence. And I can look at it and say, "Okay, one way I'm going to do this is I'm going to take a brain and I'm going to grow by having this metric of neurons or synaptic connections." Filling your blank. And I'm just going to keep getting that number bigger, right? 

If I want to make a smarter brain, I'm just going to get a bigger brain. And so, if I want to get a better economy, I'm just going to get a bigger number of dollars, right? Those two are – I mean, they were doing those mappings. And the beauty of that mapping is you can look at it and say, "Well, that notion of just making the brain bigger is actually a really stupid way of trying to get to the thing that you actually are trying to accomplish." You can shift from metric to mode. 

In this case, you could say, "Well, what happens if instead of just adding more neurons, I do things like change the geometry? So that it has in-foldings and it has synaptic connections that are plastic and all kinds of different shifts in the nature of the underlying thing that I'm looking at." That has nothing to do with trying to increase the sort of stupid ass metric that is sort of the simplest thing to grasp a hold of. And also, catastrophic if I don't understand when to precisely let go of it. 

But then you have – from metric to mode, you then go to essence, which is to say, "Well, I can actually have this move." Or I can move this technical to insight. Or I can actually move from a domain. In this case, "Oh, okay. I've learned how to build a particular human brain that is now using all these cool new approaches to increasing its capacity. But I'm bound by the limit of, frankly, the size of the birth canal and things like that. I've got the best individual human I can get."

Okay. Well, I can yet again shift to a completely different mode, which is, well, now we're going to have groups of humans be able to coordinate with each other. So I don't have to solve for increasing intelligence in a single brain. I can now have these brains networked. Oh, that's neat, right? And now the limit on that is much different than the limit on the constraints of the domain that I was operating under originally. 

If the question is economic growth with the concepts of metric, mode, and essence, then the answer is the same as Jim's, which is, as far as we can tell, there is absolutely no limit on the growth of the things that we actually care about. Which, by the way, are things more like meaningfulness and values. Limits on the growth of the metric that we happen to have been optimizing for for the past 100 years or so, we're way past that. That one should just be thrown into the reboot system and then we should at the very least engage in a modal shift. 

But ideally, we can actually be grounded in an orientation towards the essential. And that allows us to be maximally creative with the kinds of modes and metrics that we play with. 

[01:18:17] JS: Thanks so much, both of you, for being so generous with your time and engaging in this conversation. Again, it's been one I've been eager to have for a very long time. And obviously, there are so many rabbit holes to go down. So I really enjoyed digging into this with you. And thank you so much. 

[01:18:34] JR: Thank you, Jenny. This has been a great forum. Great questions. 


[01:18:37] JS: Thank you so much for listening. And thanks to Scott Hansen, also known as Tycho, for our musical signature. In addition to this podcast, you can find resources for each episode on our website,, including transcripts and background materials. 

For our most essential topics like universal basic income, decentralized social media and long-term capitalism. We also have posts summarizing our research, which make it easy for listeners to very quickly get an overview of these particularly important and foundational topics. 

On our website, you can also sign up for our newsletter where we bring our weekly podcast to your inbox alongside other relevant Denizen information. Subscribers are invited to join our podcast recordings and engage with the Denizen community in our online home, The Den. We're partnered with some incredible organizations at the forefront of the change that we talk about. We share announcements from them in our newsletter as well. 

Finally, this podcast is made possible by support from the Denizen community and listeners like you. Denizen's content will always be free. Offering Denizen as a gift models a relational rather than a transactional economy. Enabling Denizen to embody the change that we talk about on this podcast through the reciprocity of listeners like you that we are able to continue producing this content. You can support us or learn more about our gift model on our website. Again, that's 

Thanks again for listening and I hope you'll join us next time.


( Scroll to Explore )

Join us!

Our weekly newsletter is the best way to stay abreast of our inquiry and ongoing events.

Oops! Something went wrong while submitting the form.
The Denizen logo.