with Spencer Greenberg
the podcast about ideas that matter

Episode 066: Why do civilizations collapse? And is ours next? (with Samo Burja)

Enjoying the episode? Want to listen later? Subscribe on any of these apps or stores to be notified when we release new episodes:

August 20, 2021

What is "long" history? Why don't historians usually focus on what happened before recorded human history? What (if anything) is special about agriculture when it comes to the development of civilization? How far back does human civilization go, and why should we care? Have humans always been gardeners? What factors cause civilizations to crumble or thrive? Should we reboot standardized tests and college admissions every few decades so that measures don't become targets? Which destructive factors are particularly salient to modern human civilization? Why is there such a disconnect between our intuition that progress is inevitable and our knowledge that virtually all civilizations have collapsed in the past? In other words, what makes us think that we'll succeed where others have failed? How does a functional social institution differ from a failing one? What is the "great founder" theory?

Samo Burja is the founder of Bismarck Analysis, a consulting firm that investigates the political and institutional landscape of society; a research fellow at the Long Now Foundation; and a senior research fellow in political science at the Foresight Institute. Samo's studies focus on the social and material technologies that provide the foundation for healthy human societies, with an eye to engineering and restoring the structures that produce functional institutions. He has authored articles and papers on his findings. His manuscript, Great Founder Theory, is available online. You can find him on Twitter at @samoburja, on YouTube at @samoburja, or on his website,

JOSH: Hello, and welcome to Clearer Thinking with Spencer Greenberg, the podcast about ideas that matter. I'm Josh Castle, the producer of the podcast. And I'm so glad you've joined us today. In this episode, Spencer speaks with Samo Burja about long history, civilizational collapse, the influence of social capital on economies of scale, and features of functional institutions.

SPENCER: Samo, welcome. It's great to have you on.

SAMO: It's great to be here with you.

SPENCER: Now, I'm especially excited for our conversation today because it feels like a blind spot to me. I really feel like I haven't studied history enough, and that there's a lot to learn from it. And in the best case scenario, history should be able to help us better understand our world today. And perhaps even more importantly, better understand how we can kind of steer the future to a better world. So I'd love to dig into that with you through a number of different topic areas, starting with the topic of long history to tell us about what is long history and why does it matter?

SAMO: Long history is my approach. And this general approach to studying the prehistory of society as much as we study recorded history, that is taking things such as the archaeological record, and genetic evidence seriously, in our understanding of how human culture has developed over the last few 1000 years, but also over 10s of 1000s of years, hundreds of 1000s of years. There are all sorts of interesting things that we can now learn about the past, thanks to new technologies, such as, Neolithic migrations, when the Americas were peopled, when agriculture first arose, and so on. In particular, I think that we will find that our civilized history is much longer and predates these written sources that are usually considered the barrier between history and prehistory. And I think that's going to be the most fertile area of development in the next few decades and might change our whole conception of what humans are.

SPENCER: So what's going on here? Why is it that traditionally when we think about history, we're not going back as far as as you're talking about?

SAMO: I think a little bit of it is due to an institutional gap. Historians are people who do brilliant work, but they usually rely on sources, primary written sources, and they work within an existing field and existing special. There are several new societies that we know from the archaeological record, where the archaeologists are sort of doing the work of the historians trying to piece together what happened when trying to produce theses of why this or that happened. But let's be honest, that's not truly the specialty of archaeologists. I greatly appreciate it, Klaus Schmidt's hypothesis on the history of religion and the origin religion at the site he discovered, Gobekli Tepper, but he was an archaeologist, he wasn't a scholar of the history of religion. He wasn't a historian. So as interesting and fruitful as his hypothesis is, we really can't ask archaeologists to rethink the social systems, even if they are the ones discovering evidence for them.

SPENCER: I see. And do historians and archaeologists collaborate? It seems like that might be ideal in this case.

SAMO: It would be ideal. I think that such collaborations frequently do happen, when there is already a known quantity, when there's sort of this society that you want to study, say, if you wanted to study the Mississippi mound builders, it wouldn't be hard to bring together historians and archaeologists, but it starts getting a little bit tricky when you're pushing on the very edges of recorded history, right? When you're looking at something like the finds in southeastern Turkey, or finds Qatar or the Neolithic and Mesolithic monuments of Europe. These are places where there's not really established literature even. And to start a new literature is often very difficult. And you're sort of bootstrapping from the archeology side of things, rather than the historian side of things.

SPENCER: So give us some idea of the timeframes we're talking about your range is usually considered civilization and then what timeframes are you discussing when we talk about long history?

SAMO: Think that the normal timeframe, the consensus frame that you might find in a popular history book is about 10,000 years. Now, this itself had been revised further back into the past, when in the 19th century, people discovered evidence of civilizations such as the Hittites and the Sumerians. It matched until recently, the oldest evidence of agriculture. However, we now have new finds, such as the hollow site in modern day Israel, it seems to show that there is evidence of agriculture 23,000 years ago, and that really should push back our assumptions about the world. Now, let me be clear, 23,000 years ago, we might have evidence of this proto-agriculture of the simple cultivation. We don't yet have evidence from that time period of complex society. But there's a different site that we'll discuss later that is much older evidence of complex society. So I think when we're talking about this long history perspective, I think we're actually talking about the last 50,000 years of human civilization. One pithy way to put it is to say, even our prehistory had a prehistory.

SPENCER: Hmm. So it seems like our intuitions today are that if something is invented, eventually people all around the world will have fear about it, and not too long. Could part of what's going on here be that things were very unevenly distributed in the distant past. In other words, you have one civilization that was much more advanced than other civilizations. And then, because information spread so slowly, if that civilization failed, others may not even ever have gotten the technology from it.

SAMO: I mean we don't need to speculate. We know the world was like this until quite recently. This is obvious when comparing new world civilization and old world civilization, right? Like all of the advances of the old world including things that we might think of, is very basic, like the use of the wheel, and so on. These were not present among Native American societies, they didn't work with iron, for example, though, they did have middle working with silver, with copper, and so on in Mesoamerica. And in antiquity, of course, travel was only much harder than it was in the 14th century, or the 15th century, traveling from one end of Eurasia to the other only became easy. And I put these in quotation marks after the domestication of the horse. Right before that, the cultural exchange between far west and far east was very limited to this. We have to add natural barriers also, such as deserts, mountains, and then the difficulty of ocean navigation. So yeah, I think the exchange of information and technology is certainly a big reason, barriers to this exchange is certainly a big reason that we see deeply uneven development even in recorded history, and much more so in the sort of eras of 10,000, 20,000, 30,000, 40,000 BC.

SPENCER: Now, you mentioned agriculture, talk for a moment about what's special about agriculture when it comes to civilization.

SAMO: Yes, the consensus theories of civilization put the origin of agriculture side-by-side with the origin of complex society or what we usually call civilization. The assumption is that, once you have agriculture, you can have large population centers emerge. It's necessary to store the progress of agriculture usually assumed to be first grain, it's necessary to protect it, it might lend itself easily to things like taxation. And so you could imagine that the origin of the modern city states is usually considered a straightforward consequence of agriculture, right? You start seeing these city states like where an oil can emerge very quickly after we have evidence of agriculture. Now, I have to say this was the theory until recently, but new evidence has come to light that suggests that agriculture long predates the cultivation of these grains and long predates the origin of these city states.

SPENCER: So what's going on there? How is it that you could get these two things being decoupled?

SAMO: What I think is happening is that there is a long transition from hunter-gatherer societies to agricultural societies, where it's not the case that they immediately switch over to some very different mode of living. It's rather the case that early agriculture kind of sucks, it's kind of difficult, you have to do a lot of work. It has some predictability to it. But actually, the food that you're eating is perhaps somewhat impoverished compared to the food you could be eating if you supplement it with hunting and gathering. Forests, in particular, might be much richer than we usually imagine when it comes to providing sources of food, and particularly chestnut tree forests, I think in temperate climates, we know from accounts of Native Americans living in California could provide quite a few calories, like a lot of calories. And we don't necessarily know how many There's also evidence of gardening and symbiosis with the forests of the Amazon rainforest, as well. And the creation of artificial soils, such as terra preta, which was much more fertile than naturally occurring soil in the Amazon. So, basically, early agriculture is not immediately 10x or 100 times better than hunter-gatherer lifestyle. So think of it as agriculture and especially gardening, perhaps being just part of a toolkit of people in a culture that was still primarily relying on on hunting and gathering, and also fishing.

SPENCER: I've also heard it argue that it may be not enough to have agriculture. Generally, you might actually need very specific crops. For example, you might need crops that have predictable seasons that can be stored for a long time. And part of this might relate to sort of taxation, that if you're going to have a big city, you're going to need most likely some form of taxation. And in order to tax people, you need a way of figuring out well, how much grain is this person producing, and then you have to have a way of knowing when they're going to harvest it. And they have to be able to store their brands, and so on. And I'm curious what your reaction to that theory is.

SAMO: I think there's a lot of truth to it, where you could even argue that say, a system of irrigation, right, a system of irrigation especially, lends itself very naturally decentralization, the reason is because an system of irrigation has to be maintained centrally, you are dealing with the entire course of the river, an alteration made upstream affects water flow, flooding probability downstream, we see this as very central in Egyptian civilization, and to some extent Sumerian civilization. So perhaps it's the unpredictability of rivers, large rivers, especially both the destruction and bounty they can provide, that might be an even greater factor, even more so than the domestication of grain, though, of course, there's fun arguments to be had about the grain. I especially like the argument that bread is basically a drug that it's mildly intoxicating. So once you start eating bread, you keep craving it over and over again. So it'll be kind of amusing, if chopping down forests, planting grain extending an irrigation system is sort of done because you want to politically expand your control over the local peoples, rather than it being just the best means of sustaining the most people.

SPENCER: Over that work for political control, you're saying people become addicted to bread, and then you've got to give them their fix. And so they have to keep coming to you.

SAMO: Exactly in the very system of producing bread is centralized, at least in a desertified region that requires irrigation to grow a lot of grain.

SPENCER: I don't know how to feel about the theory. But I do have to say, you know, if someone gives me tasty white bread, I will definitely overeat afterwards. Like, why did they tell me rolls to this?

SAMO: Yeah, there's something almost compulsion like about it. I don't necessarily think this theory is the theory of what's happened. But I think we have to consider these alternate theories. Another example is brewing barley at scale, things such as the introduction of alcohol, these are, then substances that we do know can be addictive. And there's some evidence near the Gobekli Tepe, a site and other sites in southeastern Turkey, that, you know, they were brewing beer in the Neolithic.

SPENCER: You mentioned that there's new evidence that there may have been a more advanced civilization, what 20,000 years ago, something like that. Do I tell us about that?

SAMO: Well, to be precise, I think we can infer that there were civilizations 20,000 years ago, evidence exists of agriculture 20,000 years ago, evidence of I think, a different complex society that pushes aside the 10,000 year date is the site of Gobekli Tepe in southeastern Turkey. That site is currently dated to eat 11,500 years old. Now, why does this matter? This matters because, again, the normal story of origin of society is the ice age. Climate becomes predictable. We can engage in agriculture, after we have agriculture, we start doing things like monumental construction, and settled permanent living, decided Turkey has evidence of what is assumed to be a set of temple complexes, concentric circles, built with these massive pillars, each wing 10 to 20 tons. The current estimates are that the needed labor force to assemble such buildings was about 500 people. Now, we don't have evidence of agriculture from that particular site. But it is a very old site. It's 11,500 years old, much much much older than Stonehenge much, much, much older than the pyramids. But then I have to ask you, how were those 500 laborers fed, looking at the climactic data of the region, it was as dry then, as it is today. And it's far away from any a bountiful source of fishing. So if it's not rich wildlife, like, say, the North American plains once provided, if it's not, you know, these sort of chestnuts style forests, if it's not fishing, what exactly are the laborers eating? Right? How are they being fed? How does economics of their communities work? So I think it's indirect evidence of agriculture. And then if we flip it around, I think agriculture is then like, also indirect evidence that somewhere in the world where we haven't yet dug probably, there are older, permanent In settlements, there are places where fully relying on agriculture likely was the correct choice, even 20- or 30- or 40,000 years ago. Once you go past any sort of default null hypothesis that you had, right? When we had this null hypothesis of it all developed after the last ice age, your next best guess, is not on the basis of justifies it is on the basis of what's the new null hypothesis? Where should we dig? What's the new hypothesis of When did humans start behaving in this novel way? Right? When we find beaver skeletons, we have no problem assuming that they build beaver dams. When we find human skeletons, we're much more reluctant to assume that they have an associated extended phenotype of a certain kind. And I think that's a little bit of a bias. That's a mistake. And it would be very fruitful. To explore this further, partially because the most recent fines are the floor, not the ceiling of past achievement, right, we're getting a very, very attenuated sample from the past. But on the basis of seeing what human beings are, what kind of behaviors we have, which technologies traveled together, we can infer more than what just defines themselves show.

SPENCER: Presumably only a very small fraction of civilization development ever gets recorded in history. Is that right?

SAMO: Yeah, the vast majority is lost. I mean, we know even with written sources, we have an interesting estimate. There's an estimate that of all the ancient Greek authors and Roman authors, we know by name for 94% of them, we have no complete works, we don't have fragments of the writing. These are people who are named dropped in the books and scrolls that we do have, that are mentioned as authors. But we never found those books. So if written material that can be easily copied is attenuated in this way, I think we should really consider what are the odds that we're going to have preserved always the peak of human achievement? I think it's really unlikely.

SPENCER: So why should we care about whether history is 10,000 years old or 20,000 years old or longer? And what is the kind of consequences of this?

SAMO: I think one of the biggest consequences is that perhaps we have to reevaluate what Homo sapiens is. Now I have to say that, again, this is not something that's been proven, but it's a hypothesis worth exploring. The hypothesis being that maybe homosapiens has always been a gardener. Maybe we have always occasionally lived in villages. The consequences for any attempt at say, an evolutionary psychology would be significant, right, it would suggest that, oh, perhaps we do have evolved behavioral adaptations to societies at even a scale that's larger than Dunbar's number. Another important one is, we might come to understand, or we might see further evidence of something that I think is already established, but neglected, which is the fragility of complex human societies, we know just from recorded history, that's the last 10,000 years, not counting any of these extra millennia, that we've had about, I don't know, let's say 10, identifiable Dark Ages, just in the Eurasian continent, social collapses, like the late bronze age collapse and eastern Mediterranean around 1000 BC, or much earlier, the end of the Mohenjo Daro, based in this Valley civilization, which was the first civilization in India, or later examples that, you know, they're much better known such as the fall of the Roman Empire, this should be a warning for our own society, right. And I think the more data we have, the more complete our data set is of these civilizational collapses, the better we can understand the common threads behind them if they are any, and the better, we can prepare and prevent our own societal collapse. Right? After all, we we have grander aspirations than just to be another set of ruins for future archaeologists to study.

SPENCER: Well said. So I want to get into civilizational collapse with you both. Before we do. You mentioned that this idea that agriculture might go back much further, maybe town dwelling might go back much further than people realize. This is a theory, we don't know for sure that it's true. What kind of probability would you put on it if you had to bet on us?

SAMO: Oh, I'm very happy. In fact, to set up that I would put the odds that humans have lived in villages, at least occasionally for a very long time at 70 or 80%. And by a very long time, I mean, many 10s of 1000s of years. I recently wrote an article titled "Why civilization is older than we thought" and after reviewing all of this evidence, I conclude the article by saying that I'm happy to take a long bet with a qualified challenger. You know, what's a long bet? It's a bet that's resolved over a longer period of time. So the best terms are 20 years from today, we will know of at least one permanent settlement older than 20,000 years, 20,000 years because that's comfortably on the other side of this ice age transition, right? That's usually referred to, it means that the old scientific paradigm is broken. And if we are in a different paradigm of history, so I'm comfortable betting that we will find something within the next 20 years.

SPENCER: When was this ice age?

SAMO: It depends on how exactly you date the end or the start of it. But there's a general broad consensus that the last glacial period occurred from the end of the Eemian to the end of the Younger Dryas. So the Ice Age is considered to have lasted from 115,000 years ago, to about 11,700 years ago.

SPENCER: This would suggest, according to the that you want to make, that there were towns.

SAMO: There are permanent settlements 20,000 years ago. Yeah, exactly.

SPENCER: Right. So that would be during the Ice Age, right?

SAMO: Exactly during the last Ice Age. And I think then that conclusively breaks this sort of old hypothesis and the history books have to be rewritten, then people have to say, well, you know, humans started settling down in permanent settlements sometime during the last Ice Age. We're not sure when maybe 100,000 maybe 50,000, maybe 20,000 years ago, our best evidence is 20,000 years ago, right? It wasn't the climate shift that pushed us.

SPENCER: And the alternative, just to make sure I understand, is that what humans were in bands, but they were roving bands if they didn't have kind of permanent settlements.

SAMO: Yes, they were humans were in roving bands, these bands were small, let's say 50 people, maybe a hungry bull, they never settled permanently, that wasn't economically or socially sustainable. They never made use of agriculture or gardening. Because you know, that requires you to stick around. Right? It requires you to be in the sort of same range. So yeah, the alternative is that there were only Paleolithic cultures all around the world, rather than cultures that, in my opinion, reached a Neolithic level of technology.

SPENCER: Got it. Okay, so now going back to civilizational collapse, which is obviously really important when we think about our future, as you said, we don't want to be one of these ruins that's eventually found by future people. What do you think we can learn from past civilization collapse about potential future civilization collapse? I know, it's a broad question, but just to hear your thoughts.

SAMO: First off, there are several interesting books that can be read on this right? There's a Tainter's book, The Collapse of Complex Societies. There's Jared Diamond's book on civilizational collapse, which focuses primarily on climactic causes. I think, generally, I'm willing to say that societies dissolve mostly from within or from external conquest. Climate can be a stressor, climate can be a significant issue, but looking at fully developed society, so not peripheries, right? Like whatever we think of say, Vikings in Greenland. They are the periphery of Norse civilization. They're not sort of the core of it. They're not the center of that society. They're just this far flung outpost, so them being pushed out by climate the way Jared Diamond argues, well, okay, that's not really saying much about society as a whole. The same I would say, actually applies to the Easter Island, which is this far flung element of Polynesian civilization that's more complex society that's, of course, inherently fragmented. It's an island hopping culture. Whenever we looked at these examples, if you go through history, I think you see, sometimes frontiers being pushed back and forth by climate. But really, usually, the society itself has ceased to be adaptive to novel circumstances. And that's the real cause. It's kind of like if I take a fall, when I'm 20, and I break my hip, I'm very unlikely to die. If I take that same fall at 80 and break my hip, I'm actually know a lot of trouble. Right, I might not recover. So that's how I would describe climactic effects. I think if a society is adaptive, vital, it will perhaps take a hit, but it will adapt to new climate conditions, it'll come to live with them. However, if the society is already in trouble, such as say the Mayan civilization was right, then I could think climate could be a decisive blow. And I'm using here climate to be a stand in for all sorts of other factors, right factors, such as resource scarcity, or other environmental shocks, such as plagues, and so on. Mayan civilization, by the way, of course, ended long before or reached terminal decline, long before the arrival of the Spanish and other Europeans from the old world. It was, you know, 12th, 13th century, it was already an area in the Yucatan peninsula that the jungle almost fully reclaimed. Even today, it took new techniques, such as the use of Lidar from planes to uncover the full extent of the jungle, right Lidar is basically laser radar. The laser beans can occasionally bounce through the canopy, that's where you can sort of take a picture of the landscape that strips away the vegetation. And there's some very interesting work done on that. And it showed that Mayan civilization scale was greatly underestimated by an order of magnitude, when it comes to the number of settlements, the likely number of cities and population estimates and all of that. So I think that we are going to find more surprises in the jungles of the world and the deserts of the world of this type. And I actually think that this kind of societal collapse and failure is probably inherent to a long term attenuation and rugged application of institutions, where elites of a society tend to mostly lose any sort of caretaker or directing function and assume a mostly parasitic character, when they're sort of protecting their existing patronage networks, revenue streams, or rents, right in the economic sense of rent collection. So if you have elites vested in keeping things the same, and you encounter something like a climactic shock, well, then your society might be in a lot of trouble. Because if the elites are in favor of keeping everything the same, in a given society, but the physical environment has changed, well, you know, that might be resolved through the collapse of your society, not through adaptation.

SPENCER: Okay, so stepping back for a second, it sounds like you think there are two broad classes of reasons why civilizations fail. The first is from external threats, like they're just invaded by a neighbor or something like that. The second is sort of this internal rot. Am I right so far?

SAMO: Yes, yes. And of course, the external conquest can be facilitated by internal rot. Do I believe that there are basically healthy societies that have been destroyed by external conquest? Right? For example, I have no reason to believe that the Persian Empire that Alexander the Great conquered, was decadent in any real sense. It all the other indicators were very positive. They just were very unlucky to be fighting Alexander the Great.

SPENCER: Okay, so then if we think about this sort of internal rot that can happen, you're describing this as sort of having to do with institutions did you want to kind of give us the trajectory and the way you view it, of how institutions get started? And then what happens to them to kind of makes them go sideways in the way that they kind of search rotting civilization from the inside?

SAMO: Well, there are many perspectives on institutions, I take a somewhat functionalist view, where, I tend to look at institutions and ask what role is this playing in broader society? One way to think of societies and civilizations is that they are these interdependent ecosystems of institutions that are often through the services they provide in the externalities, the positive externalities that they can never fully capture, even if they tried to capture them, subsidizing all together institutions in a cluster. Like I think these societies are better thought of as forests, not as trees, right. So the death of any individual tree is not the death of a forest, the death of many of the trees to never regrow. Well, I think that's the end of that particular civilization, even if you have later, you know, different shrub growing or a different biome emerges in that area. So if you think of something like the university system, right, I would easily class universities, as a basic institution type in our civilization. Universities are some of the longest lived organizations in western civilization, especially the University of Oxford, predates the rise and fall of Aztec civilization. For example, it's 1000 year old organization.

SPENCER: It's wild.

SAMO: It is, it is. And you might think, okay, Oxford, it's on an island, it's isolated. It's probably an exception. But the University of the Sorbonne in Paris is just as old or compatibly old, or see universities like Bologna, they're also compatibly old. And then, of course, we all know, you know, Italy, France, these are places with frequent wars. So why do these universities persist? You know, what are the longest lived institutions in the world is itself an interesting topic? Maybe we'll touch on it later. But to go back to answering your question, I think what happens is, these institutions play a symbiotic role. They are often say, places where knowledge of a particular kind is transmitted. They provide some kind of economic or political value over time, there are three factors, one of them is, situations change, right? Just because something was built had an origin was designed at certain way, doesn't mean that it's appropriate 500 or 600 years later. The second reason is you can have deep and profound succession failure, where say, if you compare in classical civilization, the philosophers of the classical era with the philosophers of the Hellenistic era who had started dabbling in engineering and science right, Archimedes is better thought of as a scientist. He's producing exact laws running experiments than just a philosopher, right? He's crossing this gap. And then you compare this with the Neoplatonic philosophy that exists during the decline of the Roman Empire. And you realize, wow, these people call themselves a Platanus. They call themselves philosophers. They call themselves engineers, and a radically misinterpreting the texts that they have. They are producing stuff that's much less impressive, much lower quality. The scientific and engineering works that exist in the Roman Empire are greatly inferior to the ones that exist in Hellenistic Greece, right? If you go and read someone like plenty, the correct way to think of him is that he is basically writing pop science in Latin, for a Latin speaking audience that's never read the original Greek works. Well, what happens when only the pop science is around, but not the original scientific works when the original scientific works are lost on the culture of sciences lost. That's a pretty severe succession failure, a succession failure of knowledge. You can also have other problems such as a succession failure when it comes to power, where you can have political crises that sort of break down a society more and more, because central authority fails, the new set of authorities they emerge, have smaller scale, therefore have worse economies of scale. Those go on to fight each other more and more and more. And that would say be the civilizational crisis we saw in China, after the end of the late Zhao dynasty, with the warring states period emerging from it. And that's also a different way in which these societies can sort of self cannibalize. So again, the first one is just changing circumstances. The second one is this failure of knowledge. And the last one is this sort of profound and fractal failure of political coordination.

SPENCER: You elaborate on the political coordination one, I'm not sure I understood that.

SAMO: Right. It's basically where the conditions of political crises are no longer a passing crises but become a permanent political crisis. I could give the Egyptian civilization example where Egyptian civilization underwent several transitions from the Old Kingdom to the Middle kingdom to the New Kingdom. Sometimes these were driven by foreign invasions. And other times there was this almost like fractal aspect of Civil War, where you basically have the replacement of central power, with really, really fragile, local power that's just as tyrannical that's just as extractive with eventually just roving bands of people. We saw some of this also in Mesopotamia, where, during the late bronze age collapse, which used to be these very developed cities, eventually became empires. The empires start crumbling and eternal warfare, and it's driven all the way to the abandonment of agriculture. Large parts of Mesopotamia, are no longer worked as fields, but become pastures for essentially nomadic people, nomadic herders. And these herders, you know, they still claim amazing titles like the king of war, but the city walls, you know, they're crumbling, they're grazing their sheep among the ruins. So it's sort of this endemic warfare, right fractal endemic on terminating warfare that's driving down economies of scale, rather than driving them up.

SPENCER: Got it.


SPENCER: So if we reverse these criteria for a second and think about what makes a civilization succeed over a really long time. It sounds like one you need adaptability, like as the world changes the civilization has to change to that changing world. Two, it sounds like you need a way of maintaining quality so that you don't just get a degradation. We're an institution doesn't just get become a worse version of itself. You know, every 20 year, 50 years, you need a short succession planning. And then the third one is it sounds like you have to have some way of preventing the study from just falling into extreme polarization where it becomes just zero sum conflicts all over the place. Does that seem like a right way to think about it?

SAMO: I think so. The difficulty is keeping these three in a balance, easily overshoot on say, trying to provide peace and stability by setting up a sort of central authority that resolves all political disputes, and in the process, end up killing adaptability, right, a big factor of adaptability is middle powers, doing what high powers, central power cannot, in a way, possibly Roman conquests are to be as politically useful as they were for the Roman Republic, might have actually been the cause of intellectual degradation when Rome conquers the city states but more importantly, also these Hellenistic states like Ptolemaic Egypt, right? That are supporting efforts, such as the Library of Alexandria and the Romans themselves, no longer support these efforts, or no longer know how to tell the good philosopher from the popularizer. Well, that leads to a degradation of knowledge, even if it stops, you know, what otherwise would have been a very chaotic and war torn region. But it was a war torn region that was also growing economically, there's lots of evidence of economic growth there. So I think it's it's a difficult balance to strike. When it comes to knowledge succession, I think it's not just succession planning. It's also a matter of figuring out well, how do you recognize when someone has the knowledge. In China, almost every single dynastic cycle throughout its history kind of reboots the civil service, and comes up with a completely new civil service exam. It's not because the previous civil service exam was flawed or wasn't designed quite right. It was basically that over decades and centuries, people when they were trying to enter bureaucratic work, would study for the test, rather than to try to get the generativity and knowledge that the test used to be an imperfect correlation for. So essentially, good hearts law, for example, is a really significant barrier to our attempts to bureaucratically measure and replace human talent. And I would argue that it's good hearts law, which is that as soon as a some kind of social measure becomes a target, it ceases to be a good measure. I think that was actually the fundamental foe that each Chinese dynasty had to struggle with, you kind of had to set up a new test. So people wouldn't guess the passwords, wouldn't guess the fake version of the knowledge, right? And would actually end up being competent civil administrators and competent scholars and all of that.

SPENCER: In your view, should we reboot the SAT and ACT in the US as entrance exams in colleges every, you know, 20 years or something like that?

SAMO: I mean, I think it might be very useful. I haven't studied those in particular, closely enough. And I know that there is a competing hypothesis that of GE that the roboticist being the sort of like, general factor of intelligence, and that these tests actually, you know, correlate well with it and stuff like IQ and all of that, I would say that we would do well to completely reboot university admissions every 20 or 30 years, because I note that, you know, in current America, it's the sort of intangibles that people are prepping in their application, as much as it is the prepping for the test itself. And I do think I for what it's worth, with the caveat that I have not studied them that closely, I do think these tests can be prepped for. But even more important than that, I think, you know, when someone comes in with a CV, and they created a whole new nonprofit, during high school, and the nonprofit was not fashionable cause and they have beautiful photos of all the work they did in high school on that nonprofit. How staged is this? And in fact, the answer is yes, it's fairly typical for very staged high school experiences, where people are building up essentially a fake resume to hand to the university administrators and the admissions officers. And they have to write a good story about it, which Okay, that's a real test of skill. But you know, the choreographed high school experience? Well, you know, that's just a reflection of a certain kind of class privilege. That's not a reflection of intelligence at all.

SPENCER: So it seems to me that one of the most important questions that we could ask about this framework for why civilizations fail, is why is our civilization going to fail? Right. So let's take a moment to kind of do an exercise. And let's imagine that the United States fails at some point in the future. What is your best prediction about which of these elements leads to fail and serve? What's your most predictive story on that?

SAMO: I think it's always hard to imagine a civilizational failure from the inside. Historically speaking, Roman sources and poets, they never really say, "Oh, our civilization is in decline." They say everything is great. Li Zhao civilization in China, Li Zhao civilization they are quite aware of their own decline. And possibly that's why they managed to reverse it. And why Chinese civilization experiences another Golden Age fairly quickly after the Warring States period, right? The Han era especially, and the Han hegemony. For own society, I think each civilization is, again, it's a different ecosystem. It's a different set of institutions. I think there are many deep similarities between our cities and the cities of antiquity. But they're also massive, massive differences in how we organize production, to predict a fall of a society or how that might look need a hypothesis as to what are the core social technologies of that society? What are the core principles of organization? What are the core institutions of it? I think for our own current society, the best guess is probably industry. I think the Industrial Revolution looks fairly interesting, might be historically unique, might have some mild predecessors and Hellenistic era states. But if we take the Industrial Revolution, and the new system of production that emerged in it to be the core of our society, I think what we would see is a failure where, over time we lose the capacity to organize material production. And at no step, can we acknowledge that this is what is happening. When the industrialized, the factories move to China? What if in fact, China will eventually have to move the factories elsewhere? The positive version of this is that, oh, it's no longer their comparative advantage, right? They're going to outsource to Africa, India, and those places in turn, are going to industrialize. Well, okay. Let's say they do that. But what if the negative version of the story is true? What if there are difficult to measure types of social capital, such as public true speaking, work ethic, high social trust between, you know, small teams of people that in turn enable things such as collaboration on the factory floor, accurate reporting of profits and losses, accurate, logical understanding of how to organize material production and improve your product, it could actually be that we are exhausting the social capital needed to maintain industry. In that case, maybe the Industrial Revolution goes into a post industrial society, not because we transcended industry, or because we offshored it. After all, it's a finite planet. Eventually, we'll run out of shores. At that point, prices might begin to rise, products might become worse. As soon as products are worse, and prices are rising, globalization ceases to have good returns. Once globalization ceases to have good returns, it's not like a static system. If it was one predicated on growth, you might see it shrinking. As globalization shrinks, recedes, the beautiful economies of scale that allows us to build all of the world's CPUs on a tiny island like Taiwan cease to be so beautiful, right, the economies of scale break down, and you have to perhaps revert to factories serving more local markets that make slightly worse chips. I don't know how far this could unwind. But if this hypothesis is true, followed by also this assumption about the social capital hypothesis that in fact, we don't know how to have disciplined, well-organized workers in Western societies anymore, and that in a few decades, China won't either and a few decades after that, Africa won't either. And if we won't know how to reward, excellent engineering, and improvement, or even accurately materially give it its due and reinvest. If all of that is the case, then I think our society doesn't recede completely back into the dark ages. I do think we fail to complete the information revolution, I think we start rolling it back, all of the beautiful AI and machine learning innovations that we're seeing, these are all predicated on massive, massive amounts of compute. So as the price of compute those up, these innovations go away, right, they have to be radically downscaled algorithmic progress, and so on, has been much slower than just progress and computation. I don't think we quite lose cars. But we might lose everyone having a car, cars might be rare. And this, in fact, could be even rationalized in public discourse. You know, every year, your washing machine would be a bit worse, every year, your car would be a bit worse, it would be hard to find replacement parts, you would have to be relatively scrappy and improvised, buildings would look like they need a fresh coat of paint. Plane tickets would be something that you can't really afford anymore. This imperceptible slow decline in living standards together with the machines coming shadier. I think it would stabilize that sort of like, kind of living standard comparable to the 1930s. Except, I think, for example, we would probably keep a cell phone network then maybe our phones wouldn't be quite as smart. And we would probably keep something like the internet, though maybe only 5% or 10% of the population could log on. So I do think that there is some amount of progress that isn't wiped out by such a fall by such a intellectual and economic Dark Age. That is absent external shocks, right? The only way that I think something like currently ongoing as a human driven climate change could prove to be truly disastrous, is if we experienced such a dark age, just as there are these massive, massive disasters on the folding or massive deep climactic transitions such as transition to a new ice age. In that case, it it could get much worse, it might go all the way back to agricultural society.

SPENCER: I imagine that many of you listening would be surprised about this prediction, because we have such a strong sense that things move forward and not backwards, you know, the technology progresses and doesn't regress. So I wonder you to just talk a little bit about that. And also, what it would actually look like lose these capacities, because you're like, what more specifically would be lost? Like, you know, why couldn't you just put set people up in the same way that they used to be set up? What would actually be gone now?

SAMO: Yeah, I mean, we are used to everything getting better, and nothing getting worse, because that's the trajectory of our current civilization is on we can be optimistic and hope that it's inevitable law of history for the future. I'm here not commenting about progress over all of human history, right, I'm just commenting on progress in a particular human society. You can easily imagine a thesis where progress is true, but there are setbacks. And again, I think the history of the world really does demonstrate this right? Population did go down. After the end of civilization in classical antiquity, our best population estimates for the population of the Roman Empire are 60, to about 250 million people. That's a population size is quite comparable to modern states, you had, you know, seven storey apartment buildings in ancient Rome and so on, you had engineers such as Heron of Alexandria, building steam engines, and so on. If the belief is that, oh, yes, our technology is so much better, it's so much more advanced, that it's sort of impossible for us to lose it. I would counter this by saying this much more advanced technology is much more complex, and has much greater material demands than say, ancient from a technology that, you know, Hellenistic society didn't need to conquer the planet and open all of the oceans to trade routes in order to sustain its economies of scale, right, it only needed to stay secure the Mediterranean. Modern society really does need it, right. Like, the only way, the real complexity of a CPU, or a phone or a car isn't the car, it's not the CPU, it's not the phone, it's the factory that built the phone, I guarantee you that if you look at your car, the factory that built your car has 100 machines, each at least as complex and difficult to make as the car itself. If you look at those, right, each of those machines is probably made in a factory. But when you look at the factory, making the tools that build their tools to build our tools, it stops being so big, it starts being this weird workshop in Germany, or Japan, making very specialized equipment, maybe putting out a small volume, maybe like 50, or 60 machines a year, or just a small amount of these very precise, very high quality tools such as machine tools. So if you start hitting the basis of our society, that's let's say, let's for the sake of argument, say that that is mass production, where production goes down, it becomes more expensive to make each individual car and then it becomes more expensive to make each individual robot that's making the car and then it becomes etc, etc. all the way back to that little weird workshop in Germany or Japan. At the same time, all of the existing power centers remain in their place, medical debt continues to go up, education continues to eat up more and more. If you have a society that's stagnant, that's not growing, or even is actually its economy is slowly shrinking. And instead of growing, it's shrinking. All the zeros some races have things such as Oh, I have to stay in the middle class. So it's, it's okay for me to go into debt because as long as my kid makes it to Stanford, they'll be able to get an okay job and then one day they'll be able to afford a house and medial have grandkids that way, or at least grandkids that are not downwardly mobile, all of those dynamics become so much more brutal and worse, at no point, say, a university degree become cheaper in a world where cars have become rarer, or more expensive or where cell phones cease to be an everyday good. At no point do you have power centers just step aside and cease rent extraction. So at the most theoretical level, this would be rent extraction, exceeds economic growth. All that would be needed for that is for economic growth to stall, cause the sort of rent extraction, the sort of zero sum type competition is in fact spreading is in fact metastasizing. It's just been outgrown by industry every step of the way for many, many decades, if not many centuries. And you know, I feel these, these are arguments about plausibility. They're not necessarily arguments about likelihood, though, I'm trying to pick the most evocative examples I can. And really, once we think about something like, you know, current civilization, why in the world do we think that we've reached escape velocity, sort of metaphorical escape velocity, right, where we shoot off into space, and we grasp infinity, rather than we are just another society, just as all the other societies were, and we operate at a bigger scale. But does that really mean our curve is going to end up looking that different, every single pass civilization has eventually failed, and has experienced progression of technology of population of wealth? I think it's an extraordinary argument that we are different. So I feel I've yet to hear that extraordinary argument.

SPENCER: So when you talked about certain capacities being lost, it sounded to me like they were social capacities that could be lost. And then later, you mentioned sort of what I thought of is it sort of a different idea that there's two competing forces, one is forces a service extract from other people. And the other is a sort of growth force for like, your productivity increases. And if you have the extraction force of like, basically middleman like that, or not of any value, but that actually suck away money from people or other attractive forces. And if those exceed the growth force, then then you actually start getting declined. So those seem to me like two separate ideas, losing these kind of the social technology, and then that separate thing about extraction, or are you doing those the same thing?

SAMO: I agree that they are separate, but they are mechanically connected. To say a little bit more about this. Social technology being things such as law, organized religions, basically elements of culture, but I prefer to use the term social technology because they think we kind of today tend to think of culture, it's just, you know, it's a different skin, you put on the same sort of mechanical structure, right? We assume it's just a different flavor, right? It's sort of like, oh, yeah, culture, that's like, you know, the difference between different cultures is like the difference between vanilla and chocolate and strawberry ice cream, it's kind of important, but kind of irrelevant. It's all ice cream. It has the same caloric content, it delights, different people in different ways. Meanwhile, social technology is like no, actually, there is social machinery in operation. And different societies have different social machinery running, not bad, not worse, often difficult to evaluate from within the value system of any one given society. And I think the evidence is really strong there that there exists different ways of doing things, different modes of social organization, different types of organizations, I don't have much tolerance for the person that says the law you might have experienced in ancient Athens is the same as the law and legal enforcement mechanisms you might have experienced in medieval London, is the same as the legal enforcement mechanisms you're going to experience in modern day London. And it's not just technology. Think about something like policing, right? The existence of a police force is a feature of modern London, not of medieval London and not of ancient Athens. Right? So this is like a social organization issue, not just a technology issue, right? It's not just the CCT, the closed circuit cameras all over London, that make law different in modern London, than medieval London, or ancient Athens, or Rome, or whatever. So really, first, I just wanted to clarify a little bit what I mean by social technology. But now to answer how the two are related. If you have a social technology of a norm of public truth speaking, where you are embarrassed to lose face, and are shunned from economic opportunities, for saying untrue things, or things that are not backed up. And then this is replaced by a norm where you're perhaps even actively rewarded for saying untrue things that are, however tactically expedient with economic and political rewards. Well, it was intense competition and a zero sum competition that made it so that it was more profitable to be less than truthful, to be a propagandist to be a salesman at all costs, but it resulted in a change of social technology. So I sort of feel this growth between an extractive and a productive force, this ratio between the two, I have this piece of writing called Empire Theory where I dive into this and I propose that we can think of in a simplified way, a two-by-two matrix of expanding and declining in centralized or decentralized you might have an expanding centralized society, where it's the center of society and the central point hours that are engaged in the forces of production. Or you might have a decentralized expanding society where it's the periphery that's growing, and the center is kind of stagnant. Or you might have a centralized declining society where the center of society is drawing in all the resources into itself in a very vampiric parasitic way, it's easy to argue that the late Roman Empire was of this type, and so on and so on, right. The different regimes of growth and extraction result in a different landscape of applicable, viable social technologies. And after a social technology, live leaves, living memory, it is very difficult to reconstruct, people in medieval Europe saw no fundamental break between their society and the society of antiquity. That's why when they refer to people by the title of Dukes write a hereditary feudal title, it came from an ancient Roman appointed title, Dukes, right, a kind of administrator, you can have something still call itself a republic call itself a monarchy, call itself, the Persians or whatever, or the Egyptians, even if all the practices have changed. So once you no longer have that living memory of how the social norm operated, and you can never fully put this into words, right? This is very tacit knowledge, it's people's behavior, a written description will be less than helpful, at best, you can use it as inspiration to try to re-engineer the past. But unless you have mentors or people who are skilled at this mode of behavior, that's going to be very difficult to reconstruct. So I claim that the social technology just will reliably be lost. So I think that if you have extracted force, in a society overpower force of growth, you will reliably start losing certain very pro-social social technologies, they might be replaced by other social technologies, right, you might have a society that's in a very extractive orientation, it might really perfect, for example, some aspects of propaganda, or some aspects of political control, or some aspects of warfare and so on, it's just on that, I think they're gonna start losing social time.

SPENCER: I guess the idea would be with loss of certain sorts of technology, that could actually have implications like losing the skill to make the machines that make the machines that make cars or something like that. And so with the erosion of that sort of tech, you could have erosion of like, the ability to actually build useful technology. Is that the idea?

SAMO: Well, yes, I can even make it more specific than that. It might be perhaps, that there is not a single good engineering school left in all of Canada, or not a single good engineering school in all of Europe, or America or Russia, and that you find it really difficult to hire a good engineer. Also, when you try to hire a good engineer, you're not quite sure how to tell who is a good engineer and who's the slightly worse engineer. So when you hire and you start making products, to your surprise, you realize that the error rate is unacceptably high, and that the profit you're making isn't as high as you would expect. Again, I really want to hammer home that this isn't purely hypothetical. We in fact, saw a version of this during the late decline and eventual collapse of the Soviet Union, you had a decline in technical capacity. You had also, to some extent a decline in living standards, a massive drop in fertility, a massive drop in expected lifespan. Now, of course, the collapse of the Soviet Union was not a full civilizational collapse. But I do think you would probably agree with me that it was a kind of like a catastrophic deindustrialization. And of course, it was an oppressive society, etc, etc. Media was better to end that experiment. But it was a deindustrialization. It was a de-skilling. The few remaining skilled people, left for the United States, in Western Europe. And today, Russia struggles to reindustrialize I mean, I guarantee you that Putin has probably done his very best to try to reindustrialize Russia over the last 20 years, and has completely failed. And one story is that okay, Russia was just economically out competed. Another story is the Russia of today is not the Russia that industrialized, right? And the social norms are too different from those that would enable industrialization.

SPENCER: You know, when you mentioned you kind of this particular theory of how American civilization could collapse, I was a little bit surprised, because I expected you to focus more on institutional decay, like our institutions were built for specific purposes. And then the world changed and they didn't keep up or they became sort of parodies of themselves over time increasingly, where maybe the people that founded them, really were focused on solving certain problem. And then their successors were a little bit less focused and you know, three generations out, they're kind of they've -

SAMO: Completely lost the plot.

SPENCER: Exactly. Do you have thoughts around that form of decay and how that could affect America?

SAMO: I think that form of decay is it's obvious but from any particular institution decaying. It's hard to establish that the society as a whole is decline. I think that form of institutional decay can eventually then result in all of these other economic and technological problems that I described. I hope it's obvious that if you had institutional decay in the Federal Reserve, or the University of Harvard, or the US Federal Government, some of its branches such as, imagine a DARPA, that's much worse at its job, what's the compounding loss to yearly GDP growth. So, again, that's sort of the other side, that's the more institutional perspective, the only reason I focus earlier on mass production is because I was running with the hypothesis that industrial civilization is a unique phenomena. But the more general theory of institutional failure is, again, a theory where the institutions become extractive, they no longer fulfill their purpose, they best maintain themselves in the resource distribution, they no longer deploy these resources, well, they no longer attract or cultivate the right talent, and they are not replaced by other institutions. The sort of perspective I would give is that there are organizations that prevent us from making better, much more functional organizations. If you imagine the set of contractors that NASA was working with, for the past few decades, it seems very obvious that these contractors were actually an impediment to aerospace progress, rather than an asset. And it took extraordinary efforts for the creation of something like SpaceX. Why was that an extraordinary effort? The engineers were lying around? Well, yes, the engineers were lying around, they might want to build some things, they might even want to build useful vehicles, much cheaper rockets, which who exactly was going to be the client for SpaceX, Elon actually had to maneuver the company and his successors as well. They've had to maneuver SpaceX so that it was capable of winning the right contracts, contracts that would make production of rockets like the Falcon 9 and later ones like starship worthwhile, they wouldn't make them worth the investment. So old institutions can crowd out new institutions, even when they cease to be as functional. And then, of course, sometimes extraordinary teams or individuals find ways to circumvent those dying old institutions. But other times they don't, right. And the society really rigidify in a profound way.

SPENCER: That makes sense. I imagine the situation where if universities gotten much worse, and they were still producing sort of the people in charge at a lot of other institutions, that could have kind of a follow on effect, right, where now you're getting people coming out of the universities that are not qualified to run these other institutions. Or another example might be, if newspapers and the media continue to get worse and worse, then eventually, it becomes harder and harder for everyone to sort of figure out what's really going on and what's just political spin. And you can see that sort of exacerbating many of these other problems simultaneously.

SAMO: Right, you could see, in fact, over time, a decay and poisoning of the epistemic commons, right? Because we do, I think, rely always when it comes to truth seeking on other people, we have to rely on them, because there's just too much stuff to figure out. And you have to trust your ability to evaluate experts directly, or you trust whatever system that exists such as university that evaluates experts for you. Now, the fun fact about institutions is nothing stops them from lying to you. One of my favorite anecdotes when growing up was when I asked my dad about the transition from Communism to Capitalism, because this is super notable to me, I grew up in the 1990s. And he said, you know, he sometimes he wouldn't be as eloquent as I am, let's be polite, and call me eloquent rather than verbose. And all he said was, "SAMO:, no one asked me to redo my economics degree". So yeah, like, isn't that like, deeply important somehow, if it's just about the content, he learned, all of the socialist economics should have been falsified? Right, it should have been the Marxist stuff should have been replaced by I don't know, like, not and ran with Hayek or Keynes or whatever. No one asked him to redo his economics degree. He just continued in his role as a CFO or whatever. And I often think back on that, and sort of like, okay, we trust this credentialing system. And we trusted even after the falsification of the core theory, the credentialing system was measuring happened.

SPENCER: Yeah, it seems like, in general, any corruption of a credentialing system that really is trusted by people has these very severe knock on effects. Once the credential can no longer be trusted, but people don't realize it can't be trusted. Now you just have misinformation being spread. And then it's like, how do you go back from there? Once the credential doesn't mean what it used to? Of course, the people doing the credentialing are not going to be like, oh, yeah, you should stop listening to us, right. They're going to find ways to justify why you should still listen to them.

SAMO: Exactly. I mean, at the end of the day, one of the reasons, the word authority in modern English has a mild negative, almost pejorative sting to it is because Protestant critiques of the Catholic Church in the 16th 17th and 18th century focused on its claim of authority over questions of salvation, right. So there are all these big tracks like decrying the supposed authority of the Church and so on. Kind of over time, the term authority acquired this negative connotation were originally kind of all it means this kind of otoritas is that you have the relevant knowledge, and people should probably listen to the person with the relevant knowledge. It's an easy claim to make, it's difficult to substantiate though, it's harder to substantiate.


SPENCER: So we talked about ways institutions can go wrong, right, tell us a bit about functional institutions and what they do differently.

SAMO: Yes, I mean, there's no single unifying theme to functional institutions, I would broadly say that they are overproduced radically. They're not difficult to spot, right, a functional institution will be this organization, and almost on any metric you choose, radically outperforms, given its resource base, let's say an organization nominally of the same types of this happens if you look at religions, organized religions, such as churches, it's there's very clearly. There's one church that's kind of the biggest, it's like Catholic Church. And you might say, so what's the point of organized religion? I'm not going to comment on that. I sort of functionality here is a little bit immoral. It's just noting that there exist these really massive, profound outlier organizations, the same can be done for states, the same could be done for companies. In fact, it's routinely done for companies, right people note that, in their heyday, there are some firms that seems to just have a golden touch, just have this economic magic to them. And they persists sometimes for years for decades for the lifetime of founder. And then generationally, they usually are some sort of flaws, some sort of problems with it. So again, one feature of the institution is its radically over producing its peers out producing its peers on something important, right, either number of believers amount of territory captured, widgets produced whatever. The second sort of identifiable sign of it, this one, I think, is probably also mechanically relevant. So it's not just a symptom, it might be, in fact, like a cause for the functionality. I think the founder plays a very significant role. Now, the founder doesn't necessarily need to have dictatorial control, maybe they set up a very different system of discussion, debate, and work where they only hold a tiny portion of that power. But initially, they still were the one that created the system that put it into motion. Again, a comparison here could be made if we're gonna reference political theory, to say America's founding fathers. They had a lot of political power in early colonial America, but they set up a system where none of them individually, necessarily held all that power. Still, it's a fact that they did set up that system, the Constitutional Convention was 300 of America's most powerful and rich people gathering together without public oversight to determine what the next government of America should be. You know, I really think that the Constitutional Convention at that moment was the designer, right? A collective designer, if not an individual one. And of course, it was a unique historic circumstance. Still, if the designers are still present And usually the maintenance of a system goes, well, if a designer is not present, if that founder, cohort or founder is not present, then maybe you have solved the succession problem successfully. And you correctly identify new individuals who have what it takes. Or maybe you didn't, right. So, basically, when founders are present, I think things go better. When founders go away, maybe it's okay, maybe it's not, it's like, lower probability. Another feature of a functional institution, I think, is there is internal differentiation, where, in fact, people develop very deep skills. And they exist in social symbiosis with the other individuals. So people's roles are perhaps very different. But they're not necessarily like that status differentiated, if that makes sense. They sort of know what to expect from each other. It's almost like the sort of union archetypes or to put a more modern frame on it. It's like a superhero team, right? It's like kids on the schoolyard will always debate which superhero is the best and could beat all the other ones. And the point really isn't about that, the team, each of them is very distinct, you know, Hulk, Thor, whatever, Spider Man, they're not however competing, they're not fighting, they're not like reproducing excess capacity, they're each doing something quite different. So let's call this functional specialization, right? Functional specialization, which can very naturally lead to hierarchies, but doesn't necessarily lead to hierarchies.

SPENCER: So what's a good example of that in a successful institution?

SAMO: I think that there are examples from the history of engineering and the history of war. But since I've talked so much about the industrial process, let's maybe talk about the Person General Staff system. In the Person, General Staff system, a general was not the person that concocted the entire strategy of a battle, rather, worked much more similar to take a fictional example, the bridge of the enterprise and Star Trek, The Prussian general, we have a particular question something about the battlefield, something about the contingencies of the battle, he would consult staff officers, who are in charge of planning for these contingencies, as what the situation is, in sort of work as a this type of extended brain, right? There's this team of people, they exist in this super hierarchical environment where all of military culture is about following orders and so on. Yet they kept it as a necessary prerequisite of work there, that if the general picks the wrong plan, you as a staff officer could still register and write down your protestation of this choice, right, you could object to the choice that was made. You couldn't overturn it. However, later on in the course of the battle revealed you were right. Well, those very important, and almost as a matter of habit, the person generals did defer to the decision of the sort of staff officers.

SPENCER: So was there any other properties of functional institutions you wanted to mention or the story cover?

SAMO: Yeah, I think one of the more important ones might be, also the functional institutions tend to be somewhat unique. I think that they don't necessarily engage in that close in imitation or mirroring of other organizations. They tend to be, for lack of a better term, intellectually sovereign, they feel qualified and justified to think about their area of human activity, and don't really seek out external authorities or dependencies. So that may be one more important feature.

SPENCER: So you want to just give us a few examples of kind of classic functional institutions just so people can have something in mind when I think about this.

SAMO: I think that, for all its flaws, 1960s. NASA is actually a great example. It's a very mission oriented organization. It achieves remarkable feats. Another example is the British Royal Society, which I think was absolutely vital to the creation of scientific culture. And I think it probably stayed functional for most of its history, I think that plausibly after the 1900s or maybe as early as the 1850s. I think it falls out of functionality. But before that, I think it has a good 200 year run, where it is consistently connecting scientists and thinkers around the world, not just in Britain itself. Another example, that's, I think, political example, right functionality, I think would be the Hellenistic states that were created after Alexander the Great conquests, I think that say the government of Ptolemaic Egypt was an incredibly functional institution, where it supported all the classical functions of the state right, such as defense and in Egypt's case public irrigation, but uniquely also managed to support a healthy, state funded scientific enterprise. This being the great library of Alexandria. The museum was more than just the library was also a gathering of philosophers and engineers and scholars, he's we're all paid by the Crown. Another example that's I think contemporary of a functional institution is I would say DARPA is relatively functional. And it's the American defense apparatus is probably, let's say, the best at facilitating the development of breakthrough technology, it could perhaps be argued that it used to be more functional, that up close, it doesn't work that well. But by the standard of functional institutions, it outperforms its competitors.

SPENCER: So you this idea you write about sometimes, which is great founders theory, you want to talk about how that connects to functional institutions. And then tell us more about how great founder theory works?

SAMO: Yes, I tend to have like a very founder oriented perspective, this doesn't necessarily mean that a great founders, just a normal founder, who happens to be exceptional, right, great founders theory is, after all, a theory of history, it's a theory of civilizational development, is the core thesis is that social technologies are actually more similar to material technologies than we might assume. They have an inventor, an implementer, or popularizer. Right, someone that brings it to society. And I think that when you look at the broad course of human history, it's the social technologies, ultimately, that end up facilitating, or getting in the way of the development of material technology. So social technologies have creators, social technologies seem to me to be a significant bottlenecks and gatekeepers, at least of which material technologies come to be, therefore I sort of believe that it is this particular kind of deep social reformer, that ends up being the most important figure in history. So a random general, great founders theory differs from just a normal great man theory of history, in that it focuses on these institution creators who produce new, unique and innovative institutions. So a great founder would not just be a successful general, right? Most generals would not be great founders at all, no matter how important the battles they won. So no one Patton, right, no one Marshall, Zhukov, and so on. However, Charlemagne specifically, who happens to be a conqueror, but also happens to be a deep social reformer of law, and education, and most importantly, setting up the feudal system, as is understood later, could be seen, as a great founder, can be argued, is a great founder, and I think is likely relatively likely to he defines this core of Western civilization, around which later other institutions are built in parallel. A different example from Chinese civilization not if a social reformer would be Confucius right where I think Confucius successfully created a school of thought where he himself failed to execute the plan, Confucius, his plan was that you're going to figure out how to cultivate human virtues, how to cultivate human capital, you're going to gather together all the thinkers you can find in China, and then you're going to go advise a prince of a particular small warring state in China and implement these social reforms that never worked out for him. He was always politically maneuvered. However, his successors who were trained in the same school of thought came to be very influential advisors, to the later emperors of China, and in fact, have reshaped Chinese civilization many times over right many times over the Confucian project, after the end of the old dynasty was taken up again, once more. So Confucius is a very strong candidate for a great founder. In say, Egyptian civilization, I actually think that a particular architect and statesman impotent might actually be the best candidate a better candidate than some of the other ones. And then, when we start viewing history, that's more recent, right? More recent history, it becomes trickier partially because we don't yet know the score, right? We don't necessarily know which innovations are going to last and which are not. If we were speaking in 1970, I might decide to be provocative and call Marx, a great founder of a new civilization. In fact, the list of the top 10 Most Influential human beings in world history made in about what was it 1976, I think, listed Muhammad first, Jesus second, and Marx third, and only then started talking about people like Napoleon, right? I don't think anyone making such a list today would put Marx up there with, you know, Jesus and Muhammad. But at the time, it looked pretty much like a new type of civilization was created a Marxist based civilization, and it might endure for many hundreds of years and might redefine Russia and China. And honestly, you know, maybe if the Chinese system continues to exist continues to thrive. Maybe in two or 300 years, it will be considered that Mao or Gang or great founders of a new, very different kind of Chinese civilization when not focused on Confucianism, not focused on the Emperor, but rather focused on this materialist theory of history.

SPENCER: So would you say great founder theory is kind of a refinement of great man theory, because great man theory points to the specific people that had either sort of undue influence like these great generals, or presidents or whatever, but prefer monetary, you're going one step further and saying, No, it's not enough to be in a position of power, you have to actually create a sort of new type of society or a new type of social technology that then kind of spreads widely.

SAMO: Exactly. And it's not the case that you're engineering the full future, of course of a civilization, right. Psychohistory is not real. But it is the case that you have to fully implement a deeply transformative social technology of this type. And I think that it's not necessarily, even though, the series called great founder theory, I think that very small groups of people could be doing this collectively as well, I do believe in Senius, the genius of seems, I think something of that type is responsible for the flourishing of Renaissance Italy. But essentially, I do think it has to be relatively small, it's sort of like, these cultural changes happen too quickly and too completely where the working prototype emerges too fast for it to be this slower process of cultural evolution, or this mimetic selection, and so on. You have these really deep radical transformations like the spread of Islam, right, where Muhammad batches together a whole new system of law, a whole new system of emotional regulation, all of this stuff, right? And of course, if one's a once a believer, you say that, and that's evidence that he was divinely inspired, because there's no way a mere human being do that. My theory is, at least that sometimes, yes, human beings can do some of that. And of course, they're very much outliers. And they aren't unstoppable. It's somewhat down to chance. And you might say, "Oh, actually, if history saturated with this, then these great founders don't matter. They're inevitable." And I'm going to say no, actually, they're not one sort of economics framing of of this might be is that I would say that great founders are radically under supplied, you only have these individuals occur in the right place and time to do what they're doing once every few centuries, once every few decades. It's not the case that everything that can be done, has been done. It's not the case that if Charlemagne was counterfactually, killed, and so on, the Frankish state would end up becoming this kind of feudal society that it was, it's not the case that if we had eliminated Confucius, Chinese civilization would have gone on the same course and so on. These are really idiosyncratic and unique individuals. At the end of the day, of course, these individuals are still who they are, because of their environment, but they're not a necessary product of their environment. They're a very rare event.

SPENCER: Would you say that there are examples of this in science, or these tend to be more around things like religion?

SAMO: I would say that they definitely exist in science, because they consider science to be a whole space and deep well have its own set of social technologies. And I actually think that the study of science would do very well to like, closely examine what it is that is unique about scientific culture, right? The self perception of scientists is valuable, but not necessarily accurate. You know, at the end of the day, we don't ask a bird to teach us aerodynamics, even though we don't question the birds expertise at flying. So you know, scientists could easily be practicing science and be quite wrong about science. In fact, you can tell because if you ask different scientists about how science works, they tend to disagree. Yet, their unstated non verbalized intuitions as to who to trust who not to trust, who is insightful, who is not insightful, they actually tend to agree on those, they tend to agree who they respect, even though they disagree on their explicated principles of science. Now, if you're asking me who the founder of modern sciences, who among the founders of modern sciences, it's almost too difficult to question. It is something that I think would require a deep dive in its own, I could say make some arguments as to who I think is the founder of particular schools of thought. But then how civilizational transformative are those schools of thought? I do think that whoever you choose to attribute the creation of the Royal Society to and there is some ambiguity there has a very strong claim to the title of the great founder of science.

SPENCER: You have a small group be the great founder, where they're sort of riffing off of each other and in a way that creates this new social tech.

SAMO: Yes, I think that there are some of this happening with Alexander the Great where Alexander, the great tutor is Aristotle. And most of the key decisions are taken in partnership with his other key generals. And we can tell that it's not just Alexander the Great himself, because once his successors inheritor, parts of the empire note his successor isn't his son, right? There's this vicious sort of civil wars that happen. You have people like, Ptolemy, or Solutias, who are competing very strongly. If it was just Alexander, these generals would be overthrown by native elites. In Egypt, for example, the priesthood is still there, the powerful priesthood, or in the Persian example, many of the Persian nobles are still alive, or in the example of various Greek city states, the city elites, you know, be they, the wealthy citizens or priests or local commanders, they're still around, none of them actually manages to change the political system and the social system. And this new social and political system is radically different than what the kingdom of Macedon, which was this PEF, provincial, out of the way small Greek kingdom was practicing. And they're not governing these states, the way classical Greek civilizations oriented around city states. No, these are a new type of entity. So I basically think that Alexander the Great had a senior surround him, the genius of his scene.

SPENCER: Somehow one of the really interesting things about talking to you is that it feels like you've absorbed so many historical examples that you have the search algorithm, where, for any topic that's input, you can just pull historical examples out of your mind, which is pretty cool. And a pretty interesting thing to do. And basically think about a way of thinking about people's expertise as on what topics can they just pull examples immediately for whatever's being discussed?

SAMO: I think it's an interesting test. But it's not sufficient, I think one has to check for all sorts of additional constraints when evaluating thinkers. But thank you, thank you for the compliment.

SPENCER: Anything you want to share with the audience, before we wrap up.

SAMO: I would strongly encourage people if they want to understand and think about history in a way that's actually applicable, it's actually useful to do very close case studies, of either particular organizations, or particular, relatively narrow periods, and try to analogize them with what you know about humanity, like the correct way to use history as sort of like, these are human beings like you, you're acquiring a new data point on what can be done. And one of the most fruitful uses is when people have a clear, narrow understanding of a particular historical event in enough detail that they can reconstruct the social technologies or tactics that exist in that time period, and then fruitfully apply it to the present. I think the historical record isn't just a warning for things like civilizational collapse, and so on. It's also a source of hope. It's an existence proof, right? With one human being can do, another can do. And there are all sorts of things that I think today we've grown be accustomed are impossible. I think we're a little bit like you know, the elephant, where we've learned that we are going to remain chained and there's no reason to go elsewhere. I think if that elephant that chained elephant sees a different elephant, traveling freely, will maybe that breaks him free from his delusion.

SPENCER: Thanks so much for coming on.

SAMO: Thanks for having me on the show.





Click here to return to the list of all episodes.


Sign up to receive one helpful idea and one brand-new podcast episode each week!

Contact Us

We'd love to hear from you! To give us your feedback on the podcast, or to tell us about how the ideas from the podcast have impacted you, send us an email at:

Or connect with us on social media: