Songs Of Societal Collapse

I'm not joking when I talk about my fears of society collapsing.

Civilization is a fragile thing, much more fragile than people realize. So much of it is path-dependent, and the foundation is littered with load-bearing columns that look like this. It’s especially obvious in tech, but this dynamic is everywhere

Imagine living in a world where, if Ronald gets hit by a bus, all digital e-commerce stops.

To help you imagine, we’re going to do a video game review!

---

Songs of Syx is one of my favourite games. On the one hand, it has done some truly novel and unique things, and on the other hand, it follows my rule of “made by like one guy”. I’ll let Sseth explain the game before continuing

Songs of Syx is a really cool game that is an exploration of what you’d get if you bolted on a Dwarf Fortress-style¹ city builder to a Total War-style grand strategy game. The game starts with you building up a city, slowly building up wealth, scientific discovery, and infrastructure. But unlike other city-builders, where indulging in municipal planning is the point, in this game, you’re building your city for the only reason that any leader has ever built a city, in all of human history: to power your army and conquer your neighbours.

Your army in the field requires logistics. Logistics require supplies. Supplies must be generated by an economy. So your goal is to build a city that generates enough economic surplus to supply that army, or at least to supply your militiamen when raiders attack.

Which brings us to another cool thing. The combat system, as it currently exists, is a little anemic.² Most battles you’ll fight are pitched battles taking place on featureless plains. But when raiders attack you, you aren’t defending a featureless plain. You’re defending your actual city, that you built. The attackers are sieging your walls. They’re pillaging the factories and houses that you have built. So you made sure you built them to be easily defended, right?³

---

The tight integration between the city-building element and the grand strategy element is once instance of the thing, in Songs of Syx, that makes it truly unique and cool, which is that this game has done an amazing job of creating relatively simple game mechanics that combine and interact in very complex ways, and that accurately reflects the real world. I’ll elaborate on that, but first, let me sketch an outline of how a game of Syx typically goes for me.

I start with a small population of settlers, and my two immediate priorities are 1) create a warehouse for the goods in my economy; and 2) secure a food source so that people don’t starve. Warehouses are a critical backbone to this game. Since you’re trying to build an efficient economy, making sure you have appropriate warehouses for collecting and distributing supplies is half the game. I tried the strategy of giant centralized warehouses once, and quickly found that my economy collapsed because all of my artisans were spending half of their workday walking to and from the warehouse to get input goods. Further, goods in this game decay over time. This is important, and we’ll come back to it. But warehouses get involved because goods stored in a warehouse will have substantially lower decay rates. For example, if I’m remembering the numbers correctly, if your population is surviving on a diet of fish, fish will decay at a rate of 20% per year, but only 5% per year in a warehouse.

To secure my food sources, I quickly spin up some small farms on the most fertile plots of land. These farms will not scale up to larger populations, but they will make sure our existing people can survive for now. In order to build those farms, and some housing and urban infrastructure for my people, we need construction materials: wood and stone. Fortunately, you can pick stones up off the ground, and wood grows on trees. So I task some of my pops⁴ to manually collect those stones off the ground, and bring them to my warehouse, where they magically transform from rocks on the ground, into stone resources for building houses and buildings. And I task some others to start felling trees, but you have to be careful; trees take a while to grow, and if you cut down a whole forest, it’ll take longer to grow back.

As all of this is happening, I’m slowly building up infrastructure that makes my pops happier. Basic housing, wells for water access, hearths to keep warm in winter. And as long as your happiness is high, immigrants will come to your town, boosting your population. Generally speaking, you always want more immigrants,⁵ because more workers means more economy, but you have to be careful. Too great an influx of immigrants too quickly means you run out of food, and everyone starves.⁶ Further, the different races in the game have different preferences. Most don’t much care for other races, and even the races that can live in harmony have different preferences. If you’ve, for example, built a city optimized for human happiness, on perfect square angles and stone pavement, and you add a bunch of forest-loving Tilapis, who want green space, wooden structures, and organic, round shapes, those Tilapis are going to get real pissed, real fast.

---

Eventually I’ll have a stable enough food supply to expand my economy, and the first way we do this is with carpentry. Carpenters can take the wood supplies in my warehouses and turn them into furniture. Furniture has a lot of uses. You can allocate some to housing units, to increase housing happiness. Most advanced crafting buildings require furniture for construction, which represents a sort of generic pile of tools for them. And, when constructing most buildings, you must include internal components. For example, when building a bakery, you must include ovens (which directly control how many employees can work there, and how much bread they can bake), auxilliary ovens (which must be present in a certain ratio to the main ovens, and increases the efficiency of those ovens), and a storage pile (where the bakers put the bread when it’s done, awaiting a warehouse teamster to collect them). Most of the efficiency things, like the auxilliary ovens, require furniture. And efficiency is the name of this game, so you might as well start carpenting.

In the early game, you won’t really need a steady flow of furniture, though. You’ll start stockpiling it, and using it in one-off construction projects, and you’ll end up with a surplus. What do we do with economic surplus? WE TRADE. For every economic good in this game, there is a fully simulated trade market where supply and demand between you and your direct trading partners controls the current market value of that good. And so, we started selling our extra furniture, and using it to buy materials that our economy can’t manufacture right now. For me, that’s usually metal (which is needed for more advanced buildings) and basic weapons/armor (which is needed to construct guard posts and police buildings, which quickly become critical once your population grows large enough).

Once I’ve gotten the metal and basic weapons needed for my early purposes, all of the rest of the furniture surplus goes into food imports, and I will very quickly get to the point where a majority of my town’s food supplies are imported. Now, those farms I spun up seem kind of silly. I have 1/3rd of my population slaving away in the fields, only for them to generate 20% of the food that selling my furniture is generating. So I let the fields lay fallow, and start building scientific research buildings. They will in turn generate research points, which I will use to unlock advanced crafting buildings, and various upgrades to existing things.

The first things I unlock are various basic infrastructure things. Lavatories, speakers’ corners, food distribution centers, paved roads. Then, I’ll unlock some agricultural efficiency upgrades. Through all of this, I will incrementally increase the happiness in my town, leading to immigration, and the food security of my town to cover the increased population. Eventually, I’ll have enough of a population surplus to build enough scientific buildings to unlock paper, and universities.

Universities let me make my pops go to school. Pops “employed” at a school do not work, instead, they attend school, increasing their education percentage by 1pp⁷ per day. This makes them generally more effective at all tasks, increasing economic productivity, and it also is worth some research points on its own. Libraries consume paper (which must be produced at paper mills) and act as a multiplier on your laboratories’ science output. These two unlocks will dramatically increase the research points available to me, quickly allowing me to unlock whichever industries I need, and get some much-needed efficiency upgrades to my core industries.

---

Around this time, I unlock administration. Administrators work at embassies, and generate administration points. Administration points are used to build upgrades in the provinces we’re about to conquer in the grand strategy mode.

My furniture trade has been steadily generating some wealth in our vaults, and it’s time to use that wealth to hire mercenaries and conquer our neighbours. So we do that, and once we do that, we are now an Empire™. Our capital city is managed directly by us, we build the buildings, we allocate the labour, etc. But our provinces are abstracted away from the details. The provinces will have a population distribution and various multipliers to various industries, and we’re going to use our administration points to build up infrastructure and industries in these provinces. We build up enough infra to get their population to useable levels, and then I’ll typically either have them specialize as woodcutters or as farmers, depending on their geography.

With this, we now get a massive influx of food, as the tribute shipments come in once every few days, and now, we have a large, reliable source of food, at least as long as nobody conquers it, and we don’t run out of paper for our embassies to generate admin points. Further, as our city grows in population, it grows in space. Those small farms, built at the outskirts of our 50-pop town, are now smack dab in the middle of prime real estate in the downtown of our 500-pop town. And we don’t need ‘em anymore. So I’ll bulldoze them and replace them with more labs and embassies, to generate more tech points and admin points. The tech points go towards unlocking more efficiency gains, and the admin points go to conquering more foreign lands, which give us more primary resources, which let us specialize away from primary industries into more advanced industries.

After a few iterations of this, we have a stable, prosperous capital city, where half the population is dedicated to the sciences, half to service industries, half to advanced manufacturing of opiates, weapons and armor, and machine tools, and half to logistics.⁸ At this point, we can use our massive economic might to draft conscripts from our provinces, supplying them with food, clothes, and weapons, and send them off to pillage the world map. Eventually, we all become one, and are unified under my great, glorious empire.

---

As mentioned earlier, one of the coolest things about this game is how seemingly simple game mechanics have combined into a very complex system that closely mirrors the real world. I just walked you through that.

Just like in the real world, my city starts with 90% of my pops working the fields, desperate to stockpile enough food to survive the winter. As our population grows, and our efficiency grows, we eke out enough extra production to start re-investing in our economy. The efficiency unlocks from our tech tree is basically investing in capital, to make labour more productive. And, as we build up the economic surplus to start transitioning to advanced industries like carpentry, brewing, papermaking, weaponsmithing, the income we make from this generates enough food that we don’t need to have 90% of our pops in the fields all day anymore.

The game doesn’t have tech levels, it doesn’t have eras, it doesn’t have any explicit mechanics for ‘industrial/economic transformation’. Just like in the real world, all of these things fall out of the simple facts of reality.

By the time you’re in the late game and expanding your empire, shockingly few of your pops are actually economically productive. Rows of factories and fields are replaces with rows of houses and rows of services. A town of 5000 people needs plenty of restaurants, plenty of lavatories, plenty of barber shops, after all, and those all take up space. We don’t need to have 300 carpenters anymore, either; we have 50 weaponsmakers importing iron with our trade surplus and selling the weapons back to the world for a 300% markup, and this generates more than enough income for us to buy all the furniture we need from less advanced civilizations.

Yes, this game is so cool, that it created game mechanics that cause outsourcing to be optimal in an organic way. Just like in the real world.

---

But also, just like in the real world, as your supply chains become more complex, they become more fragile. And if you specialize too hard, you may not be resilient to economic disruption. The same mechanics that cause this complex, emergent economic behaviour, can cause cascading collapse if you’re not careful. This has happened to me several times.

Let me sketch you out an example. An enemy army is approaching my town. Maybe it’s a bandit raid, maybe it’s an enemy army, doesn’t matter. As soon as they enter my province, they are blocking the trade route, which stops all food imports from coming as long as they’re there. I have a good supply of food in my warehouses but, with all my farms replaces with suburbs, I have no ability to sustain for a long time.

So I rally my direct forces, the city militia that my city maintains. My pops are all required to train to serve in the army, and they’re some of the best-trained and best-equipped troops in the world, so we’ll have no problem defending the city. But…. activating my forces means taking them away from their day jobs. For the duration of the defense, my entire economy grinds to a halt, and the goods in my warehouses start to decay without being replaced.

I sally forth and defeat the enemy in the field, handily. However, we do take some losses. My 800 militiamen lose 50 guys. When I de-mobilize them back into private life, our economy is now missing 50 people. Further, I take a massive hit to happiness, because people don’t like it when their friends die, which causes another 200 people to leave my city. Further further, I don’t have enough cemetery space to bury all of my war dead, and this causes another hit to happiness, and causes another 200 people to leave.

I just lost 10% of my workforce, after eating through half of my stockpiled food and watching my warehouse goods slowly erode away. The food will be quickly replaced with tributes from the provinces, but the workforce, not so much.

I now have to quickly reprioritize my workforce, as large numbers of jobs are going to go un-filled. Obviously, I shut down the universities and force those dweebs to get real jobs that actually power the economy. But research points, like all other goods, erode over time if workers are not constantly generating more of them. The marginal decrease in my scientific output causes me to lose so many stockpiled research points, that various techs I have unlocked get un-learned until I restore my research points.

Most of those techs were efficiency upgrades. And, as luck would have it, I lost just enough of my efficiency upgrades to papermaking, that I can no longer produce enough paper to power all of my libraries (and paper, generally, is too expensive to import at scale). So now my libraries are running at reduced capacity. And remember how libraries act as multipliers on the laboratories’ scientific output?

When all is said and done, a 4-day siege that I trivially broke with only 50 casualties (to the enemy’s 3000) ends with half my population leaving, and 2/3rds of my tech tree needing to be re-researched. 50 hours of game progress, gone in an instant.⁹ And none of this was programmed into the game. It’s just a natural consequence of a simple but relatively realistic model of an economy.

---

Our real world works this way, too, and I don’t think people fully realize it.

ALL of the economic pain we’ve been facing for the past five years was a direct consequence of shutting down the economy during covid, and then printing money instead of telling everyone they’re going to have shitty lives for a while. Literally just a three-month period of bad policymaking, before states realized they can’t just shut down economies and still be alive, and half a decade later we’re still dealing with the fallout.

Things like this are why I freak out so much, all the time, about societal collapse. Normal people do not understand how fragile all of this is, and how much time, energy, and money, is invested in just keeping society going as it is. I’ll leave you with two examples.

In February, 2021, Texas got two feet of snow. As a Canadian, this was mostly just amusing to me, watching as all hell broke loose over a storm so mild that we’d just call it “Sunday” back home. Oh no, it’s 20 degrees outside. Oh no, there’s some snoh on your roahds. Jesus christ, pussy, it’s like you’ve never driven in a blizzard before.

Except, well, this is Texas. They hadn’t. This is Texas. They didn’t realize that ice expands when it freezes, but your pipes don’t. They didn’t realize that not all sources of heat fuel are the same, and if you burn charcoal in your fireplace, you’ll die of carbon monoxide poisoning. They didn’t realize that, when snow melts, it doesn’t disappear, it turns into water, and if that water re-freezes, it doesn’t re-freeze into snow, it re-freezes into ice.

The most hilarious instance I saw of this¹⁰ was a dude in a fancy sports car, stuck on black ice on I think highway 360 but I’m unsure. His car was sideways, basically perpendicular to the direction of travel, and his wheels had no traction at all. If he was Canadian, he’d know you do. Put your car in neutral, get out, push it to rotate it until it faces the direction you want to go. Then, take the little foot-mat under your driving pedals out, and put it under the wheel. The spikes on the bottom of the map bite into the ice, and the wheel gets traction on the fabric. Very slowly accelerate the car until your wheels aren’t on ice anymore. Get out, collect the mats, put them back in your car, and go on your merry way.

This dumbass, on the other hand, thought “well, pedal makes car go. If car not go, need more pedal”. He was flooring it. In a fancy sports car. That was pointed directly at a limestone cliff face. If at any point his wheels actually hit pavement, he would have been a red mist on the side of that cliff in about 0.27 seconds. But he didn’t die, so I’m allowed to laugh.

But that’s just me dunking on dumbfuck Americans who hate me. The actual point of this is that the Texas power grid almost collapsed. Wind turbines froze and stopped spinning. Natural gas burners that hadn’t been winterized got ice in the pipes and gas stopped flowing. Solar panels covered in snow can’t catch any sun. The state disconnected large numbers of people from the power grid, for multiple days, to avoid overloading the grid.

Most people don’t understand how the power grid works. For our purposes, there are four important factors:

First, there is no energy storage in the power grid. The power you consume is generated, on demand, at a power plant, when you consume it. Those plants need to have both the capacity and the agility to scale up and down power consumption in an instant, if needed.

Second, if supply and demand of electricity ever get too far out of whack, very large things start exploding. The power grid is filled with all kinds of tripwires and failsafes to forcibly disconnect things before they explode, if needed

Third, if stuff does explode, the economics and supply chains of that stuff are so complicatd that replacing individual components might take two months, and the manufacturers of those components can manufacture a few dozen a year. If 50 explode, you’re waiting two years to get your power grid back up.

And fourth, most power plants require considerable amounts of power from the grid to start up. So if you have forcibly disconnected those plants from the grid, for failsafe reasons, it’s not a case of just starting them again. There’s a whole process. It takes weeks.

My point here is that the Texas power grid came within about 20 minutes (if I recall correctly) of a complete and total collapse, which would have resulted in the entirety of Texas having no electrical power grid for several months. Can you imagine how much of society would collapse if you had to go without electricity for several months? Everyone in every hospital will die, for starters. Millions of dollars of food will spoil. All digital commerce grinds to a halt. I don’t think it’s hyperbolic to worry that that scenario results in a million deaths. I don’t think most Americans even realize that this is a possibility.

---

My final example motivated this post. One of my favourite Youtube channels, Smarter Every Day, put out a video the other day. This video catalogues his four year long experiment with bringing a useful product to market that was entirely made in America.

Turns out, you can’t. It’s impossible. Not impossible like, “well cheap shit from China will outcompete you and you’ll go bankrupt”, although that is also in play. But impossible, as in, “we literally do not know how to do that anymore”.

This video is worth watching in its entirety, but to highlight the relevant part, at one point he’s talking about how they started prototyping with 3D printers. 3D printers are pretty cool for prototyping, but they do not scale up to proper manufacturing. They take too long, and their build quality is just not as good. So, once he prototyped the correct forms of his parts, he wanted to injection-mold them. This is what it sounds like, you make a mold out of metal, and you inject melted plastic into it, and the plastic hardens into the shape of the mold.

As it turns out, America can still injection-mold stuff. But America can no longer manufacture the molds themselves. Destin got lucky enough to find some ancient greybeard with decades of experience in making injection molds, and that guy helped him manufacture the metal dies used in the molding. But, halfway through this four year project, that guy died of old age. And with his death, the only guy in the United States that Destin could find who knew how to do this, disappeared. Any knowledge he hadn’t written down is gone forever.¹¹

This is what I love about Songs of Syx’s technology system. Technology isn’t something you just achieve. Technology is actively maintained, and it goes away if you stop. By the end of the game, you’ll have 1/3rd of your workforce working in labs and libraries, just because you need that many pops doing that, to maintain your existing tech level. Any interruption to that rapidly cascades into a systemic failure.

Destin’s video highlights one particular way in which this is playing out in our current society, right now. Why is everything made in China? Not because its cheaper, that’s a lie; it’s cheaper than making it in America, but it’s quite a bit more expensive than making it in other countries. Everything is made in China, because China is the only place in the world who bothered to maintain the technological and industrial capacity to actually make the tools themselves that we need to make stuff. Americans can still injection-mold whatever they want, as long as they send the CAD files of the molds themselves to some Chinese factories,¹² who sell them the tools they need to make the thing they want. Americans can no longer make those tools.

What is the solution to this? At scale, I have no idea; I can point to specific things making it worse (eg the tax code), but removing causes is not the same thing as repairing the effects, and removing causes doesn’t stop new causes from appearing, either. Honestly, the long term solution, at the ‘human race’ level, is probably “Hegemony ended with America; China is my new best Hegemon”.¹³ At this point, I’m not even sure that would be a bad thing, especially given that I’m going to be deported to a Chinese satrap in two months. Hope y’all have fun being on the losing side of World War 3; I’ll be in the founding nation of the Greater Chinese North American Co-Prosperity Sphere. They’ll probably run all the Indians out of Canada, too.

1. The people in their discord got mad at me for comparing it to Dwarf Fortress and I still don’t know why

2. In fairness, the game is still in early access and they’ve added a bunch to it recently.

3. I have generally not bothered with such trifles as city walls, because most of the time when I play, I find it easier to intercept raiders before they make it to my village, or just pay them off. Walls take up space that could be allocated to more economically productive uses, anyway!

4. “pops” is a general term used in simulation games to refer to people, it’s short for “population”

5. Just like in America

6. Even if people don’t starve, a major factor in happiness is “food-days”, as in, how many days’ worth of food you have stockpiled. You want to always have at least 20 food-days in storage. If you have 20 food-days of storage for a population of 50, and suddenly an immigration influx grows your population to 100, even if you have plenty of food, you just cut your food-days in half, and get a major happiness penalty for it.

7. 1pp is ‘one percentage point’. Because percentages are proportions, if I were to say “increases by 1% per day”, that would mean that you’re multiplying the current value by 1.01 every day. If I want to communicate that I’m adding 0.01 to the current value every day, I say “percentage point”

8. The mathematically astute will notice that this comes out to four halves. Yes, yes it does, emphasizing one of the core lessons of Songs of Syx: you will NEVER have enough labourers.

Unless you start taking slaves. Which you will. Because you’ll have to. It also makes your own guys happier to enslave other races.

9. Well, not entirely. The buildings and infrastructure I built, remain. With no employees, I can deactivate them, which will prevent them from consuming resources on maintenance, and when the population comes back, those houses and factories are ready and waitinf for them.

10. It’s only funny because he didn’t die

11. And it’s important to emphasize that most of his knowledge is implicit knowledge, things we in tech would call “Tribal Knowledge” until some racists entered HR and assumed that ‘tribal’ meant ‘native American’. You can write down the process to build injection molds, but you can’t write down the million tiny lessons you’ve learned through trial and error about what works and what doesn’t, because you don’t even know what those lessons are. You know them intuitively from experience, you can’t describe that to someone else. You can show them, and they can learn through their own experience, but it’s a fool’s errand to try to document that stuff.

12. Who then immediately rip off the design and sell counterfeit goods on Amazon cheaper than your goods, destroying your business.

13. 🎵He-ge-mon, Hegital monsters, hegemon are the CHAMPIONS!