My understanding of the article is that complex systems don't collapse faster, they simply need a lot more time to be built and collapse faster than that
But in the end Roman Empire did not collapse very fast, it kept existing in a different form (the eastern Roman Empire) for an additional thousand years and did not fail after all.
Roman Empire heritage is huge and spans from culture to engineering, from military tactics to religious beliefs, from political systems to law systems, even languages were heavily influenced by Latin, especially in continental Europe, and the process is still going on.
So probably it's not really that complex systems collapse faster, but that they are able to undergo to radical changes without disappearing from history.
They have the ability to reshape themselves and somewhat survive even if their original form does not.
A simpler system would not be able to do that but OTOH it would be much simpler to rebuild it from scratch or reboot/replicate it.
The best example of a complex system is the human body. It's useful to juxtapose such a complex system to a "complicated machine" such as a car as far as you they breakdown (i.e. collapse).
A single part failure can easily cause a car to become completely undrivable, whereas a surprisingly amount can go wrong with human body and it works more or less the same.
Not enough fuel and a car stops. It's completely binary, either you enough gas to go, or you can't go. For a human there is an incredible range of failure modes for not having enough fuel. Humans can survive an absurdly long time when they are 'empty' of food.
However once that human body does fail, it's over. Additionally all parts of it collapse together.
A car is more or less the sum of it's parts. You can take each individual part and take it off and reuse it, like wise each failing impacts primarily it self. The engine can go find with a flat tire, you can use the headlights on car with no gas and no wheels until the batteries fall out. When an essential component for driving the car fails, all of the other components are still useful. This also means that any piece that is necessarily for the car to drive causes complete failure for the system when it fails. But it also means you can restore the system trivially by repairing a single part.
The human body is more than the sum of it's parts. You can't trivially remove or replace parts. The upside is it is wildly resistant to failure. You can lose an eye and still see, you can lose huge parts of the brain and still function, you can damage a leg running a marathon and still find a way to finish rather quickly, the entire system can be under attack by an invader and automatically defend it self.
But there are limit and when they are cross the entire system fails completely and irreversibly. And in this sense they do collapse faster because once that limit is crossed the system rapidly starts to fail and can never be restored.
> Not enough fuel and a car stops. It's completely binary, either you enough gas to go, or you can't go. For a human there is an incredible range of failure modes for not having enough fuel. Humans can survive an absurdly long time when they are 'empty' of food.
You're highly selective. Let's switch things around a bit:
Not enough oxygen and a human stops. It's completely binary. But remove the oil from a car, and it can survive an absurdly long time.
More seriously, you can in sequence replace all parts of a car and it still functions as the original one (https://en.wikipedia.org/wiki/Ship_of_Theseus), but you can't do that with a human (even if cells do that in a way). This means you can have a car last 1000 years, but not a biological human.
> But remove the oil from a car, and it can survive an absurdly long time.
I wouldn't call less than 30 minutes[1] "an absurdly long amount of time". I've lost a car due to all the oil leaking before. I don't even think it lasted ten minutes after the low oil light popped up (I was in the middle of an expressway), and that engine was toast afterwards. I had it pop up another time for another car and thankfully I was two minutes away from an exit to an oasis and got oil into it pretty quickly.
If you mean an EV car then that's not fair, it's not designed to use oil, that's like saying a human can survive an absurdly long time without eating dirt.
> I don't even think it lasted ten minutes after the low oil light popped up (I was in the middle of an expressway), and that engine was toast afterwards.
Pretty sure they were referring to storing a car for a long time (months or longer) without any oil in it. Not trying to operate the ICE car without oil in it.
> More seriously, you can in sequence replace all parts of a car and it still functions as the original one (https://en.wikipedia.org/wiki/Ship_of_Theseus), but you can't do that with a human
Actually there’s enough evidence at this point that you can, you just have to do it on the cellular level, not tissue level, and the hardest problem for therapies in living humans is not creating cancer at the same time.
Partial reprogramming (resetting methylation status on the DNA) looks like the most important maintenance to do inside the human cells, but there are other known problems with partly known solutions.
My point was that complexity is not a single face phenomenon and can't account alone for risk of failure.
I believe we should look at why complexity is there and what purpose it serves.
Human body is a complex machine, but the fact that failure can bring it down to the extreme is because human body is fragile and once single organs start failing things cascades to the point of no return very quickly.
We are in fact not build for extreme resiliency, but for extreme adaptability (not even the most extreme nature created)
A simple system most of the times is built with simplicity in mind (sorry for the tautology) and sometimes because of simplicity is more efficient.
It can also happen to more complex systems, like for example our body which is very energy efficient at the expenses of resiliency. Klingons OTOH have two livers, an eight-chambered heart, and two stomachs. They are bigger, consume more energy and need to eat (and drink) a lot more. Redundancy adds complexity, but have its purpose.
Klingons do not exists obviously, but nuclear factories are another example of complexity serving safety, not more efficient operations.
Simpler systems usually exhibit single point of failures, like for example now with the war our very complex supply chain can shield us better from the consequences than countries that don't have them or can't afford them.
Historically they died sooner and we haven't records of their sudden fall, because they never reached the point were it mattered enough.
So complexity - I would call it complex redundancy -, which is very costly, depends a lot on the ability of gathering the resources.
Going back to the Romans, at one point they stopped making new steel and warships because the huge amount of wood necessary was not sustainable and Europe witnessed the first massive deforestation of its history.
So, before collapsing, they had to add another layer: recycling. Which can be simpler as a process but also requires a longer chain of supply.
Add to that the will of their enemies to conquer them, the lost knowledge on how to reboot failing systems because they were so old that people took them for granted and things can go south pretty rapidly, but that's not an inherent property of complexity, but of fragility.
The Universe is immensely complex, but I believe it's still going strong after 13 billions years from its birth.
I agree with you completely, but the last statement kinda irks me. The universe can't 'fail'. No event that happens in the universe is a failure to the system, it's unconscious and doesn't really have a purpose that we know of.
If you believe in God then the universe is a creation of God and we are creations of God which separately adapt and create things of their own. Such as offspring. And the purpose is for God's glory.
There are a ton of systems built into living organisms that handle regulation and repair. These systems make us incredibly robust to many different types of insult. We don't build our systems like that, unfortunately humans are incredibly stupid and short sighted so we need to work harder to build robust systems. That's on top of the fact that nature has had such an incredibly long time to build these regulatory and repair systems
Yeah, the author sort-of deforms actual historical data to his own needs.
The Roman empire in the West was rather slow to collapse, even the fact that it survived the crisis of the third century [0] speaks more to its resilience than fragility. As late as 460, Western Roman commanders were able to subdue hostile barbarians and reattach their territories to the Empire proper [1].
The Roman empire in the East, as you notice, survived for another thousand years.
If the author wanted an example of an empire that collapsed really fast, it would be the USSR. That was indeed rather fast. In 1985, Moscow controlled not just the USSR itself, but several important European satellite states east of the Elbe, plus it held a lot of sway in the developing world. Six years later, the empire was gone. Not even Western Kremlinologists expected such a fast unraveling of the Soviet system.
Easy to explain. The USSR was a simplex system, the Roman empire a complex system.
he mostly explained large vs small systems, and didn't really qualify complex systems, until the very end.
a better explanation would be military vs democratic systems, the military being the simplex one, with straight hierarchies. the democracy with various complex interlinks, feedback loops and control mechanisms. Military is like a company, need to grow fast, and dies fast. commands need to spread fast.
I am not a history expert but it is interesting to hear that re: Rome, given the author's background
> Ugo Bardi is professor of physical chemistry at the University of Florence, Italy. He is a full member of the Club of Rome, an international organization dedicated to promoting a clean and prosperous world for all humankind, and the author, among other books, of The Seneca Effect (2017), Before the Collapse (2019), and The Empty Sea (2021).
The Club of Rome is not composed of specialists in Rome history though. It is a "club" that held its organizational meeting in Rome, in 1968 - hence the name.
Intuitively it makes sense to me that more complex systems collapse faster because potentially they have more points and modes of failure.
For instance, due to advances in technology, many aspects of our current modern global society depend on the availability of Cobalt as a component in battery tech. This is a relatively rare element, with concentrated extraction.
Lots of things in modern society depend on having small, cheap, powerful batteries. Lots of systems are built upon systems which have that dependency. This creates a single point of failure which can have reverberating effects throughout the whole system.
I suppose the more complex and interconnected a system is, the more likely it is that you have many of these weak points floating around.
Complex systems indeed have a lot of failure modes, but what you'll find in any complex system which has survived any appreciable amount of time is that it'll be very tolerant of them: one maxim of system design is a complex system will always be operating in at least one failure mode: even more complex systems will usually have multiple going on at once. Systems can and do deal with this without complete collapse all the time: complete collapse usually happens only after enough failures build up over time such that it cannot compensate.
That said, this doesn't stop someone from designing a complex system which cannot tolerate failure: it's just it'll tend to fall apart as soon as they start to put it together (and they'll either learn quickly how to make it tolerate failure or it'll never get off the ground), it won't generally run fine for a long time and then implode suddenly.
Complex systems tend to have complex control systems and feedback loops that compensate for perturbations and try and maintain homeostasis.
But when these feedback loops break, for whatever reason-- reaching the limits of the elasticity and controls, for instance-- they tend to abruptly de-compensate and fall apart.
This looks sudden: more and more slack is being taken up, until the actual compensatory mechanism is exhausted, and then smack, it's all over.
It's like when you get hypothermia, and then vasoconstriction to maintain core temperature stops... and you start shedding lots of heat through your skin, and shivering ends... then you feel hot and want to take off your clothes. Your body is doing many different kinds of things to try and fight getting cold, and once one falls apart "because it's too cold" -- the rest fall like dominoes.
yes and no. complex systems are not designed to be complex from the start. they are in fact successful simple systems that have evolved. Any complex system that is around for a while will have build in redundancy and, as a matter of fact, will run in degraded mode.
So I don't buy into the complex system collapse faster idea. I would say that if you were to look at a simple system and and a [working] complex system the simple system is going to collapse faster in case of a failure, while you may not notice failures as a complex system works around them). What the author here observes is the catastrophic collapse in the late stage of the system where something leads to the almost simultaneous collapse of multiple subsystems.
You can have a complicated system that spends its complexity "budget" on redundancy and/or protection mechanisms. This means that when A fails, B keeps things working, and when B fails, C does. And so the system just keeps going, with B failed, and F, and K, and Q and X and Z. And maybe nobody (or very few people) notice that there's all these failing subsystems adding up.
And then A fails, but hey, it's still running great!
And then C fails. And the system collapses, because A, B, and C all failed. And everybody thinks that it collapsed quickly, because nobody thinks of the collapse as starting when B failed.
TL;DR: A complex redundant system can run for a long time in a partially-failed state. If you measure only from the start of full failure, you can miss how long the collapse took.
It seems to me like this relates to how much emphasis there is on competition monopolizing the entire market and eliminating less efficient solutions. In the 1980s it seemed like every quirky solution would kind of survive, maybe being too bulky and too expensive (to fabricate) but not licensing X, Y or Z for an exorbitant price.. That greatly added complexity but meant redundancies.
In the current form, I feel like the highest efficiency solution maker (or maybe the one in second place) is usually trying to do extremely low licensing with the idea of winning the whole market. That's great in terms of efficiency and actually lowers complexity but means monoculture with exactly identical dependencies.
>need a lot more time to be built and collapse faster than that
I think this is a very good judgement.
It's a lot easier to deplete resources than to allow them to retain their value, or with even more difficulty achieve growth..
Sometimes things really fly off the shelf (viral) but mostly it takes a lot of work with a focused mindset for growth to be initiated to begin with and continue to take place for any length of time.
For a large multi-generational system like that the tremendous strength required to get it flying and keep it going will often have to result in some remarkable upward momentum.
The initiative required can sometimes be lost over one or more generations and the average person may not notice as the upward momentum alone prevents things from faltering immediately.
Until it's too late because the collective talent's focus needed to be retarding collapse as soon as the elusive initiative was lost, but that event did not register against the macro upward momentum still remaining.
So the whole thing comes crashing down while presided over by those who seemed OK at stewarding upward trends, while they were actually not capable of continuing the mindset or making equivalent contributions over one or more generations, and completely out of their element when downward pressure arises that would otherwise be overcome.
Agreed. When I saw Pisa Cathedral from 11th century in person it made me think ... There was no dramatic collapse in 5th century. Yes, a lot of upheaval, but I guess a lot of engineering knowledge from the Roman Empire period was preserved which in turn means that complex society in Italy remained quite sophisticated through the so called "dark ages".
We still don't know what actually happened, but the scholars studying it are discovering piece by piece that it was a "perfect storm" and no single event can be assumed as the root cause of the collapse.
The most interesting part of it, for me, is that it was a cascade failure caused by "globalization" of the times.
Maybe if we really understand what happened it can help us prevent a collapse of our current society (assuming it's not already too late)
Wow, this takes me back to 8th grade science fair in probably 1982 or 1983. I was casting about for a science project, and came across an article in Creative Computing magazine about this model, and the Limits of Growth and the Club of Rome, and I was like, "yeah, I'll make a population simulation on my Ti/994a in Extended BASIC and that'll be my science project." And so I did. I remember my model predicted population to fluctuate in a kind of sine wave pattern with some major crash in the early 2000's. I remember explaining the concept of GIGO (garbage in, garbage out) to the judges, and suspecting that my program was suffering a bit from GIGO. And I won in my category and went on to the state level science fair in Jackson, Mississippi, where I did win some scholarship from the Army (well, I'm not built for the army) but got beat out (justifiably) by this kid who had built a TRS-80 controlled robot. I remember being very impressed by his robot, of course, but maybe even moreso by the fact that all the silver finish was worn off his keyboard from where he'd been typing so much. It was also on that trip, during a visit to the university that I saw WordStar for the first time, and I was impressed by the automatic line breaking moving words to the next line, and I also saw some implementation of Lunar Lander on a vector graphics display for the first time.
I confess I haven't read the article yet but it gave me the feeling of being related to limits to growth.
It intrigues me that growth is one of the goals of sustainable development (the 8Th), when we have a group of brilliant MTI scientists warning of its limits since the 1970s.
At a time when all we hear about is hyper growth, I hope the work of Jay Forester and Donella Meadows is revisited by new engineers. It's time to question the goal of the system.
I'll be honest, I find most of the stuff from that group to be a bit vapid and typically doesn't reflect deep familiarity with the background literature. As an example, the author offhandedly proposes a just-so story about why Rome collapsed. If we go to our trusty 210 reasons for the decline of the Roman Empire [1] (compiled by a historian frustrated that everyone and their pet had different, often contradictory explanations), we see that "depletion of mineral resources" is #53. So why #53 as opposed to #191, or #67, or #3? Similarly, there are entire fields of studies about how complex systems evolve created since Forrester's work, yet the headline here is all about how complex systems aren't resilient (which flies in the face of most observation and literature).
we see that "depletion of mineral resources" is #53. So why #53 as opposed to #191, or #67, or #3?
Most causes in the list are just idiotic, so it doesn't matter. There are a few that are more convincing than "sexuality" TBH.
Anyway, yes, the article and everything around it is just weak, malthusianism all over again. The first wtf for me was the "every civilization has fallen, why not ours". Exactly, why not? Don't they see any difference? And did all previous civilization fall for the same reasons?
> we see that "depletion of mineral resources" is #53. So why #53 as opposed to #191, or #67, or #3?
I think the list is on alphabetic order.
Regardless of the reasons above the real problem is that we are putting all the eggs on one basket. When we listen to Elon Musk saying we need to be an interplanetary society we think "the guy is right", but we have difficulties on accepting that having a single global economy is dangerous.
Resilience is the new buzz word but anti-fragility is what we should strive for. Nassin Tabel give Switzerland as an example. Again you see the pattern, instead of one big and fragile system you have 26 cantons with some degree of independence.
Queue theory suggests the more nodes you add to handle your queue then the longer your queue can survive (facing rising traffic) but the faster it will collapse when it finally runs out of capacity. Zach Tellman went over this in one of his talks and he had a nice graphic to demonstrate what this looks like:
Attempted to find critical counter points to the “seneca curve” — but was unable to find any via Google, Wikipedia, etc.
As is, worth noting that human’s analysis of complex systems is very limited and likely will never realistically be of any truly significant state prior to the collapse of humanity; no formal proof of this myself, but to me, it is clear relatively speaking humanity’s cognitive capacity, observations of universe large, small, over time, etc — are extremely finite.
While it’s possible I have misunderstood the claims made by the seneca curve, the core issues I take are that:
— most man made complex systems likely do follow the seneca curve, though in my opinion, so do most man made systems, not just complex ones.
— many organic systems though do not follow this pattern. For example, the human body reaches peak complexity, that is full development, early in the average life span, then slowly decays and is very resilient to failures within its system.
Guess not having read the original research, to me the seneca curve feels like both literal & semantic cherry picking.
________
As it relates to the narrow topic of civilizations covered by the article. Yes, humanity has created & labeled various civilizations, but if an alien race was observing humanity, would they really see any meaningful use to these labels in understanding humanity? If not, I would argue neither should humanity and that the true concern should be the collapse of humanity, Earth as we know it, etc.
Isn’t Rome itself a counter example? It peaked in expansion and size in 117AD but didn’t really collapse until hundreds of years later (or a thousand of you consider Constantinople through Istanbul).
My recent example is Novell Networks that peaked in like 1995 but didn’t die until 2014. It took almost two decades to actually collapse.
You have noticed the problem with humans. They don't like modeling decay. So decay is simply ignored and assumed away. Meanwhile biological systems model decay. There is no obsession with growing forever. People's height is limited by nutrition and genetics, nothing else really.
If humans actively modelled decay in their society, their civilizations would last as long as ancient Egypt.
Ancient Egypt was conquered by Romans aka outside forces. Meanwhile the Roman empire collapsed from within.
> Perhaps the first person who reasoned in scientific terms about how to avoid collapses was the American scientist Jay Forrester (1918-2016). He was one of the main developers of the field known today as “system science.” To him we owe the idea that when people try to avoid collapse, they usually take actions that worsen the situation. Forrester described this tendency as “pulling the levers in the wrong direction.”
This is an interesting insight but I wonder if it is sort of an observational artifact. A complex social system that has developed formidable power, like the Roman Empire, probably has the ability to pull itself out of most tailspins if its superior resources are directed in a retrospectively-optimal way. So whenever you see a collapse of such a formidably powerful system, you'll see ways in which it could have averted the collapse, and attribute its decline to not doing those things.
Retrospectively optimal control isn't possible in general, so this insight might not be as useful as it sounds.
That appears to be the point of the comparison, no? If you are operating a machine so complex and unpredictable you don't truly know how it works (a Roman Empire, in this case), you'll never know which way to pull the lever until you have already done so.
Just speculation, but I think complexity provides decisive advantages, at least initially, and that's why we see complex societies dominate. Human groups that can avert catastrophe by adding complexity will do so, and those who do not may be unknown to history.
If civilization survival were as simple as KISS, we probably wouldn't struggle with it as much. We need smart, strategic complexity, and humans are limited in their strategic smarts.
Also short term beneficial choices may be long term bad choices.
The industrial revolution was great for humanity for well over a century. The standard of living in industrial countries advanced farther and quicker than ever before. But that same industrial activity had long term effects that we (as a society) are only just starting to accept and are still in disagreement about how to handle.
We have a cognitive bias that undervalues “negative design” — removing things from a design — and so our social structures ratchet towards more complexity until they break.
We encounter that in software or mechanical design as well.
This. But I'd also suggest couching it in different psychological
terms.
We have a "cognitive bias" (fear) against relinquishing control.
Therefore our social structures ratchet towards more bureaucracy and
rigidity until they snap. A skidding car, careening this way and that,
needs the driver to stop over-steering. Alternately pulling the wheel
the wrong way in a desperate frenzy to take control is the problem.
Many things just work out when you take your hands off the levers.
But we also have a political and commercial class who are way too
proud to do so. That would seem to invalidate their "scientific
management" ideology [1]. The Wizard of Oz would be exposed as a
fraud.
I think what's going on in the digital realm shows we've already
entered that phase of overcompensation, because so many technologies
simply go against common-sense survival instinct, and we seem stuck in
cycles of solutionism, adding ever more layers of sticking plasters to
fix the stuff we broke last week. We know we are building dangerously
brittle and irresilient systems such as a "cashless society". It is
this fetish for control that contains the seeds of breakdown.
[1] pretty much the thesis of Adam Curtis's "All Watched Over...",
"The Trap" and "Hypernormalisation"
We don't like negative interest rates because of loss aversion. We would rather be fooled into believing that we have something that doesn't decay, rust or rot and then be surprised that it is gone at some point instead of seeing and being fully aware of much we are losing over time.
For example, suburban sprawl tries to cover up the negative yield of suburban roads, at some point growth is no longer possible, suddenly, the negative yield of underutilized and overbuilt infrastructure rears its ugly head. The local government is now bankrupt.
It is also the same with capitalism. We need endless growth of the population and the economy. To paper over the fact that each individual ages and their body decays over time, the same applies to a lot of physical capital which rusts, rots or decays in another way. The moment growth stops you will have an aging population, your short lived physical capital will disappear.
The ancient Egyptians understood decay, that is why they tried to make their civilization immortal, that is why they invested in extremely long lived capital, capital that survived four thousand years. The pyramids.
The article seems to argue that the capacity to do optimal control was there but the gold standard resulted in hoarding of currency and therefore required deficit spending by the state which was driven by the state. At some point the government could no longer dig out more gold to paper over the deficiencies of a gold standard and was forced to adopt suboptimal control instead. There are common stories about soldiers simply switching over to those with money in the Roman empire.
I don't think it's just pulling the levers the wrong way. The problem is that the same attributes that made a system successful initially are the ones that bring it down. One example might be the ancient Maya: they had a king who was also the head of their religion (and an athlete, and a musician.. he was a big celebrity). That worked great for a while because it helped to bring people together to have a leader like that. Anyways the Maya were having issues with deforestation, and trade routes moving, and so forth. Now the king was supposed to be talking to the gods to avoid this sort of thing. If he was to say, "we need to go back to more sustainable agriculture, and deal intelligently with having less wealth for a while" his subjects would probably kill him. He's not supposed to tell them to make sacrifices, he's supposed to be perfect and he's supposed to arrange things with the gods so they don't have to make compromises. So he has to say "the gods are angry, they demand temples and sacrifice" and so the Maya went on a construction spree while also launching costly wars against their neighbors. This did not improve the situation.
Take a look at capitalism. It's predicated on competition in which the company that can expand the fastest, grow the fastest, wins and eats the other companies. And profits must increase every year or else capital just packs up and goes somewhere else. It's great if you want development at maximum speed. But not so great if you grow so much you threaten the earth you live on. But, our whole society is structurally unable to do anything else than this. Faced with economic problems, the only answer our system can give is to double down, do more capitalism faster and more aggressively. The bigger the problems, the more we double down on what worked before. Not the best idea. But since power and wealth in our society comes from holding capital, and since holding capital only brings you that power and wealth if it can move and grow- that means to do anything else than double down would require all the people to hold power and wealth in our society to lose that power and wealth. They'll see any threat to their power as a more immediate problem than global warming, which is slow moving and will affect everyone else first before they suffer at all.
> And profits must increase every year or else capital just packs up and goes somewhere else
Close the loopholes. Implement negative interest on cash. Break up monopolies. Tax land value. If there are no god like safe havens of capital that no actually productive investment can win against, the pressure to do the impossible will greatly diminish.
Netflix became popular around 2004-2005, but it took another 5 years after that for blockbuster to fail, and it was sudden.
A similar trend was observed with Myspace and Facebook. Myspace was doing well all the way up to early 2008, 4 years after the creation of Facebook, which was already becoming quite popular, but then in late 2008 Myspace suddenly became worthless.
A similar 4 year lag time is observed with the iPhone and the sudden decline of Research and Motion.
In both of those cases a switch of sorts was suddenly flipped.
At least in the case of Blockbuster, I'd say the lag can be accounted for by the fact Netflix got into streaming in 2007. Prior to that Netflix started eating Blockbuster's lunch, but the service really took off once they managed to mainstream streaming. This was helped, in no small part, by the rapid rise in broadband penetration of that same time period.
The key thing is the same as in the coronavirus pandemic, it's exponential growth - think of Facebook as the virus, Myspace as the immune system and word-of-mouth being the infection way.
The problem with all the cases you mentioned (and the current fight of Facebook/Meta vs Tiktok) is that the established entity has become entrenched in its ways and too unflexible to adapt, so competitors exploit the weakness and eventually take over.
a lot of times the trigger is usually some kind of massive payment that needs to be made, but then all of a sudden whatever was enabling these payments to happen before is no longer happening and the company goes into default.
Everyone seems to be focusing on the crazy assertion that the Roman Empire collapsed quickly. But what stood out to me was citing Blockbuster as a complex system. They didn't die because an interconnected web of dependencies started to rot, they died because their entire reason for existing disappeared.
It's an interesting topic but the examples included are really weak.
> they died because their entire reason for existing disappeared.
That's only true if you characterize Blockbusters reason to exist as "distributing videotapes/dvd via physical stores" rather than just "distributing media".
Netflix also began in the physical media space but was able to pivot to streaming, helped along by their pre-existing remote business model.
Blockbuster tried to get in on the streaming game but it was a behemoth that couldn't swiftly disentangle itself from its old business model, helped along by several bad bets. So, it limped along for years past it being needed until it disappeared practically overnight.
You might think that this outcome was inevitable but that just means you've already taken it as a given that hulking, complex enterprises can't shift their main operating principles on a dime given a massive change in conditions.
Very true! I'd be curious to learn why they failed to make the shift to digital. I imagine having to support hundreds of storefronts was a huge distraction.
I don't think this article is very convincing, as far as the complexity of a system having anything to do with its potential speed of collapse. The article's first example (the Romans) collapsed due to resource shortage (mining becoming harder). Their society WAS complex, but that isn't why it collapsed.
Human grown societies tend to under-invest in redundancy, for the sake of more growth, so our societies end up rather brittle. But their complexity is not itself to blame, in my view.
The Roman Empire example that the article goes into is more interesting than it makes out, and can be an example of the wrong sort of redundancy being applied to a problem.
The reforms in the 3rd century were not just an increase in taxes, but a restructuring of society to keep the army strong. Price controls were placed on goods to make sure that people and the state would always have reliable access to goods. Tax collection on private property was formalized and made much higher. Peasants were bound to the land to make sure they always were able to produce. This makes sense based on the view the Emperor had of the empire: an empire is like an army, so this strict control should boost resilience and redundancy.
The tax collection was incredibly onerous and led to a lot of small estates going under. The price controls didn't work at all, in part because it didn't understand that price differences can be caused by things other than greed. The remaining landowners increasingly pulled out of the money economy and the cities and instead focused more on their own estates and the newly bound peasants. It is often stated as the starting point of feudalism, and ultimately did more to weaken the empire's structure than sustain it.
When you ask for redundancy, it's important to think about what kind of redundancy you want, and who it serves.
"When you ask for redundancy, it's important to think about what kind of redundancy you want, and who it serves."
Well, of course that's the case. Though I must say, none of the things you listed sound like redundancy to me. They sound like an illustration of the fact that a complex system has a lot of ways to fail, and the result of a certain action can be hard to predict.
The article doesn't really drive at this though - it seems focused on irresponsible use of resources. TBH the article's whole thesis strikes me as a bit weird.
Also, I bet any serial entrepreneur can explain why we tend to invest in growth over redundancy: we don't even know the names of those who didn't, because they died in the cradle.
Exactly this. In any sort of survival-of-the-fittest environment, a redundant system can always be outcompeted by a highly optimised, brittle one, that dodges a few bullets.
Biology is one of the most ancient and cutthroat examples of "survival of the fittest" imaginable. Yet it's filled with highly redundant systems outcompeting better optimized and brittle systems. Cells spend huge amounts of resources on redundant DNA, transcription error checking, and redundant organelles. Animals have redundant organs, complicated immune systems, hugely expensive neural systems, and so on.
Human designed systems tend to be less redundant and comparatively fragile by contrast.
Very true! Though biological systems are also vulnerable to resources running out. Also, biological systems are frequently stupendously complex, as well as being robust (contrary to the article's claim).
At a guess, I'd think the difference is to do with the shape of the fitness landscape. If things are failing, or getting attacked and disabled all the time, robustness will result. Whereas a rare but serious failure is much less likely to be manageable.
Then again, perhaps evolution / god is just a better designer than us!
I imagine this can even be made mathematically precise. You need some minimal amount of robustness depending on the number of bullets you need to dodge, but conditional on that, someone who is just robust enough to be lucky will win, and if the population is large and diverse enough, a lucky, minimally robust, growth optimized individual will survive and win.
Nice! So excess robustness will cause failure through inefficiency (with "excess" being a situation dependent variable) - but you need some robustness. Evolution of course, optimizes for this, hence biological systems' wonderful balance.
I do think this is the part of the equation that is left out by the "precautionary principle" types.
Seems like any highly complex system can have elements that are chaotic- in the sense that they are very subject to small changes that can lead to a threshold. You may be moving a number of levers and all is going well. Then you make a change or fail to adapt to a change that in the past was not of consequence - but now it is. Only in this case you quickly find yourself at a threshold you did not expect and no amount of effort can keep you from it. If you had full understood the situation you could have easily prevented the unraveling but you did not have compete knowledge due to the complexity.
Junk writing with cherrypicked examples. As mentioned by others, the collapse of the Roman empire was not slow. Blockbuster may have fallen quickly, but it wasn't any more complex than other companies of our time that have risen quickly or fallen slowly.
Complexity can be fragile or it can be stable. A system with redundancies and ways to adapt can be quite complex and yet stable. And many, many simple systems collapse quickly, for instance most start-ups. It's just not as noticeable as a huge system collapsing quickly.
Geoffrey West has some really good work on this. What happens most of the time is that new alternative technologies and ways come in and change the whole context around. That is, the old civilization really does collapse in a sense but a new and different one takes its place. A good recent example of this is the "green revolution" which for better or worse created a burst of productivity in agriculture that has so far sustained ongoing population growth that previous agricultural traditions could not have supported.
The Santa Fe Institute (of which Geoffrey is a member) is explicitly dedicated to complex systems research. They tend to take some theoretical stances certain fields might consider controversial, but there's a lot of smart people there nonetheless. I'd also recommend Marten Scheffer's work on the stability of complex systems as particularly relevant to this thread.
I think it’s less a matter of complexity and more a matter of dependencies. In the case of our society, we have enormous dependency on centralized systems for energy, heat, water, medical, and more. That is where the real risks exist.
I think you nailed the problem. I once had a car that had a single chip control various subsystems. You couldn’t easily get to the chip because it was buried behind the dash, and you had to remove the entire dash to replace it. Once the chip went bad, the collision detection system stopped working, which is kind of important to the modern driver, since you want to know if you are too close to objects or if something is too close to you. Then, the radio stopped working, which is kind of important if you like to listen to news or get weather updates. Then, the various stereo components stopped working, which is kind of important if you like to listen to music, the CD player, podcasts, or use the phone hands free to talk with people as you are driving. Etc. All because of one bad chip buried in the vehicle dash that nobody could get to in order to replace it.
Why complex social systems collapse faster - the article doesn’t talk about general complexity but the kind found in social structures.
It also suggests a kind of simplistic mathematical analysis which doesn’t sound super insightful or accurate. Otherwise, Asimov’s psychohistory could already be a thing, if it was possible to mathematically model with accuracy the behaviour of societies.
This model of complexity just doesn't make sense to me. So I looked it up on Wikipedia:
> Complexity characterises the behaviour of a system or model whose components interact in multiple ways and follow local rules, leading to nonlinearity, randomness, collective dynamics, hierarchy, and emergence.[1]
Is a star topology wih 50 nodes more complex than a star topology with only 5? To me, there are concerns of size and scaling, but not complexity only given that as a problem. And according to Wikipedia, the difference in leaf nodes wouldn't increase complexity.
But that's exactly what the article is calling complexity, more nodes in a star topology!
And then it calls other things complexity that make even less sense:
> Blockbuster, too, was a network: one formed by the links that the company had with its customers [star topology again, like I pointed out above]. But by the early 2000s, those links started breaking down: The company was losing customers because its business model had become obsolete [and not because the business was too "complex," so let's blame complexity].
Maybe complexity works differently in physical chemistry. Maybe that just doesn't translate well outside of chemistry. I don't know. I just don't get it.
There's no formal theory behind complexity. There's qualitative and scientific studies but I have seen any strict mathematical formalism around it that fully fleshes out that system A is more complex then system B definitively.
I remember reading that currently there is a movement take intuition of of complexity and transform it into a formalism. If you squint it looks like it's possible, however this isn't always true.
For example the term "life" can't really be formalized because the intuition encompasses a set of things that are over complicated and even blurry at times. It's a loaded word that spawns philosophical debates and nobody realizes they're debating about a poorly defined arbitrary concept. It could be that "complexity" is the same thing... personally it doesn't look that way to me, but I'm not sure.
In some models, this picture of slow gain followed by rapid decline is actually the worst possible situation for happiness, with the inverse (rapid gain followed by slow declines giving optimal happiness). See here:
"Already in the time of Seneca, the first century CE, the problem of mine depletion may have been felt. By the third century CE, Roman mining production collapsed. At that point, gold started vanishing from the empire; most of the gold that was in circulation went to China to pay for luxury items, like silk. Eventually, no gold meant no money. No money meant the troops couldn’t be paid."
???
'Gold' is not 'money'. We just use it as such.
We can use anything.
If there is 'less gold' to go around, we can just value it in higher denominations, i.e. deflation.
Of course that can have practical problems, and an expanding economy, you generally want more 'currency' ...
... but you can just as easily say '1 ounce of Gold' is '1 plot of land' ... and then make it worth more.
Or use salt. Or silver. Or whatever.
And oddly, the Roman Empire took forever to really come undone.
>If there is 'less gold' to go around, we can just value it in higher denominations, i.e. deflation.
Imagine borrowing $300k for a house denominated in gold and then the price of gold doubles. Now you owe $600k even though you haven't agreed to this amount in the original contract. You have taken something less valuable and now you are returning something more valuable. In any other business context one would expect to get paid for providing this service. The illogical nature of gold demands compensation instead, as the lender could just do nothing with his gold and get the same benefit despite the fact that doing nothing is actively harmful to the economy.
Ancient Egypt simply used grains as money. You deposit the grains at a local grain bank, get a clay tablet in return. They never had a "not enough" money problem. Because borrowing grain and returning it in pristine condition was actually considered the service it is. If the lender does nothing, his grains will spoil in front of his eyes.
The roman empire seems like an especially terrible example of complex systems failing quickly given it existed in a partially collapsed state for longer than most major empires existed period.
Exactly. Even considered on its own it's a really bad example of rapid collapse. Roman civilization lasted for about 1000 years until the collapse of the Western Roman Empire in 476 AD (250 more if you include the pre-republican period). One could argue at length when its decline started, but a reasonable starting point would be ~250 years prior to that, with the great crisis of the third century. That's a quarter (fifth) of its entire existence, and a lot of human generations, who would mainly have noticed significant up- or down-swings during their own lifetimes (with somewhat equal probability). Only the most historically literate would maybe have perceived a general downward tendency over the centuries.
I have yet to see one comparison to the fall of Rome that wasn't complete horse shit - and this is no exception. It's just madness from the beginning - "Rome" began to fall as soon as it was relevant, it lasted for more than a millenium, and if you just pick the right point in time, you can make an argument for almost anything.
Except in this case the author didn't even bother to pick the right time, instead opting for some quote "A few centuries" before the "rapid" descent. Most empires last only a few decades.
Even the fall of the republic to "tyranny" / dictatorship took a pretty long time. You could pick a number of different starting points I guess but I think it'd be hard to argue it wasn't in progress post Sulla, which is like 30 or 40 years before Caesars whole thing. I guess on the scale of history that's pretty short but that's still multiple lifetimes for most people back then.
the Roman Empire being brought up over and over again makes sense in the current context where people like to compare the USA with the Roman Empire. Apart from certain small similarities IMO the comparison is nonsense and does not make sense.
The human body is a very complex system. It's an incredibly complex system with 30 trillion moving parts (cells). And it doesn't stop there. Layer upon layer of abstraction culminating in organs, sub-systems (cardio, nervous, GI, ..), and finally a thinking hairy biped. It operates in a hostile environment subject to ceaseless assault by germs, viruses, and other organisms. Complexity clearly can not be the issue, otherwise evolution would favor simpler organism, but we're the ones lording it over simpler systems.
I think our design thinking is lagging. We need to work on that front.
Scale is another big part of it. Over longer time periods resilience comes to mean something different. How many web sites from the 1990s are still up, accessible, and relevant?
This guy/gal is writing sensationalist propaganda. Spectacular collapses of systems are rare, which is why it's noteworthy when they happen. Rather, systems transform. Even his/her example of the Roman Empire wasn't a true collapse but rather a transformation: most of medieval Europe's institutions were continuation of Roman institutions.
The corollary to this might be that linearly increasing complexity requires a non-linear amount of input energy. That’s consistent with what we’ve seen from human civilization development and evolutionary development.
The observations about complex systems declining faster than they grow was interesting. The rest of the post kind of boils down to "complex systems fail because they're complex".
Complexity can stabilize an unstable system, allowing it to stick around long enough for us to notice how unstable it really was all along when the stabilization eventually fails.
I'm not sure what you mean by "simple systems are easier to dismantle". It seems easy to think of counterexamples if I'm understanding what you mean. Even with technology, the bitcoin blockchain (simple) seems way harder to dismantle than layer 2+ web3 DeFi projects running on smart contracts (complex)
You definitely do not want a simple system if your singular goal is resiliency. The most resilient systems we know about are all complex, from ecosystems to organisms to cities to the internet itself. The main problem is that our current toolkit is so primitive that understanding anything about them, from how to intelligently intervene to how resilient any particular system is at any particular point in time is damn near impossible. There's a lot of research trying to change that, but it's obviously a difficult problem.
Complex systems are always in a state of maintenance. Some part is
always failing, but we can repair it. Systemic failure happens when we
can't keep up with the necessary maintenance. It's very easy to look a
a big complex system, see lots of failure, and conclude the whole
ship is going down, but that may not be the case.
Funny thought: then we may characterize a complex system as a system that's (most of the time) too big to fail all at once, and whose resilience to failure simply arises from continued previous failure. Which sounds like another way of saying "I don't know what I'm doing, but it kinda works"... until it doesn't. Maybe when it has inevitably grown too big to be successfully maintained anymore.
But "failed state" simply means that some of the controls have been disabled, at least according to John Gall. That just translates into working without quite knowing how they work.
Perhaps the solution to preventing complex systems from collapsing is to built in a reset button to rest and restart. Most of us became used to regularly restating our early PCs, which is how we delt with these complex systems collapsing. I speculate the human body does something similar when we sleep, which is an example of a highly complex system
A lot of the relevant researchers basically define complex systems as the systems that you don't or can't understand well enough to simply reset and restart, nor would that be an appropriate intervention strategy most of the time. Should we drain lakes and put in new soil and water every time there's an algal bloom? Similarly, you can't truly reset an economy, or a polity with anything we'd call "reasonable methods".
Invalidating money every n years could work because it forces people to make an arrangement with people in the future. However, it would be simpler to introduce a three month expiry date to cash bills that demands you pay 1% once it expired, resulting in -4% interest per year. People will be glad to lend out money at 0% as they avoid the -4% interest on cash.
I agree with the comments that the article is misdirecting fluff. Cutting through...
All organised systems (anything that we would describe as a system as opposed to random or chaos) require additional energy inputs to maintain organisation over and above those energy inputs required for performing input-output function. Maintenance requires work.
All life on Earth depends on the thermodynamic gradient supplied by the Sun being appropriately steep enough for that additional organisational margin to be provided within tolerances, but not too steep to make everything excessively random, or too shallow to stop stuff moving.
Failing to provide that extra margin of energy for organisation for any part of the functionality of a system results in breakdown and halting.
E.g. Quite a lot of the extra margin of human organisation energy is provided via being omnivores _at the top_ of the food chain. If all the plankton and birds and bees die -> Humans don't get enough food -> Disorganisation -> Dead humans.
It is possible to build forms of organisation that are inherently unstable outside of limited tolerances. Stepping outside of those tolerances exponentially raises the extra costs required to maintain organisation.
Needless to say Nature doesn't keep many such systems around, if any, particularly when under significant stress. They get weeded out fast.
The extra kicker in human economic organisation is money as a mediator for controlling energy allocation.
Organisations retain function if they can afford to support the additional cost of being organised.
Within monetary systems, profits have to remain high enough to pay off loan interest and support organisation.
But...
If people without enough jobs or money starve -> The latent labour pool dies -> No support for growth, no maintenance -> breakdown.
So the labour force loses a margin of value-add generated in order to support organisation.
The hidden labour force that is Nature loses out once human economies/populations grow too fast or too large.
So modern humans do this "supporting organisation" without considering Nature as part of the organisation, and frequently fail to support the labour pool properly.
Obsolete monetary/value system designs are a big problem.
Globally, monetary systems to support sustainable societies should provide for: equilibration against destructive monopolies - especially at the ForEx level; enough liquidity - supplied through limited liability loan failures under capitalism - to meet aggregate interest on loans and organisational costs AND latent labour pool support; maintenance of key functional inputs, like food, sunlight,...; and zero cumulative pollution;
Why shouldn't every viable economic or social unit operate monetarily locally according to Kirchhoff's current law, if money is an adequate proxy for energy?
Why do we not price in the extra energy input Nature requires to remain coherent and still balance supply and demand?
Humans previously kept building inherently unstable civilisation structures with inherently unstable monetary systems that cannot cope with externally imposed constraints arising.
They kept failing to preserve and maintain the inputs sensibly, and they designed monetary systems that are typically destined to create and exarcerbate those failures.
They could collectively: stop breeding like rabbits; design better monetary systems; stop shitting on the environment.
All of it is technologically feasible already.
How much of that is politically feasible right now?
People are holding people back by being people. Wicked problem. Difficult to handle kindly.
But in the end Roman Empire did not collapse very fast, it kept existing in a different form (the eastern Roman Empire) for an additional thousand years and did not fail after all.
Roman Empire heritage is huge and spans from culture to engineering, from military tactics to religious beliefs, from political systems to law systems, even languages were heavily influenced by Latin, especially in continental Europe, and the process is still going on.
So probably it's not really that complex systems collapse faster, but that they are able to undergo to radical changes without disappearing from history.
They have the ability to reshape themselves and somewhat survive even if their original form does not.
A simpler system would not be able to do that but OTOH it would be much simpler to rebuild it from scratch or reboot/replicate it.