For those unaware of the Dark Forest Hypothesis (DFH):
> “Dark forest theory” holds that civilizations fear one another so much that they don’t dare to reveal themselves lest they immediately be considered a potential threat and destroyed.
> In The Dark Forest, a 2008 novel which Discover Magazine calls "a dark answer to the Fermi paradox",[138] Chinese writer Liu Cixin envisions aliens as being paranoid and wanting to kill life because it might be future competition. Other aliens dare not reveal themselves because they might be considered a threat, and they stay quiet as if in a dark forest.[139]
This is not a particularly accurate representation.
It's a game theoretical problem. Because of the huge delay in communication over interstellar distances, you can not adequately communicate with other in diplomatic terms.
due to the speed of light limitation, if you contact another civilization to e.g. propose to establish diplomatic relations, you will only find out if they sent a big ass rocket as a reply in about the same time it would take to receive a message as an answer.
Making certain assumptions about the technology available, like: No instant communication possible, solar system or at least planet destroying weapons possible, etc., then in purely game theoretical terms if a civilization seeks to minimize the chance of being destroyed, it then becomes a good strategy for a civilization to hide and not reveal themselves, and immediately attack and destroy any civilization that should make the mistake to reveal themselves by e.g. sending out a message. thus giving them no chance to harm you before they are destroyed themselves.
it's only a hypothesis, and even at a superficial glance there are alternative strategies that might end up being better on average, much better if each civilization is willing to take some risk, etc.
I appreciated the write-up but i think it runs into one of the issues I've encountered myself messing around with netlogo, where for complex agent-based stuff, you sort of end up baking the result into your modeling choices. For the three civ variables, there are no constraints or trade-offs between them, and all comparisons are monotonic (more power + detection always better, less visibility always better). It's cool to see the visualizations and quantify the progression, but I don't think you need to run the sim to predict the result.
"Powerful civs that are good at detecting others seem to always do well, even if they are themselves visible."
Power is important, but not the most important factor. He states that he needs to collect more data, but within his current dataset there is a clear and visible preference in the surviving population.
The words might be, but the context and approach goes back to at the very least
to The Killing Star by Charles R. Pellegrino and George Zebrowski published in 1995, who wrote:
We ask that you try just one more thought experiment. Imagine yourself taking a stroll through Manhattan, somewhere north of 68th street, deep inside Central Park, late at night. It would be nice to meet someone friendly, but you know that the park is dangerous at night. That's when the monsters come out. There's always a strong undercurrent of drug dealings, muggings, and occasional homicides.
It is not easy to distinguish the good guys from the bad guys. They dress alike, and the weapons are concealed. The only difference is intent, and you can't read minds.
Stay in the dark long enough and you may hear an occasional distance shriek or blunder across a body.
How do you survive the night? The last thing you want to do is shout, "I'm here!" The next to last thing you want to do is reply to someone who shouts, "I'm a friend!"
What you would like to do is find a policeman, or get out of the park. But you don't want to make noise or move towards a light where you might be spotted, and it is difficult to find either a policeman or your way out without making yourself known. Your safest option is to hunker down and wait for daylight, then safely walk out.
There are, of course, a few obvious differences between Central Park and the universe.
There is no policeman.
There is no way out.
And the night never ends.
But of course, even Stephen Hawking was talking about this for decades. I find it bit of hilarious social commentary, that some Chinese author copied it and now people claim he invented it.
IMO Liu steelmaned the argument with respect to game theory, relativistic diplomacy / suspicion chains, tech explosion etc. He elevated it from commentary to hypothesis.
Yeah, it would be interesting I think to add an aggression variable. If two species meet and their combined aggression is high enough, they fight (with some large bonus going to species with some combination of aggression and ability to detect, as it is assumed they can get off a devastating first strike).
If the species are not aggressive, they can presumably at least the weaker of the two should get some large bonus.
Of course then we could play with the first strike bonus and the alliance bonus, and get whatever result we want from the simulation.
There's also a huge damping variable due to communication latency between star systems, and of course travel time if you want two armadas to gang up on a third system. In fact this is exactly the two generals problem: you really don't want to launch an armada without confirmation from your ally, and if you do launch it, you will have the two generals problem recalling it.
The model is flawed and fails to account for the activation of the XCOM project: humanity's first, and last line of defense against the alien threat.
In all seriousness though, if the universe is indeed populous then there's probably been innumerable alien XCOM projects.
I also can't help but think that we usually fail to appreciate the technology gaps involved. Human beings in a span of a few hundred years managed to go from the advent of electricity and industrial society, to artificial intelligence—all in what amounts to not even a blink of a cosmic eye. Yet recursively improving intelligence is conceivable to us.
2001: A Space Odyssey did a wonderful job illustrating such gaps. I think there's a good chance most civilizations (ours included) would not be equipped to even comprehend how to detect or fight an advanced civilization, let alone understand their very nature.
It is probably the greatest moment of the second book. I was giddy reading it, because it feels so logical, so realistic, and our reaction so congruent.
I love Xcom but it very much portraits the absolute opposite of the 3 body problem trilogy.
> Human beings in a span of a few hundred years managed to go from the advent of electricity and industrial society, to artificial intelligence—all in what amounts to not even a blink of a cosmic eye.
Also keep in mind that Humanity is a latecomer to the universe. We weren't the first species to evolve on Earth. First came the dinosaurs, about 230m years ago, and lasted till about 66m years ago, then went extinct. Then came humanity.
But there may be planets in our galaxy or other ones where intelligent life happened to be the first species to evolve on that planet. If so, they have a multi-hundred million year head start on us, evolutionarily, scientifically, technologically.
Yeah, there's no possible XCOM project against that. We just have to hope they're benevolent. (And if they've solved scarcity with energy->matter conversion, as in Star Trek, they may very well be - there's nothing we have that they could need or want)
For me, it’s rather probable that advanced civilizations will collapse or/and kill themselves. It happened with civilizations on earth, it certainly can happen with other civilizations in the universe. Destroying other makes no sense because the risk is much higher than the reward. There is more than plenty of resources out there for any possible civilization.
> There is more than plenty of resources out there for any possible civilization.
It looks that way to us because of the limited rate at which we consume resources relative to the scale of the cosmos. But our rate of consumption isn’t anywhere near an upper bound for a civilization. We’re not even a Type 1 on the Kardashev scale.
I think that the speed of light limitation is insurmountable. Advanced civilizations become insular and stay close in space to themselves to minimize communication latency. They all become addicted to Internets and never go exploring very far, if at all.
The problem is self-replicating technological mechanisms undermine this hypothesis. A civilization which is being efficient, and as such Matroska brain's it's local star would still find value in launching Von Neumann explorer probes, since that's practically zero-cost at that resource-scale but extremely high reward: without reconnaissance you can't know if you should defend yourself or not.
Also even if it takes millions of years, having your autonomous machines cut up the local stellar bodies and ship materials back to you is an obvious move. You're immortal, so who cares how long it takes.
> self-replicating technological mechanisms undermine this hypothesis
The problem is observed reality contradicts this hypothesis, so far there are no observed techno-signatures. Some call this a paradox because they're very attracted to their hypothesis, but it's not a paradox if the explanation for not seeing anything is there being nothing to see. I expect there to be gnomes in my garden, but I don't see any gnomes in my garden. Is that a paradox? Maybe I'm just wrong about gnomes.
As for recon probes, maybe. But we wouldn't see them would we? If a recon probe is looking for threats to the civilization that created it, it wouldn't do to have the recon probe seen. Recon is about seeing while not being seen.
Disassembling other star systems? It's a neat idea, but there's no evidence for it. Maybe it's not such an obvious idea to the sort of civilization that might be capable of it and it only seems obvious to primitive apes who can't do it anyway. Or maybe there are no such capable civilizations in our neighborhood at all.
Maybe if a civilization evolved on a world extremely poor in heavy metals, they'd never get around to fission. Maybe fusion bombs will forever be impractical without a fission kickstart?
Maybe this just reflect poorly on my numerical intuition, but it seems to me the forest may be dark because it's more of very sparse forest? As in, no matter how many trees there are, they are so far apart from each other that it's really more of a steppe. When "nearby" is tens or hundreds of light years, and most things are not nearby, it seems there's plenty of both time and space to make it very believable that there's just little to no overlap between when detectable civilizations and civilizations able to listen coincide.
The problem is exponential growth destroys all. You can't explore the universe very quickly on human timescales...but presuming the trend lines of technological civilization hold up, eventually immortal sufficiently-intelligent Von Neumann probes can rapidly and easily (hundreds of thousands of years, but a blink of the eye geologically) explore the entire galaxy.
Even if it took a few million years to do, that's still far less time then the Earth has been habitable and had life. So the question of course is still where are they?
Civilizations will sort out neatly into expansionist and stable. Stable civilizations will be, generally, much, much older than expansionist neighbors, and so incomprehensibly more advanced.
Thus, each expansionist civilization will spread out to encompass an exponentially increasing number of advanced, stable civilizations, and antagonize an increasing number of them. Soon, they will encounter one that chooses to put them in their place, and maybe just wipe them out.
So we may think of the galaxy as an archipelago of stable civilizations that interact little, with splashes of expansionism in between that encounter and annoy them, and are swiftly contained. Such expansions are exceedingly unlikely to get to us before such an event curtails them.
I think that model only works if the stable civilizations are advanced enough to survive arbitrary threats, otherwise they'll just get wiped out by chance over time and replaced with expansionist ones which are more robust against anything other than being systematically destroyed by something they irritated. ... plus to destroy an expansionist civilization you'd have to have significantly superior expansionist technology.
I.e. how do you destroy an expansionist fleet which has scaled up to the point that it's furthest point from you is increasing close to the speed of light?
But they are, remember, incomprehensibly more advanced than you or me. And they know a bunch of other incomprehensibly more advanced civilizations all the way on the other side of the expansionist noobs. And, just because they don't usually see much point in going out there doesn't mean they can't, or won't, when the time seems right for it.
Not much point in reasoning about magic though, might as well say "God will provide" as it gives similar insight. I think assuming that the laws of physics, as we understand them, and reasoning whats possible and likely within them is a lot more interesting.
I can come up with one respons within the believed laws of physics at how a hypothetical stable civilization could suppress exponential ones even though exponential ones would quickly be expanding their radius at close to C: the stable civilization would need to pre-seed the visible universe with idle hidden outposts of a sufficient density to detect and suppress exponential expanders.
The non-expansionists don't need to fight all possible expansionists. They only need to fight the ones that come around. If B can't beat A, A will soon encounter C, and then D and E, with a fresh roll of the dice each time. The bigger they get, the more adversaries they cross.
We are not predicting the future, but trying to account for an observation. For the scenario to remain on the list of reasons why we haven't been visited, it suffices that it is possible. There are of course lots of other possible reasons, among them that there is nobody to visit.
> The non-expansionists don't need to fight all possible expansionists. They only need to fight the ones that come around.
That come around us since we're trying to explain why we haven't seen an expansionists. So this line of thinking predicts that we're in the shadow of a non-expansionists, I guess?
> Stable civilizations will be, generally, much, much older than expansionist neighbors, and so incomprehensibly more advanced.
This is not necessarily true. A roiling cluster of ephemeral, competing life forms and civilizations may very well develop warfare and defence technology better and faster than stable civilizations. This is especially true if you consider the cost of stability: all new technology is a potential threat to stability that has to be weighed carefully, so I expect stable civilizations to evolve much slower than unstable ones. In the aggregate, instability is not really a problem: a civilization's progress will be scavenged or stolen by its successors, retaining the advantage, and for that reason I would argue the unstable regime will almost systemically overwhelm the stable one.
You can kind of see this with humanity: a lot of the big problems we have to face as a civilization (but not all of them) result from the careless adoption of technology. Climate change would not be a problem if we had industrialized slowly enough. The Internet spread too fast for us to understand its effects on the propagation of (dis)information. A truly stable civilization would have taken a thousand years to do what we did in a hundred. Perhaps it would be happier for it, but if we met them, we would eat them alive.
The stable societies still standing after confronting and defeating numerous upstarts will be the ones that really are more advanced. You could say there will be a survivorship bias in the extant population.
Both conflict and technological advancement are destabilizing influences, though. How do these societies remain stable? My worry here is that any confrontation with an inventive upstart will require swift change and adaptation in order to fend off the threat, but such fast change is likely to cause internal instability. I don't think there is such a thing as a "stable society" that cannot implode and turn into a bunch of expansionist upstarts, unless they are so careful about progress they are practically defenceless.
Stability and not-expanding are not necessarily the same thing. ("Stable" is easier to write.) They could be extremely dynamic, but if their population isn't exceeding their resources, why do they need more room? Expansion is what brings cultures into contact and conflict.
Puny humans are confined to one solar system, and even just the one planet, but are pretty dynamic. Population in wealthy countries is in decline.
Remember we are not just talking about humans with bumpy heads, here. Social universals on Earth do not apply. Expansion might be limited just by a physical inability to reproduce beyond their home planet, without giving up any dynamism.
The premise of the DFH is just so bizzare. Essentially godlike aliens go around killing other aliens for hell of it. It'S fOr SeLf ProTeCtIoN! Please. If you're a godlike alien, you don't actually have to care about anyone else. Hell, not only do you actually have to deal with intergalactic distances, we're talking about a plan that requires at this point quadrillions of individuals to be dedicated to multiple 100,000 year plans of preêmptive genocide against literally unknown civilizations.
I'm sorry. But I'm not supposed to suspend disbelief for serious discussions.
The older I get, the more I find these debates as tedious as medieval demonologists wondering about how many angels can dance on the head of a pin, and who is the fifth viscount of the third circle of Hell, and just what exactly is his bailiwick.
> Essentially godlike aliens go around killing other aliens for hell of it. It'S fOr SeLf ProTeCtIoN! Please.
Isn't the point the point that they aren't godlike? They're glass cannons, capable of creating weapons with incredible power, but incapable of defending themselves against that same class of weapon.
Other civs -> wasps, mosquitos, cockroaches. Not a real existential threat, but enough of a nuisance. Wasps, even though they live in a complex culture from their point of view, have no chance to understand human culture. Not wanting to be accidentally stung sort of has a chance to be understood by both sides, very-very approximately.
Maybe the disconnect is the word "godlike". A truly godlike power wouldn't care, because they're beyond threat. But imagine something more like the cold war: Powers capable of mutual destruction, but not godlike.
In the aftermath of World War II, John von Neumann advocated a preemptive nuclear first strike on the Soviet Union before they could develop their own nuclear weapons. It's monstrous, but it's the same cold logic that the DFH proposes: Destroy others before they have a possibility of destroying you.
I think that's a useful analogy but I would say that it also implies limited DFT.
Von Neumann's plan was based on an incredibly short time scale. It wasn't long before the Soviet Union would develop its own weapons. The power asymmetry was only briefly held.
The thing is at a universal scale that would imply that the only threats we would honestly face are ones in close power level, and we can barely leave the earth at the minute. Another species with more advancement than us would be n-years more advanced than us even after we reach the level of what they were at on first encountering them (unless they decide to share technology which would be the opposite of DFT). Technological stagnation could cause a catch-up but then the theory has to assume technological stagnation is incredibly common for enough length of time to allow upstarts to catch up.
Now history has shown that more powerful nation's have conquered others. The 19th century is choked full with examples. That primarily was driven by competition by world powers seeking ever more resources for that competition. The universe is abundant in resources however, we have a lot to pick from once we are space faring species.
>Now history has shown that more powerful nation's have conquered others. The 19th century is choked full with examples. That primarily was driven by competition by world powers seeking ever more resources for that competition. The universe is abundant in resources however, we have a lot to pick from once we are space faring species.
We can model this using just exponential growth and fixed speeds. If civilizations with interstellar capability existed prior to now, we'd see them. The galaxy would be literally teaming with them. That's the Fermi paradox. That's what the Grabby Aliens paper explored.
I use the word "godlike" because of just the magnitude of the energies and timescales we're dealing with. Drawing analogies with contemporary technology and behaviors really understate the magnitudes we're talking about.
Let's assume these aliens are dealing with just well understood Einsteinian physics. Nothing theoretical like wormholes, or thing. Their weapon of choice, is the same weapon of humans million years ago -- throwing big rocks.
Now, finding the mass of asteroids online is pretty difficult. It seems like most asteroids aren't actually measured, because the distribution of mass is highly skewed towards the Ceres, but two I'll look at are: 10 Hygiea, with 8.32E19 kg and 101955 Bennu (of OSIRIS-REx fame), at 7.329E10 kg.
Now we don't need to blow up the Earth Alderaan style. Instead, we'll aim for a Chicxulub impact, which apparently released something close to 4.2E23 J.
We'll also be pretty conservative about where aliens are, and assume they share our galaxy. This puts them at most around 100,000 light years away (the diameter of the Milky Way).
Let's say that we're accelerating our rock to 1% the speed of light. That means it would take 10 million years to hit your target. To put 10 million years into perspective, that's roughly how long ago chimpanzees and humans diverged. You're literally shooting at a species that probably won't exist, to protect your own species, that also probably won't exist.
Accelerating Bennu to 1% the speed of light would require 3.74E32 J and give an impact energy of 79x of Chicxulub. Hygiea at 1% would require 3.74E32 J be 900 million times the size of Chicxulub (8x the Great Impactor Event that birthed the moon). To put this in perspective, the Sun generates 3.8E26 J per second. You'd need to capture the entire output of the Sun for a day at 100% efficiency to accelerate Hygiea to 0.01c. Bennu takes a modest .0009 seconds of solar output.
10% the speed of light? 3.77E34 joules for Hygiea (9E10x Chicxulub) and 3.32E25 joules for Bennu (8x Chicxulub).
Just for giggles, 90% the speed of light gives 9.68E36 joules (800 years of solar output) to accelerate Hygia (2.3E13 Chicxulub, or 2 million times of the energy released by Great Impactor Event that formed the moon) and 8.52E27 joules (2.25 seconds of solar output) for Bennu (20000x Chicxulub). In this case, we'd be atomizing the planet in around just over 100,000 years.
On the other side, if we yeet these rocks at the speed of Voyager 1 (0.00005c, or 17000 m/s), it would take 2 billion years hit, and we'd get 20,000 Chicxulubs from Hygiea at the cost of 9.36E27 joules (2.5 seconds of solar output), and Bennu wouldn't even come close to Chicxulub, but rather a mere 4000 Tungustas at a cost of 8.25E18 joules.
If we want to cause 1 Chicxulub in 10 million years, we need to fire a mass of 1E11 kg, using 4.5E23 J of energy.
This is of course assuming that we're able to hit this planet without any course corrections. A planet mind you, we probably couldn't even theoretically image given the vast distances, infinitesimal angles, and the paucity of reflected photons that hit our telescopes. But hell, maybe we can develop some massive explosive or something that can last 6 million years to redirect our rock as we get closer to the system. The fact that we have to design an autonomous command and control system to last 10 million years in interstellar space is quite a lift, but let's just assume we get lucky.
In the Three Body Trilogy, the hyper-advanced aliens use projectiles accelerated to very near the speed of light. The premise is that there's enough energy in these that their impact is an extinction-level event for the planet they hit.
There's a fun thread on Quora[1] about so-called "relativistic kinetic kill vehicles", with the top comment making the good point that a small projectile moving very fast might just punch a clean hole in the planet and keep going, thus not transferring most of its energy.
I had been thinking about aliens, von Neumann probes, and even giant super computers to simulate the universe recently. I've come to the conclusion they're all absurdly dumb like a pulp scifi story.
People always think these shots are one and done, but they’re not. You have to keep hitting the planets with extinction level events, and you have to do it before you detect a civilization simply because of the transit times. If for sake of argument, we assume the dinosaurs were murdered, how did that work out for the trisoleans?
>There's a fun thread on Quora[1] about so-called "relativistic kinetic kill vehicles", with the top comment making the good point that a small projectile moving very fast might just punch a clean hole in the planet and keep going, thus not transferring most of its energy.
I saw something similar about a microblackhole or strange matter or something like that. Basically, you wouldn't even notice unless you standing right under/over it. Then you'd have a bad time.
The alien's desire to destroy other civilizations is what drives the plot forward. The message that I took away from the novel was not what aliens care about, but that blind reliance on technology will not solve all problems.
Likewise debating angels dancing on the head of a pin, as if it were literal, wasn't the point. Depending on what sources you check it was either a made up story to mock scholars wasting time or a practice topic for debating.
Demonology is a “study” conducted by true believers for hundreds, if not thousands of years.
Someone sat down in the 13th century, and earnestly questioned whether Forneus was a marquis or earl, and whether 30 legions of demons under his command was too many. (It was. He has only 29.)
>plan that requires at this point quadrillions of individuals to be dedicated to multiple 100,000 year plans
I viewed it as more like developed nations funding WHO to erdicate outbreaks in developing back waters. Dedicating fraction of a civilization's resources to genocide viruses that no one wants to think about. Also at some point it's just another jobs program for a sufficiently large civilizational bureaucracy finding things to do and threats to mitigate for KPIs on their powerpoints.
>energy expenditure and timescales that are the problem
Feels like that's anthropomorphizing human motivations (which Q at at some level was written to embody). IMO even for humans, not really an issue, wealthy countries spends fraction of discretionary budget to entertain long term projects. Last month Switzerland just commited 20B/3% of GDP on nuclear waste storage. Frankly, it could just be passing hobby of the nephew of some K1-2 civ energy sheikh sending out off the shelf grey goo genocide drones from his local cosmic mart for clout. Space twitch chat asked him to do it, and they do because it's a nonchalant activity, they're not thinking 1 minutes into the future let alone 100,000 years. They're not spending entire cycles of local sol energy to shoot expensive relativistic kill vehicles. They're spending pocket credits for a laugh, and 100,000 years later trillions of organisms end up dying. I think part of charm is recognizing the potential for lonewolf /small group of actors to be more dangerous than ever as societies move up tech tree.
I’ve already done the math about how hard it is to chuck a relativistic rock. Assuming solar output, you could arguably maybe do it as a Kardashev I, but it’s more a Kadashev II thing. It gets harder if you find yourself born around a common red dwarf.
If Zaphod Beeblebrox is chucking asteroids for lolz, while he waits for Slartibartfast to finish building a new pleasure planet (with space blackjack, and three boobed Martian hookers from Total Recall no doubt), we’re dealing with residents of Mount Olympus, with Zaphod playing the role of Zeus.
Come to think of it. That would certainly make the ancient alien conspiracies more entertaining —- and more terrifying.
The DFH doesn't claim that the aliens would be "god-like" per your definition. They just have a first-strike capability and motive to use it. Could be a low risk-tolerance for coexisting with other aliens. Or something else. Either way, you don't have to suspend disbelief because you're arguing with a strawman you created.
I don't think you actually considered the energies and timespans required (let alone the political will) to yeet a rock at an unseen planet on the other side of the galaxy.
At 1% the speed of light, it will take 10 million years just see if you hit anything on the other side of Milky Way. If you have to wait 10 million years, why do you even care? You'll be extinct, as will your target.
Even if you do decide chuck a dinosaur killing sized boulder (which, really is probably the smallest you'd want to use), you'd have expend 4.5 * 10^23 joules just to accelerate it. That's equivalent of all the solar radiation that hits the Earth over five days. That's a huge lift for even a supposed Kardashev I civilization. It's something more practical for a Kardashev II civilization. Although to be fair, the practical threshold is probably somewhere between I and II. And let's be clear here. If you're dismantling planets to build mirrors, you're Galacticus.
Anyway, how would the decision to throw a rock across the galaxy even work? A civilization obtains Kardashev II, and immediately starts looking for Kardashev Is or earlier, and starts tossing rocks at them, just in case they don't go extinct in 10 million years, and somehow don't also become a Kardashev II in the next 10 million years, because as soon as they do, they're going to hit you with a rock in 20 million years? To put that in perspective, 20 million years ago was when apes and monkeys split on the evolutionary tree. 10 million years ago was when humans and apes split. At this point, you might as well toss a rock at every planet that has a biosignature, because 10 million years is a long time. Better keep tossing rocks in that direction, in case something else evolves. Maybe you better toss a planet at it, just be sure. That's only 2.19 * 10^37 joules, or 200 years of the sun's output. Shit. You better be Kardashev III. Wait! We know there aren't any Kardashev IIIs in the observable universe! Double shit!
If this casually genocidal Kardashev II civilization exists on the far side of the Milky Way, and wanted kill humanity, they'd be looking at homo erectus. They'd have wait another 100,000 years just to get a radio signal, and then what? Throw the rock immediately, and hope that in the ensuing 10 million years humanity hasn't learned how to redirect a relativistic rock?
Or maybe they just threw the rock without even waiting for the radio signal, and it's on its way right now! Maybe the asteroid that hit 65 million years ago was a warning shot! Maybe they were trying to kill off a potential Silurian civilization before it even got really started, and ironically spawned us!
Replicator machines could outlive their creators. Then maybe most civilizations think it's safer not to risk being discovered by one, just in case it exists. That seems reasonable and doable.
You’ve gone from accusing me of a straw man argument, to postulating that not only do the Stargate Replicators exist, but every other advanced civilization knows they exist but somehow hide from them, even though by all conservative estimates, if Von Neuman probes did in fact exist, more than enough time has elapsed that the entire galaxy should have been converted to a grey goo by now.
Replicator machines “seem doable”?!?
Explain. Seriously. Explain what technologies and energy sources would create a complex autonomous machines from unrefined raw materials.
If it’s easier, explain how you’d make make a paper clip from first principles in the middle of a Greenland glacier. What are you going to do? Use electrolysis to split the water, and then have a fusion device fuse hydrogen into iron? Scrape across the surface to find trace atoms, and hope you find enough iron? Even if you have the iron, are you going smelt it, or what? How do you plan to forge it into steel, and then shape it? Are you going to use a tunneling electron microscope to place individual atoms? What’s the energy budget for this? How are you collecting and storing it? What’s the mass of this machine?
At this point, we’re talking about a World Devastator from Star Wars: Dark Empire.
“Seems doable”! Pshaw! EVERY ONE of these thought experiments from the 1940s are just magical premise to base a much simpler analysis on. In the case of Von Neuman proves, it’s determining a limit on how fast you could settle the galaxy. It’s not a serious technical proposal. It never was.
Still a strawman because I didn't say sci fi replicators. You could make an automated system that travels, mines, and manufactures. Like a 3D printer that prints copies of itself. That's not unbelievable.
The problem with von Neumann probes isn't the proverbial 3d printer. It's the filament. You have to build an entire supply chain for every individual component from first principles. That's why I asked how do you even make a single paper clip in Greenland. That's literally just a stainless steel wire. That's trivial compared to making an excavator, let alone an interstellar spacecraft.
Think about it. Mineral deposits aren't uniformly distributed. You're going to need to combine elements from thousands of kilometers away, and that's assuming their reachable from the surface with relative ease.
The only plausible way to build a "replicator" is actually to send entire premade factories, harvesters, and transports to a planet in a giant cooperative swarm, along with literally tons of preprocessed materials for repairs until you get a self sustaining supply chain running. And that implicitly assumes that all the requisite materials are even available. Land on a planet without plate tectonics, and your heavier elements may forever be locked under tens of kilometers of solid rock, or even better 60 kilometers of solid ice, liquid water, and rock. (Good luck with your Europan gold mine!)
Oh yeah, and it all of it has to work after a million years in hibernation.
So there's no benefit to collaborating because of distances and time scales. A system of machines is plausible, so there's a non-zero risk of encountering something like that. Civilizations decide to hide because only bad things could happen if they don't. And now you have a dark forest. The hunter doesn't have to exist. The prey just has to be fearful.
83 comments
[ 2.8 ms ] story [ 161 ms ] thread> “Dark forest theory” holds that civilizations fear one another so much that they don’t dare to reveal themselves lest they immediately be considered a potential threat and destroyed.
* https://techcrunch.com/2019/01/20/technologys-dark-forest/
> In The Dark Forest, a 2008 novel which Discover Magazine calls "a dark answer to the Fermi paradox",[138] Chinese writer Liu Cixin envisions aliens as being paranoid and wanting to kill life because it might be future competition. Other aliens dare not reveal themselves because they might be considered a threat, and they stay quiet as if in a dark forest.[139]
* https://en.wikipedia.org/wiki/Fermi_paradox#Communication_is...
then again, given the technology necessary for traveling that far, it wouldn't be a fun fight for them, so no point in coming.
We could spend a million years building a single civilization destroying weapon, and it would still be worth it.
Since we’re not likely to encounter another civilization much more often.
Of course, humans couldn’t really stay interested in something that long, so the practical range is probably a few decades :P
It's a game theoretical problem. Because of the huge delay in communication over interstellar distances, you can not adequately communicate with other in diplomatic terms.
due to the speed of light limitation, if you contact another civilization to e.g. propose to establish diplomatic relations, you will only find out if they sent a big ass rocket as a reply in about the same time it would take to receive a message as an answer.
Making certain assumptions about the technology available, like: No instant communication possible, solar system or at least planet destroying weapons possible, etc., then in purely game theoretical terms if a civilization seeks to minimize the chance of being destroyed, it then becomes a good strategy for a civilization to hide and not reveal themselves, and immediately attack and destroy any civilization that should make the mistake to reveal themselves by e.g. sending out a message. thus giving them no chance to harm you before they are destroyed themselves.
it's only a hypothesis, and even at a superficial glance there are alternative strategies that might end up being better on average, much better if each civilization is willing to take some risk, etc.
https://www.goodreads.com/book/show/23168817
Power is important, but not the most important factor. He states that he needs to collect more data, but within his current dataset there is a clear and visible preference in the surviving population.
We ask that you try just one more thought experiment. Imagine yourself taking a stroll through Manhattan, somewhere north of 68th street, deep inside Central Park, late at night. It would be nice to meet someone friendly, but you know that the park is dangerous at night. That's when the monsters come out. There's always a strong undercurrent of drug dealings, muggings, and occasional homicides.
It is not easy to distinguish the good guys from the bad guys. They dress alike, and the weapons are concealed. The only difference is intent, and you can't read minds.
Stay in the dark long enough and you may hear an occasional distance shriek or blunder across a body.
How do you survive the night? The last thing you want to do is shout, "I'm here!" The next to last thing you want to do is reply to someone who shouts, "I'm a friend!"
What you would like to do is find a policeman, or get out of the park. But you don't want to make noise or move towards a light where you might be spotted, and it is difficult to find either a policeman or your way out without making yourself known. Your safest option is to hunker down and wait for daylight, then safely walk out.
There are, of course, a few obvious differences between Central Park and the universe.
There is no policeman.
There is no way out.
And the night never ends.
But of course, even Stephen Hawking was talking about this for decades. I find it bit of hilarious social commentary, that some Chinese author copied it and now people claim he invented it.
You assume that he knew about it:
* https://en.wikipedia.org/wiki/Independent_invention
IMO Liu steelmaned the argument with respect to game theory, relativistic diplomacy / suspicion chains, tech explosion etc. He elevated it from commentary to hypothesis.
If the species are not aggressive, they can presumably at least the weaker of the two should get some large bonus.
Of course then we could play with the first strike bonus and the alliance bonus, and get whatever result we want from the simulation.
https://en.wikipedia.org/wiki/Two_Generals%27_Problem
I question power and visibility being independent variables.
I question the idea of needing to be more powerful to destroy each other when a weaker society could still kill the other simply by attacking first.
A kinetic care package could spend as little as 7 hours inside Pluto's orbit and it wouldn't have to be mountain sized.
It might indeed only follow it's own light by a small increment leaving even a much more advanced tech with few options.
Overall entirely pointless.
In all seriousness though, if the universe is indeed populous then there's probably been innumerable alien XCOM projects.
I also can't help but think that we usually fail to appreciate the technology gaps involved. Human beings in a span of a few hundred years managed to go from the advent of electricity and industrial society, to artificial intelligence—all in what amounts to not even a blink of a cosmic eye. Yet recursively improving intelligence is conceivable to us.
2001: A Space Odyssey did a wonderful job illustrating such gaps. I think there's a good chance most civilizations (ours included) would not be equipped to even comprehend how to detect or fight an advanced civilization, let alone understand their very nature.
I love Xcom but it very much portraits the absolute opposite of the 3 body problem trilogy.
I dont want to commit to some sort of a 3-tomb marathon ;(
Also keep in mind that Humanity is a latecomer to the universe. We weren't the first species to evolve on Earth. First came the dinosaurs, about 230m years ago, and lasted till about 66m years ago, then went extinct. Then came humanity.
But there may be planets in our galaxy or other ones where intelligent life happened to be the first species to evolve on that planet. If so, they have a multi-hundred million year head start on us, evolutionarily, scientifically, technologically.
Yeah, there's no possible XCOM project against that. We just have to hope they're benevolent. (And if they've solved scarcity with energy->matter conversion, as in Star Trek, they may very well be - there's nothing we have that they could need or want)
It looks that way to us because of the limited rate at which we consume resources relative to the scale of the cosmos. But our rate of consumption isn’t anywhere near an upper bound for a civilization. We’re not even a Type 1 on the Kardashev scale.
https://en.wikipedia.org/wiki/Kardashev_scale
Also even if it takes millions of years, having your autonomous machines cut up the local stellar bodies and ship materials back to you is an obvious move. You're immortal, so who cares how long it takes.
The problem is observed reality contradicts this hypothesis, so far there are no observed techno-signatures. Some call this a paradox because they're very attracted to their hypothesis, but it's not a paradox if the explanation for not seeing anything is there being nothing to see. I expect there to be gnomes in my garden, but I don't see any gnomes in my garden. Is that a paradox? Maybe I'm just wrong about gnomes.
As for recon probes, maybe. But we wouldn't see them would we? If a recon probe is looking for threats to the civilization that created it, it wouldn't do to have the recon probe seen. Recon is about seeing while not being seen.
Disassembling other star systems? It's a neat idea, but there's no evidence for it. Maybe it's not such an obvious idea to the sort of civilization that might be capable of it and it only seems obvious to primitive apes who can't do it anyway. Or maybe there are no such capable civilizations in our neighborhood at all.
The Road Not Taken by Harry Turtledove is a fun take on something like this concept.
I don't see how nuclear weapons are necessary for destruction if that is your point.
Even if it took a few million years to do, that's still far less time then the Earth has been habitable and had life. So the question of course is still where are they?
Thus, each expansionist civilization will spread out to encompass an exponentially increasing number of advanced, stable civilizations, and antagonize an increasing number of them. Soon, they will encounter one that chooses to put them in their place, and maybe just wipe them out.
So we may think of the galaxy as an archipelago of stable civilizations that interact little, with splashes of expansionism in between that encounter and annoy them, and are swiftly contained. Such expansions are exceedingly unlikely to get to us before such an event curtails them.
I.e. how do you destroy an expansionist fleet which has scaled up to the point that it's furthest point from you is increasing close to the speed of light?
But they are, remember, incomprehensibly more advanced than you or me. And they know a bunch of other incomprehensibly more advanced civilizations all the way on the other side of the expansionist noobs. And, just because they don't usually see much point in going out there doesn't mean they can't, or won't, when the time seems right for it.
I can come up with one respons within the believed laws of physics at how a hypothetical stable civilization could suppress exponential ones even though exponential ones would quickly be expanding their radius at close to C: the stable civilization would need to pre-seed the visible universe with idle hidden outposts of a sufficient density to detect and suppress exponential expanders.
We are not predicting the future, but trying to account for an observation. For the scenario to remain on the list of reasons why we haven't been visited, it suffices that it is possible. There are of course lots of other possible reasons, among them that there is nobody to visit.
That come around us since we're trying to explain why we haven't seen an expansionists. So this line of thinking predicts that we're in the shadow of a non-expansionists, I guess?
This is not necessarily true. A roiling cluster of ephemeral, competing life forms and civilizations may very well develop warfare and defence technology better and faster than stable civilizations. This is especially true if you consider the cost of stability: all new technology is a potential threat to stability that has to be weighed carefully, so I expect stable civilizations to evolve much slower than unstable ones. In the aggregate, instability is not really a problem: a civilization's progress will be scavenged or stolen by its successors, retaining the advantage, and for that reason I would argue the unstable regime will almost systemically overwhelm the stable one.
You can kind of see this with humanity: a lot of the big problems we have to face as a civilization (but not all of them) result from the careless adoption of technology. Climate change would not be a problem if we had industrialized slowly enough. The Internet spread too fast for us to understand its effects on the propagation of (dis)information. A truly stable civilization would have taken a thousand years to do what we did in a hundred. Perhaps it would be happier for it, but if we met them, we would eat them alive.
Puny humans are confined to one solar system, and even just the one planet, but are pretty dynamic. Population in wealthy countries is in decline.
Remember we are not just talking about humans with bumpy heads, here. Social universals on Earth do not apply. Expansion might be limited just by a physical inability to reproduce beyond their home planet, without giving up any dynamism.
You described barbarians. But barbarians don't need Von Neumann probes, they can do everything they want with a stone hammer.
I'm sorry. But I'm not supposed to suspend disbelief for serious discussions.
The older I get, the more I find these debates as tedious as medieval demonologists wondering about how many angels can dance on the head of a pin, and who is the fifth viscount of the third circle of Hell, and just what exactly is his bailiwick.
Isn't the point the point that they aren't godlike? They're glass cannons, capable of creating weapons with incredible power, but incapable of defending themselves against that same class of weapon.
Godlike aliens -> humans.
Other civs -> wasps, mosquitos, cockroaches. Not a real existential threat, but enough of a nuisance. Wasps, even though they live in a complex culture from their point of view, have no chance to understand human culture. Not wanting to be accidentally stung sort of has a chance to be understood by both sides, very-very approximately.
1. Wasp-Aliens are everywhere in the universe.
2. Therefore they know how to travel with some speed v.
3. Therefore they know how to accelerate objects to speed v.
4. They can launch said objects towards your planet.
5. Therefore they can destroy you, so assumption 0 doesn't hold.
In the aftermath of World War II, John von Neumann advocated a preemptive nuclear first strike on the Soviet Union before they could develop their own nuclear weapons. It's monstrous, but it's the same cold logic that the DFH proposes: Destroy others before they have a possibility of destroying you.
Von Neumann's plan was based on an incredibly short time scale. It wasn't long before the Soviet Union would develop its own weapons. The power asymmetry was only briefly held.
The thing is at a universal scale that would imply that the only threats we would honestly face are ones in close power level, and we can barely leave the earth at the minute. Another species with more advancement than us would be n-years more advanced than us even after we reach the level of what they were at on first encountering them (unless they decide to share technology which would be the opposite of DFT). Technological stagnation could cause a catch-up but then the theory has to assume technological stagnation is incredibly common for enough length of time to allow upstarts to catch up.
Now history has shown that more powerful nation's have conquered others. The 19th century is choked full with examples. That primarily was driven by competition by world powers seeking ever more resources for that competition. The universe is abundant in resources however, we have a lot to pick from once we are space faring species.
We can model this using just exponential growth and fixed speeds. If civilizations with interstellar capability existed prior to now, we'd see them. The galaxy would be literally teaming with them. That's the Fermi paradox. That's what the Grabby Aliens paper explored.
They don't exist.
Let's assume these aliens are dealing with just well understood Einsteinian physics. Nothing theoretical like wormholes, or thing. Their weapon of choice, is the same weapon of humans million years ago -- throwing big rocks.
Now, finding the mass of asteroids online is pretty difficult. It seems like most asteroids aren't actually measured, because the distribution of mass is highly skewed towards the Ceres, but two I'll look at are: 10 Hygiea, with 8.32E19 kg and 101955 Bennu (of OSIRIS-REx fame), at 7.329E10 kg.
Now we don't need to blow up the Earth Alderaan style. Instead, we'll aim for a Chicxulub impact, which apparently released something close to 4.2E23 J.
We'll also be pretty conservative about where aliens are, and assume they share our galaxy. This puts them at most around 100,000 light years away (the diameter of the Milky Way).
Let's say that we're accelerating our rock to 1% the speed of light. That means it would take 10 million years to hit your target. To put 10 million years into perspective, that's roughly how long ago chimpanzees and humans diverged. You're literally shooting at a species that probably won't exist, to protect your own species, that also probably won't exist.
Accelerating Bennu to 1% the speed of light would require 3.74E32 J and give an impact energy of 79x of Chicxulub. Hygiea at 1% would require 3.74E32 J be 900 million times the size of Chicxulub (8x the Great Impactor Event that birthed the moon). To put this in perspective, the Sun generates 3.8E26 J per second. You'd need to capture the entire output of the Sun for a day at 100% efficiency to accelerate Hygiea to 0.01c. Bennu takes a modest .0009 seconds of solar output.
10% the speed of light? 3.77E34 joules for Hygiea (9E10x Chicxulub) and 3.32E25 joules for Bennu (8x Chicxulub).
Just for giggles, 90% the speed of light gives 9.68E36 joules (800 years of solar output) to accelerate Hygia (2.3E13 Chicxulub, or 2 million times of the energy released by Great Impactor Event that formed the moon) and 8.52E27 joules (2.25 seconds of solar output) for Bennu (20000x Chicxulub). In this case, we'd be atomizing the planet in around just over 100,000 years.
On the other side, if we yeet these rocks at the speed of Voyager 1 (0.00005c, or 17000 m/s), it would take 2 billion years hit, and we'd get 20,000 Chicxulubs from Hygiea at the cost of 9.36E27 joules (2.5 seconds of solar output), and Bennu wouldn't even come close to Chicxulub, but rather a mere 4000 Tungustas at a cost of 8.25E18 joules.
If we want to cause 1 Chicxulub in 10 million years, we need to fire a mass of 1E11 kg, using 4.5E23 J of energy.
This is of course assuming that we're able to hit this planet without any course corrections. A planet mind you, we probably couldn't even theoretically image given the vast distances, infinitesimal angles, and the paucity of reflected photons that hit our telescopes. But hell, maybe we can develop some massive explosive or something that can last 6 million years to redirect our rock as we get closer to the system. The fact that we have to design an autonomous command and control system to last 10 million years in interstellar space is quite a lift, but let's just assume we get lucky.
Here's a spreadsheet of my work https://docs.google.com/spreadsheets/d/1_W_mJ1ev1mkvZZMkwVlJ...
References Wikipedia for masses and diameters https://www.omnicalculator.com/physics&#x...
In the Three Body Trilogy, the hyper-advanced aliens use projectiles accelerated to very near the speed of light. The premise is that there's enough energy in these that their impact is an extinction-level event for the planet they hit.
There's a fun thread on Quora[1] about so-called "relativistic kinetic kill vehicles", with the top comment making the good point that a small projectile moving very fast might just punch a clean hole in the planet and keep going, thus not transferring most of its energy.
[1] https://www.quora.com/What-are-Relativistic-Kinetic-Kill-Veh...
Oh god. I made a fucking spreadsheet! @_@
I had been thinking about aliens, von Neumann probes, and even giant super computers to simulate the universe recently. I've come to the conclusion they're all absurdly dumb like a pulp scifi story.
People always think these shots are one and done, but they’re not. You have to keep hitting the planets with extinction level events, and you have to do it before you detect a civilization simply because of the transit times. If for sake of argument, we assume the dinosaurs were murdered, how did that work out for the trisoleans?
>There's a fun thread on Quora[1] about so-called "relativistic kinetic kill vehicles", with the top comment making the good point that a small projectile moving very fast might just punch a clean hole in the planet and keep going, thus not transferring most of its energy.
I saw something similar about a microblackhole or strange matter or something like that. Basically, you wouldn't even notice unless you standing right under/over it. Then you'd have a bad time.
Likewise debating angels dancing on the head of a pin, as if it were literal, wasn't the point. Depending on what sources you check it was either a made up story to mock scholars wasting time or a practice topic for debating.
Someone sat down in the 13th century, and earnestly questioned whether Forneus was a marquis or earl, and whether 30 legions of demons under his command was too many. (It was. He has only 29.)
https://en.wikipedia.org/wiki/Forneus
I viewed it as more like developed nations funding WHO to erdicate outbreaks in developing back waters. Dedicating fraction of a civilization's resources to genocide viruses that no one wants to think about. Also at some point it's just another jobs program for a sufficiently large civilizational bureaucracy finding things to do and threats to mitigate for KPIs on their powerpoints.
Even Star Trek's Q, doesn't care want to wait 100,000 years, and he's immortal, all powerful, and exists outside of space and time.
Feels like that's anthropomorphizing human motivations (which Q at at some level was written to embody). IMO even for humans, not really an issue, wealthy countries spends fraction of discretionary budget to entertain long term projects. Last month Switzerland just commited 20B/3% of GDP on nuclear waste storage. Frankly, it could just be passing hobby of the nephew of some K1-2 civ energy sheikh sending out off the shelf grey goo genocide drones from his local cosmic mart for clout. Space twitch chat asked him to do it, and they do because it's a nonchalant activity, they're not thinking 1 minutes into the future let alone 100,000 years. They're not spending entire cycles of local sol energy to shoot expensive relativistic kill vehicles. They're spending pocket credits for a laugh, and 100,000 years later trillions of organisms end up dying. I think part of charm is recognizing the potential for lonewolf /small group of actors to be more dangerous than ever as societies move up tech tree.
If Zaphod Beeblebrox is chucking asteroids for lolz, while he waits for Slartibartfast to finish building a new pleasure planet (with space blackjack, and three boobed Martian hookers from Total Recall no doubt), we’re dealing with residents of Mount Olympus, with Zaphod playing the role of Zeus.
Come to think of it. That would certainly make the ancient alien conspiracies more entertaining —- and more terrifying.
At 1% the speed of light, it will take 10 million years just see if you hit anything on the other side of Milky Way. If you have to wait 10 million years, why do you even care? You'll be extinct, as will your target.
Even if you do decide chuck a dinosaur killing sized boulder (which, really is probably the smallest you'd want to use), you'd have expend 4.5 * 10^23 joules just to accelerate it. That's equivalent of all the solar radiation that hits the Earth over five days. That's a huge lift for even a supposed Kardashev I civilization. It's something more practical for a Kardashev II civilization. Although to be fair, the practical threshold is probably somewhere between I and II. And let's be clear here. If you're dismantling planets to build mirrors, you're Galacticus.
Anyway, how would the decision to throw a rock across the galaxy even work? A civilization obtains Kardashev II, and immediately starts looking for Kardashev Is or earlier, and starts tossing rocks at them, just in case they don't go extinct in 10 million years, and somehow don't also become a Kardashev II in the next 10 million years, because as soon as they do, they're going to hit you with a rock in 20 million years? To put that in perspective, 20 million years ago was when apes and monkeys split on the evolutionary tree. 10 million years ago was when humans and apes split. At this point, you might as well toss a rock at every planet that has a biosignature, because 10 million years is a long time. Better keep tossing rocks in that direction, in case something else evolves. Maybe you better toss a planet at it, just be sure. That's only 2.19 * 10^37 joules, or 200 years of the sun's output. Shit. You better be Kardashev III. Wait! We know there aren't any Kardashev IIIs in the observable universe! Double shit!
If this casually genocidal Kardashev II civilization exists on the far side of the Milky Way, and wanted kill humanity, they'd be looking at homo erectus. They'd have wait another 100,000 years just to get a radio signal, and then what? Throw the rock immediately, and hope that in the ensuing 10 million years humanity hasn't learned how to redirect a relativistic rock? Or maybe they just threw the rock without even waiting for the radio signal, and it's on its way right now! Maybe the asteroid that hit 65 million years ago was a warning shot! Maybe they were trying to kill off a potential Silurian civilization before it even got really started, and ironically spawned us!
This strains credulity.
Replicator machines “seem doable”?!?
Explain. Seriously. Explain what technologies and energy sources would create a complex autonomous machines from unrefined raw materials.
If it’s easier, explain how you’d make make a paper clip from first principles in the middle of a Greenland glacier. What are you going to do? Use electrolysis to split the water, and then have a fusion device fuse hydrogen into iron? Scrape across the surface to find trace atoms, and hope you find enough iron? Even if you have the iron, are you going smelt it, or what? How do you plan to forge it into steel, and then shape it? Are you going to use a tunneling electron microscope to place individual atoms? What’s the energy budget for this? How are you collecting and storing it? What’s the mass of this machine?
At this point, we’re talking about a World Devastator from Star Wars: Dark Empire.
“Seems doable”! Pshaw! EVERY ONE of these thought experiments from the 1940s are just magical premise to base a much simpler analysis on. In the case of Von Neuman proves, it’s determining a limit on how fast you could settle the galaxy. It’s not a serious technical proposal. It never was.
The problem with von Neumann probes isn't the proverbial 3d printer. It's the filament. You have to build an entire supply chain for every individual component from first principles. That's why I asked how do you even make a single paper clip in Greenland. That's literally just a stainless steel wire. That's trivial compared to making an excavator, let alone an interstellar spacecraft.
Think about it. Mineral deposits aren't uniformly distributed. You're going to need to combine elements from thousands of kilometers away, and that's assuming their reachable from the surface with relative ease.
The only plausible way to build a "replicator" is actually to send entire premade factories, harvesters, and transports to a planet in a giant cooperative swarm, along with literally tons of preprocessed materials for repairs until you get a self sustaining supply chain running. And that implicitly assumes that all the requisite materials are even available. Land on a planet without plate tectonics, and your heavier elements may forever be locked under tens of kilometers of solid rock, or even better 60 kilometers of solid ice, liquid water, and rock. (Good luck with your Europan gold mine!)
Oh yeah, and it all of it has to work after a million years in hibernation.