Grabby Aliens & The Fermi Paradox
This Episode is brought to you by world Anvil A new model suggests that nearly a quarter of the Universe is now colonized by a few thousand grabby alien mega-civilizations, each one controlling not just millions of worlds, but millions of galaxies, and that Humanity will likely join their number shortly, as the heir to quadrillions of stars and planets. So, just over a year ago, Robin Hanson and some colleagues released a paper in the Astrophysical Journal titled: “If Loud Aliens Explain Human Earliness, Quiet Aliens Are Also Rare”, which we’ll link in the description, as well as a website, grabbyaliens.com, and almost as soon as it came out folks started asking if I’d do a video on this new solution to the Fermi Paradox, and I joked that we had already done several. Grabby aliens is a cool mathematical model but we need to say from the outset that it is not a new Fermi Paradox Solution, it’s modeling a trio of variables we have some data on that are related to civilizations getting out to the stars, and it seeks to show us that for any plausible value of those variables, the only one that makes sense is that alien civilizations don’t pop up much, that the kind that remain fairly quiet from an observational standpoint aren’t vastly more common than the big loud expansionist ones, and that therefore, humanity doesn’t see aliens because we’re among the earliest. It suggests that there might not be any of what we would consider to be alien civilizations in this galaxy and that it’s probably half a billion light years, if not more, much more, to the next high-tech alien civilization. Alien life isn’t really rare, it’s just rare right now, and any alien civilization that emerged in its local corner of the Universe is very likely to be doing as we do, wondering where all the other life is, and concluding that someone had to be the first, and for that region of space, it was them. The concept being that life spreads quickly, so, much
as the first amoeba on Earth might see a whole empty ocean, it and its offspring would probably fill that whole ocean before life independently evolved elsewhere in that ocean again. Channel regulars know that’s the default position of this show and it’s the argument I advanced in Season 1 episode 1, way back in 2015, the Dyson Dilemma, which isn’t a Fermi Paradox solution either, but supports this same solution which is the Rare Intelligence camp of the Rare Earth Hypothesis. This is no surprise, Robin Hanson is best known in these circles as the gentleman who gave us the term “Great Filters”. These describe the various hurdles that life has to pass through to get from dead young planet to building rockets, like evolving life, cellular nuclei, brains, tool use, etc. all while not being wiped out by asteroids, climate shifts, supernovae, or older civilizations who come and take the planet for their own uses. Now, that wasn’t quite a new idea then either, Brandon Carter – an astrophysicist best known on this show for his discussion of the Anthropic Principle and Doomsday Argument – had discussed this in the 1980s but had borrowed the term ‘Hard Steps’ from the medical field. I have rather
intentionally not reached out to Dr. Hanson before writing this script, as I’m trying to be properly critical of the theory; which is hard enough when it’s strongly backing my own preferred stance on the Fermi Paradox and it’s written by a man whose work I respect, so I didn’t want to make that worse by reaching out for a dialogue, instead, just reading the paper. As a result, I don’t know if they opted to use the term Hard Steps instead of Great Filters, because he had coined that term, but it has resulted in Hard Steps showing up more in Fermi Paradox discussions these last several months and I rather regret that such is the case. I mentioned Hard Steps was borrowed from medicine,
and there, it is discussing the probability of mutation on the road to cancer and I hate when anyone compares humanity to either cancer or a swarm of locusts when talking about us on this Planet or heading out into the Galaxy. So I’ve always avoided the term and prefer Great Filters and rather hope Hard Steps doesn’t catch on. I’m also not terribly fond of the term “Grabby Aliens” as to me at least that puts a negative connotation on the entire concept of space colonization, but, fundamentally, both are just nitpicks by me that have nothing to do with the concept at hand. It needed a name and it was clearly catchy, and brought more interest to the topic, so no complaints, and it’s much more compact than Aggressively Expanding Civilizations, in keeping with Jay Olson’s Fermi Paradox paper from 2015, “Homogeneous cosmology with aggressively expanding civilizations” it helps make the case that we can’t just limit our notion to aliens colonizing one galaxy, but also moving beyond, and as we’ll see today, the assumption is that grabby aliens grab a lot more than one galaxy or even thousands. The paper also explicitly rejects the Zoo
Hypothesis, the idea that Earth is a zoo and surrounded by aliens who keep themselves hidden from us, and while that’s fine for the running of the model, I don’t believe that’s justified as a conclusion. For the model, it's kind of necessary, as otherwise, you could get alien expansion spheres erupting inside existing ones or you could have aliens inside another civilization that didn’t know they were inside of one, and we could easily be one. See our episode on the Zoo Hypothesis for more explanation of the pros and cons of that solution.
That’s the basic intro though, for channel regulars, this isn’t a new Fermi Paradox solution, it’s a new model that adds support to the Great Filters and Rare Earth or Intelligence solutions. Today we’re going to look at that model, its assumptions, and some of its strengths and weaknesses, but overall it’s quite strong, and it follows from Carter’s own earlier reasoning on the Anthropic Principle that even though space travel is slow compared to human lifetimes, it can be incredibly fast compared to planetary and stellar lifetimes, so the first species in a galaxy – or supercluster even – who get intelligent and gets expansionist is going to colonize everything in that area before anybody else arrives on the scene, because intelligent life and the spawning of an intelligent, cooperative civilization from it, isn’t very likely to occur on worlds in any given galaxy, at a frequency that is faster than the travel time of colony ships across that galaxy – or supercluster. Thus, every species that colonizes a galaxy started by asking why the place was so empty, and found out that it was because it actually was empty before them. Nobody else ever gets the chance to wonder that because they can’t evolve on a planet that someone else has already colonized or disassembled for building materials. They can’t peer into the night sky and wonder if
anyone lives around those stars because those stars aren’t visible to the naked eye any more, they’re surrounded in Dyson Swarms, and at most, your world is a protected nature reserve they left behind or uploaded into a computer simulation or has been pushed into a pocket Universe of its own by higher technology. And we’ve discussed those options and many others in our Fermi Paradox series and our Alien Civilizations series, and to save time, we’ll reference some of those episodes as we go for those wanting more discussion on the matter. So, the paper mentions Loud Aliens and Quiet Aliens, and for general context, Quiet Aliens are those who contribute to the Great Silence, The Great Silence is the observed condition of the Universe by modern astronomy and SETI – the search for extraterrestrial intelligence – that we don’t see anyone, and the basis for the Fermi Paradox, which is the question of why we don’t see any civilizations in the vast and ancient universe. Why we see this Great Silence? Quiet aliens are, by definition, the only kind we currently observe, since they preserve the Fermi Paradox, and Loud Aliens are what we are looking for with SETI.
For the purpose of the paper though, loud aliens have 3 characteristics: they expand fast, they last a long time, and they make visible changes to the places they go, like terraforming planets and building megastructures and Dyson swarms and so on. Quiet aliens are those that don’t do at least one of those three, maybe they don’t expand or do so very slowly even on cosmological timelines, maybe they ceased to exist, blew themselves up for instance, or they don’t really leave a footprint where they go, like a non-technological species of giant whales that lived on gas giants or inside stars which might migrate between them at a galactic scale and still not leave anything behind we would see. And of course, there are dozens of different methods or exceptions to each of these conditions, but critically, none of them really discuss technology, motives, or culture of aliens involved, only the most general condition and one that we can see, a species that expands fast, lasts long, and leaves marks. Making them easy for us to notice, or loud. Grabby Aliens in the paper are those which, quote “(a) expand the volumes they control at a common speed, (b) clearly change the look of their volumes (relative to uncontrolled volumes), (c) are born according to a power law in time except not within other GC volumes, and (d) do not die unless displaced by other GCs ” unquote. Let’s unpack that.
First, a GC is a galactic civilization, and we may extend that to include Intergalactic ones too, or even multiverse-traveling ones. Indeed based on the model’s results, a GC in this context is likely to be a multi-million galaxy empire rather than a civilization inside a galaxy or controlling an entire one, what GC usually means and what we call a Kardashev 3 or K3 civilization, and I’d suggest maybe the civilizations in this model should be considered K4 instead. Second, this is all about setting up a computer model of various origin planets of various aliens randomly distributed at a given density, and each with a random expansion speed and duration, in order to see how likely we are to be first in our region of space, and if so, what that indicates about how often that loud, grabby aliens occur and run into each other. From the model they typically expand to get to a million or so galaxies before running into neighboring expansion waves.
That quote from a moment ago is them setting up the parameters. So, we have expanding volumes under their control at a common speed. What does that mean? In this case, it means that every GC, once it starts expanding, will move at the same speed the whole time – until it bumps into a neighbor – and that all those neighbors move outward at the same speed. That’s got some problems we’ll get to in a moment, but models always need to simplify some variables. We’ll be pointing out some of those simplifications today too,
and some might matter for the conclusion but probably not this one. We assume ships will get faster with time and thus colonization rates outward would speed up but at the scale involved, that probably would be maxed out long before it really mattered. A few thousand years of spaceship propulsion improvements to get from 1% light speed ships, up to 80%, matters a lot in a system where civilizations all start popping up within a few thousand years of each other and expanding a few thousand light years, like we see in the typical space opera of hundreds of aliens in a galactic region, but it really doesn’t matter at extragalactic scales of millions of years of travel time or more. We assume the same physics applies Universe-wide and that loud aliens are curious ones too, where science is concerned, and thus work to improve their ship drives until everyone ends up with that same optimized engine concept. From a practical standpoint, that means that all the expanding bubbles of new GCs expand at the same rate from their homeworld even if not from the same start date. In practice it would seem
like what will actually matter for colonization will be if there’s a maximum safe speed to travel between galaxies, as most of our expansion time in this model is intergalactic colonization and K3 civilizations, or even K2, really shouldn’t have any problem building powerful and long enough acceleration ramps to let ships reach 99% of light speed for that voyage, but that might be too much for safe travel on voyages lasting several million years. Barring that, it would be my expectation that at the intergalactic scale colonization occurs at .99c, a tiny fraction slower than light speed. A possible caveat, however, is that intergalactic space is orders of magnitude emptier than interstellar space, so it may just allow for higher cruising speeds after all, owing to the lower risk of dust collisions. See our generation ships series and Intergalactic Colonization episode for more discussion of that, including hurling planet-sized ships out of the galaxy, or moving galaxies back here for harvesting.
Second bit, B, that grabby aliens clearly change the look of their volumes (relative to uncontrolled volumes), is easy enough to understand. Uncontrolled volumes or natural volumes of space not yet colonized will be recognizably different by astronomy than those the aliens have reached and colonized or otherwise utilized, like building Dyson Swarms around every star or giant beacons or any number of other megastructures, or even just sucking bits of space dry for bringing back to their homeworld – or home dimension, we don’t want to take for granted that our Universe is the only board in the game, though understandably the paper does this, for simplicity I assume. And condition B basically amounts to saying when you colonize a place it looks different then when it was untouched. Third, grabby aliens are born according to a power law in time, except not within other GC volumes. Which means the older the Universe gets,
the more likely any random spot not already occupied by a GC is going to produce intelligent life. More stars have formed, more heavy elements of big planets, older planets with more time for life to have emerged and evolved. And the model will effectively argue that no new GCs would be emerging after about 4 billion years from now, as the whole Universe should be full of other GCs, even in very pessimistic occurence and growth rates. In terms of the model,
it means the longer it runs, the faster new intelligences are appearing in unoccupied space, but that they do not appear inside already occupied space. Or where they might, that they would just expand at the same speed as part of that wider GC from a practical perspective. And again, they reject the Zoo Hypothesis here, for the purpose of the model at least.
There are a lot of good objections to that, but they mostly should not matter for that model. For instance, us getting killed off by Artificial Intelligence of our own devising in the next few thousands years; doesn’t matter to this model anymore than if it takes us a century or a hundred centuries to perfect our spaceship drives, in terms of maximum speed. Us getting replaced by AI or by a human-descendant clade or by some terran or alien species we found and uplifted - supersmart dolphins for instance – just changes that GC marker to either shift it to be quiet aliens or still loud aliens but with a different profile pic, so to speak. Same business, new owners. Lastly we have D, that a grabby alien civilization or empire does not die unless displaced by other GCs, which is that, GCs, once they get rolling off their homeworld, just aren’t going to evaporate away. They might have that AI rebellion that replaces them but if you’re spread out to a thousand worlds and growing, only an AI that wanted to do the same would really be a threat, as opposed to parking on the colony world they rebelled on. Indeed, you getting chased out by pursuing waves of angry AI
or uplifted chimps is a pretty good way to encourage fast and loud colonization too. As we discussed in our Extragalactic Sanctuaries episode, where we had folks colonizing whole star systems and turning them into giant fortresses and death traps, merely because they are running in retreat across the galaxy, hurling quintillions of ships and troops at the invader just to slow them down and distract them. An expanding GC is not implied to be unified here, just that any given elements near its current frontier edge tend to be prone to expanding that frontier, even if on various sides of that GC’s bubble, none of them have ever spoken to each other or know much about the other side of the frontier, nor do they even necessarily still resemble one another. A million years of evolution, genetic engineering, and cybernetics, might mean that the ‘humans’ expanding into the Western part of the galaxy resemble those on the Eastern Fringe no more than a cougar resembles a shark, for all they are ancient relatives. Now, once they began running these simulations with different values for the parameters and in random setups over and over, the evidence tilted to us being about in the middle of Galactic Civilization emergence, which would seem to be 4 billion years ago to 4 billions years from now, and that GCs already controlled about a fifth to a quarter of the observable Universe and generally colonized outward at nearly 80% of light speed. That the light of expansion evidence has
reached 35% of the known Universe and we’re in the roughly 65% that can’t see another GC yet. Most GC Bubbles, as they expand and maximize like a big bubble in a sink or bath touching many others, will be bigger than our GC bubble, with 62% being bigger than humanity’s eventual GC. Not that we’ll be crowded, we’ll probably have a volume of space at least as big as our whole supercluster of 100,00 galaxies before we bump into another GC. Roughly 7000 GCs will appear in an Observable
Universe-Sized Volume. As usual we do not know how big the entire Universe is, possibly it is infinite. We assume the Observable Universe is relatively small compared to it and in this case, that portion would have 7000 GCs. Note that this is essentially the approach of looking for the most crowded Universe that fits the data and model, nothing in this says that GCs can’t be even more spread out than that, nor does it imply that Earth isn’t the only place life ever appeared, we may be be 1 in gajillion, this model just seeks to say we are not less than one in a quintillion, in terms of stars giving rise to GCs. We might be civilization 100 to appear, not 4000-or-something, or even #1. Though we expect the model to roll out the same, it would just shift from 10-18
billions after the Big Bang to more like 13 to 20. We had reached that parallel conclusion in the Dyson Dilemma discussion way back when the show started and it’s a logical extrapolation of prior work, all the way back to at least Michael Hart in the 1970s, who coined the term the Fermi Paradox, and argued that since interstellar travel seemed increasingly plausible and biological life should be prone to wanting to expand, intelligent life had to be very rare or we should see it way more often. In our case, we looked it at with as little math as I could get away with, just as today I’m trying to skip as much as we can, and we also looked at it from the angle of how far off we could see expansion occurring given that a species that emerged 4 billion years ago as a GC needs to be less than 4 billion light years from us for us to have seen that emergence; Indeed, a good deal closer due to Hubble expansion, but Hanson et al does it much more rigorously and mathematically than we did and it's very nice to see this properly modeled at the intergalactic scale. And as is normal for these lines of reasoning, it is assuming that faster-than-light travel is not possible, which bugs many folks, but as a reminder, FTL only exacerbates the Fermi Paradox, as it means that a civilization appearing 10,000 years ago in the Andromeda Galaxy could already have colonized our galaxy, as opposed to the multi-million year journey otherwise required.
You should be able to run this same model at various superluminal expansion speeds and the result would just be to expand those individual GC bubbles and thus imply GCs were even less common. Now, in a lot of the discussion around Grabby Aliens I see folks bringing up how it mentions that our star and planet have a much shorter window for life than normal, with the majority of stars being red dwarfs that can live a trillion years or more, and with our planet likely to become uninhabitable inside another billions years or so. Channel regulars know we’ve discussed ways to extend a star’s lifetime, and that we wouldn’t expect GCs to really be bound by star lifetimes or types much once they got rolling, but honestly I didn’t see the relevance of star lifetime to the discussion at all. We don’t know how likely life-bearing worlds are to form around red dwarf stars compared to yellow ones, but, longer-lived stars being able to spawn their first civilization far later, say 100 billion years after forming, really would only matter if it turned out that it is less likely for intelligent life to form on worlds around those longer-lived stars than it is for intelligent life to form on worlds around yellow, orange, or white stars.
We could stretch that, our Sun has been getting hotter every year we’re around it, as stars do, growing slowly brighter over time, and thus the habitable zone of a system will change. Red Dwarf’s planets really don’t have this issue, they are slow to age and brighten, but it might be that stars much bigger than our own sun, while likely having far more planets and bigger habitable zones, may brighten too fast to allow life to be sustained on them, so that even stars that live 5 billion years couldn’t have intelligent life emerge on them even though our own sun only is 4.6 billion years old. The obvious problem of course is that, at no fault to the authors, we just don’t know much about alien life yet, and we probably will get to find some examples of unintelligent life before we come across what is indisputably-intelligent life or other GCs. Indeed, under this model you would expect to have the leading edge of your GC run into more primitive worlds or even smaller proto-GC bubbles far more often than it runs into full GCs, as it still works under the usual assumption that the time between now, and us colonizing every star in this galaxy, is on a similar order of magnitude, as how long since we discovered fire. A million years to go from fire to rocket ships, and a million years to go from them to galactic dominance. To me, that’s a bit more permissible for the Zoo Hypothesis, as, first off, we are not really assuming any real relation or rapport between elements of a GC besides maybe locally within a few hundred light years.
Nobody really sits around waiting thousands of years to get an email back from all the mutants living in their tiny corner of the galaxy for instructions on how they should handle this new planet they found with smart slugs on it, and by that point in time there would probably be accepted branches of humanity that bore less resemblance to you than those smart slugs. In that regard, I don’t think expanding GCs are necessarily bad for younger worlds near them. In a system with many competing and rival interests – like a GC or even a small piece of its expanding frontier – extermination policies don’t work well, on top of being monstorus. Nothing can be
done in secret, and now all your neighbors know you’re a bunch of genocidal maniacs and even if you’re all derived from the same homeworld and that initial spacefaring civilization was entirely fine with exterminating aliens, you’ve diverged beyond that and you also know the universe has other GCs at that point and have to worry about them knowing what you did when you do inevitably run into them and also while knowing that you cannot do it to them quite as easily, since they will likely to be expecting it even more so if they’ve seen you do it once before. In a non-FTL and expanding universe, there’s no real winning percentage for an eternal expansion into your neighbors at the Supercluster scale, even if you began with a billion year headstart over them. So, I don’t see rolling over other younger worlds and burning them really happening universally, and more importantly, I just wouldn’t expect a single GC to have a universal policy on encountering primitive worlds, let alone every GC having an identical policy. Probably just absorb and add into the local interstellar ecosystem, and if there’s a tendency for GCs to go post-human, or post-alien and inorganic, then it's entirely possible nobody would care if their neighbor had evolved from smart slugs rather than a monkey like them, just one more icky option they don’t like to dwell on. For the math-inclined I really do recommend the paper, and I want to emphasize that you should not be taking away from this that Hanson et al have gone and proven it is possible to travel at 80% of light speed economically or that there are 7000 intergalactic empires out there, and that roughly 4000 have already emerged and we’re next. That there’s a little under three
thousand whose galactic capital homeworlds are busy somewhere between developing photosynthesis and discovering fire, while some of the older ones already have empires a billion lightyears wide. Relatively minor changes to the numbers or additions of other parameters, or more or less hard steps could alter the dynamic a lot. And for all that it's essentially extrapolating off the Anthropic Principle, the Mediocrity Principle is still in a lot of the basic concepts and it seems rather normal that we’d end up in the middle of the distribution like that. Each GC presumably has run this same model and gotten similar results, adjusted a bit for the age of the Universe when they ran it, whether they were #1 or 4000 or 7000 in truth. But if the thousand oldest and biggest GCs had already settled 10% of the Observable Universe between them, then that means they’d average 20 Billion-Billion Stars each, and given that we were talking about the Kardashev Scale last week and I’m sure that’s brought up the idea of K4 civilizations, it always does, we are going to go ahead and semi-officially dub a GC of this size, a K4 civilization, which is a bit ironic since again GC is short for Galactic Civilizations and those are the definition of a K3 or Kardashev 3 civilization.
Maybe we should call them an Average Hanson Bubble or GAC, Grabby Alien Civilization, K4, though if anyone has a better and appropriate acronym, let me know in the comments below. In the meantime, while again, it’s not conclusive proof of anything, and we already mentioned some exceptions and have discussed other ones in other episodes, I do think it shows us a fairly plausible model of what our Universe and its civilizations look like. And it’s one that (on paper) almost guarantees us at least a few galaxies and trillions of stars to ourselves, if not a few million galaxies, which is about as far as we can probably expand at this point in our Universe’s age and expansion. The one big take away from the model though, and the Rare Intelligence solution to the Fermi Paradox, is that us and any other alien who’s smart enough to be able to look up at the night sky and see stars and wonder what they are, has a good chance of becoming the ancestors of a civilization which will control every one of those dots of light, even the smudgy ones that represent neighboring galaxies. That if they proclaim that all that they can see is theirs, then they will turn out to be right if they only put the effort into earning it, and aren’t so foolish as to destroy themselves first. And if they do, there are many worlds waiting in the wings to rise and take their forfeited place a few million years down the road.
And hopefully, we won’t forfeit our place among the stars. So today we saw a theory that makes a strong case for humanity having at least as many worlds free for us to claim as there are living humans, and indeed many times over, even if that future is still centuries ahead and trillions of miles away. Thankfully, we already have endless worlds at our fingertips that we can behold through the mind’s eye, and one of our great assets as a people is our ability to share those with each other. I think it is easy to forget sometimes just how awesome it is to be able to share stories and tales and imagined worlds with each other, and I was reminded of that recently as my gaming group had been on hiatus for some months and we finally had a chance to start a new campaign.
If you’ve ever had occasion to roll some 20-sided dice around in some pen and paper RPG or their computer and online version, then you probably know what I mean. And those shared settings can become amazingly complex, the game we just restarted returns us to a campaign setting we’ve been playing on and off again now for almost 20 years and I was very glad to have World Anvil, the award winning worldbuilding toolset, to help. It made it easy to transfer our various remaining notes in a dozen different formats into a collected and easily accessible format. It was also the first time one of our players had seen World Anvil in action and they said almost word for word the same thing I said when I first started using it, “I wish this existed when I first started gaming”. I love the ability to upload many different setting maps and link them together along with articles, whether they’re text and images I’ve carefully written or even just a photo snapshot of some old notes I had but don't have time to retype and reformat, it just makes it so easy to tie all my notes together and add more and do it in a way that’s not only easy for me to work with, but easy for my players to navigate through, and easy for me to lockout materials they shouldn’t be able to see yet or which only certain players should see. And while its highly versatile, world anvil also makes lots of standardized gaming needs like timelines, genealogies, map-connecting, character sheets, and articles on people and places very easy. But the newest feature of World Anvil that
I loves is their whiteboards, it's a canvas for connecting articles, diagrams like family trees or hierarchies, tavern boards, mindmaps, and character arc development. There are just so many features that make worldbuilding and sharing easier and better, with world anvil, but the very best feature is all their easy online tutorial videos that not only show you how to do everything, but give endless ideas for uses you might not think of on your own, that will just keep making your setting better and better. With world anvil, you can elevate your setting, whether it's for a game, podcast, or a novel, to a whole new level. If you would like to give World Anvil a try and let it help you forge new worlds, just click the link in this episode’s description! So that will wrap us up for September and the eighth year of SFIA, and we’re heading into October to start year 9 by looking at the idea of colonizing planetary rings, like Saturn’s, and if life might be able to evolve in such places. Then we’ll ask the question of what we do if all these options for space travel never pan out, and we are stuck here on Earth, and then we’ll have our Scifi Sunday episode to look at what happens if space travel does indeed pan out and what strange alien environments we might encounter.
Possibly on your own Personal Spaceship, which we’ll look at on October 20th. If you want alerts when those and other episodes come out, don’t forget to subscribe to the channel and hit the notifications bell. And if you enjoyed today’s episode, and would like help support future episodes, please visit our website, Isaac Arthur.net, for ways to donate, or become a show patron over at Patreon. Those and other options, like our awesome social media
forums for discussing futuristic concepts, can be found in the links in the description. Until next time, thanks for watching, and have a great week!
2022-10-03 05:02
