Robin Hanson, George Mason University: A Simple Model of Grabby Aliens
2:28PM Feb 9, +0000
Speakers:
Keywords:
civilizations
galaxies
power
point
aliens
universe
question
robin
model
expanding
early
grabby
key
gravity
stars
diagram
deadline
assumption
volumes
cosmological constant
Welcome to this is now the third meeting of the intelligence cooperation group. We're doing something a little bit different this time, which is brought about by Robin having had a eureka moment lately, and being willing to share that with us and all of you. So I'm really, really, really happy to have Robin here. This is going to be a little bit of a of a satellite topic to the rest of it, even though we will be discussing with Andrew Sandberg, who I think just joined here as well. And later this year, also how to cooperate best with and with aliens and potential future civilizations. So perhaps we can already we can already preload some of that to today's discussion. Today, we have Robin Hanson on to discuss his simple model of grabby aliens. I will post a little bit more info about Robin in the chat and about the model in the chat to not take away too much of office airtime. And are we meeting more people as they join em, but Robin, if you want to start already, then please feel free to do so.
Alright, somebody put a thumb up if they see the screen. Alright, so let's get started. So as Allison indicated, I basically had this eureka moment in the middle of December, and I've been working crazy on it. And I think this is going to be one of my main accomplishments in my lifetime. And of course, you don't have to agree, but I want you to hear it. So let's see where it goes. The This is joint work with several co authors listed here, some of whom are in the room, and could be also the alliance of questions. The main place to go to find out about this is now grabby aliens comm where you can get the preprint and videos and other sorts of things. All right, so let's get started. Okay, so most of you have heard through most of your life about what I'll call quiet aliens. These are the aliens in Star Trek. And in most science fiction stories, these aliens are the ones estimated by the Drake equation, which is very hard to estimate things which gives you huge uncertainty. So what are these pilot aliens? Well, on a cosmic scale, they don't last very long. And they don't do very much and they don't expand very much, which is why it's so hard to say anything about them. It's why we don't know so little about them is because they're so damn quiet. That's the idea of quiet aliens. So I decided instead to analyze what I'm going to call loud, aliens. Big, loud, noisy aliens. Well, I'm not going to give them the time grabby. What's the difference between loud aliens and quiet aliens? Well, loud aliens. You can see in this diagram in this world, that little quiet aliens, we start with this little yellow origin of the loud ones, and then they just keep growing, they don't die soon and go away and disappear. They keep growing and making a visible difference to what they can see. And key ideas. Even if there's only a very tiny number of loud aliens in the universe compared to these quiet ones, they would make a disproportionate impact, eventually, you would notice them eventually, they would seem different. And so I wanted to analyze loud, which I'll call grabby aliens. I'm having trouble getting this. Okay, so all right, there we go. So I'm going to show you a few example, stochastic samples from this model in the next few slides. And then we're going to get into sort of where all the math comes from and how I set everything. But first, let's look at some pretty pictures. So this is in one spatial dimension in one time dimension. And as you can see, they start and originate at random points in space time. And the rate at which they appear gets faster in time. So you can see later on, they appear to pop up very quickly. And then eventually, relatively quickly, all of space is taken up so there's no more place for them to appear. And that's a deadline by which any of them must appear before you know if they are to appear at all. And that's purported to explain why we are so early as humanity and universe's history, which I will try to show you in a few minutes. These diagrams, as you can see, basically have these grabby alien civilizations spread out at a constant speed from when they originate. And then when they meet, for convenience in these diagrams, I have them retaining their original volumes without enormous conflict and war and they're not taking over each other volumes. That doesn't, isn't really a central part of this. That's just the easy thing to do in these diagrams. You know, a simple peaceful scenario here. But the key point is that once something has taken over these volumes, you know, one of them has it Which one isn't really very important. Once a gravity civilization has taken over the volumes, then some gravity civilization from that on controls them. Sorry. So this is a one time dimension to spatial dimensions simulation. And so now Time goes down into the diagram. And so you can now again see that these things start out at points and spread out as cones here. And again, they fill up the volume relatively quickly. And there's more of them toward the later part of the time, because the probability will go up, as we'll see. And basically, it all gets filled up. And I have, just as of yesterday, I'm moving now to show you
and keep on the show me will be okay. And
it's spacey music. So hey, but you'll see a sphere of activity and the sphere will basically start out with small numbers that grow and then they all fill in. And it's rotating just so you can see different angles. And again, for convenience, we assume they don't like fight and take over each other's areas. And will again see it now but now with a transparent volume and somewhat so you can see somewhat through them to see what things are happening behind them.
And as it's been going on, there's we've had the diameter of this volumes have been changing due to the expansion of the universe, the number of galaxies in the volume has not been changing due to the expansion of the universe. We show you how many civilizations there are in this that have originated in this volume. And we have some parameters we're going to discuss later. So that that's the key idea here of this simulation.
And
that's all pretty and of course you're nodding and going Yes, but what does this mean? And where's the where's the math come from? And that's what I'm about to explain. All right, so
stop that.
Next slide. There we go. All right. Now. So now you got an A slide with words, numbers, and not so pretty, right? Okay, so the key idea here is got a simple mathematical model to describe grabby alien, this model has only three free parameters. And we're going to get all three of these free parameters from data. So we're gonna fit a model of where the aliens are and what they're doing to data in the sense that we're going to be able to say this is a data driven estimation of where these kind of aliens are. Now, again, it's not the quiet ones, the loud ones, but maybe this is the important part, because it'll be the ones that you should be worried about meeting someday. quiet ones aren't really much of a problem because they go away. Alright, so three free parameters. So basically, first of all, there's a power law on time as to K to the N. And that's that they arrive in time with power loss, that's that over time, they arrive faster and faster. And we're going to explain why that is. So the two key parameters, there are the power n and this constant K. In addition, there's the speed parameter, the rate at which they expand. And those are the three free parameters of a model. That's it. So now I'm going to tell you briefly where the numbers come from. But then later in the talk, you're going to see this in more detail. But just to give you the idea right up front, first of all, the rate at which they appear the overall constant at which they appear, that's going to be set by the assumption that we might turn grabby in our future. That is, it's not clear that we should or that we will, but there might only be one in a million chance. But the idea is, if it will happen, it'll probably happen sometime in say the next 10 million years, not a very short deadline, but on a cosmic scale, plenty short enough. And so if we might become grabby in 10 million years, then our current IP space time event is a sample from the origins of brevi civilizations. Now, we might be at the beginning of the distribution, we might be toward the end, we don't know. But we think there should be a uniform prior over that in the sense that we are equally likely to be any of the different origins of gravity civilizations we are at the date at which could be one of them, we don't know which. And so that's going to set our estimates of this constant K. The assumption that we are a random sample from gravity, civilization origin dates. Secondly, the power law is determined by the process of evolution which has sequence of hard steps that has to go through. I'm going to discuss that in more right now in a minute. But the key idea is we can look at Earth's history, and the timing of events in Earth's history, at least for a crude indication about how many hard steps there might be in the process of evolution. And roughly somewhere in say, the range of three to 12. That's a plausible range. And so that's going to set this power, which is how quickly things start to Russian appear toward the end. And again, I'm gonna discuss that a lot more detail. But that's the second parameter, the power, the constant comes from our date, the power comes from the details of our Earth's history, and when things happen. And the third parameter is the rate at which they expand. So that's a free parameter. But it turns out that the model predicts that say, at a random appearance date, like ours might be roughly 40% of the universe, volume has already been taken over by these gravity civilizations, we assume that they make a visible difference, because that's the kind of thing we're looking at loud and dif something you could see. And so you should if you were looking at from a random position in random directions, you should be able to see that now. But we don't see that. And the reason we don't is arguably a selection effect, that is, if they are going fast enough, then most of the volume in our backward light cone, if they had started there, we wouldn't exist, they would have precluded us. And therefore, because of that, there's a selection effect. And that selection effect gets stronger as the speed goes up. And so that allows us to conclude that they're probably expanding at a speed, say half the speed of light or faster, because of that data and of the fact that we don't actually see anything out there, and they would be really big, far larger than the full moon. If there were gravity civilization volumes in the sky, they would be really obvious is the point. And we don't see them. And so that's why we must conclude they expand quickly. So that's my quick outline of the three parameters, and how we set the three parameters. And now I'm going to sort of go through the logic in more detail here.
Alright, so to start with the key concept here of luck picking. And so this is a metaphor for evolution. But let's, you know, stick with the metaphor for a moment. The idea is, imagine you have a one of these 24 terrorism shows, and in order to save the world, you have to pick a bunch of locks, and you've only got an hour, you've got to pick a whole bunch of locks in and say you just have to do it at random and the probability of succeeding is low, your odds are the world will die and you won't save everybody. But hey, you're gonna try anyway. And so each lock has some expected time it would take to pick that lock. And say, for most of them, it's longer than the time you have. So you know, a lot might be a one hour lock and a 10 hour lock and 100 hour lock. And you still gotta sit there and try them. And we're interested in this scenario where you get lucky. And in that hour, you managed to figure out all the locks. So that's the lock picking scenario. And in that scenario, we're going to call a lock hard if the average time to pick that lock would be longer than the window we have here. And we're going to call ourselves success if we nevertheless, manage to pick each lock. And the sum total of all the time it takes to pick locks is less than our deadline. And it turns out to be a hard statistical fact that you can derive for yourself if you have doubts, but I'm not going to walk through the derivation for you here that conditional on success here picking all these locks and this deadline, then the duration for each lock to be picked turns out to have been drawn from exactly the same distribution, no matter how hard these are, as long as they're all hard. And that's also true of the remaining time that you have at the end before the deadline. After you've picked the last luck. All of those are going to be drawn from the same distribution. And in this, in addition, the chance you have of succeeding by this deadline goes as a power law of the time that you had raised the number of locks. That's this T over k raised to the n. So that's the key power law that I've been talking about here. And the key idea is that evolution on earth, and wherever else that happens is like picking locks. That is, as you may know, there were lots of hard stuff that had to happen on earth. And Earth had a deadline and that sense that apparently, according to our best estimates, roughly within 1.2 billion years from now, birth will become too hot and inhospitable to fragile life like us. So all these things happen to have had to happen within this window. A lot of them are complicated, and we're not sure exactly which of them are which kind of hard steps in which way but at least two of them seem the clearest example of clearest to clearest examples are the very first period between when life was hospitable at all on earth, and the very first appearance of life that seemed to be about point 4 billion years. And the remaining period we see now roughly 1.2 billion Yours, both of those, according to this theory, should have been drawn from the same distribution. And because of those two distribution just plugged in an equation and say roughly, that gives us an estimate somewhere between, say, three to 12, hard steps happened on earth history, and evolution reaching where we are now from the beginning. Now, it could be that this power that actually is relevant is larger than three to 12. Because in fact, there was panspermia and other stuff happened before Earth on some other eating. And then life moved there. And then there was more time for more hard steps to happen, in which case the power would be larger. But we're fine here for the for our purposes to say, roughly three to 12, a and middle estimate of six hard steps happened in the history of life, which would then mean this power has a power of six. That's the power law that we'd have to be playing. And so the key point is to remind you, if you look too lightly, what a power law looks like. Power laws really concentrate later on. Right, so here are powers from one to 10. And as you can see, the power of ones increases gradually. But the power of 10 is very low at the beginning and all of a sudden boom. And so this is the dependence at which say, gravity, civilizations would appear. Because they're appearing accordingly as parallel for a long, long time, the universe is just empty and nothing's happening. And then they start to pop, pop, pop, pop, pop pop up, and then toward the end, they're popping very quickly, except time runs out, because the volume fills up, and there's just no more place for them to pump. Alright, so
let's talk a little astronomy. Because we need to actually think more in detail about are we early. So to set up the discussion for are we early now we have to ask, Well, when things should have happened, and of course, the key things that had to happen is first stars had to appear. And then they had to have planets around them that might have been habitable. And then those planets would have had to have durations during which this power law grew and evolution, various sorts of locks were picked and things happen. And then at the end of that period, on some planet, there was some success. And the key point is there's this literature on what's called a galactic habitable zone literature, which in part looks at where but in for our interest happens in when when these things would have been habitable. And the key point is that this first blue curve shows the star, a typical star formation rate, which peaks about 4 billion years after the Big Bang. But during this early peaking, and when all the stars happening, apparently they were, they were blowing each other up all the time. They were creating supernova, and gamma ray bursts, and all sorts of things were happening, boom, boom, boom, right next to each other, making it very difficult, apparently, for life like ours to actually live on those sorts of stars. So the habitable zone literature tends to find that most of the early stars were inappropriate for habitability. And so this gray line near the bottom is a best reconstruction of what the typical habitable zone literature finds, maybe peaking around 12 via years, not too much before our moment. And that means like most of the stars were in, you know, we're just too crowded and too much stuff going on. But we can have the simple functional form. Sorry, the vertical is just a rate. Just, there's no absolute scale on the vertical, it's just how fast things are happening. So as you can see, the peak, say above 1.4 is just lots of stars were being formed at that date, of roughly 4 billion years after the Big Bang. And by now, of course, most of the star formation rate has drastically declined. But now there is also far fewer gamma ray bursts and supernovas and things like that, so that stars can actually have planets around them that don't get blown up quickly after they form. And so this is, we're going to basically rely on this black hole habitable zone curve, the gray curve here to be talking about when planets like ours could have formed and could have created life. So now we get to a key equation, sorry, it's mathy. But I'm not going to make you solve it. But this is the key equation that would represent calculating when we should appear in history. So the alpha here is the chance per unit time that life as advanced as ours might show up somewhere. And it's going to be an integral over time of from the origin of the universe up to that moment in time. And it's going to be an integral over this variable x, which represents when a star was born. So we're going to integrate over all the different stars that were born at different dates. And we're going to ask for each one. Well, how What was the rate at which habitable stars were being formed at that date? And then we're going to ask well, and what's the duration from that date until this time t that we're asking about, for life to have originated. So we've got this x to the n minus one, that's the p power law here that says, the chance of succeeding and picking these locks goes as a power law. And then we've got this row, which is the habitable star formation rate. This is the curve we were trying to look at on this previous graph showing when stars were forming. And then what we have on the right term is this h L, which is a CDF distribution function over lifetimes of stars and planets. And so basically, there's a lot more smaller mass stars, the larger mass stars, and the smaller mass stars have much longer lifetimes.
And so that creates a lot of interval in there for larger lifetime stars. Now, you know, it turns out that many people have made complaints suggesting that maybe longer lifetimes, stars are not good places for habitable planets to be. And they've talked about title locking, they might have more solar flares that might come from the larger from the smaller stars and things like that. So we're going to try to represent the possibility of disagreement here through two ways one is going to, we're going to set some sort of maximum lifetime parameter. And we're also going to have this possibility, but we're going to call a mass favoring prior, where we say, well, maybe there's just a weight we should give to the larger masses. And we consider a cubic weight as a possibility there. So but this is as much of the math I'm going to make you look at. And now I'm going to show you the results of this sort of calculation. So this is an example diagram, I have a bunch of other variation diagrams to show you if you want to ask me about them. But this is a nice big, sharp diagram. So here's the key two dimensions here. on the y axis, we have this key power. So remember, best estimate about six, maybe a range three to 12 is roughly what the power would be looking at, or we're going to show you a wider range of things for that, on the x axis here is what is the maximum lifetime that a planet is allowed to have and still be considered habitable. So this starts all the way on the left with 5 billion years of basically, Earth's lifetime. So there are definitely stars out there that lasts less than Earth long and even shorter, but we're going to assume at least Earth's lifetime is long enough, because hey, here we are. So that's why we cut it off on the left at five. And we go all the way over to the right. And the right is basically what we know to be the maximum stellar lifetime out there, which is 20 trillion years. So there really are stars out there right now that will last 20 trillion years, and therefore the planets around them will also last 20 trillion years.
Okay, so now
in this space, the key message is G, the numbers in that diagram are really low, unless you go way over down to a low power and over to the left for a very short lifetime. That is otherwise we are very early. This is basically an earliness graph that shows how early we are in this distribution, depending on the parameters you're assuming here. And so you can see that say for a power of six, even at twice the Earth's lifetime of say 10 billion years, you're still at a 10% earliness. And you know, you get over to 1%. If you go to twice that long in terms of a planetary lifetime, that could be like what it could last and you need to go down farther and powers relatively unlikely powers according to the history of life on Earth, before you really can save, allow you to loud it to have long, you know, a substantially longer planetary lifetime and still not say we're crazy early. So this is a diagram for the for some other combinations is that combination of the 12 billion year peak and when the Galactic habitable zone, it's also for in middle decay rate of a decayed every stars form and then decay every 4 billion years, we can switch that to two or eight. I can switch this mask for everyone. klaris does have some other graphs to show you. But basically the key answer is this doesn't change that much no matter how you change these other numbers. And the key story is we look early. Humanity looks early in the history of the universe. And our explanation is going to be because there was a deadline because grabby aliens were going to come along and grab everything unless we showed up. And that's why we are early because everybody who shows up in time has to be early relative to this, these calculations are assuming that the universe is just always going to remain empty and it's going to be there for you whenever you're ready to evolve and finally show up at the end of the power law. But if that's not true, then that's why we're early. Right? So I'm sorry, let's um so as I indicated before, Typically, the fraction of the universe that's controlled, you know, as you know, as civilizations show up in time is, on average, say over a third, which you might think is puzzling, because hey, we don't see anything. And here is a graph showing you four different speeds, say going from 10%, or, and, or to 50% of the speed of light, sort of how big is the biggest civilization you would see in the sky. And so what you can see is, the full moon is the red line. So anytime you see anything, it's almost always much bigger than the phone. So this is not at all an issue of trying have bigger telescopes to see farther away to maybe see grabby aliens, this is just not remotely the case. If you would see them, they would be huge in the sky, they would be enormous you would couldn't miss them is the key point here. And this is the key graph we're going to give you for doing up the your Basie and updating, because this is the likelihood ratio in the two parameter space. So the two parameters are again, this power, remember, six is a reasonable middle number. And the other power parameter is what is the rate at which these gravity civilizations expand? Obviously, they probably want to expand as fast as they can, and how
fast can they expand, you might
think expanding even a 10th of the speed of light is just like maybe impossible, maybe that just can't be done. But here we're going to look at so you'll want to have some prior in your mind over this two parameter space. But we're going to give you the likelihood ratio to update your prior with to give you your posterior. And this likelihood ratio says Well, what's the chance that if you had this parameter combinations that you would see zero aliens in the sky, out it grabbing civilization origin date, and you can see that say, we can look at the point one curved line on here. And you can see that you know, four speeds of Route above a third of the speed of light for many different powers, there's a 10 to one ratio of the chance that you wouldn't be you know, 90% of the time, you wouldn't see one and only time for sign some of the time, you would not see one, which would make us pretty unlucky, you know, pretty lucky or rare to have been one who happened not to see one. So basically, this likelihood ratio is saying unless your prior is really strong for low speeds here, you kind of got to conclude that they're expanding at a high speed, because that's what the likelihood ratio is pushing you to believe. You got to believe that they somehow can expand at over half, maybe over three quarters of the speed of light, that looks technically difficult source. But remember, we've got another 10 million years, maybe before we would get to this stage and actually be doing this wholesale. And so sometime in the next 10 million years, arguably, we might figure out how to do this, even if we can't figure it out today, it's got a long time that we can do that. Alright, so I want to sort of walk through the selection argument. So here is in spacetime, us now. And here is our backward light cone. And so we see all the events along this red curve, that's the those are the events that we see from here. Now. Everything within this yellow region are the origin events, where if an alien civilization had originated there, then they would have expanded and taken over our current spacetime event. And therefore we could not be at a potential gravity civilizations origin event now, because they would, they would have precluded that by the assumptions of gravity civilization. So we can only see something that shows up in this green region where they it's between the speed of light and the speed of expansion, and we just happen to see them in that difference. So if they have an origin there, then we might see them and we look in the sky. But when we look in the sky, we'll see something really huge, it won't be tiny. And then we show where on the diagram is when we'd eventually meet them, you know, some of our descendants meet some of their descendants. Alright, so all I have left to do now show you and the rest of this talk Asides from doing q&a is just show you some diagrams of things we've noticed. So the key idea here, remember is there's only three free parameters. There's the power and the power law, there's the constant of the power law, and there's the speed of expansion. We're fitting each of those three parameters to data. That means we can now give you data derived distributions over all the questions you could want to ask about gravity aliens. And so that's what I'm about to show you now is these sorts of graphs and distributions, right? So here's the first one. When do they show up in time. So on the x axis is sort of the time in the universe's history. And the the point at which all these things are crossing is our current date, because we're basically making an assumption that we were randomly sampled from origin dates, and that means we might be really we might be light but basically, they're all going to line up at that date. And so you can see as the power goes up, they get crammed more and more to a narrow range of time and when they show up for a little Power, they could have shown some of them showed up really early and some of them show up really late. But the higher the power, the more they all show up at the same time. This is when we might meet them. And so so now you can see that, mostly, we won't meet them for maybe a billion years. But the higher the power, the more they collect and show up at the same time, and so the shorter in time until we would meet them. And so you know that and there is always going to be a small chance where we might meet them very soon. As you can see in this simulation, some of them end up showing up at the last minute, and they're show up really close to each other. And so the distances between those aren't so big. The, this is one of the one graph that I hope to have a better version of soon, but I haven't quite gotten it yet. But so I want to have that whole distributions over this. But all I have is averages at the moment. But the key idea here is that these things are huge in traditional, you know, quiet alien terms. Basically, we're looking at grabby alien civilizations that encompass many millions of galaxies each on average.
And so this means that say, if there's a one in a million chance that a symbol that a non gravity, civilization turns into gravity, that means there can only be a million times more, you know, quiet aliens than there are the noisy ones, in which case, each of the quiet ones still happens less than one per galaxy, ie they're pretty rare. Because again, on average, these even gravity civilizations end up controlling on average, hundreds of millions of galaxies each. So we're talking very large spatial scales, we're not talking the aliens next door that you might be in Star Trek, the number of aliens you would see again, as the speed goes down, you end up seeing a lot of them. Because you know, they're at any one random time the universe is full of them. But at least when they're arriving, you can calculate when you would be until you saw them. If they're expanding the speed of light, you won't see them until they get there. But if they're expanding lessons speed of light, then you could see them before you finally meet them. And this gives a distribution when you see them. This is a whole bunch of these graphs all together. That you can see all at once in browse here. And so I think at this point, I will stop and take questions and comments. And I have some other graphs in reserve. Should you have questions about those. But let me therefore stop the screen sharing temporarily here, although I can bring it back. Answer questions, but let's hope there is interest in discussion here. And let me ask Allison to take charge of how you want me to do this.
Thank you, Robin, thank you so much. Okay, there's a lot to discuss, I think and we already have a queue of I think about five to six people waiting. And I'm guessing that we'll have more as we continue. And I would love to ask everyone who's asking a question, just give a brief background of either where you're working at or something that pinpoints you, Robin, how are you coming at that question? Even if he knows you, not everyone else in this chat knows you. So give some context. Okay. Okay. Alan Carr, we have you first. Oh, yeah,
I have a GPA, excuse me a PhD in astronomy, but I haven't done any in 40 years. But I still have that as my background picture, if you can see the galaxy there. In 1982, I attended a triennial International Astronomical Union meeting, where there was a full day session on finding intelligent life in the universe. And at the end of the day, Frank Drake gave the closing talk. And he put up a slide during that talk, he put up a slide with a big red.in, the middle. And he said really interesting day, folks. What I've concluded is what you're doing is arguing about the slope of the straight line through your data point. And I'm wondering, in what way, are you not doing the same?
Well, as I've said, I have a model with three few parameters. And each one I'm fitting the data. So I am not just making up a model and saying, but there is no data there for I'm just gonna make it up. That's the whole point of this is, I've given you three parameters, each of which I am fitting the data, I'm showing you the consequence of fitting back to actual data. So to remind you, what data do we have, we have the data of our origin data at the moment, we are at a date, it has a particular date. It's not a unknown date in the history of the universe, and we can calculate things relative to that date. Secondly, we do not see aliens at the moment we look out in the sky and we do not see them. That is a data point. You might not think it's surprising until you reflect on it but you should be a bit surprised because there you see nothing out there and maybe you should. The third data point is a set of data points which is the history of life on Earth. history of life on earth is a structured data, it happened at different times. And that's the process by which on here, life went from something very simple to something very advanced. How hard was that? How many steps there were there, the history of life on earth is big clue to that. And elsewhere in the universe, if they're going to reproduce our success, or you can go farther, they have to go through some set of analysis steps, and therefore our data is relevant for predicting what they will do.
I wish that that thing had been that session had been now because it would be online. And I could find the arguments, I recall, several of the things that you talked about at least one or two of them as being part of that day. But my notes are lost to history. And probably the sessions are too. So thank you. But we want
from you your memory of what their critique is, what's wrong with
the argument? One word, one of the biggest ones that struck me was it may not be possible for life to survive interstellar travel, that there's too much radiation, too much or too many hazards for it to be possible.
That's a common model. But then you have that side. But then why are we so early? So our earliness is the key data point that I am using to argue for why we would need to resort to a model like this. And of course, look, stuff seems to move around the universe. We've seen asteroids and stars and you know, there's definitely stuff that moves around. And so it's mainly a matter of how fast you can move. And you know, how reliably But definitely, we seem to have stuff moving in the universe.
Oh, yeah. And then one of the other arguments was that you fill up the galaxy and 100 to 100 million years, which is instantaneous. So again, we're early, but it's not millions of years, again, to fill up the galaxy.
But so the model here is the power law. So I'd say that the key thing that the literature so far has neglected is this power law, the power law is very well established statistically. But the Galactic habitable zone literature is simply neglected to use or include this power law, this power law directly predicts that you'll have a long period of not getting filled up, and then a rapid period of filling up as the power a lot, you know, cranks up?
Yeah, I don't it's not an argument. I remember hearing before. But as I said, it was a long time ago.
All right. Thank you, Alan. And next up, we have Tom.
Yes, I was thinking that your your graphics were basically showing civilizations growing hugely, for the most part. But isn't there an effective limit on the size of a civilization on Interstellar terms, based on the communication lag from the speed of light max maximum, it would seem that after you get to a point where communication takes a certain amount of time, a civilization would just naturally split. And that is no longer relevant to hear what someone thought 20 years ago, and they can't get to you anyway.
So the word civilization isn't intended to convey very much of a constraint and our story here, there's an origin, and then things grow from the origin. And then they go out and they fill up. And we're making no presumption about any degree of coherence or governance or structure within that region, the key thing is just that they are taking control. So you know, similarly, life on Earth, spread and took over the earth, Earth up until very recently did not have global coherence, or global control or global communication. Each, you know, spore or bacterium in one part of the earth did not effectively communicate with other parts of the earth. Nevertheless, they did fill up the earth. So we're talking about a scenario where they fill up and take over and control. But we're making no assumptions about coordinating or, you know, communicating or doing anything interesting across those timescales.
So by civilization, you effectively mean a species of life form as it were.
Right? Like the key constraint is they fill it up, and then if you show up too late, they have taken over and you can't, from the point of view of the model, that's the key constraint. You have to show up before because otherwise, if you show up after it was too late, that's that's the key thing the model is doing.
Lovely, thank you, Tom. Next up, Anders, and please, perhaps a sentence or so about your background to to give some context?
Yep, under Sandburg, a future humanity Institute at University of Oxford, so I have been looking at the Fermi question and expanding civilizations for some time, and they also recently had a paper out where we tried to do it by ECM fit of the heart steps model. So this talk is totally right up my alley, and I really like it because I've been doing very similar things but Robin, the his core precept, kind of taking it to the next level. There is only one thing that I'm really annoyed with that is that claim about the full moon. I think that's wrong. And I think you should look into the Steven J. Olsen's papers about rapidly expanding civilizations. Because when you try that, when you try to look at the angular size, you need to do a full relativistic treatment. And the effect is basically, if you think about that 2d diagram, when you have two phones, that the intersect draw the closer you get a very elongated teardrop trade. So basically, when you see the gravity civilization coming towards you, you're mostly going to see the closest Edge of the World, our life is going to be rather remote, it's going to look way smaller than it actually is. It's actually almost upon you, but it looks like a very tiny pinprick, I think you just need to adjust. That's my only complaint about it. Otherwise, I kind of have very much fun.
So. So you're you're picking on a point that I didn't elaborate in this talk, but that we do elaborate in the paper, again, we have a archive paper that you can find, you know, from grabby aliens.com. And that paper does have a cosmology section, where we try to get this stuff, right. So let me elaborate a bit because Anders gave me the excuse here. So in our simple cosmology, things, you know, show up at points and then they stay at points. And then they grow at constant speeds relative to those points. Now, as we know, the actual universe is expanding. So in order to have our model apply to the actual universe, what we do is a change of coordinates. That is, we have our model space really be describing conformal time and co moving space. And in conformal time and co moving space. In fact, our model directly applies in the sense that things show up at positions and they stay at positions. And interestingly, we'd argue that the constant speed assumption actually makes sense in terms of the actual co moving space, uncomfortable time, because what we expect to have happen as a kind of expansion process is that in general, something sends out probes some speed, which then lands somewhere, produce, you know, grow resources locally to reproduce, and then send things out again, so that the actual expansion process isn't like some sort of momentum process or kinetic process, we just throw something out that and then drips forever until it hits something, the process is going to be more like how an airplane flies, an airplane flies at a speed relative to the air it's in because the engine is pushing relative to that air. So if the air is moving faster, the plane moves faster, or vice versa. Similarly, this spatial expansion process that we're postulating here is a process relative to the local co moving stuff in the universe. it lands near stars and grows and then moves out, etc. So that, in fact, what we do expect to see is a constant speed relative to co moving stuff. And in conformal time. And so given that assumption, in fact, our coordinates we do we say, do apply that as our model applies exactly in CO moving space, and conformal time, as long as the universe's expansion continues to be a power law, which unfortunately, in the last few billion years, it has not been. So the major way in which we are an error is that deviation where that is, until recently, the the, basically, the scale factor has been growing as the power two thirds of time. And in that period, then this has been a exact equation. But you just have to do a power law transformation to look at the other coordinates.
So that's not much of a problem. I actually did it with a full cosmological model. And we're approaching kind of exponential expansion, which unfortunately means that there is a final, final point of conformal time, that kind of corresponds to infinitely far in the future. But it's not terribly hard to apply that to your model, I don't think it changes much to the real problem is, the angular size, however, behaves weirdly when you you can't just convert straight from cool moving this bounce into normal, but that's how you actually need to do an adjustment. And that's
the moment I disagree, because co moving space has angles, and they do describe the angles of light. And so, you know,
we should probably bring the maps over a session where we can go to the whiteboard on this.
But in any case, I mean, you know, that I think the robust thing is the angles are going to be large, how exactly large they are, but but basically, you're going to see you know, you're not you're not to squint in the in a long distance to see see that basically for any small substance, you know, after their initial origin. Once any substantial size, they're going to be pretty large in the sky.
Yes, the final thing, you know that the expected number of seats you eventually meet are going to be 48 divided by 45 times pi squared plus two, that is 15. Point 54. There is an analytic expression for it, which is just plain awesome.
So in our story, that's going to depend on the power. So the power will affect how densely they appear.
Okay, I feel another one of those in the making. Great. And next one up, we have bread, and then Adam and Brad, perhaps say a few words about YouTube. And then next year, I'm
sorry about that I was just visiting my mother, she just got out of the COVID word. And so I'm doing it virtually. And so I missed a section of q&a that we're in, but I had a couple of questions. The first was about this obviousness of the full moon sighs Where are we, I mean, they're going to you're just talking about something where entropy is highly increased. Because if they were perfectly efficient civilization, all of their sphere would look like would be a sphere of extremely high entropy, red, light, and maybe even barely detectable to us. But I did just seems you're making a lot of assumptions about what the simulations are like. Your other light cone suggested to me that you're making assumptions about all civilizations, the civilization that happens to be master of this piece of space. Our negative light cone could be different from the assumptions you've made about what a general rabbit civilization looks like. And the other question I had was a technical one, do all galaxies meet the same pattern about stellar age and how long it was where stars were exploding too much for civilizations to form, or just smaller or less densely populated galaxies not meet that criteria, in which case, you would see these grabby spheres in those galaxies. As you probably know, Eric Drexler has done a hunt for those and didn't find them.
Alright, so several points there. First, the times, when I originally started doing this, I was hoping to take the actual spatial distribution of galaxy sizes and spatial correlations and actually do the modeling within that spatial structure. But it turns out, the spatial scales we're looking at are just larger than the scales on which galaxies have interesting structure. Remember, the average gravity civilization here has millions of galaxies, within its scope, hundreds of millions of galaxies. So within that scale, actually, you know, individual galaxies are not much account. So we're averaging basically over different kinds of galaxies and their different consequences for our purposes here, because the scales we're looking at are that large? Well, what I what I'm,
what I'm suggesting is that if you take if there is if there are galaxies out there a subset of galaxies, where it's the curve is different than the curves you showed where civilization, capable planets emerged a billion years into the existence of that galaxy, as opposed to as you suggesting, it's just starting to emerge 5 billion years into the no sorry. galaxies. Yeah, sorry, it's 10 billion years into the existence of our galaxy that we're finally ready to have a star form with a planet like ours. And then 5 billion more years for that to result in life is what you're saying is the that's an early did time point in your story, there are other galaxies, or are there no other galaxies for galaxies for which early would be earlier
is what I'm so
we are basically trying to summarize and inherit a literature that is still somewhat spotty on this galactic habitable zone. So there are now maybe a dozen papers, most of them do model our galaxy, but some of them do look at other galaxies. And we are trying to basically put, you know, an estimate in the middle of the distribution, and what they see that there is one paper that seems to suggest that larger galaxies are more likely per unit star to have habitable civilizations
earlier them earlier,
not earlier, it's just more likely to have them. But I don't know, I didn't see anything about different galaxies at different dates. If that wasn't the literature, I missed it. But, you know, basically, my our key point, though, is that our calculations here are pretty robust to substantial variations in those dates. And so we're, we show you in the paper, a lot of those variations so that the earliness claim, if you're going to contradict it, you really need to go with this low power or very low limit on the lifetime of planets. I don't think any of the other parameters are really going to save you. It's not going to save you to say well, what if different galaxies, you know, have a different peak, there are a different decay. Honestly, if you look at our different graphs, as we vary those things, it doesn't make enough of a difference. You're really gonna have to go with the lifetime Limit as or power limit as as a, as a as a way to save.
And of course, there's this opposition described as fake.
So I'm, again, remind lets me I want to remind everyone of the key working attitude or stance here analytically is people have forever talked about all these sort of quiet aliens who could be doing all sorts of things. And who knows, because hey, they could be hiding. And we decided to say yes, but let's focus on the loud, noisy ones, who we could be talking about who we could do know things about. So I'm going to define the loud, noisy ones in a key set of ways. And one of the ways I define them is that when they control something, you notice, you can see the difference,
as I said, all the big requires is that the ones who happen to be in charge of this part, don't want us to see the noisy ones. And that's all it's necessary for us to not see them.
Well, they would have to have the ability for us to see past them, to prevent us seeing past them into the rest of the universe. That is correct, the universe would have to be full of all sorts of things out there. And somehow they would have prevented us from seeing all those other things happening.
And hence, the sky is fake.
That's Yeah, that's basically what is it? It is a possibility. It's also, you know what I mean, it's also an extreme supposition. Yes.
Just as a as an editorial here. There's lots of other things I study in my life, like sickness or work, or all sorts of things where you could say, Well, you know what, nobody really knows who's ever really sick, because some people can see and well and not really be Well, some people could seem sick, then actually be Well, some people could be working, but pretend not to work. And some people could be not working by pretending you know, the world does all sorts of full of things that could be not the way they seem. But once in a while, it's worth just analyzing what what if things are what they seem, I agree with this
supposition. But what I distinguish in from answers between the ones that assume something is true for all of the universe, and something is true just for our area of space. And there's a I classify them differently. But I appreciate I appreciate the approach.
All right. Thank you. And, David, before you, I have Adam in the queue first. So Adam, Please say a few words about yourself. And good.
Guy, listen, I'm Adam Brown. I'm a theoretical physicist and cosmologist. Hi, Robin. I have a question. But before we get to that, I want to clarification first. So in your model, we are asked, we're not the grabby aliens, which seems like a despite the fact that in your model, grabby aliens vastly outnumber us. So that seems like you're just hypothesizing that or like, the question is why? It seems like if we if we
don't really carefully, because obviously, this is an a key point, I want to make sure you get it right. So you are my standard, I want to make sure you understand it right. And I will stick with it until you say you understand this. So the key point is, here we are now we could potentially become aliens. For others. Maybe we are in some sense, potentially aliens for others now, but we are not yet grabby. That is we have not yet reached this stage where we are expanding and nothing would stop us etc. And then we change our volumes. But the key assumption is that we might become great grabby. And if we do that would happen within say 10 million years. Yeah. And that is the key us working assumption because that means our date now is then pretty close to a date of origin or grabby aliens. Because if it happens, it happens that soon. That's the key working assumption.
In your model, almost all of the sentient beings that currently exist are other much larger grabby aliens rather than early civilizations such as ours. And we need to explain why one of those,
we're trying to be agnostic about the ratio between gravity and non gravity aliens. So like we said at the beginning of the talk, it potentially there could be all these quiet aliens out there at vast numbers, unknown density and numbers. And we're it's hard to say much about them because you can't see them. We're focused on the grabby ones because we can say things about them. But we are agnostic about the relevant ratio there.
But the grabby aliens are occupying large fractions of the universe. So it sounds like there should be some reason they're not sentient. It sounds like they should be more new. Well, the
prediction is they will they will eventually occupy a large fraction, the universe what they do that isn't where agnostic, we don't know. Exactly.
But the observers who would ever exist, were extremely atypical in your in your model, which is seems to be a point against
that that wouldn't be a crazy conclusion to draw. But you would have to make further assumptions about observers. So we're not in our analysis, making assumptions about observers. We're not saying grabby aliens are observers or that they will produce a density of observers. We're just not saying anything about them other than they make a visible difference, and then you would see them that that's all we're saying.
Okay, well, let me move on to my main question, then, which Perhaps portraying the my day job. Let me present an alternative theory for the resolution of the Fermi paradox that sounds very different from yours, but I think, ultimately is sort of quite similar, which is that the reason we don't in your grabby alien model, the reason we don't see them is that they essentially excise a large fraction of their future light cone, so we don't, we don't see them. Another model is, when civilizations get sufficiently advanced, they run stupid science experiments. And those stupid side science experiments cause vacuum decay in the Higgs sector, for example, in which there'll be a vacuum bubble of new bubble that will expand out at the speed of light and really excise the future light cone of of those advanced civilizations. That actually, that theory has a lot in similar, a lot in common with your theory in the sense that both revolt and advanced civilizations excising that future light cone. And all of the evidence that counts in favor of your theory also accounts in favor of that there are at least all the evidence of sort of, you know, the these n steps in evolution, why are we so early all of those questions and and has the sort of additional advantages that you don't need to explain why it expands at the speed of light, because that's the sort of input from theoretical physics that it'll definitely expand at the speed of light if you make a new bubble of vacuum. And also has the advantage that you don't need to explain why we don't live in one of those bubbles. Because there's nothing alive in those bubbles, you've destroyed the Higgs, the Higgs vacuum. So there seems to be sort of some commonality between the sort of vacuum decay bubble literature and and what you're saying. And it'll be interesting to sort of look back at the bubble nucleation literature in the light of your comments and see whether they bear on that too. The other thing I was going to say is, if you want to play that, and maybe I'll then go offline after I've asked this question, if you really want to play these non profit games, there's another explanation for why we're so early, which involves a variable that you haven't discussed, but some other people do discuss, which is you take the cosmological constant as a fixed input. But if you imagine that in the multiverse, we scan over all values of the cosmological constant that would also tend to push us earlier than naive reasoning would think because you, you want to live early, in order that you can get away with a larger cosmological constant, which is statistically preferred. So that's another variable if you really wanted to make your life difficult that you could add to your model. And I think you kind of do have to add that variable, because I think if you want to play games around profits, you just need to take in to everything account, and it's very hard to isolate any one part of the question from any other part of the question. Okay, I'll mute myself now. You can respond.
Thank you so much, Adam, for your comments, and observations. So I'll comment first on the last point you make, which is, we are assuming that we know at the moment what the value of the cosmological constant is. And with respect to that actual constant, we are still early. So the constant can't fully explain our earliness if we are still looking early, relative to the cosmological constant. Now, you know, you still might want to say yes, but why are the numbers even in the same ballpark, you might use selection effects to try to at least explain why they're similar magnitude, times. But roughly, the cosmological constant by itself can't explain why we are as early at least as the calculations I've just shown to just I fully agree that, from the simple observations point of view, the theory that a something like a gravity civilization extend of expanding and taking control, you know, accidentally destroys the similar volumes of the universe has all the same predictions. That is it predicts which volumes we have to show up earlier before because otherwise, it's too late. It explains the expansion rate. And it presumably also explains a rareness in the sense that they don't destroy the universe in this way until they've gone through the hard steps of evolution and show up at a certain point in time and have the capacity to do so. It's a tragic story. In the sense that force for some terrible reason, all of these really quite advanced civilizations, who again, by assumption, say maybe 10 million years past where we are finally reached the point where they can expand and somehow they do not foresee this terrible tragic consequence of their local experimentation, which is to destroy the space time they live in.
So
you know, I gotta think at least some of them don't do that. So a more complicated model would have some of them not destroying the universe and expanding and then others destroying the universe and expanding and then this mixed model, then you at least have some volumes where in the gravity civilizations control the volumes for a while. until their neighbors destroy the universe. That that is also a tragic scenario. But at least it's a scenario with a lot more observers and potential value of the the scenario where there is only they they all quickly and reliably destroy the universe near them. And then basically, nothing else happens, because they do. You could see why I might at least want to point to hopeful variations on the scenario, as opposed to focusing entirely on the worst worst case.
All right, thank you very much, Adam. We have David Byrne. Next, I want to do a quick note. Robin, do you still have a few minutes? Or Yes, of course? Absolutely. Absolutely.
They are for a long time.
Okay, great. I just want to make sure that we're now minutes past the hour in case you have to drop off. And we are still going to take the questions that we had for those who want to stay on. Afterwards, we're going to meet, move to gather again, which is this more decentralized, facade virtual lounge, which you consider tables in a more decentralized manner, talk to each other. And that's just for what is going to happen from now on. And I'm going to take the next questions one by one. And next one up. I have David Britton, David, a few words about yourself, and then many of your comments that you made in the TED, please.
Hi, thank you, Alison. And, Robin, whatever it is, you're taking to stay looking like it like a guy in his 20s. I'd like to you to ship me a bottle. I'm glad to see you fixed your visuals, got rid of the function effect. And it's much clearer now. I was happy by the way, I was at the 1983 Interstellar migration conference in Los Alamos. And that led to my Quarterly Journal paper about that was talked about the then at possible theories for the Fermi paradox. This is a very interesting approach. It's very similar to what Anders takes and it's a it's a spin off from the basic notion of the exceptions make the rule. And so you know, you might have 100 modest, you might have 10,000, modest quiet civilizations, but the one that's gracious is going to make the rules in its in its region. By the way, Adams theory as an aside Adams theory about causing a change in the physics in your region, rendering your area extinct is exactly discussed in the dark forest by Lucy Shan, which is the second book in the three body problem series. Okay, so in in there, you have the visible versus invisible species elucidations. notion is that, that even very powerful races make themselves less visible, given that they are going to be encountering other bubbles. And that taking them by surprise might give them an advantage, there is certainly an an incentive for trying to develop even if you're aggressive and less visible lifestyle. But that's not really what I want to discuss. Number one, a discrete power laws, they generally are severable, into ever smaller and ever smaller thing, sub power laws. And when you do that, you get the probability effect where point nine times point nine point times point nine gets small, a lot faster than just point nine. So if you break up your probabilities into a lot of different power laws, they're going to become more powerful. Now, the assumption here is that you have conquest. And and that when you meet another bubble, there's either conquest or exclusion or a wall, if I understand that this is a simplifying assumption. But such types of prevention don't happen in nature. And I refer back again and again and again to ecology. What happens when you have the situations that you're describing Robin is an ecosystem. And ecosystems can have a top predator, they can have a top Paramount race that is like we are on earth. And that can have effects, major effects, like when they meet each other biomes meet each other. But generally speaking, there's a whole pyramid underneath and these spaces are going to have health within their bubbles. That is going to be determined by whether or not they have a healthy ecosystem. And that means having some kind of provisions for alternatives for diversity, and those who are just below them who might be heirs which of course brings up my whole notion upper left. So those are all some comments on this. And but I think it's I think it's wonderful stuff. And I think it's terrific contribution to the conversation.
Thank you, David. So David mentioned, I basically tried to quickly present these results to him in an early form, you know, five weeks ago, or many weeks ago,
I didn't, I didn't introduce myself, David brand science fiction author, and another astrophysicist, who sometimes is involved in the science.
So, you know, in case, it's not obvious that we're talking about gravity, civilizations that control hundreds of millions of galaxies, surely, there's a lot of internal structure and variety going on, where we are completely just not talking about all the internal details that surely would exist, we're just abstracting them from a distance thing, they control the volume. And you know, that the key point is, then they would then prevent other gravity civilizations from appearing in that volume that they might well tolerate other kinds of things. And, you know, if we were another kind of thing, maybe somebody would tolerate us. I don't understand the point about power laws in the sense that surely, like, there's a selection effect for whatever the easiest way to produce advanced life is, then it'll most likely show up in the easiest way. And so that was probably with the lowest power at which that is, if you you know, if you have a low power way of doing it, and a high power way of doing it, I won with small number of steps, hard steps, number of larger number of hard steps. Well, most likely, the one with the smaller number of hard steps would be the one that succeeded more often earliest. But it does look like in fact, the universe is pretty empty at the moment. And we have arrived here at some modest level of success. And we does look like it took us a number of hard steps here. So that suggests that there is no way around, there's at least a substantial number of hard steps, which then produces a substantial power in the power law. That That just seems to be the way it has to be. And I guess, I guess I shouldn't go on. I should, I should stop. Because we're out of time. And we're gone to another part of the meeting.
No, I, I'm happy to stay on and
I will stay on and discuss more with whoever has more.
Okay, in case you want to write, like, and we can also use, but I think I'm
happy to stay here for 10 hours. I expected the rest of you guys who don't want stationaries. Yeah,
we have next we have Robbie, then bill and then ship. And yeah, where are we just introduce yourself maybe with a sentence or so and then go ahead.
Yeah, Ravi Pandya, longtime foresight member, currently Microsoft Research known Robin through a very long time.
I guess that question, Anders partially answered it, but you know, is just the fact that this sort of matter is not very uniformly distributed effect is very much like, you've got galaxies, and then it's a long way to the next galaxy, and then galaxy clusters, and along with the next cluster, mean, that is that that's sort of both the incentives and the difficulty of expanding. It's not it's sort of more like a step function than just a uniform spread. And do you think that would make a difference.
So like I said earlier, initially, we thought that would make a difference. And so initially, we were concerned to get like some actual data on galaxy correlation functions, and, you know, the shape of filaments, and the pancakes and that sort of thing, so that we could try to model this in that sort of a space. But it turns out, that the spatial scales that we're looking at here are just substantially larger than the scales in which galaxies are correlated. So, you know, filaments, you know, are up to maybe 50 million light years, you know, the distance between us and Andromeda is 2 million light years, there are these strips of filaments of galaxies that are maybe 50 million light years, maybe 100 million light years. But you know, we're just looking on larger scales than that for the most part. Now, if you want to go into the the tail of the distribution, so we showed you these graphs with CDF, and they, you know, the lower left of the CDF is how close they might be at how lower probability and if you want to get that lower left tail of the CD of the distributions, right? Yes, you'll probably have to actually analyze those correlation functions of galaxies, because that will probably actually thick in those tails that will make it slightly more likely that we meet them relatively early. Because of you know, they followed along with filament and, you know, we're gonna meet in the middle of filming or something like that. But for a first cut here, it turns out the models are just at larger scales, then galaxy correlations, scales, and therefore, it's University is actually pretty uniform on scales we're looking at.
Well, thanks.
Thank you so much. Next up, and Bill. Hi, there,
Robin. I think Allison is asking me to comment about about some of the things I stuck in the chat. One of which was that, if it turns out that Interstellar spaces to a hospital for us to exist in, obviously, we transform into machines. And we can build as much error correction and redundancy in those as we need to, to allow them to survive in that kind of environment. But it also occurred to me listening to comments from other people, that it's not really us or something like us. It's the entire ecosystem that supports us. And you know, I hear people talk about the top of an echo system. And I'm reminded that I live in an area that has mosquitoes, and we have a COVID vaccine running around here. So I'm not sure where the top is. That is one of the things but if you want to comment now, go ahead.
It You know, I'm happy to emphasize the point you made that is, when we're talking like something that might happen in 10 million years, we're talking about civilizations that control hundreds of millions of galaxies, we're clearly not talking about us now. And anything close to our current capabilities or technologies, we're talking a substantial time into the future with substantial potential increases in all sorts of abilities and in the character and nature of who we are and how we relate. And whatever our descendants do. So that's the kind of thing we're trying to talk about here. Usually, we would just like give up and presume that's impossible to talk about. But my claim here is that our earliness date is a key date. And that says, you know, you should face the fact that maybe there will be this deadline and something will be out there. That is why you are early. And you need to think about what that is, is it, you know, gravity civilizations that occupy volumes? Is it mistaken physics experiments that destroy the universe? It's something you know, it's something you should think about.
The other thing was, we have a long history of all competing struggling with each other up to the point of war. And do we can we get far enough from the other things that are next to us that we no longer have this urge? Or is that going to get worked out of a somehow? or How was this going to be dealt with?
I of course don't know. But I would say, you know, the most likely scenario is a lack of Central coordination. That is, since we have rarely ever had very strong central coordination in, in the history of humans and the history of biology and earth we've had far less, the most likely thing that would propel a gravity civilization to expand outward is internal competition and conflict. In a scenario that lacks central coordination, or Central, you know, division, and in fact, you know, that might be well, why they wouldn't hide is because they can't coordinate well enough to hide, even though they wish they could. They simply don't.
That's fine. The
vast, the vast majority of the members of that civilization are in the interior. And so what's also worth considering is the selection effect, that after millions of years, the the forward edge of the expanding sphere has been selected for the fastest and the most impatient and the most eager to move ahead. In which case, perhaps at what, there comes a point in time where they have nothing whatsoever in common with or any need to interfere with the planetary civilizations that they're blowing past.
You can think you can make that argument about it hypocrites in human civilization. Anyway, I think I finished it.
I so I have a paper from years ago where I tried to model do a game theory model of interstellar colonization. And certainly, in my simplest model, there's a race at the frontier to rush past the frontier and continue on the ones that the frontier do grab whatever resources are especially useful to their race, but other resources that aren't useful. They don't grab and they leave alone, but that's only reassurance. If you are sure that the third and fourth and 25th wave of colonization after the first will also leave you alone, which I'm less sure of.
Thank you. All right, and then we have to admit, um,
this is a great, great presentation. I think I just want to make sure I captured sort of the higher order bit correctly. Just one of the one of the sort of traditional answers to the Fermi question as well, were the first two, which a lot of people, you know, reaction as well, that doesn't seem very likely, I think what you presented is an argument based on some data that says not only is it likely it's, it's, it's a virtual certainty. And now, the, the the other piece, which I think was just alluded to, which is sort of the internal logic of evolution, you know, suggests that aggressive expansion is, is kind of a biological imperative, which makes this this as I said, in my comment, you know, perhaps the most extreme formulation of manifest destiny that I've ever encountered. I know, it's just a comment. But okay.
Thank you, Chris. Yes. So just to try to make clear the top bit. We look early relative to a calculation for when we could show up if there were no deadline. And we could show up at any time. That is, if life were so rare that anytime we showed up, the universe would be there and ready for us, and we could appear into it and do what we wanted to until we killed ourselves, then we would predict we would show up much later than we are. And so the question is Why so early? Now, you could just say, well, somebody's got to be early and we happen to be it. But that, you know, as a Bayesian explanation has to, you know, be hit with its unlikeliness. The more unlikely you posit in your hypothesis, the less plausible we should consider that if we have any other hypotheses to work with. So here the other hypothesis that to work with is, well, we're not early if there's a deadline. And there could be a deadline, because others who show up preclude later ones by grabbing the volume or killing everyone.
Thanks, and you thought you got away Roman. But I think we have another comment from bread.
It's just an amusing comment. I want to posit the crazy idea that civilizations decide it's much better to just upload their brains into emulated computation, and then form civilizations and find that far superior to traditional matter. And in particular, that there's some way of doing computation that doesn't depend on the physics we see. And that if that were the case, it would not mean the gravity civilization would look the way you described.
So the key
idea is there is this universe, and that something like a gravity civilization could go grab it and use it. That isn't to say that every possible motivation would produce that behavior. Nor is it the claim that there that's the only possible kinds of resources you could have or things you could do. So the key idea is that even if you're going to create a civilization to branding relations, it needs physical resources to sport, it needs mass, it needs negentropy. And there are many things that
are those are understanding of the physics of computation that it needs those things. Yes,
yes, indeed. So, and I'm going to work with our understanding of these things. And even if you know there's a standard start, well, what if they're all navel gazing? What if they found some like, small hole? And they're just really fascinated about their little small holes, and nobody wants to go anywhere else. But what if only a small fraction of them ever go somewhere else, then? There's still the question Why?
It has to be something that's so appealing, that says, This is so much better than using traditional matter, traditional energy, but it's such a thing exists and it would be attractive to everyone. I think. Well, there's
Jim Bennett's hypothesis about the Amish inheriting the earth. Yeah, you know, if you have some offshoot of a civilization that decides they don't like that, for reasons which are maybe crazier, irrational, nevertheless, they're the ones who populate the galaxy.
That's right. No. So the real question is, of course, yeah, we do have to assume that the physics of computation and the physics of energy, that we understand them sufficiently, which I'm not sure we do, in order to make these posits about gravity simulations, and what they would look like, to our instruments.
You know, as always, one has to say that any analysis is going to be based on our current understanding of something. And it's always possible we completely misunderstand something in our purview of analysis such that we go completely wrong and that's true of every analysis we ever do. And it should be a footnote on every analysis, but no more than a footnote on every
tedious dimension,
that every analysis could be very wrong, if you are very wrong about something.
And there is one there is one particular counter that which is versus if there was an early so it was like a much earlier civilization, which was inclined to not be interested in ordinary matter and energy. But nonetheless has decided to spread over the entire universe. It's got to be very, very early to have done that, and then set out a nature park. That's also a reasonable possibility.
And notice that Robin did not cite his book showing that He is an expert on the very scenario that this question.
Of course, I assumed everyone in this room knew I was saying that in an ironic sense,
hold it up. Hold it up, Robin, come on. Oh, sorry. We
need to have it we need to, we need to show the book. Oh, ah, ah,
damn. It. uploaded uploaded people.
The irony is of course that you have famously wanted to describe intelligent civilization possibly inevitably moving towards a form, which is almost worried might even be happening on your planet, you don't even know what's happening on your planet, other than you see all these, you see some things you don't understand. So anyway, it's just an ironic comment. Don't you don't need to? I do understand what you're saying about the danger of too much more phrasing here.
Okay, lovely. Well, thank you so much for joining, do you want to have any last words or
I will again, say thank you all for listening and engaging this. And this is the first forum in which I have been able to give a talk on this topic. So I think for site for being able to host the first time and maybe once this becomes eventually accepted, this will be acknowledged as the first place that it was presented into a critical audience, and yet our critical audience, which is what
you would be us, for having
chosen us, we feel very, very, very fortunate, I think, thank you, everyone for your fantastic questions. I think it was really great discussion. And thank you, Robin. It's really a fantastic effort and pulled together very, very
well. I have to I have to
tell everyone, the good news is I am now apparently COVID recovered. All right.
Thank you, Robert. Very much. Good. Good.
Good news. So
you know apparently I've been another two days without any fever or symptoms after my initial you
know, I just illness and so now I'm short. I just read that you should still get vaccinated.
I can't get vaccinated for a while because I for a while and would respond to strongly
Yeah.
And and keep your eye get a full and I mean, truly full medical checkup and a couple of months. I mean, the full deal, because some things some of these things lurk in play.
Unfortunately, it's not that the numbers are known because they haven't had enough COVID survivors then later die for other reasons and be autopsied, but from the small number they've had that they've seen, lasting damage, microphone was opposed, and stuff like that in the body and you want to be wary of it.
Yeah. Zinc.
Even in a my two and a half week experience, I gotta tell you, it's a little worse than the flu. Okay.
Oh, man, well, over the hump, and I shared in the chat, the room for gather for those who want to continue check chatting. And I'll switch over there. I'll see you there. And I'm going to share the video afterwards. This one is all recorded. And I'll make sure that you get the link to this. So thank you everyone, for joining. And for those who want to stay on, there's a gather link in the chat plus the password that you have to use to enter the gather. So I'll see those of you who are there in a second. All right. Bye, everyone. Thank you very much, Robin.