I've always thought that life is likely incredibly rare. I don't buy the argument that because life appeared only a few hundred million years after earth's formation, that must be approximately the mean time to life forming on an earth-like world (law of large numbers and all that).
Given that the universe is infinite, which we don't know for certain but strongly suspect from the part we can see, there's really no lower bound for the probability of life arising from the muck. Whether it was very unlikely, so there was only a 10^-10 chance of happening on earth, or so unlikely it defies the human capacity to grasp, at 10^-1,000,000 it would arise on earth, there will still be planets where life arises, given the infinite number of planets in an infinite universe. It could be as unlikely as 10^-Googol for arising on earth, but such incredibly unlikely odds still happen an infinite number of times out of an infinite set.
If it was this unlikely, we'd still only find ourselves, or life would still only find itself, looking up at an unintelligent universe. Essentially, the soft anthropic principle, which is we should only find ourselves to exist after the conditions that allowed us to exist, so we shouldn't assume that those conditions are common, any more than we should assume it's common for a sperm to fertilize an egg just because we were the unlikely winners, without a memory of the preconditions or all the failures.
But that still leaves the question, if life was incredibly unlikely to exist, why did it arise on earth relatively soon after earth cooled down? Wouldn't we expect it to show up somewhere in the middle of the time it had to arise? That would be true if we took the entire geological history of earth as the time when life had equal odds to arise. Take alkaline hydrothermal fields as a precondition for life. Perhaps the earth had a large number of these in a short window after our Hadeon Eon where there was still enough volcanic activity to have a large number of unique alkaline fields, while the earth had cooled down enough to have persistent oceans. If the conditions for life arising for life only had any real likelihood for a few hundred million period (lots of alkaline fields while still being cool enough for biology to happen), then the arising of life could have been right in the middle of the period when it couldn't arisen at all. If the window passed by, and earth cooled down, the odds could have dropped from 10^-Googol to 10^-Googol^Googol odds on earth specifically.
To sum it up, we don't really have a grasp of the conditions necessary for life even still, so I don't think life arising relatively early in earth's history is evidence that it's likely to arise on any planet with the right conditions. There is no lower bound, due to the infinite size of the dataset, for how unlikely life can be. Coming up with specific unlikelihoods to solve the Fermi Paradox is, in my view, far more likely to miss out massive filters than find them, and for it to settle at a probability where we'd expect to find life in many other places, seems like motivated reasoning.
Of course you could accuse me of the same motivated reasoning, of wanting earth and humanity to be "special" somehow, but when we're reasoning about unknown probabilities, about unknown preconditions for a low likelihood event to happen, I don't think there's much basis on which we can justify our conclusions.
To be honest, in this piece, I wasn't even worrying about "0 to 1," I was more interested in figuring out why prokaryote -> eukaryote is so hard.
To your "infinite" point, doesn't it change things that even if the universe IS infinite, we're all in finite, shrinking pockets of it?
It's easy to say (like Tegmark), "give me an infinity and I'll give you immortality and any wonder you can imagine." But practical, on the ground infinities seem to be in short supply. We live in a ~13b light year bubble, and it's shrinking as the outer edges of it expand away faster than anyone can reach.
So now it's no longer "of course even with 10^-bignum odds, it'll happen somewhere" because the number of planets on which it can happen can be much smaller than bignum in a finite volume, and there is a lower bound.
> To your "infinite" point, doesn't it change things that even if the universe IS infinite, we're all in finite, shrinking pockets of it?
Presumably the observable universe is only a matter of perspective. Look as far as you possibly can in every direction, and the CMB is exactly the same distance away no matter what direction you look. In that sense we are smack dab in the middle of the observable universe, just like how every boat on the ocean is right in the middle of its observable ocean.
Finding only one instance of life on another planet would set a lower bound, but only finding ourselves here already doesn't give us that lower bound, since no matter how likely or unlikely life is, so long as its possible there would still be life somewhere in the infinite universe. We have to exist in order to observe ourselves here, so whether life was so likely it happened 100 times every Hubble-volume, once in every Hubble-volume, or once in every Googol Hubble-volumes, we'd still find ourselves in the same position we currently do. Existing at the center of the observable universe, looking up at quiet skies.
I think the analogy is a boat that spontaneously found itself drifting on the open ocean. It can't see any other boats nearby, but maybe that's because it isn't that good at looking yet. If the seaman had a reason to believe the ocean infinite, until he saw another boat he would have no grounds to estimate how far away the next boat was. It could be a mile away, just in front of the horizon, a trillion miles past the horizon for all he could know. Him being where he is doesn't tell him whether it's likely there will be another boat within his horizon or not, since a boat with no others on the horizon is exactly what he would see no matter how unlikely it is that boats pop into existence. Whether the average was 1 per horizon, or 10^-Google^Googol^Googol per horzion, when the boat popped into existence, it would see the same thing.
> We have to exist in order to observe ourselves here, so whether life was so likely it happened 100 times every Hubble-volume, once in every Hubble-volume, or once in every Googol Hubble-volumes, we'd still find ourselves in the same position we currently do.
Yeah, but I think this is jumping the gun a bit.
I think there's two models here - my model, as articulated in the post was "simple life probably isn't that hard, but complex multicellular life is pretty hard, hard enough we'd need bignum shots-on-goal to get there, and the Milky Way isn't really big enough."
Your model seems to be "even simple life could be that hard, you might be overestimating how easy it is, so you don't even need to bother with how hard the prokaryote - eukaryote jump is. Just philosophically, we've woken up in a barren universe, so just by inspection it's pretty hard and we can't be sure why."
Am I framing this fairly?
And here I think I'd have to defer to the book - Nick Lane makes a pretty good case (in my own opinion) that due to basic chemical abundance (all life relies on the CHNOPS elements, which will all be common and abundant on any planet with water and vulcanism), how chemosynthesis works, and how geological structures like alkaline hydrothermal fields work, we should have a fairly decent chance of simple life arising anywhere all those things conjoin. And personally, I do buy that overall picture - like I'd love for us to get up to Enceladus and look for some, and if found, that's obviously one potential resolution of the question.
There could be LOTS Of planets with simple microbial life, is what I'm saying. Our current tools and resolution and understanding doesn't allow us to know this either way.
That said, I do agree with you overall - maybe rocky planets with water are more rare than we think. Ditto vulcanism, or the conjunction of the two. Maybe we really DO need abundant oxygen, for the 10x metabolic boost, and it's rare, and there could be many many more unknown Great Filters making simple life rare, too, and we just don't know at this point without more data and time.
I've always thought it interesting, on the one hand we're newborns in the crib, who have just woken up and opened our eyes, and are peering out there, and in geological / galactic time this happened milliseconds ago.
But on the other hand, the universe is so big and time so deep, it really IS odd that we can't see anything out there, even with our newborn eyes and limited faculties and scope of vision.
If we were a smarter species, we'd probably take Dark Forest scenarios more seriously, instead of beaming sitcoms and COPS and Jerry Springer out there to all and sundry. But that's humanity for you - by the time it matters, we'll probably no longer care about the problem, or be some sort of ASI hive mind.
That's about right. The prokaryote --> eukaryote jump might very well be as hard as you describe, and probably is, but it seems like everywhere we look, from the first jump to life, to eukaryotes, to multicellular life, then to complex life and finally intelligence (all within the mean time an planet like earth has to exist) we find plausible enough very low odds in each category that the Fermi Paradox isn't much of a Paradox to me. We don't really have a grasp on the odds, just that they're very low, and there's no lower bound to how low it can go.
> I do buy that overall picture - like I'd love for us to get up to Enceladus and look for some, and if found, that's obviously one potential resolution of the question.
Me too. I really hope we get low cost high mass to orbit rockets sooner rather than later. All of a sudden every deep-space mission becomes 100x easier, and cheaper, when we don't have to shave down every milligram from probes just to make them light enough.
> If we were a smarter species, we'd probably take Dark Forest scenarios more seriously, instead of beaming sitcoms and COPS and Jerry Springer out there to all and sundry.
I've never really bought this theory. There's on ~100 Billion planets in the milky way, and each one is marked by a very bright star. I can't imagine it would be that hard for an advanced civilization capable of wiping us out as we are to have a single planet-sized telescope check on each one every few million years or so for biosignatures, then bomb the place before the muck starts getting uppity.
Either way this was an interesting post. Every time I see an in-depth dive into the Fermi Paradox or the origin of life, I leave convinced that in the way analyzed, it's very unlikely. I think the universe would be a lot cooler if there was life on most planets, with intelligence being the final barrier that only happens rarely, but I'm skeptical it's not a previous step that's just significantly less likely to happen then we expect, including possibly the origin of life itself.
I also wonder if early abiogenesis could be probabilistically favoured because the extra time it allows for the evolution of consciousness within the habitable window is beneficial enough to outweigh the lower probability of life emerging early.
I've always thought that life is likely incredibly rare. I don't buy the argument that because life appeared only a few hundred million years after earth's formation, that must be approximately the mean time to life forming on an earth-like world (law of large numbers and all that).
Given that the universe is infinite, which we don't know for certain but strongly suspect from the part we can see, there's really no lower bound for the probability of life arising from the muck. Whether it was very unlikely, so there was only a 10^-10 chance of happening on earth, or so unlikely it defies the human capacity to grasp, at 10^-1,000,000 it would arise on earth, there will still be planets where life arises, given the infinite number of planets in an infinite universe. It could be as unlikely as 10^-Googol for arising on earth, but such incredibly unlikely odds still happen an infinite number of times out of an infinite set.
If it was this unlikely, we'd still only find ourselves, or life would still only find itself, looking up at an unintelligent universe. Essentially, the soft anthropic principle, which is we should only find ourselves to exist after the conditions that allowed us to exist, so we shouldn't assume that those conditions are common, any more than we should assume it's common for a sperm to fertilize an egg just because we were the unlikely winners, without a memory of the preconditions or all the failures.
But that still leaves the question, if life was incredibly unlikely to exist, why did it arise on earth relatively soon after earth cooled down? Wouldn't we expect it to show up somewhere in the middle of the time it had to arise? That would be true if we took the entire geological history of earth as the time when life had equal odds to arise. Take alkaline hydrothermal fields as a precondition for life. Perhaps the earth had a large number of these in a short window after our Hadeon Eon where there was still enough volcanic activity to have a large number of unique alkaline fields, while the earth had cooled down enough to have persistent oceans. If the conditions for life arising for life only had any real likelihood for a few hundred million period (lots of alkaline fields while still being cool enough for biology to happen), then the arising of life could have been right in the middle of the period when it couldn't arisen at all. If the window passed by, and earth cooled down, the odds could have dropped from 10^-Googol to 10^-Googol^Googol odds on earth specifically.
To sum it up, we don't really have a grasp of the conditions necessary for life even still, so I don't think life arising relatively early in earth's history is evidence that it's likely to arise on any planet with the right conditions. There is no lower bound, due to the infinite size of the dataset, for how unlikely life can be. Coming up with specific unlikelihoods to solve the Fermi Paradox is, in my view, far more likely to miss out massive filters than find them, and for it to settle at a probability where we'd expect to find life in many other places, seems like motivated reasoning.
Of course you could accuse me of the same motivated reasoning, of wanting earth and humanity to be "special" somehow, but when we're reasoning about unknown probabilities, about unknown preconditions for a low likelihood event to happen, I don't think there's much basis on which we can justify our conclusions.
Yeah, I get where you're coming from.
To be honest, in this piece, I wasn't even worrying about "0 to 1," I was more interested in figuring out why prokaryote -> eukaryote is so hard.
To your "infinite" point, doesn't it change things that even if the universe IS infinite, we're all in finite, shrinking pockets of it?
It's easy to say (like Tegmark), "give me an infinity and I'll give you immortality and any wonder you can imagine." But practical, on the ground infinities seem to be in short supply. We live in a ~13b light year bubble, and it's shrinking as the outer edges of it expand away faster than anyone can reach.
So now it's no longer "of course even with 10^-bignum odds, it'll happen somewhere" because the number of planets on which it can happen can be much smaller than bignum in a finite volume, and there is a lower bound.
> To your "infinite" point, doesn't it change things that even if the universe IS infinite, we're all in finite, shrinking pockets of it?
Presumably the observable universe is only a matter of perspective. Look as far as you possibly can in every direction, and the CMB is exactly the same distance away no matter what direction you look. In that sense we are smack dab in the middle of the observable universe, just like how every boat on the ocean is right in the middle of its observable ocean.
Finding only one instance of life on another planet would set a lower bound, but only finding ourselves here already doesn't give us that lower bound, since no matter how likely or unlikely life is, so long as its possible there would still be life somewhere in the infinite universe. We have to exist in order to observe ourselves here, so whether life was so likely it happened 100 times every Hubble-volume, once in every Hubble-volume, or once in every Googol Hubble-volumes, we'd still find ourselves in the same position we currently do. Existing at the center of the observable universe, looking up at quiet skies.
I think the analogy is a boat that spontaneously found itself drifting on the open ocean. It can't see any other boats nearby, but maybe that's because it isn't that good at looking yet. If the seaman had a reason to believe the ocean infinite, until he saw another boat he would have no grounds to estimate how far away the next boat was. It could be a mile away, just in front of the horizon, a trillion miles past the horizon for all he could know. Him being where he is doesn't tell him whether it's likely there will be another boat within his horizon or not, since a boat with no others on the horizon is exactly what he would see no matter how unlikely it is that boats pop into existence. Whether the average was 1 per horizon, or 10^-Google^Googol^Googol per horzion, when the boat popped into existence, it would see the same thing.
> We have to exist in order to observe ourselves here, so whether life was so likely it happened 100 times every Hubble-volume, once in every Hubble-volume, or once in every Googol Hubble-volumes, we'd still find ourselves in the same position we currently do.
Yeah, but I think this is jumping the gun a bit.
I think there's two models here - my model, as articulated in the post was "simple life probably isn't that hard, but complex multicellular life is pretty hard, hard enough we'd need bignum shots-on-goal to get there, and the Milky Way isn't really big enough."
Your model seems to be "even simple life could be that hard, you might be overestimating how easy it is, so you don't even need to bother with how hard the prokaryote - eukaryote jump is. Just philosophically, we've woken up in a barren universe, so just by inspection it's pretty hard and we can't be sure why."
Am I framing this fairly?
And here I think I'd have to defer to the book - Nick Lane makes a pretty good case (in my own opinion) that due to basic chemical abundance (all life relies on the CHNOPS elements, which will all be common and abundant on any planet with water and vulcanism), how chemosynthesis works, and how geological structures like alkaline hydrothermal fields work, we should have a fairly decent chance of simple life arising anywhere all those things conjoin. And personally, I do buy that overall picture - like I'd love for us to get up to Enceladus and look for some, and if found, that's obviously one potential resolution of the question.
There could be LOTS Of planets with simple microbial life, is what I'm saying. Our current tools and resolution and understanding doesn't allow us to know this either way.
That said, I do agree with you overall - maybe rocky planets with water are more rare than we think. Ditto vulcanism, or the conjunction of the two. Maybe we really DO need abundant oxygen, for the 10x metabolic boost, and it's rare, and there could be many many more unknown Great Filters making simple life rare, too, and we just don't know at this point without more data and time.
I've always thought it interesting, on the one hand we're newborns in the crib, who have just woken up and opened our eyes, and are peering out there, and in geological / galactic time this happened milliseconds ago.
But on the other hand, the universe is so big and time so deep, it really IS odd that we can't see anything out there, even with our newborn eyes and limited faculties and scope of vision.
If we were a smarter species, we'd probably take Dark Forest scenarios more seriously, instead of beaming sitcoms and COPS and Jerry Springer out there to all and sundry. But that's humanity for you - by the time it matters, we'll probably no longer care about the problem, or be some sort of ASI hive mind.
> Am I framing this fairly?
That's about right. The prokaryote --> eukaryote jump might very well be as hard as you describe, and probably is, but it seems like everywhere we look, from the first jump to life, to eukaryotes, to multicellular life, then to complex life and finally intelligence (all within the mean time an planet like earth has to exist) we find plausible enough very low odds in each category that the Fermi Paradox isn't much of a Paradox to me. We don't really have a grasp on the odds, just that they're very low, and there's no lower bound to how low it can go.
> I do buy that overall picture - like I'd love for us to get up to Enceladus and look for some, and if found, that's obviously one potential resolution of the question.
Me too. I really hope we get low cost high mass to orbit rockets sooner rather than later. All of a sudden every deep-space mission becomes 100x easier, and cheaper, when we don't have to shave down every milligram from probes just to make them light enough.
> If we were a smarter species, we'd probably take Dark Forest scenarios more seriously, instead of beaming sitcoms and COPS and Jerry Springer out there to all and sundry.
I've never really bought this theory. There's on ~100 Billion planets in the milky way, and each one is marked by a very bright star. I can't imagine it would be that hard for an advanced civilization capable of wiping us out as we are to have a single planet-sized telescope check on each one every few million years or so for biosignatures, then bomb the place before the muck starts getting uppity.
Either way this was an interesting post. Every time I see an in-depth dive into the Fermi Paradox or the origin of life, I leave convinced that in the way analyzed, it's very unlikely. I think the universe would be a lot cooler if there was life on most planets, with intelligence being the final barrier that only happens rarely, but I'm skeptical it's not a previous step that's just significantly less likely to happen then we expect, including possibly the origin of life itself.
You summed up my take on it quite nicely.
I also wonder if early abiogenesis could be probabilistically favoured because the extra time it allows for the evolution of consciousness within the habitable window is beneficial enough to outweigh the lower probability of life emerging early.