We Might Be Alone in the Universe

[Doctor Steuss]

To me, it seems that it all relies heavily on assumptions that have an incredibly anthropic bias to them.

How common is life in our solar system? Ultimately, we don’t 100% know, do we? We’ve got a pretty good idea that it’s only earth, but even that requires some (deductive) presuppositions.

Let’s assume though that we had physically surveyed all planets and moons in our solar system to a degree that we can definitively exclude the possibility of any type of living organisms on any of them (keep in mind, that it was only 2 decades ago when the existence of APMV was discovered on our own planet). That only tells us that, definitively (based on the data we have direct access to), life has about a 1:154 chance of developing on a planet or moon. Let's say 50 years from now, we are able to survey the 850-ish currently known multiplanetary systems. We find no life. Well... now we can say there's a 1:850 chance of life developing in a multiplanetary system, without making assumptions. To get to 10^Potatoe odds is just making educated guesses based on an extremely limited and isolated data set.

We ultimately don’t know if there’s a “Goldilocks” zone and planetary structure for germanium-based life. Or for silicon-based life. Or sulfur-based. We can make educated guesses, and complex models, but ultimately we have a single reference point that we are basing all assumptions on.

Even if we add in the “intelligent” modifier, we are still left with a minefield of anthropic assumptions. It assumes that an intelligent civilization would utilize similar technologies and leave similar traces to those that we have and/or are leaving.

[Res Ipsa]

To me, it seems that it all relies heavily on assumptions that have an incredibly anthropic bias to them.

How common is life in our solar system? Ultimately, we don’t 100% know, do we? We’ve got a pretty good idea that it’s only earth, but even that requires some (deductive) presuppositions.

Let’s assume though that we had physically surveyed all planets and moons in our solar system to a degree that we can definitively exclude the possibility of any type of living organisms on any of them (keep in mind, that it was only 2 decades ago when the existence of APMV was discovered on our own planet). That only tells us that, definitively (based on the data we have direct access to), life has about a 1:154 chance of developing on a planet or moon. Let's say 50 years from now, we are able to survey the 850-ish currently known multiplanetary systems. We find no life. Well... now we can say there's a 1:850 chance of life developing in a multiplanetary system, without making assumptions. To get to 10^Potatoe odds is just making educated guesses based on an extremely limited and isolated data set.

We ultimately don’t know if there’s a “Goldilocks” zone and planetary structure for germanium-based life. Or for silicon-based life. Or sulfur-based. We can make educated guesses, and complex models, but ultimately we have a single reference point that we are basing all assumptions on.

Even if we add in the “intelligent” modifier, we are still left with a minefield of anthropic assumptions. It assumes that an intelligent civilization would utilize similar technologies and leave similar traces to those that we have and/or are leaving.

I think that's a pretty sensible assessment.

[doubtingthomas]

"Nobody is rejecting the possibility of winning a multi-billion dollar PowerBall in many states. However, if winning is possible everywhere, why do I happen to live in a very unusual state where there actually was a winner?”

Answer, fully vetted by multiple Vegas odd-setters:

“Just lucky, I guess.”

I'll try one more time. Imagine 1,000 cities, but only one city is very unique and different from the rest.

What's the most likely outcome?
A) someone wins the Powerball in one of the 999 cities
B) someone wins in that unique city

[Morley]

I'll try one more time. Imagine 1,000 cities, but only one city is very unique and different from the rest.

What's the most likely outcome?
A) someone wins the Powerball in one of the 999 cities
B) someone wins in that unique city

Obviously, the Powerball will be won by (A), someone in the 999 cities. Folks from (B), the unique city of duck people, have a difficult time filling out the lottery forms with their feathery duck hands.

[Physics Guy]

Suppose, for the sake of argument at least, that Earth is an unusual planet. Can we infer, from the fact that we are on Earth, that life is likely to require an unusual planet like Earth?

The logic of that question is like that of this analogical one. Two kinds of animals inhabit a large desert that includes one small oasis. Goldfish can only live in the oasis; geckos live just as well anywhere. Here is an animal that I found in the oasis. Is it more likely to be a goldfish, or a gecko?

doubtingthomas's argument seems to be like saying that this oasis animal is surely more likely to be a goldfish, because if it were a goldfish, then of course it would have to have been in the oasis, but if it were a gecko, then that gecko would only be in the oasis by a huge fluke. It could have been anywhere in the whole desert, so why would it be in the tiny oasis? Bayesian inference—which is to say, simple logic—says that the most likely scenario must be the one that makes the data least fluky. For a gecko to be in the oasis is a fluke, while for a goldfish to be in the oasis is normal. So, the argument seems to run, the oasis animal is much more likely to be a goldfish, because the goldfish being in the oasis would be normal rather than fluky.

That does sound persuasive at first. But it's wrong, and here's why.

First of all you can tell that it's wrong by just forgetting about flukes and thinking like a wildlife biologist. If you found an animal in the oasis, then the only thing that determines how likely it is to be a goldfish or a gecko is the relative numbers of goldfish and geckos in the oasis. Their outside-oasis population densities do not matter one bit, be they zero or high. End of discussion: it really is just this simple. Imagine you're doing the study, counting animals in the oasis. You will never need to ask about life outside it. Goldfish might as well thrive in sand dunes, and geckos die in the sun, as far as your inside-oasis study is concerned.

So then what went wrong with that seemingly persuasive argument about it being a great fluke for a gecko to be in the oasis? The problem is this. If the desert is N times the size of the oasis, then it is indeed a 1-in-N fluke for any one particular gecko to be in the oasis. But the total number of geckos is also N times the average number of geckos in the oasis, whatever that is. That's a lot of geckos, and the chance that some few of them are in the oasis, out of so many, is not a small chance at all.

In fact the very same factor N that makes it fluky for any one particular gecko to be in the oasis also makes it less fluky for some of them to be there. The flukiness factors of N cancel out exactly—and as soon as we look at it this way, it's obvious that they have to cancel exactly. Whatever the average number of geckos in the oasis is, having that number of geckos there is just normal, and not a fluke at all—no matter how fluky it may be, because of how large N is, for any one particular gecko to be in the oasis. All that really matters is the average number of geckos in the oasis, and the 1/N flukiness of any one gecko being there is a sheer red herring that has no actual bearing on our question at all.

So to return from goldfish and geckos to the intelligent life problem, we can either think about the problem planet-by-planet, asking how likely it is for the planet to host intelligent life, or we can think intelligent species by intelligent species, asking how likely it is for each species to be on any given planet.

If we think planet-by-planet, then we're thinking like the wildlife biologist, and it's immediately obvious that no observations that we can make on this planet will tell us anything about probabilities on other planets. Maybe we can try to make educated guesses about probabilities of intelligent life on other planets, based on our knowledge of chemistry and physics if not of alien biology. But we cannot draw any logical inferences about probabilities of life elsewhere just from observations of life here.

If we think species-by-species, then the problem is a bit trickier. If we consider the hypothesis that each species is gecko-like, being equally likely to live on any planet, then it would be a fluke for any one particular species to appear on our unusual planet (and be us). But it would not be a fluke for some species to be here (and look around and be surprised at how unusual the place was, as we do). The factors of N cancel out. So the species-by-species perspective ends up agreeing exactly with the planet-by-planet perspective, when we're careful enough. Logical inferences from observations on Earth can only ever tell us about the probability of intelligent life to be here; they can say nothing about its probability anywhere else.

[Doctor CamNC4Me]

Great post, PG. I suppose a clever person could insert a Great Filter example or two into your explanation to explain why there aren’t noticeable oasises (sp?) as the goldfish peep out into the night sky with their goldfish telescopes. Or perhaps there were great oasis civilizations that came and went millennia ago, far away, so that they’d never be noticed. Time and distance could be the very things that separates advanced oasises.

Perhaps we’re in a polycosmic situation (as described by the book Anathem I recently read) where ‘intelligent’ life is just a small variation of ourselves, and the way to find it is traverse space and time [insert mathematical explanations here that are beyond me].

In other words, there could be endless reasons why we don’t see signs of great civilizations existing, colonizing, or leaving behind artifacts. Considering human nature, if alien races were to exist, and if they experience the similar environmental and genetic pressures that produced us, it’s probable they never leave their oasis and thus will never be noticed by other goldfish.

- Doc

[huckelberry]

h more likely to be a goldfish, because the goldfish being in the oasis would be normal rather than fluky.

Toasis study is concerned.




So to return from goldfish and geckos to the intelligent life problem, we can either think about the problem planet-by-planet, asking how likely it is for the planet to host intelligent life, or we can think intelligent species by intelligent species, asking how likely it is for each species to be on any given planet.

If we think planet-by-planet, then we're thinking like the wildlife biologist, and it's immediately obvious that no observations that we can make on this planet will tell us anything about probabilities on other planets. Maybe we can try to make educated guesses about probabilities of intelligent life on other planets, based on our knowledge of chemistry and physics if not of alien biology. But we cannot draw any logical inferences about probabilities of life elsewhere just from observations of life here.

If we think species-by-species, then the problem is a bit trickier. If we consider the hypothesis that each species is gecko-like, being equally likely to live on any planet, then it would be a fluke for any one particular species to appear on our unusual planet (and be us). But it would not be a fluke for some species to be here (and look around and be surprised at how unusual the place was, as we do). The factors of N cancel out. So the species-by-species perspective ends up agreeing exactly with the planet-by-planet perspective, when we're careful enough.
,,,,,,
Logical inferences from observations on Earth can only ever tell us about the probability of intelligent life to be here; they can say nothing about its probability anywhere else.

Physics Guy.
You obviously enjoy laying out the details, I separated your last summarizing statement because it is so clear and I think accurate. You have said it more than once in this thread.

I wanted to ask you a related technical question. Radio waves from a particularly loud rock and roll station carry quite a ways. I am wondering how far from earth would it be possible to listen to this station and recognize there was a musical pattern? Using the best devices now in operation (tandem sets of radio telescopes I think) at about what distance would Chuck Berry fade away?

The question comes to mind because the question of why we do not hear other civilizations out there keeps coming up.

[Doctor Steuss]

I wanted to ask you a related technical question. Radio waves from a particularly loud rock and roll station carry quite a ways. I am wondering how far from earth would it be possible to listen to this station and recognize there was a musical pattern? Using the best devices now in operation (tandem sets of radio telescopes I think) at about what distance would Chuck Berry fade away?

The question comes to mind because the question of why we do not hear other civilizations out there keeps coming up.

I hope you'll forgive me for dropping an anecdotal answer.

I once had quipped to a friend who was a broadcast engineer about aliens being introduced to humans via Nickelback. He told me that most radio stations aren't discernable in space, even if you're up in orbit, as antennas are generally designed to broadcast parallel to the earth. Even if we were to blast a station upwards (on FM frequencies), you probably wouldn't be able to get much of a recognizable signal (against all of the background noise) even as close as Mars.

This obviously wasn't based on using the best type of transmission device (or a very focused spectrum like the S-band, or X-band), but it's brought me cosmic comfort in moments of cringe cruising the radio dial.

[Res Ipsa]

Thanks for another thoughtful and reasonable post, PG.

The answer I would give to questions about the potential for life in the observable universe depends heavily on how the question is stated. For example, I have no objection to the statement "It is possible that we are and will be the only intelligent life in the observable universe and it is also possible that there are other universes that have intelligent life." I interpret possible as meaning something like, using a very high degree of proof (virtual certainty), there is insufficient evidence that the proposition is impossible. It's a very low bar.

I think it's self-evident that we don't have sufficient evidence to rule out either possibility. Given that we have viable mathematical theories that predict the existence of other universes and that our universe is part of a meta verse, we have no reason to believe that either is virtually impossible. On the other hand, those predictions also imply that other universes are inaccessible to us (Just as something on the order of 1/250th of our own universe is inaccessible to us), the existence of other universes cannot be tested by evidence. In my view, that puts them in the same category as "God did it."

If the question is, do we have sufficient evidence to conclude that there is some other intelligent life in the universe, I'd give an emphatic hell no. We haven't gathered enough evidence to conclude that there is other life in the universe, let alone intelligent life.

But if the question is, if you were going to bet, what would you bet on?" I'd put my money both on life and intelligent life, simply based on the law of large numbers. Our tools for finding whether life exists or ever existed on other planets or moons are extremely crude, and we've barely scratched the surface of possibilities in our own solar system, let alone the observable universe.

Based on information known today, I'd personally go further and bet that both are a virtual certainty. That's based on the fact that we know there is at least one example of both life and intelligent life in the universe and that there are, using Ethan Siegel's estimate, 10^30 planets (including those within and outside of solar systems). We can bump that up another order of magnitude to 10^31 by including moons as possible locations for life to form.

In doing so, I'm acutely aware that I'll have to adjust those odds as we discovery relevant evidence that allows us to rule out entire classes of extra-solar bodies. But we don't have that evidence, because we have no good evidence as to which conditions are necessary for the formation of life or intelligent life. We have a data sample of one.

I don't think my position is inconsistent at all with a general statement that, in terms of a bet, "life is common but intelligent life may be rare." Neither "common" nor "rare" are equivalent to "absent." Both terms are consistent with life and intelligent life being present somewhere in the observable universe, which is exactly my bet based on current evidence.

My disagreement with DT is about his attempt to use evidence to argue that the theoretically possible (no other intelligent life in the universe) is likely or even plausible. He fails to address a fatal logical flaw in his argument: that characteristics of our solar system that may be unique (based on a sample size that is virtually zero) reduce the probability of finding life somewhere in the universe. He's cited absolutely no scientific literature that makes the same logical mistake. He also cherry picks snippets out of papers to argue that they mean things that the authors clearly did not intend to say, which is highly misleading, whether he intends it to be or not.

From what I've read by and about Dr. Kipping, I don't see that our "bets" on the existence of life and intelligent life elsewhere in the universe are substantively inconsistent. If he's betting life is "common," in the universe, then he's not betting that it's absent. And if he's betting that it's rare (although rare v. common in his study is a close call), he's also not betting that it's absent. I'm betting that both are present.

[Res Ipsa]

h more likely to be a goldfish, because the goldfish being in the oasis would be normal rather than fluky.

Toasis study is concerned.




So to return from goldfish and geckos to the intelligent life problem, we can either think about the problem planet-by-planet, asking how likely it is for the planet to host intelligent life, or we can think intelligent species by intelligent species, asking how likely it is for each species to be on any given planet.

If we think planet-by-planet, then we're thinking like the wildlife biologist, and it's immediately obvious that no observations that we can make on this planet will tell us anything about probabilities on other planets. Maybe we can try to make educated guesses about probabilities of intelligent life on other planets, based on our knowledge of chemistry and physics if not of alien biology. But we cannot draw any logical inferences about probabilities of life elsewhere just from observations of life here.

If we think species-by-species, then the problem is a bit trickier. If we consider the hypothesis that each species is gecko-like, being equally likely to live on any planet, then it would be a fluke for any one particular species to appear on our unusual planet (and be us). But it would not be a fluke for some species to be here (and look around and be surprised at how unusual the place was, as we do). The factors of N cancel out. So the species-by-species perspective ends up agreeing exactly with the planet-by-planet perspective, when we're careful enough.
,,,,,,
Logical inferences from observations on Earth can only ever tell us about the probability of intelligent life to be here; they can say nothing about its probability anywhere else.

Physics Guy.
You obviously enjoy laying out the details, I separated your last summarizing statement because it is so clear and I think accurate. You have said it more than once in this thread.

I wanted to ask you a related technical question. Radio waves from a particularly loud rock and roll station carry quite a ways. I am wondering how far from earth would it be possible to listen to this station and recognize there was a musical pattern? Using the best devices now in operation (tandem sets of radio telescopes I think) at about what distance would Chuck Berry fade away?

The question comes to mind because the question of why we do not hear other civilizations out there keeps coming up.

Hi Huck,

Ethan Siegel (theoretical astrophysicist) wrote about this in December. https://bigthink.com/starts-with-a-bang ... etc-earth/