Szydagis' point 3: Interstellar travel is too hard

[Scaramanga]

Wrong.

Back that up with evidence. If you are claiming that alien life is more likely than not...where is your empirical evidence for such ?

 

[Ann K]

It's odd how often on this site it is argued that there's zero evidence aliens could get to Earth, yet few seem to want the equally scientific stance that there is zero evidence aliens even exist to come here in the first place. Alas a generation of watching Star Trek has clouded rational debate on the matter, and given the totally bogus impression that aliens are 'likely'.

Let's look at it from the angle of earth. (1) Do we exist here? Why yes; yes we do. That means it's possible for intelligent life to form. (2) Have we ever traveled to a different solar system? Nope, we have not. That doesn't mean we will never do so, but we do know enough to find the possibility vanishingly small. I don't see those two as having an "equally scientific stance", when one is known to be possible and the other is not.

 

[John J.]

It's odd how often on this site it is argued that there's zero evidence aliens could get to Earth...

There's zero reliable evidence aliens have got to Earth. This thread discusses whether it's possible for a hypothetical extraterrestrial civilisation to traverse interstellar space to get to Earth.

...few seem to want the equally scientific stance that there is zero evidence aliens even exist

You are of course totally correct that there is zero evidence that alien life (let alone intelligent alien life) exists.
It's not just a scientific stance, it is objectively true. But absence of evidence is not always evidence of absence.

Prior to 1992, we had no confirmed evidence that exoplanets existed; there were suspicions from at least the 1970s that the "wobble" / radial velocity measurements of some stars indicated planetary companions, but ironically some of the best candidates identified then have been refuted (e.g. 61 Cygni, "Detection limits from the McDonald Observatory planet search program", R.A. Wittenmyer, M. Endl et al. 2006, https://www.semanticscholar.org/reader/3cdeac2df0ff1877b61c3afb3658a0dd963f474b).

We now know exoplanets are common, some with similar masses to Earth and possibly at distances from their star which might allow liquid water to exist.

If G-type stars (like the Sun) are 7% of the Milky Way's approx. 400 billion stars, that's 28 billion Sun-like stars in our galaxy.
(More conservative, often older estimates of star numbers in the galaxy are around 100 billion, therefore a naïve calculation might give us 7 billion Sun-like stars).

According to UBC astronomer Jaymie Matthews: "Our Milky Way has as many as 400 billion stars, with seven percent of them being G-type."

"As many as six billion Earth-like planets in our galaxy, according to new estimates", University of British Columbia cited by Phys.Org website, https://phys.org/news/2020-06-billion-earth-like-planets-galaxy.html
Not sure about this article, even the figures used in it give 5.04 billion, not 6. There are more conservative estimates:

In The Occurrence Rate of Earth Analog Planets Orbiting Sunlike Stars (2011), PDF below,
Joseph Catanzarite and Michael Shao used data from the Kepler space telescope to estimate the number of Earth-like planets (in terms of mass) in our galaxy whose orbits would allow liquid water at the surface.

We estimate that 1% to 3% of stars like the Sun are expected to have Earth analog planets, based on the Kepler data release of Feb 2011.

In more detail,

We present a calculation of , the occurrence rate of Earth analog planets orbiting FGK stars, based on the February 2011 Kepler data release.

depends on the adopted definition of the HZ. For the conventional HZ boundaries (Kasting, Whitmire, & Reynolds, 1993), we find

For the less conservative HZ boundaries given in the ExoPTF Report, we find

Using the lower (possibly dated) estimate of the number of G-type stars, and excluding the F and K stars included by Catanzarite, Shao we have 7 billion stars similar to the Sun.

Taking the most conservative estimate from Catanzarite, Shao, 0.8% of FGK stars possessing a near-Earth mass planet within a habitable zone, 7 billion /100 x 0.8 = 56 million G-type stars with a planet of mass similar to Earth in the habitable zone.

That's an awful lot of "test tubes" in which to attempt abiogenesis, which we know is possible. Many will be of similar age or older than the Earth.

We have no evidence of extraterrestrial life, but we don't have enough knowledge at present to scientifically rule it out.

 

[JMartJr]

That's an awful lot of "test tubes" in which to attempt abiogenesis, which we know is possible. Many will be of similar age or older than the Earth.

Yes, but call the number of test tubes X. If the chances of life arising are 1 in (X*10^10), say, then that is not nearly enough test tubes to give us good odds. ARE the odds that bad? Heckifiknow. Nobody knows, all we know is that the odds are greater than 0, since we know it has happened once.

We have no evidence of extraterrestrial life, but we don't have enough knowledge at present to scientifically rule it out.

Absolutely agreed. Or even to roughly estimate odds that there is any. (Or if there is any close enough to us to matter, in time or space. If there are bacteria-analogues on a world in the Pinwheel Galaxy in Ursa Major, we'll never know it.)

 

[Mendel]

Back that up with evidence. If you are claiming that alien life is more likely than not...where is your empirical evidence for such ?

@FatPhil was not claiming "more likely than not" (which means over 50%), he was claiming 42%:

It is highly improbable that it's 1 in 10^527 *because* we have a sample of one.
The true neutral stance is a poisson one, which says that the chance of there being life on other planets is 42%. And no, that's not a joke, that's 1-e-1​/(1-e-1​). Everything else is assuming something.

And that's before we consider the probability that this alien life exists at the same time as we do (whatever "time" means in this context).

 

[Mendel]

2) So why should the fact that a copy of you was made while you were asleep make any difference ?

3) Sure, the copy wakes up thinking he is you. But, in fact nothing has actually changed from the situation prior to copying being invented. OK so there's a copy out there who thinks he's you. So what ? It does not alter YOUR status at all. Nothing about you yourself has changed one iota from prior to copying. You wake up thinking that you are you...because you are !

4) The crux here is that people confuse what the copy thinks with your status..as if it somehow affected it. The fact that some copy thinks he is me is no more relevant to my status than some person thinking they are a re-incarnation of Napoleon or are Jesus.

The change in situation is that there was only one person who thought he was me before I went to sleep, but there are two persons when I woke up.

My existential problem is that I might be the copy.

And the person who thinks he's Napoleon is clearly a lunatic, but the other person who thinks he's me does look like my identical twin and knows all of my passwords. It's much worse than identity theft is now!

Fly out preserved fertilized eggs. Have the ship incubate and teach them at the destination. Saves hugely on fuel, and doesn't break physics.

 

[FatPhil]

Back that up with evidence. If you are claiming that alien life is more likely than not...where is your empirical evidence for such ?

There's no "evidence" for mathematical deductions. Mathematics isn't an experimental science.

I've already given a reasoned argument why I consider the non-existence of alien life to require fine tuning, and thus to have lower credence than the alternative hypothesis, but that's utterly irrelevant, that wasn't the point I was making.

 

[Scaramanga]

@FatPhil was not claiming "more likely than not" (which means over 50%), he was claiming 42%:

The claim is ridiculous. You cannot apply a Poisson distribution to a single data point, where you do not know that an event can occur more than once in a given time frame. You have to already know the average number of events in a given time frame to create a Poisson distribution. You have to already know the lambda...which is precisely what one is trying to work out. Thus Phil's claim has to be a case of circular logic.

 

[Scaramanga]

The change in situation is that there was only one person who thought he was me before I went to sleep, but there are two persons when I woke up.

My existential problem is that I might be the copy.

And the person who thinks he's Napoleon is clearly a lunatic, but the other person who thinks he's me does look like my identical twin and knows all of my passwords. It's much worse than identity theft is now!

I could wake up and think I am you. Does that make you suddenly not be you ?

Your existential problem is non-existent. Why should the fact that a copy has been created mean that YOU might wake up as the copy ? OK so some copy thinks he is you. So what ? I am right now thinking I am you. Are you having an identity crisis ?

The fact that the copy 'looks' identical is irrelevant. It is a red herring. I could use the same technology to make myself look like you and have all your thoughts. Does that mean I am suddenly you ?

There is absolutely nothing that makes me suddenly be you...not least because due to the Pauli exclusion principle all your atoms can only occupy specific places in space, and that is precisely how objects are unique. The fact that a copy cannot occupy the same space as you ( just as identical cars coming off a production line cannot occupy the same space ) means that a copy of you physically cannot be you.

People muddle their thinking by asking 'how do I know I am not the copy ?' But it is a bogus dilemma. You know that you are you by the exact same process by which you knew that you were you prior to any copy being made. Nothing has changed ! The fact that some copy out there thinks he is you changes nothing about your personal status. I guess it just requires a certain amount of lateral thinking to grasp this.

 

[John J.]

Fly out preserved fertilized eggs. Have the ship incubate and teach them at the destination. Saves hugely on fuel, and doesn't break physics.

An obviously flawed idea.

Spoiler

How do you expect the shipmind to teach chickens?

 

[Mendel]

There's no "evidence" for mathematical deductions. Mathematics isn't an experimental science.

When you want to use a mathematical model to say something about reality, you need to anchor the model and its inputs in reality.

 

[Mendel]

People muddle their thinking by asking 'how do I know I am not the copy ?' But it is a bogus dilemma. You know that you are you by the exact same process by which you knew that you were you prior to any copy being made.

The dilemma is that when the copy executes this process, it will also determine that I am I.

You are correct that old-me is going to keep existing. But new-me will also think that he is old-me, so the hypothetical copy process leaves us with two people who share a common history (and physical characteristics) with old-me—and no way to tell which is which.

Realistic cloning, as practised today (and for ages in horticulture), takes a tissue sample and grows a new organism from the identical genetic code, resulting in distinct entities.

 

[FatPhil]

@FatPhil was not claiming "more likely than not" (which means over 50%), he was claiming 42%:

And that's before we consider the probability that this alien life exists at the same time as we do (whatever "time" means in this context).

Technically, that was a different argument based on different premises. I admit I've not been particularly clear in stating all my assumptions each time. That's the "the universe is big enough that there's no reason to think the conditions here are unique or fine tuned"-based argument. That lets you just sample star systems (it's easier that way, rather than sampling planets) as independent (<-- because of this) random variables, and this lets you model it as a discrete poisson process.

However, my disagreement was not over values of probabilities, it was over *changes to* values of probabilities - the updating of priors given new information. Which holds for the first bit of information as much as it does any subsequent bits.

 

[FatPhil]

When you want to use a mathematical model to say something about reality, you need to anchor the model and its inputs in reality.

All you need in this case is to accept that "you know absolutely nothing" is equivalent to "P(p<x) = { 0 if x<0; x if 0<=x<=1; 1 if x>1 }". However, I'd still say that that was definitionally true. My actual statement was true independent of the "reality" you're trying to model, as long as you accept that it can be modelled using probabilities at all. If you bring an omniscient omnipotent entity into your universe of discourse, for example, then that wouldn't be a premise you could automatically accept.

 

[Mendel]

All you need in this case is to accept that "you know absolutely nothing" is equivalent to "P(p<x) = { 0 if x<0; x if 0<=x<=1; 1 if x>1 }". However, I'd still say that that was definitionally true.

I don't follow.
But if that's true, p obeys the standard uniform distribution. You'd need to reason why you chose this distribution.

 

[Scaramanga]

The dilemma is that when the copy executes this process, it will also determine that I am I.

You are correct that old-me is going to keep existing. But new-me will also think that he is old-me, so the hypothetical copy process leaves us with two people who share a common history (and physical characteristics) with old-me—and no way to tell which is which.

Unfortunately one cannot make anyone think laterally to grasp that the 'dilemma' is an illusion. It's one of those things where one just has to have that light bulb moment and grasp it.

I think the non-existent 'dilemma' is caused by trying to put yourself in the mind of a copy who thinks he is you, when what you really need to do is grasp that nothing about yourself actually changed. Imagine that our hypothetical machine could re-arrange all my atoms so as to be an exact copy of you. Does that suddenly cause me to become you ? No. Nothing about you changes. There is simply a copy out there who thinks he is you. But so what ? At no point did you cease being the real you.

Why should the fact that some copy out there thinks he is you alter anything about your reality ?

If no copy of you was made, you would not wake up and ask whether you are you.

If a copy of you was made and you did not know it existed, you would not wake up and ask whether you are you.

So why would your knowledge that there is a copy out there alter any of that when nothing about your status has changed ? You don't need to 'tell which is which'. The fact that some imposter out there thinks he is you is irrelevant.

 

[Asherah]

A common misperception. None of the visible stars we can see in the sky are anywhere near as distant as a million light years. Alpha Centauri is only four light years away, Sirius is eight - even a really distant star like Deneb is only 1500 light years way, and it is a relatively simple calculation to find where it will be 1500 years from now.

If we have to travel a million years to find another civilisation we might as well be alone.

The issue I was alluding to is that there's a practical limit to the distance aliens could travel to come to earth- irrespective of how close to the speed of light they are able to get. That's because they would need to establish a destination at the beginning of the journey.

Yes, there are stars within a practical distance, but how many have the full set of conditions needed for life to emerge?

Life (of any kind) is improbable. Intelligent life would be even more improbable. Intelligent life that pursues the relevant technology, and with the inclination and resources to travel the large distances seems exceedingly improbable.

 

[Ann K]

Life (of any kind) is improbable. Intelligent life would be even more improbable. Intelligent life that pursues the relevant technology, and with the inclination and resources to travel the large distances seems exceedingly improbable.

I agree about the improbability of travel to excessively distant places, because we have a grasp of the necessary physics behind such a journey.

We are only taking the first baby steps toward identifying "Cinderella" planets. But given that candidate planets have been found, we really don't have any basis for judging the probability of the formation of life. All we really know is that it has been done before, right here on earth.

 

[Mendel]

So why would your knowledge that there is a copy out there alter any of that when nothing about your status has changed ? You don't need to 'tell which is which'. The fact that some imposter out there thinks he is you is irrelevant.

You are sailing right past my point that, in this hypothetical situation, there are two indistinguishable people who, according to you, should think of themselves as original and the other as imposter.

 

[captancourgette]

you are talking about a von Neumann probe, or self-replicating spacecraft. A von Neumann machine is just a computer.

Yeah thanks for the clarification though according to wiki
https://en.wikipedia.org/wiki/Von_Neumann_machine

Von Neumann machine may refer to:

• Self-replicating machine, a class of machines that can replicate themselves
  • Universal constructor (disambiguation)
  • Von Neumann probes, hypothetical space probes capable of self-replication
  • Nanorobots, capable of self-replication

Though I agree probe is a better word in the sense I used it, as that implies going out

 

[FatPhil]

I don't follow.
But if that's true, p obeys the standard uniform distribution. You'd need to reason why you chose this distribution.

Because you know nothing about the probability, and have no reason to favour any probability over any other. And the mathematical way of expressing "I have no reason to favour any probability over any other probability" is the above. As I said - this is almost *definitionally true*. This really ought to be a universally shared premise, you're looking a bit weird objecting to it.

Let me turn this around:
If you accept that knowing nothing about the probability is the same as having no reason to favour any probability over any other, then how would you mathematically express that?
And if you don't accept that knowing nothing about the probability is the same as having no reason to favour any probability over any other, how do the two concepts differ?

 

[Mendel]

Because you know nothing about the probability, and have no reason to favour any probability over any other. And the mathematical way of expressing "I have no reason to favour any probability over any other probability" is the above.

You are not going to convince me that "I know nothing about the probability density, therefore the probability distribution must be uniform" is a logical step.

Even the simplified case, where there is a fixed probability q=P(there is life on a planet in the habitable zone), leads to a non-uniform p=P(there is life on at least one other planet in the habitable zone).

 

[FatPhil]

You are not going to convince me that "I know nothing about the probability density, therefore the probability distribution must be uniform" is a logical step.

I'm gonna have to stop you right there - you're not even getting my argument straight - in fact you've misquoted me.

What I said was "you know nothing about the probability, and have no reason to favour any probability over any other"
and what you've put in quotes is "I know nothing about the probability density, therefore the probability distribution must be uniform".

This is nothing but a straw man - the above is not my argument, which doesn't need repeating as it's so recent in the thread.

Bayesian statistics deals with the probability distribution of an unknown probability (or probabilities in the more general case). When I say you know nothing about the probability, I am clearly referring to the rightmost term, the variable. Jumping from that to the "probability density" part, the model, is a category error.

To a bayesian, there's nothing odd about probabilities of probabilities, or even of probabilities of probabilities of probabilities (you could present the Dark Matter vs. MOND argument as an example of the latter - both models could be expressed as a typical 2-layered bayesian analysis of the measurements, and as you want to work out which of the two models is more credible, you add a third layer). Frequentist balk at this, because they prefer counting to integrating, and you can't count the continuity of probability distributions. You have expressed frequentist tendencies in the past, this might be your sticking point.

 

[Scaramanga]

You are sailing right past my point that, in this hypothetical situation, there are two indistinguishable people who, according to you, should think of themselves as original and the other as imposter.

No, you are raising a false dilemma. You keep confusing what a person 'thinks' with what actually is.

I could 'think' I am Napoleon. I could have sub-atomic level plastic surgery and make myself look exactly like Napoleon. Do I suddenly become Napoleon ? No.

OK so some experiment imposter 'thinks' he is me. So what ?

The entire 'how do you know you are not the copy' line of reasoning is fallacious. The fact that some impostor out there 'thinks' he is you has changed nothing about your actual status...which is identical to what it would have been had no copy been made. I emphasise....nothing about you has changed....whether a copy is made or not. The minute one grasps that, the 'dilemma' vanishes.

This is one of those 'dilemmas' that really does require lateral thinking.

 

[Scaramanga]

However, I'd still say that that was definitionally true.

But it seems to me you are plugging '1' into a formula that was never designed to have just '1' as an input. Even just looking at the basic definition of Poisson distribution it requires an 'average number of events in a given time frame'. You cannot sensibly have an average of just 1 item. It seems to me to be a case of limited input leading to a nonsense output.

One might try to argue that the average probability across all unknowns was 42%.....but then would you be happy that the chance of one of bigfoot, mothman, UFOs, ghosts, fairies at the bottom of the garden, Atlantis, spontaneous combustion, chemtrails, and a host of other pseudo-science being true was 42% ?

 

[FatPhil]

But it seems to me you are plugging '1' into a formula that was never designed to have just '1' as an input. Even just looking at the basic definition of Poisson distribution it requires an 'average number of events in a given time frame'. You cannot sensibly have an average of just 1 item. It seems to me to be a case of limited input leading to a nonsense output.

Not "nonsense", just with error bars that doesn't make it particularly useful. But it was based on the smallest possible set of premises, you shouldn't expect too much from it.

One might try to argue that the average probability across all unknowns was 42%.....but then would you be happy that the chance of one of bigfoot, mothman, UFOs, ghosts, fairies at the bottom of the garden, Atlantis, spontaneous combustion, chemtrails, and a host of other pseudo-science being true was 42% ?

You're conflating a situation where we have one positive data point with a whole bunch of scenarios where we have zero positive data points. These are not the same. Any conclusion from the former, no matter how weak, does not transfer to the latter.

 

[Scaramanga]

Not "nonsense", just with error bars that doesn't make it particularly useful. But it was based on the smallest possible set of premises, you shouldn't expect too much from it.

You're conflating a situation where we have one positive data point with a whole bunch of scenarios where we have zero positive data points. These are not the same. Any conclusion from the former, no matter how weak, does not transfer to the latter.

The only thing we can say with certainty is that if the overall universe is beyond a certain size ( it doesn't actually have to be infinite ) then there is other life out there. There's a finite size ( Max Tegmark worked it out...it is huge, like 10^10^240 ) at which observable universes repeat due to there being only a finite combination of distribution of atoms. This actually places an upper limit on the odds against life, but it is such a huge number that even within the nearest quadrillions of observable universes we could still be alone.

I would, however, draw two distinctions....

1) There is truly alone...i.e no-one else out there at all.

2) There is 'effectively' alone....i.e there are others out there in the observable universe but they are too far away to ever detect.

My own personal belief is that (2) is the reality.

 

[Ann K]

We are only taking the first baby steps toward identifying "Cinderella" planets. But given that candidate planets have been found, we really don't have any basis for judging the probability of the formation of life. All we really know is that it has been done before, right here on earth.

Addendum: new genetic research has pushed back the date of our last universal common ancestor to 4.2 billion years ago, meaning that it occurred faster than we had formerly estimated. But of course that means it took longer for us to reach our technology phase.
(For those keeping score: +1 to create it, -1 to develop it. )

The last ancestor shared by all living organisms was a microbe that lived 4.2 billion years ago, had a fairly large genome encoding some 2600 proteins, enjoyed a diet of hydrogen gas and carbon dioxide, and harbored a rudimentary immune system for fighting off viral invaders. That's the conclusion of a new study that compared the genomes of a diverse range of 700 modern microbes and looked for commonalities to identify which features arose first.

https://www.science.org/content/art...ion-years-ago-perhaps-hundreds-millions-years

 

[John J.]

new genetic research has pushed back the date of our last universal common ancestor to 4.2 billion years ago

While not in any way qualified to make an informed critique of the relevant paper, I think it's worth remembering that the estimates are based on theoretical molecular clocks and rates of mutation, whose speed and accuracy must be debatable.

"The nature of the last universal common ancestor and its impact on the early Earth system", Edmund R. R. Moody, Sandra Álvarez-Carretero, Tara A. Mahendrarajah et al., Nature Ecology & Evolution 12 July 2024
accessible here https://www.nature.com/articles/s41559-024-02461-1

Maybe 4.2 Bya should be seen as a possible early date for LUCA (last universal common ancestor of all known life on Earth) as indicated by the methodology used.
The authors believe early life on Earth could have arisen (and obviously survived) the Late Heavy Bombardment, briefly casting doubt on the intensity, and even occurrence ("veracity") of any LHB.
-Must admit I thought the LHB was a broadly accepted hypothesis with consensus support, but this might be wrong (according to Wikipedia anyway, https://en.wikipedia.org/wiki/Late_Heavy_Bombardment).

There is at least one major assumption by the authors that lay-people might question, particularly in light of the author's conclusions (that LUCA existed in a diverse ecosystem and had anti-viral defences):

...the LHB hypothesis should not be considered a credible maximum constraint on the age of LUCA. We used soft-uniform bounds, with the maximum-age bound based on the time of the Moon-forming impact (4,510 million years ago (Ma) ± 10 Myr), which would have effectively sterilized Earth's precursors, Tellus and Theia.

(My bold).

As LUCA is very unlikely to be representative of the first life on Earth, and probably existed alongside other early organisms, allowing the maximum age constraint to be the time of a Theia- proto-Earth collision (not even the time by which Earth's crust had re-solidified) must be unrealistic, although in the event their estimate for the age of LUCA is c. 300 million years later.

 

[Ann K]

As LUCA is very unlikely to be representative of the first life on Earth, and probably existed alongside other early organisms, allowing the maximum age constraint to be the time of a Theia- proto-Earth collision (not even the time by which Earth's crust had re-solidified) must be unrealistic, although in the event their estimate for the age of LUCA is c. 300 million years later.

As they give a value of 2600 proteins, it is presumably the product of much earlier organisms. I agree with you about the reliability of the "molecular clock" analysis.

 

[FatPhil]

Addendum: new genetic research has pushed back the date of our last universal common ancestor to 4.2 billion years ago, meaning that it occurred faster than we had formerly estimated. But of course that means it took longer for us to reach our technology phase.
(For those keeping score: +1 to create it, -1 to develop it. )

The last ancestor shared by all living organisms was a microbe that lived 4.2 billion years ago, had a fairly large genome encoding some 2600 proteins, enjoyed a diet of hydrogen gas and carbon dioxide, and harbored a rudimentary immune system for fighting off viral invaders. That's the conclusion of a new study that compared the genomes of a diverse range of 700 modern microbes and looked for commonalities to identify which features arose first.

https://www.science.org/content/art...ion-years-ago-perhaps-hundreds-millions-years

Does that even make sense? Horizontal gene transfer breaks the concept of ancestry.

 

[Scaramanga]

Addendum: new genetic research has pushed back the date of our last universal common ancestor to 4.2 billion years ago, meaning that it occurred faster than we had formerly estimated. But of course that means it took longer for us to reach our technology phase.

Professor David Kipping has dealt with the 'early biogenesis' issue quite nicely. I like his argument, which is essentially that life had to form early on Earth for it to have stood any chance of reaching advanced technological life before life becomes impossible on the planet ( which may be as little as 600 million years time ).

This throws into disarray the whole notion of life 'inevitably' starting early on planets, or life being easy to get going, and shows that ( as so often happens in science ) we have an anthropic data selection bias. If life hadn't started early on Earth...we wouldn't be around to talk about it.

https://pubmed.ncbi.nlm.nih.gov/32424083/

 

[John J.]

This throws into disarray the whole notion of life 'inevitably' starting early on planets, or life being easy to get going

Maybe that's putting it a bit strongly? Kipping's paper is a Bayesian analysis by one man with a very limited data set
(age of Earth, how long before life arose, how long before intelligence arose, how long life might continue), and with no input whatsoever regarding exoplanets (we don't have that information).

However, we caution that our analysis purely concerns the Earth...

From (link) An objective Bayesian analysis of life's early start and our late arrival, D. Kipping (2020) Proceedings of the National Academy of Sciences of the USA 117 (22) (link gives access to full paper).

I guess a researcher could do a Bayesian analysis using approximately-known dates of when one person's parents met, when that person was born, when they held their most senior position at work and when they retired, but I'm not confident that it would generate a strong model of other people's life progression.

Though the idea that life will arise wherever conditions permit is sometimes expressed, I don't think it's a common view.
(IIRC Carl Sagan was rather keen on life being commonplace- I could be wrong though. Fred Hoyle and Chandra Wickramasinghe "saw" life pretty much everywhere, and despite their important less contentious work, were somewhat marginalised).

Professor Kipping's 2020 paper argues that alien life might be commonplace, but intelligent life rare (not non-existent):

If conditions sufficiently similar to the early conditions exist and sustain on other worlds for 1 Gy or more, then our analysis would then favor the hypothesis that life is common, by a factor of K>3. However, the alternative is clearly not discounted and our Bayes factor does not cross the threshold to which it would be conventionally described as "strong"...

Overall, our work supports an optimistic outlook for future searches for biosignatures.

The slight preference for a rare intelligence scenario is consistent with a straightforward resolution to the Fermi paradox. However, our work says nothing about the lifetime of civilizations, and indeed the weight of evidence in favor of this scenario is sufficiently weak that searches for technosignatures should certainly be a component in observational campaigns seeking to resolve this grand mystery.

There's a couple of sentences that I think are questionable:

The early emergence of life on Earth is naively interpreted as meaning that if we reran the tape, life would generally reappear quickly. But if the timescale for intelligence evolution is long, then a quick start to life is simply a necessary byproduct of our existence

(1) Life starting whenever it did is of course necessary for our current existence, but it certainly isn't a byproduct of our existence.

(2) Whatever happened, happened. For a different outcome, initial conditions would have to be different.
The "rerunning the tape" phrase confuses metaphor with reality. If (somehow!) suitable monitoring equipment had been recording the history of the Earth starting from c. 4.6 Bya, "the tape" would show what happened. "Replaying the tape" any number of times wouldn't show a different result. If there had been different initial conditions- perhaps a single molecule in a different location- then the outcome might be very different. Kipping is positing a counter-factual.
Of course, on a different planet, even one very similar to Earth, we don't really know what the odds of life arising are, or if it does, in what timescale.

(3) Kipping thinks that life might not arise as rapidly as it did on Earth- or at all- on other Earth-like planets.
This doesn't seem an unreasonable view. However, he seems to take it as axiomatic that if life arises on an alien world, intelligence will only emerge in a length of time similar to, or longer than, that of our own evolution. We don't know this.
It is implied that if life has a later start than on Earth, it is less likely to have time to evolve intelligence.

However, Kipping doesn't consider that K-type stars are known to have planetary systems and stay on the main sequence 17 to 70 billion years: (link)"Goldilocks" Stars May Be "Just Right" for Finding Habitable Worlds, William Steigerwald (Goddard Space Flight Center) 7 March 2019 on the NASA website.

These stars are of particular interest in the search for extraterrestrial life due to their stability and long lifespan.... K-type main-sequence stars are about three to four times as abundant as G-type main-sequence stars

Wikipedia, K-type main-sequence star.

It's believed that K-types might have higher X-ray and Far-UV output for a longer period early in their main sequence,

This prolonged radiation saturation period may sterilise, destroy the atmospheres of, or at least delay the emergence of life for Earth-like planets orbiting inside the habitable zones

(Wikipedia, link above)
- but their main sequences can be seriously long, there's a lot of them, and there's the possibility of inward migration of planets into a habitable zone after the star has "settled".

NASA scientist Giada N. Arney thinks K-types are worth a look, and that they might allow easier imaging of spectral biosignatures than G-types (Sun-like stars) while being better candidates for life-bearing planets than the more physically unpredictable M-types (red dwarves): The K Dwarf Advantage for Biosignatures on Directly Imaged Exoplanets,
The Astrophysical Journal Letters 873 L7, 2019.

Broadly I agree with you (@Scaramanga), there's no evidence of ETI, and there is the Fermi paradox: I've a horrible feeling that if space-faring civilisations were commonplace, we simply wouldn't be here. Functional starflight must be difficult (if it's possible at all), I'm not persuaded that an ETI would invest in it because they've got a really well-funded equivalent of an anthropology department or a desperate urge to nurture mindfulness across the galaxy.
Not sure Prof. Kipping's 2020 paper gives us much that is new, though.

 

[Eburacum]

...there's the possibility of inward migration of planets into a habitable zone after the star has "settled".

Main sequence stars gradually get more luminous as they get older, so the 'habitable zone' moves slowly outwards over time. This can mean that a cold planet orbiting a K-type star may eventually get warmer even without migrating inwards.

The same thing happens with M-type red dwarfs, but this is a very slow process, and peak habitability for red dwarfs might not occur for hundreds of billions of years.

 

[beku-mant]

The issue I was alluding to is that there's a practical limit to the distance aliens could travel to come to earth- irrespective of how close to the speed of light they are able to get. That's because they would need to establish a destination at the beginning of the journey.

Yes, there are stars within a practical distance, but how many have the full set of conditions needed for life to emerge?

Life (of any kind) is improbable. Intelligent life would be even more improbable. Intelligent life that pursues the relevant technology, and with the inclination and resources to travel the large distances seems exceedingly improbable.

For the sake of the thought experiment lets suppose that 10% the speed of light is doable. That's the speed Sagan claimed was possible for us to achieve with fusion rockets, and it's relatively slow enough to make radiation and dust collisions manageable.

The oldest star systems that could be candidates for intelligent life go back about 10 billion years. A civilization that sent Von Neumann probes out 10 billion years ago that travel at 10% the speed of light would have been able to colonize all stars within a sphere with a radius of 1,000,000,000 light years. That's a diameter about 20,000 times the width of the Milky Way galaxy. If you drew the milky way on your computer screen with a 4k monitor, the entire Milky Way wouldn't even be the size of one pixel.

There would be a lot of stars in that sphere.

 

[FatPhil]

A civilization that sent Von Neumann probes out 10 billion years ago that travel at 10% the speed of light would have been able to colonize all stars within a sphere with a radius of 1,000,000,000 light years.

The leap from "being able to accellerate to 10% of the speed of light" to "being able to colonise" isn't a gimme. Even "being able to slow down from 10% of the speed of light" isn't a gimme. The Voyager probes aren't even capable of slowing down from 0.005% of the speed of light, for example. Note that assuming the feasibility of Von Neumann probes is assuming the conclusion you wish to draw.

 

[Scaramanga]

(1) Life starting whenever it did is of course necessary for our current existence, but it certainly isn't a byproduct of our existence.

I think you are getting the cart before the horse.

Kipping is responding to the argument that life starting early on Earth somehow means life is inevitable. But...Kipping is well known for pointing out that many such arguments are anthropic in nature rather than genuinely Copernican or 'principle of mediocrity'.

So Kipping is making the counter response that if life hadn't started early on Earth....we may well not be here to talk about it. In other words, given how long it takes to get from primitive life to advanced life, if life had not started early on Earth then advanced life might arise too late.

Thus the observation that life started early becomes not an inevitability but a pre-requisite for us even discussing the issue ! On all the planets where life didn't start early there's nobody around to wonder why it didn't start early.

 

[FatPhil]

Thus the observation that life started early becomes not an inevitability but a pre-requisite for us even discussing the issue ! On all the planets where life didn't start early there's nobody around to wonder why it didn't start early.

I view this as insufficiently abstract reasoning. If a hypothesis about advanced life forms can be expressed mathematically, then it will be one of provable, undecidable, or false *whether or not* there are advanced life forms to ponder over it. I don't see the need for the limitation that the system can only be examined from the perspective of something in the system. A simulation scenario would be an alternative one to the mathematical one, too. Why shouldn't the ultrabeings give the same attention to the sim runs that produced no advanced life as to the ones that did produce it? That we're inside one of them pondering it too isn't necessary for that pondering to take place.

 

[John J.]

Kipping is responding to the argument that life starting early on Earth somehow means life is inevitable.

Where does David Kipping say this?

Kipping believes the chances of life arising are good:

Our results find betting odds of >3:1 that abiogenesis is indeed a rapid process versus a slow and rare scenario

Overall, our work supports an optimistic outlook for future searches for biosignatures

David Kipping, 2020 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7275750/

He thinks it is worth actively looking for "technosignatures",

...the weight of evidence in favor of this scenario is sufficiently weak that searches for technosignatures should certainly be a component in observational campaigns...

Agreed that he is saying that if life starts relatively late, there is less time available for a biosphere to exist, the evolution of intelligence is probably a slow process, and so a late-emerging biosphere might prevent intelligence evolving.

Life emerged on Earth within the first quintile of its habitable window, but a technological civilization did not blossom until its last.

For intelligence evolution, it is found that a rare-intelligence scenario is slightly favored at 3:2 betting odds. Thus, if we reran Earth's clock, one should statistically favor life to frequently reemerge, but intelligence may not be as inevitable.

However, we caution that our analysis purely concerns the Earth...

(All ibid).

Taking four data-points from Earth's / life's history, Kipping's Bayesian analysis indicates
(1) life probably arose relatively quickly (but we knew that- else Kipping wouldn't have that data point!) and
(2) "a rare intelligence scenario" is favoured at 3:2.

So in terms of predictive validity, (2) fails for the one planet Kipping has data for.

I'm not sure that Kipping defines what a "rare intelligence scenario" means in quantitative terms and whether he means per hypothetical Earth analogue/ "Earth with the clock rerun", or for the universe en bloc.

...if life hadn't started early on Earth....we may well not be here to talk about it.

Yes, agreed. But we know life on Earth got going relatively quickly, and we are here now.
Whenever life started (e.g. 4.2 Bya, or 3.8 Bya) that is exactly how long it took for us to be here today.
It seems a bit tautologous.

[Kipping:] But if the timescale for intelligence evolution is long, then a quick start to life is simply a necessary byproduct of our existence

Maybe this is just clumsy phrasing by Kipping, but "byproduct" implies a causal relationship- it isn't sensible to say that a quick start to life is a byproduct of (i.e. is caused by) our existence.
We are a byproduct of life starting whenever it did.
If we had become extinct in the past for some reason, the date on which life started wouldn't be altered one jot.

Thus the observation that life started early becomes not an inevitability but a pre-requisite for us even discussing the issue !

Totally agree- but you put it so much better than Kipping does!

Kipping thinks intelligence might have "only" another 500 or 600 million years left for it on Earth.
We've done fairly well in just 200 - 300 thousand years.

 

[Eburacum]

The entire history of multicellular eukaryotic life on Earth is only about 600 million years, starting with the Ediacaran period. So another 600 million years would be a significant extension to the history of life on Earth.