If I understand this study correctly, their conclusion is that it is likely that intelligent species are very rare in the universe. This still implies a "Great Filter", but it would have to be behind us.
"We substituted some guessed numbers for our own" isn't so much a "reasonable claim" as a thought experiment.
Even accepting their numbers, it's hardly "dissolved".
> When we update this prior in light of the Fermi observation, we find a substantial probability that we are alone in our galaxy, and perhaps even in our observable universe (53%–99.6% and 39%–85% respectively).
The key insight isn't the substituion of new numbers, it's the use of probability distributions with error bars throughout the calculation, as opposed to picking a single number. The canonical paradox is that even if you pick very small numbers for all the unknown terms, the resulting expected number of communicating civilizations is still high. This work shows that it isn't a paradox when viewed through the lense of a probability distribution, because even if the expected number is high, that's because of a long tail of super high, low probability numbers, while most of the probability is actually on 'alone in galaxy'.
I would have liked to hear about how they are now accounting for the inherent bias of our detection process to find large, close-to-star planets, because those are the ones that produce the best signals. I remember that that used to be a big issue in drawing any statistical conclusions about expolanet data. Has there just finally been enough data collected that we can model our own bias?
Yeah exactly this. We’re not finding any solar systems like our own because we _can’t_. It’s possible that the majority of stars that we don’t see planets around look exactly like ours and we wouldn’t know.
I have read in other articles that, yes, this very obvious issue has been taken into account and yes, there is still a surplus of solar systems not like ours even so. Hard to link one, though, since there's a lot of noise from articles that don't mention this.
The article says that half of the nearby stars have these large close-in planets, which would suggest that even if there are biases in the detection mechanism it's still at least as common a type of system as ours.
Along the lines of this "detection bias," I've long wondered if there's another bias: the orbital angle of the exoplanets with respect to the plane of ours. If the detection is based on transit across the star of the system, what if we're looking at the axis of the planetary system, and not its equator? All systems similarly oriented toward our instruments would come up as a false negative, no?
(I'm not an astronomer, please correct my jargon).
Another possibility is detecting planets by the wobble that they introduce in the star's position. I'm not well read on the subject, but I wonder if Gaia's position measurements via parallax are accurate enough for this.
The space-based observatory Gaia, launched in 2013, is expected to find thousands of planets via astrometry, but prior to the launch of Gaia, no planet detected by astrometry had been confirmed.
I wonder if there is any models exploring the possibility that our system is uncommon due to 2 systems merging? IOW, Sol ejected a companion star and assimilated the planets from the other.
Off-topic, but I really appreciate that they offer "Cookie Settings", giving me the option to disable ad-tracking cookies. Yes, the button is almost invisible gray-on-gray, but it's there. Is this the GDPR showing its usefuless already?
> Half of nearby stars have at least one (and often several) worlds with masses substantially greater than Earth and orbital periods ranging from mere days to weeks
Perhaps it just seems like such large inner exoplanets are so common because they're easier to find with our current techniques?
They're also easier to find, sure. Maybe the other half are exactly like Sol but we can't tell yet.
But nothing can make 1 out of 2 stars near enough to make out into something uncommon, except our local cluster of stars itself being weird in some way. Which is possible.
To me, that sounds like the Star forming process often results in masses below critical orbiting formed stars. I’m wondering whether they’re all gaseous. It also seems a little too uniform to believe without much more study. We could easily find out these sorts of systems have asteroid belts and planets at a farther distance than we can easily detect.
Or maybe the failed stars failed because they suck up the metals in the system (anything past helium in this sentence).
So, this is actually a big deal, in the "centuries-long arc of science" sense. For hundreds of years, science has operated under the Copernican Principle [1], which in a nutshell is that there is nothing particular special/unique about us. Our position on the planet is not special. Our planet is not special. Our solar system is not special. Our galaxy is not special, etc. It has been a scientifically-fruitful tool for a very long time. It has even embedded itself into the "default worldview" of the scientifically minded as a law of the universe.
But it was never more than a heuristic, and the data is increasingly coming in that the Copernican Principle breaks down at the solar system scale. In fact our Solar System is at least somewhat rare. It is certainly possible that the more we study the rarer it will become. (It can't really go the other direction at this point.)
It's going to take decades for this to properly percolate through the philosophy of science and its implications to be properly chewed on for a lot of the Big Questions.
(Note I'm not making any claims about the outcome of that process here; any such things you read into this are your own worldview poking out. I'm just saying, this is a really big event in the history of science. The sort of result that 23rd century textbooks are going to be calling out in the primary text as an important event.)
For you, that is a personal decision about the nature of "special".
But I feel confident in saying, having watched the debate about the Great Questions for a couple of decades and seen at least the last ~150 years of it, and seen how firmly the default science position has entrenched itself on the hill that the Earth is completely not special [1], it's enough to falsify that position. And that's a big deal on its own.
Is it possible the Earth is just a 1-in-10000 special case? Sure. But even that is enough to open the door to the question why that is, instead of the question just getting shouted down.
[1]: For a modern day manifestation of this idea, consider how casually it is accepted in the "default scientific worldview" that it is very likely that we will find life in the Solar System, and virtually certain we will find it in the rest of the Universe. I'd submit that a rational look at the data gives you no reason to expect life in the rest of the Solar System at this point, with a lot of reasons to expect it won't happen, and I'll be generous and say that we simply have no data about other solar systems. (A less generous jerf would point out that to the extent we have data, it certainly seems negative for large-scale civilizations, which means all the data we have is negative, but I will concede that's somewhat weak for various reasons.) This confidence does not stem from evidence, or even any particularly well-backed scientific theories. It comes from the history of the debate over the Great Questions, and the general trend/desire to ensure that nothing is special about Earth, because the default science worldview would very much like to stuff the commonness of life and the non-specialness of humanity in the religious people's faces.
You'll note I keep carefully referring to the "default science worldview", rather than simply saying "science", because I do not believe the two are the same. The default science worldview takes a number of steps that science qua science can not actually justify. Coming from me, this is merely an observation, not a criticism, since I believe any worldview has no choice but to step beyond science qua science, because "real science" underspecifies the universe by quite a bit.
The default assumption is that we are normal in all ways except what is required by the anthropic principle.
The converse of that is that as we find ways in which we are special, we should immediately start thinking about why those things are necessary for us to be here. The classic example is that a large moon is necessary to keep our axial tilt from random flipping us too far every few tens of millions of years. Another good example is that Jupiter has played "comet sweeper", seriously reducing how many extinction level collisions we suffer.
This new research just underscores the fact that these characteristics truly are essential for advanced life, and are rarer than we knew.
> It can't really go the other direction at this point
What makes you say this? As I understand it, we've only detected a tiny fraction of the planets out there and it could be that there are plenty of systems out there similar to ours, we're just not capable of seeing them yet. Is there some reason to believe we're unlikely to ever see another system like ours, other than the fact that we haven't yet?
The sibling comment to yours is correct in its logic in inferring what I meant.
It is true that we are ignorant of many things. It is not true that our ignorance is infinitely unbounded, and it is a denial of science and rationality, not an affirmation of it, to believe such. We have gathered enough data that we already know that our solar system is not perfectly pedestrian, and no amount of additional data short of the outright falsification of our current data (which, bear in mind, is a bigger ask than a falsification of a theory, which isn't much of an ask at all) can make it more pedestrian; it can only reveal more ways in which it is not normal.
> It is not true that our ignorance is infinitely unbounded
Well, if you define our ignorance as information we don't yet have. Then if the Universe is infinite, it means our ignorance is unbounded.
In fact, if any type of infinity exists, then it means in general our ignorance is unbounded, because we will never be able to have full knowledge of that infinity.
For example, you can make as many assumptions as you want about the properties of the rational numbers in between two other numbers, but unless you are able to somehow "know" all of those numbers, then your ignorance about them is in part bounded by their quantity, and if it there's an infinite amount of them, your ignorance about them will be unbounded.
About half of nearby stars that we can observe[0] have planetary make up that is different than our own Sun. By definition, the rest either have a make up that is like our own or not like our own. So let's assume that every single nearby solar system looks exactly like our solar system, we are just unable to detect that they are at this time. That still means that our solar system is distinctly different than half of the other solar systems. That ratio of difference will never go lower, it'll always be "The solar system is different than at least 50% of other nearby solar systems". But after we do more observations, it might be 60%, 70%, 80%, etc.
Or in other words, as the original commenter said, "It is certainly possible that the more we study the rarer it will become. (It can't really go the other direction at this point.)".
[0] - FTA, "Of order half of nearby stars have at least one (and often several) worlds with masses substantially greater than Earth and orbital periods ranging from mere days to weeks; in our system, the space interior to tiny Mercury’s 88-day orbit is entirely empty."
How useful is that though? They're not a random sampling of planets in the universe--the planets that are easiest to find are very different from our own. Most of those planets are found using methods that work best on huge planets orbiting very close to small stars.
I doubt we've even found all of the planets in our own solar system, so our data on planets is really pretty bad for statistics. Fortunately, the James Webb Space Telescope will be finished eventually[0].
The way I read this article, it rather reinforces the Copernican principle.
There are 3 facts in the article.
1. Planets in other systems tend to orbit faster, in a matter of weeks rather than years; therefore our system is special. Fine, this observation would go against the Copernican principle, but my guess is it's more likely to be an observation bias. Namely, if you have many transitions, before doing any math, you can't tell if it's one or several planets. If you have 10 planets, and you wrongly assume it's one, you get an perceived orbital speed that is 10 times higher (oversimplifying here a bit). If you do the math, and are able to tell apart some of the transitions, you may still think you only have 3 planets instead of 10. Identifying if transitions belong to the same planet is likely to have few false negatives but many false positives. We can follow this space, and see if future analyses confirm the current findings, or confirm my hunch.
2. Distribution of densities. Here the article mentions that for a given radius the density can vary by a factor of 10, but otherwise there is nothing special about the Earth. I would call this a plus for the Copernican principle
3. The distribution of planet sizes. This one is double peaked, and the size of the Earth is part of one peak, not of the valley. This is a big plus for the Copernican principle.
On 1. 10 planets equally radially spaced dont cause a wobble like 1 going fast. Also 10 planets equally spaced are quickly not equally spaced due to differing orbit vel. 10 isn't the problem, assuming aliasing is the problem.
If you are interested in exoplanets, I recommend Adam Frank's book "Light of the Stars: Alien Worlds and the Fate of the Earth". He discusses the history of the Fermi Paradox, the Drake Equation, and how what we know about exoplanets can inform us about Earth's history and future.
37 comments
[ 328 ms ] story [ 1237 ms ] thread[1] https://news.ycombinator.com/item?id=17302924
[2] https://news.ycombinator.com/item?id=17560462
Which is to say that we need to start building interstellar transport vessels soon.
Even accepting their numbers, it's hardly "dissolved".
> When we update this prior in light of the Fermi observation, we find a substantial probability that we are alone in our galaxy, and perhaps even in our observable universe (53%–99.6% and 39%–85% respectively).
There is nothing reasonable or new in either one of those links. Seems like people quoting trash is how myths perpetuate.
P.S. Wrapping a quoted quote in the same kind of quotation marks is kind of liberating after years of " 'ayyy' " or "\"ayyy\"".
However, it's super cool to note that we've actually directly imaged expoplanets orbiting around their stars in other planes: https://en.wikipedia.org/wiki/List_of_directly_imaged_exopla...
Another possibility is detecting planets by the wobble that they introduce in the star's position. I'm not well read on the subject, but I wonder if Gaia's position measurements via parallax are accurate enough for this.
EDIT: A bunch more information from Wikipedia on detection methods https://en.wikipedia.org/wiki/Methods_of_detecting_exoplanet...
The space-based observatory Gaia, launched in 2013, is expected to find thousands of planets via astrometry, but prior to the launch of Gaia, no planet detected by astrometry had been confirmed.
Perhaps it just seems like such large inner exoplanets are so common because they're easier to find with our current techniques?
That's common. It's half!
They're also easier to find, sure. Maybe the other half are exactly like Sol but we can't tell yet.
But nothing can make 1 out of 2 stars near enough to make out into something uncommon, except our local cluster of stars itself being weird in some way. Which is possible.
Or maybe the failed stars failed because they suck up the metals in the system (anything past helium in this sentence).
But it was never more than a heuristic, and the data is increasingly coming in that the Copernican Principle breaks down at the solar system scale. In fact our Solar System is at least somewhat rare. It is certainly possible that the more we study the rarer it will become. (It can't really go the other direction at this point.)
It's going to take decades for this to properly percolate through the philosophy of science and its implications to be properly chewed on for a lot of the Big Questions.
(Note I'm not making any claims about the outcome of that process here; any such things you read into this are your own worldview poking out. I'm just saying, this is a really big event in the history of science. The sort of result that 23rd century textbooks are going to be calling out in the primary text as an important event.)
[1]: https://en.wikipedia.org/wiki/Copernican_principle
But I feel confident in saying, having watched the debate about the Great Questions for a couple of decades and seen at least the last ~150 years of it, and seen how firmly the default science position has entrenched itself on the hill that the Earth is completely not special [1], it's enough to falsify that position. And that's a big deal on its own.
Is it possible the Earth is just a 1-in-10000 special case? Sure. But even that is enough to open the door to the question why that is, instead of the question just getting shouted down.
[1]: For a modern day manifestation of this idea, consider how casually it is accepted in the "default scientific worldview" that it is very likely that we will find life in the Solar System, and virtually certain we will find it in the rest of the Universe. I'd submit that a rational look at the data gives you no reason to expect life in the rest of the Solar System at this point, with a lot of reasons to expect it won't happen, and I'll be generous and say that we simply have no data about other solar systems. (A less generous jerf would point out that to the extent we have data, it certainly seems negative for large-scale civilizations, which means all the data we have is negative, but I will concede that's somewhat weak for various reasons.) This confidence does not stem from evidence, or even any particularly well-backed scientific theories. It comes from the history of the debate over the Great Questions, and the general trend/desire to ensure that nothing is special about Earth, because the default science worldview would very much like to stuff the commonness of life and the non-specialness of humanity in the religious people's faces.
You'll note I keep carefully referring to the "default science worldview", rather than simply saying "science", because I do not believe the two are the same. The default science worldview takes a number of steps that science qua science can not actually justify. Coming from me, this is merely an observation, not a criticism, since I believe any worldview has no choice but to step beyond science qua science, because "real science" underspecifies the universe by quite a bit.
The default assumption is that we are normal in all ways except what is required by the anthropic principle.
The converse of that is that as we find ways in which we are special, we should immediately start thinking about why those things are necessary for us to be here. The classic example is that a large moon is necessary to keep our axial tilt from random flipping us too far every few tens of millions of years. Another good example is that Jupiter has played "comet sweeper", seriously reducing how many extinction level collisions we suffer.
This new research just underscores the fact that these characteristics truly are essential for advanced life, and are rarer than we knew.
What makes you say this? As I understand it, we've only detected a tiny fraction of the planets out there and it could be that there are plenty of systems out there similar to ours, we're just not capable of seeing them yet. Is there some reason to believe we're unlikely to ever see another system like ours, other than the fact that we haven't yet?
Based on the best available scientific tools at the time, we once thought that the solar system rotated around the earth.
Based on the best scientific tools available today, we believe that our solar system and planet is extremely unique.
I think we don’t know nearly enough to say with any real confidence what solar systems outside of ours look like at this point.
It is true that we are ignorant of many things. It is not true that our ignorance is infinitely unbounded, and it is a denial of science and rationality, not an affirmation of it, to believe such. We have gathered enough data that we already know that our solar system is not perfectly pedestrian, and no amount of additional data short of the outright falsification of our current data (which, bear in mind, is a bigger ask than a falsification of a theory, which isn't much of an ask at all) can make it more pedestrian; it can only reveal more ways in which it is not normal.
Well, if you define our ignorance as information we don't yet have. Then if the Universe is infinite, it means our ignorance is unbounded.
In fact, if any type of infinity exists, then it means in general our ignorance is unbounded, because we will never be able to have full knowledge of that infinity.
For example, you can make as many assumptions as you want about the properties of the rational numbers in between two other numbers, but unless you are able to somehow "know" all of those numbers, then your ignorance about them is in part bounded by their quantity, and if it there's an infinite amount of them, your ignorance about them will be unbounded.
About half of nearby stars that we can observe[0] have planetary make up that is different than our own Sun. By definition, the rest either have a make up that is like our own or not like our own. So let's assume that every single nearby solar system looks exactly like our solar system, we are just unable to detect that they are at this time. That still means that our solar system is distinctly different than half of the other solar systems. That ratio of difference will never go lower, it'll always be "The solar system is different than at least 50% of other nearby solar systems". But after we do more observations, it might be 60%, 70%, 80%, etc.
Or in other words, as the original commenter said, "It is certainly possible that the more we study the rarer it will become. (It can't really go the other direction at this point.)".
[0] - FTA, "Of order half of nearby stars have at least one (and often several) worlds with masses substantially greater than Earth and orbital periods ranging from mere days to weeks; in our system, the space interior to tiny Mercury’s 88-day orbit is entirely empty."
[0]: https://www.xkcd.com/2014/
There are 3 facts in the article.
1. Planets in other systems tend to orbit faster, in a matter of weeks rather than years; therefore our system is special. Fine, this observation would go against the Copernican principle, but my guess is it's more likely to be an observation bias. Namely, if you have many transitions, before doing any math, you can't tell if it's one or several planets. If you have 10 planets, and you wrongly assume it's one, you get an perceived orbital speed that is 10 times higher (oversimplifying here a bit). If you do the math, and are able to tell apart some of the transitions, you may still think you only have 3 planets instead of 10. Identifying if transitions belong to the same planet is likely to have few false negatives but many false positives. We can follow this space, and see if future analyses confirm the current findings, or confirm my hunch.
2. Distribution of densities. Here the article mentions that for a given radius the density can vary by a factor of 10, but otherwise there is nothing special about the Earth. I would call this a plus for the Copernican principle
3. The distribution of planet sizes. This one is double peaked, and the size of the Earth is part of one peak, not of the valley. This is a big plus for the Copernican principle.
Talk to my when you’ve detected all planets in a random representative sample of stars.
https://smile.amazon.com/dp/0393609014