So, I read this paper and was pretty unconvinced. I think the premise of using a Monte Carlo analysis is flawed because of course multiplying together wide ~iid distributions that have big near-zero components asymptotes to zero; unless I'm misunderstanding something, this is essentially the central limit theorem. (All of the parameter samples below 1e0 drag the total outcome towards zero since probabilities are bounded >= 0) Thus, the real information is still strongly dominated by beliefs about the width and centering of the distributions, which we still don't know, and as long as enough of the parameters have high uncertainty with some potentially having near-zero values, the result will always tend towards zero.
I think what this analysis tell us is that if you don't think it's possible to get better, more specific values for the parameters, the Drake Equation is just a bad tool.
Also, the Drake Equation doesn't describe the number of alien civilizations, it describes the likely number of civilizations. Thus, extremely pessimistic estimates have to somehow correct for the fact that we know N is not zero, since we are here. That seems like a highly informative update that is not addressed in an interesting way.
Yes, the Drake equation seems quite pointless because multiplying many unknowns by one another doesn't reduce uncertainty, it increases it. I don’t know anything about the history of the Drake equation, but it looks possible it was devised to demonstrate the unknownability of the result, rather than a tool to help find it.
Also, the article seems to use "our galaxy" and "the universe" as interchangeable terms, which of course they are not.
I don’t see how the Fermi paradox is a paradox: given the distances involved and what we know about the speed of light, an alien civilization would have had to develop a very long time ago in order to send signals that we would be able to detect today. Just because, maybe, no civilization was able to do that a billion years ago doesn’t mean nothing happened since (as you say, at least we happened).
The "Fermi paradox" isn't meant to be an actual paradox in formal terms.
It is simply asking basically "if the universe is teeming with life, where is it?" and by that raising both the question of whether or not the universe actually is teeming with life, and if it is, why aren't they here? It reduces to an actual paradox only if you assume an answer to one part (the amount of life) that isn't reasonably compatible with the second part (why we haven't seen evidence of it) without a reasonable explanation.
That it simply on average takes too long for us to have seen evidence yet is in fact one of the many suggested solutions.
But overall the point of the Fermi paradox is that the problem if why we don't see aliens is two-pronged: They might never develop, or just not have developed yet, to a sufficient stage. But it's also possible they do, but that something else prevents them from contacting us.
Thus the idea of "great filters" for example. E.g. maybe the universe is full of life, but it's exceedingly rare for it to survive long enough to be able to make contact without being destroyed. Maybe the odds of getting killed off by asteroids is too high. Maybe war takes out most that get to a certain level. Maybe someone out there takes pot-shots at anyone sticking their head out? Maybe some galactic events sterilises large swathes of space too often? (intense gamma bursts have been raised as one possibility)
As well as many other options. E.g. maybe we're quarantined? Maybe we're in a simulation, and simulating aliens just isn't part of the program?
> The "Fermi paradox" isn't meant to be an actual paradox in formal terms.
I think this is a misunderstanding of the use of the word "paradox". Informally, people use "paradox" like it means "contradiction" when in actuality it means "APPARENT contradiction". See, e.g., https://en.wikipedia.org/wiki/Paradox
> A paradox is a statement that, despite apparently sound reasoning from true premises, leads to an apparently self-contradictory or logically unacceptable conclusion.
The Fermi paradox is indeed an apparent contradiction, whether or not it is an actual contradiction.
The same holds for many, if not almost all uses of the word "paradox" in standard literature. For example, Zeno's Paradox is not a contradiction, we've resolved it (an infinite sum of decreasing values can add up to a finite number), and yet it is still a paradox.
(For a more precise classification, see e.g. Quine's classification of Paradoxes)
Pretty sure we already have one answer to the paradox: the dinosaurs, and how long they were around without any sort of civilization. Simply reaching intelligence has got to be one of the great filters.
It's hard to truly grok just how long the dinosaurs were around. It's easy to say that, had life not almost been wiped out by an asteroid things would have worked out differently, but the dinosaurs had something like 200 million years and there's no evidence they developed anything approaching human-like intelligence.
> It's hard to truly grok just how long the dinosaurs were around.
One of the crazier things I've heard recently is that the time between the present and the Tyrannosaurus Rex [1] (68-66 million years) is less than the time between the T-Rex and the Brontosaurus [2] (156-147 million years ago, or 79 million years before the T-Rex).
What would such evidence look like? I don't think we can consider brain size definitive, since it's possible for humans with about ~10% of our normal brain volume to function normally. There seems to be no evidence for tool use, much less advanced technology, but again, what would these even look like after so many millions of years? I think about all we can say is that no prior species changed the planet in the same fashion we have.
They were, but mammals at that time were pretty simple and weren't doing that well against their dinosaur competitors. When the asteroid took out most of the dinosaurs, that's when mammalian life really took off, because they were able to survive the many years of lack of sunlight that ensued.
However, there are quite a few other species still around which are pretty intelligent (elephants, other great apes, whales and dolphins, octopi) so the development of intelligence may not be all that rare.
But none of them have the ability to pass learned knowledge through many generations the way that we can. For most species, what a mom can teach a kid is very limited, and special skills tend to be static between generations as opposed to continually building on each other
Several other species have opposable thumbs[1], and many mammals have elaborate long-term food storage. Other apes have been observed hunting with spears; basket making and song are also practiced by birds; cetaceans might have more advanced spoken language; sweat and bipedalism aren't common but aren't unique; etc.
But no other (non-extinct) species can start a fire.
Human language is somewhat unique in that it can describe multitude of future outcomes, not just a present state of things. Turing complete if you will. This is a bit of a wild idea I have anyway, don’t judge me harshly. :)
I’ve been hanging around with linguists, this idea sort of emerged from my eavesdropping.
In particular it’s accepted in the field (I believe) that animal languages are not generative, which is to say they do not provide for production of new forms over time. Anything an animal speakes has already been spoken before because the possible set of valid sentences is strangled by lack of variability in the syntax. Intuitively, and this is where I’m going off the rails, I feel that generativity and Turing-completeness are near the same thing - they can describe something that is yet to be. Non generative language can only describe what has already been. That is constructing a new world requires description of the machinery operating that new world, whereas describing an old world you can lean on prior knowledge of the speaker and make do with a much poorer descriptive facility.
I am sort of a hobby linguist, though I do have some years of formal linguistics (and computational linguistics) education, which is why I was interested in your wild idea and how it relates to animal language. Are you using the term "generative" to describe your own specific meaning here? I'm asking because the term "generative" has a pretty specific definition in linguistics, and I have heard it used fairly loosely by some, but never in this way. And while I think your idea is interesting, I also think it is picking nits here. That animal language lacks some feature of human language doesn't make it any less true that humans and other animals both use language. I think this would be similar to saying that because animals don't build defensive positions that they aren't actually conducting warfare, and defensive positions are what make human warfare truly unique and thus humans differentiated in the animal kingdom. Purpose-built defensive positions are as fundamental to the most basic definition of warfare, as I am thinking this feature of language you describe is to the most basic definition of language. But nobody would say that chimpanzees do not conduct warfare. And so I would say that because animal language lacks a complex grammar that isn't a language. If that makes sense.
There was a linguist who brought up those five key attributes of human language as being crucial, I can’t find his name right now :(
And I did not say that animals don’t use language. Or did I?
What I mean to say is that there is a dramatic difference in expressive power, in particular in being able to convey a plan of action. Animal language can convey “there is a tiger south-south-west 15 meters, run for your life”. But it can’t express “if there is a tiger run left, if there are two tigers run to the right” unless the animals have previously encountered that situation and survived unwittingly executing this plan of action.
To me this is akin to the difference between Turing complete and subturing languages, in that the expresssive power is so very different.
Elephants, whales, octopi, etc didn't develop their intelligence independently from humans. We're all pretty closely related species that evolved from the same origins.
What would be interesting is if we found intelligent (or any!) life that evolved completely independently from our own origins.
In the great scheme of things we're related to octopuses, but not nearly as closely as other mammals.
Also birds, the direct descendents of the dinosaurs and more distantly related to us, can be incredibly intelligent. See toolmaking in crows for example.
> octopi [...] didn't develop their intelligence independently from humans.
Are you sure?I seem to recall watching random youtube videos a few weeks / months back and saw something on octopi I vaguely recall it saying they developed intelligence differently. Although probably I have got it muddled with something else :)
The common ancestor between humans and cephalopods was around 560 mya, which isn've that closely related, and this would have been before the common ancestor developed a sophisticated nervous system.
But we are related. We have the same origins and some of the same DNA. Intelligence may have arisen independently in Octopi compared to humans, but the conditions and building blocks that made it possible were shared. All life on Earth shares a common ancestry.
Exactly. And the coming of humans as a dominant specie on Earth is pretty much coincidental (humans faced extinction several times, that we know) so there is absolutely no "fate" that leads to human-like intelligence.
- Presume that even with something like travel at 1% o speed of light, the Milky Way is colonizable in a couple of million years.
- Presume that we are not special at all and there exist more technological advanced civilizations that predate us by at least a couple million years
- Presume such a large scale colonization would be easy to observe
If you accept the 3 presuppositions, the only logical conclusion is that there are no other technologically advanced species in the Milky Way or that we're the first(very unlikely).
Or you can choose to attack any of the presuppositions, but that's also very hard in my humble oppinion.
> Presume that even with something like travel at 1% o speed of light, the Milky Way is colonizable in a couple of million years.
You're not assuming that it is, you're assuming that if it can be, it must be.
Which assumes that every part of the milky way is equally amenable to colonization, as if the galaxy were merely points laid out on a Cartesian grid, and that any species capable of colonizing one star must continue to colonize them all.
You must consider the case of technologically advanced species incapable of colonizing other stars, and technologically advanced species capable of colonizing other stars, but unwilling to do so on a large scale, or attempts at large scale colonization which fail across a scale of millions of years or even attempts to hide the obvious presence of such colonization from observers.
>Presume that we are not special at all and there exist more technological advanced civilizations that predate us by at least a couple million years
Ok.
>Presume such a large scale colonization would be easy to observe
By humans, given that we only recently became aware that extrasolar planets are even a thing, and have only been studying the universe with anything but optical telescopes for less than a century, and don't even know what the signs of "large scale colonization" would look like. No. Why?
>Or you can choose to attack any of the presuppositions, but that's also very hard in my humble oppinion.
Any argument can seem airtight when you refuse to consider more than a single case to be valid.
But it only takes one advanced civilization to do it. And we're probably only a century away from being able to spam the entire galaxy with probes, if we wanted to say take apart Mercury and use solar panels for propulsion. The probes would be small, smart and able to replicate themselves.
> And we're probably only a century away from being able to spam the entire galaxy with probes, if we wanted to say take apart Mercury and use solar panels for propulsion. The probes would be small, smart and able to replicate themselves.
1. We're really not.
2. Solar panels obey the inverse square law. Right now, we can't even use solar panels to power probes beyond the orbit of Jupiter - let alone drive an energy-hungry ion engine.
3. Even with an ion engine, you need fuel to thrust.
4. Even if we could send them, they'd need incredibly power-hungry antennas to transmit any useful data back. See 2.
People say things like 'only 1% of c' as if it's trivial. It's only 1%! It's also 3,000 km/s. No man-made machine has ever gone faster then 16 km/s. If we tried really hard, we might hit 1% of 1% of C.
And then it would still take that probe 400 years to visit our nearest star. 10 million years to cross the galaxy.
I find doubling your delta-v requirements to be far simpler then increasing them a hundred-fold. If we could get something to 1% of c, we could, with a bit more work, slow it down the same way.
I suspect your second and third points contradictveach other.
We literally have no idea what a >10^6 more advanced civ would look like.
There’s a kind of parochialism about these arguments. They seem to assume that aliens will be flying around the galaxy communicating by radio in big metal boxes or giant rocket powered rocks looking for planets to xenoform.
But in the same way that ants can’t see a human city around them, it’s possible humans will see no evidence of an advanced civ that isn’t doing any of the things we expect technology to do.
> We literally have no idea what a >10^6 more advanced civ would look like.
Yep. We don't even have much of a clue about what our own civilization will look like a few hundred years from now, it seems insane to pretend we would know what a far more advanced alien civilization would look like. Imagine showing someone from a thousand years ago a picture of suburbia and them saying, "Nope, this place must be abandoned...no one is growing crops on the land, there are no wells around, deer and other wildlife are walking around houses without being hunted. Must be no humans there." Even just a few decades ago a lot of people thought we would be building cities under the seas by now (hey, maybe not everyone would do it - but with billions of humans, it would only take a few splinter groups).
It's also worth pointing out that even a civilization at our level existing at the nearest star would be undetectable to us in most circumstances (even in the extremely unlikely event that the timing matched up).
These can be very stupid questions; but I always ponder over it
1. Is this our assumption that extra terrestrial life will have same chemistry/ biology/ consciousness as human beings?
2. What is definition of Life and Intelligence for E.T life. The phenomenon what we call life can be completely different for E.T. being. This is more or less same as que 1
3. It seems we have also assumed that our civilisation is advanced. In terms of what? Is our definition of advancement (and also civilisation) applicable to E.T life?
4. Consider a highly advanced civilisation millions of light years away from us finds Pioneer plaques. Even after finding these plaques they don't (wish to ;-)) communicate with us... does it mean that civilisation is not sufficiently advanced?
But that's the beauty of this calculation it just takes one splinter group of one splinter group and they could easily populate the whole galaxy in under 10-100 million years - which, being the conclusion that Fermi found, is nothing at the time scales of when our galaxy was formed.
Also, less than 5% of the universe is observable to us. The rest we're calling "dark matter" and "dark energy" because we have no idea what it is. There's a lot we don't know that billion-year-old civilizations may have figured out to use.
> Milky Way is colonizable in a couple of million years.
I have two takes on it:
1) We don't know the possible implication of new discoveries. Well, actually we get some glimpse of it, for example with nuclear or biological weapons around, it is fairly easy to wipe out a large part of the population. The more advanced the technology becomes it seems there is a higher chance of it blowing up in our faces and us destroying ourselves. So the galaxy could be colonizable in a few million years, but any civilization headed closer to that ability, gets closer (and faster) to its own destruction as well.
> The only logical conclusion is that there are no other technologically advanced species in the Milky Way or that we're the first(very unlikely).
Given my other pessimistic take on it (1), I would say it likely there were (are) quite a few technologically advanced species, but they've wiped themselves out or will do so before the reach the ability to colonize planets in other galaxies.
2) Another take on it is these civilization have to share the same physics and chemistry that we do. And maybe there is just a limit that gets achieved and long term colonization or fast travel is just no possible. That is there are no shortcuts like wormholes, Alcubierre drives, antigravity and other such things. We've been used to seeing huge leaps in technology and science in the last 1000 years, so it's tempting to extrapolate and but maybe we are close to reaching a wall. So there are quite a few civilizations out there but they've flown probes in their own start system, now developed internet and are sharing pictures of alien cats, hanging out on their version of Facebook and watching their alien Netflix, and of course discussing their equivalent of Fermi's Paradox on online forums. All without a chance of ever leaving their neighborhood.
1) It takes just ONE civilization not wiping itself out to colonize the galaxy; furthermore, colonizing the galaxy is a great insurance policy against extinction
2) 1% of the speed of light is very achievable even with today's technology, it's just the(irrational) fear of nuclear reactors in space and lack of political will that keeps us from exploring interstellar travel.
> 1) It takes just ONE civilization not wiping itself out to colonize the galaxy; furthermore, colonizing the galaxy is a great insurance policy against extinction
Why? See my other post. You provide no evidence for this claim. We currently have no plans to colonize the galaxy so why would other "advanced" civilization get this crazy idea? Birth control was invented in the 60s already.
We have Musk wanting to colonize Mars, don't you think it will take just ONE eccentric trilionaire (after the solar system is colonized) to colonize the Milky Way?
>colonizing the galaxy is a great insurance policy against extinction
I'm not so sure. The only difference between a colony ship and a kinetic world destroyer is that the colony ship slows down before it reaches its destination
1. If the chance for a huge colony to go extinct, raising the coefficient to galactic scale does not change the relative probability significantly unless you add constraints to the population dynamics. Basically you could get the same numbers for people in a colony and colonies in a galaxy. "only ONE" kinda obscures the fact that this one would be a mighty huge one.
2. On straight lines perhaps with ages long acceleration. Planning such a course might well take more than nuclear reactor, no?
I'm going to put on my sci fi glasses here and predict that the universe will suddenly blink out of existence once a scientist works on or develops something like warp speed.
Maybe, but I’d out my money on the universe blinking out of existence after someone begins a local escalation of privilege on the hardware simulating it.
Project Orion was supposes to reach ~10% of the speed of light by using fission bombs and could theoretically be built using 1950's technology.
There are other proposed schemes to travel at ~light speeds - light sails, nuclear fragment rockets, but I'd bet on nuclear fusion in the not so distant future.
Faster than light travel may be off the table, but I see no reason why nuclear energy alone could not get us "close" to the speed of light in the next centuries or so.
- There is no reason to believe that species want to colonize the whole Milky Way. In fact, our current value and belief systems would not recommend such an expansion either. (The same for self-multiplying robot ships - if you think it's a bad idea, why would aliens think otherwise?)
- Cyclical cultures, filters, and large time frames are not properly taken into account. If you take them into account, colonization of the whole Milky Way seems unlikely.
- Life could occur only very rarely.
So basically, the missing premise is that if all of your conditions are fulfilled, then a civilization will colonize the Milky Way and has done so far enough in the past that we can easily observe it. There is no reason to think that.
Now here is the bad news: Even if we accept your original line of thought, that would still be entirely compatible with the view that some intelligent species is just about to colonize the Milky Way and just in front of our doorstep...
I'm just kidding, of course. Like others have pointed out, this "research" is almost entirely based on speculation. We don't even know how likely life would arise if all conditions for life are fulfilled, let alone whether that also means that intelligent life will arise or with which likelihood. We know nothing.
The only thing that we can say with relative certainty according to our current knowledge is that if intelligent life develops with high probability whenever the conditions for life are met, then it is a bit harder to detect then we thought or Xenoarchaeology will be booming some day in the future.
> In fact, our current value and belief systems would not recommend such an expansion either.
This just isn't true. There are current value and belief systems, even ones held by people with a relative large amount of political and economic power, that do in fact recommend such an expansion (someday).
I would say it's the dominant value system, given that our global economy and energy consumption grows by around 2% per year, with no serious efforts to change that.
It's worth pointing out that for developed economies like the US, UK, and Japan, the amount of per capita energy usage is the lowest it has been in decades:
This may be at least in part due to the shifting of energy-intensive industries to countries like China, but it does show that countries can continue to grow their GDP (per capita) while dropping their energy usage per capita, which bodes well for the future.
> There is no reason to believe that species want to colonize the whole Milky Way.
There are definitely reasons to believe that.
You certainly don't want to be in a galaxy where someone else does it first: if you can't coëxist with them, you're probably going to be wiped out.
Large-scale projects like Dyson spheres would, I imagine, require vast resources taken from multiple solar systems. So even if you're not colonizing every star, you're probably mining from many of them.
Anyway, just a couple of reasons that spring to mind. Given our data set is exactly one intelligent civilization it's hard to draw conclusions, but expansionist pressure seems to be the safe assumption.
> You certainly don't want to be in a galaxy where someone else does it first: if you can't coëxist with them, you're probably going to be wiped out.
Well, since the problem of even figuring out if you are alone is nontrivial why would you opt to do such a astronomical operation because of a hypothetical? We most certainly would not.
On the assumption that all other civilisations are either hundreds of millions of years ahead or behind you; if they are hundeds of million years more advanced, they surely already know of your existance and if they had any hostile intentions you wouldn't even exist, while if they are behind you, they are at your mercy after you aquired all the resources in the galaxy.
> You certainly don't want to be in a galaxy where someone else does it first: if you can't coëxist with them, you're probably going to be wiped out.
Why the pessimism? We do know that we currently do not have any plans for massive expansion through the whole Milky Way, and even if we made such plans, we would definitely include safe zones and ample space for other species we might encounter. If we are that good, maybe other species are no less good? That's not an unreasonable assumption, though ultimately anthropocentric, of course.
But the point I was trying to make was different. I should have said "There is not enough reason". There are just as many hypothetical reasons for as there are against. We really don't know.
>expansionist pressure seems to be the safe assumption.
This is the first I've ever heard in my life that intelligent species wouldn't want to spread to the stars. Given the historical impetuous of our species to expand to every even remotely habitable region of Earth, and large support for living off the planet if it were possible, why would you think aliens would be the same? Every single living organism on this planet fights constantly to expand it habitable zone.
Probably the best reasons to spread out are:
a) As insurance against Earth 0 getting blown up by a passing asteroid, and/or
b) You found someplace better.
Possibility (b) seems at least somewhat unlikely, given that we (at least) evolved over millions of years to the very particular conditions of Earth 0. So finding somewhere 'better' would likely involve some gene-gineering to adapt, and maybe involve a shallower gravity well than we currently inhabit.
Possibility (a) is a really good reason to spread out a bit, but you likely don't need more than a few insurance policies... Certainly there's no need to aggressively conquer the entire galaxy: Given difficulties of communication and transport over long distances, there's likely some real question of what 'conquer' would even mean...
It occurs to me that it would be interesting to look into how much stars drift apart from one another over the course of time involved in getting from one to another. Consider that the launch window for mars comes up once every two years; what are the launch windows like for stars, where travel might take hundreds or thousands of years? Maybe it takes a thousand years to get a scouting run done, and then it's infeasible to follow up with an actual colony ship...
If you have the ability to live in space long enough to cover the vast distances between stars, why would you bother colonizing planets at all? Pretty much all the raw material you need is readily available in asteroids without having to deal with a steep gravity well or a thick atmosphere, and the star itself will give you plenty of energy to work with.
Pretty much every star system becomes habitable if you don't have to care about whether or not it has a planet that meets your species's environmental needs.
Only if gravity isn't a requirement of life. We know that bad things happen to humans when in low/zero-g for more than a month. What happens to when we are talking about lifetimes - science is exploring this topic but doesn't really have answers.
A civilization advanced enough to colonize other star systems likely
1) is advanced enough to tweak their genetics to deal with low/zero gravity
2) is advanced enough to build a ship with artificial gravity (we came up with this idea long ago, go watch "2001", it's not hard).
> We know that bad things happen to humans when in low/zero-g for more than a month.
This is mainly because humans are adapted to live standing upright in gravitational conditions, with a pressure gradient across their body. If you take animals which are not adapted for this (e.g. large invertebrates from the sea) they're crushed by their own weight when taken out of a buoyant environment.
Even so, that doesn't make it infeasible to live in space for long periods of time. Artificial gravity can be generated with a centrifuge.
However, I would say it's likely that gravity is necessary for life similar to what we are familiar with to form in the first place. Earth's gravity facilitates the existence of dense fluids, allowing for the complex reactions of primordial life.
Because controlling all of the resources of the galaxy, or at least having that option, shields your civilization from other less civilized/hostile civilizations?
A WW1 wireless operator of only 100 years ago wouldn't be able to detect modern spread-spectrum military radios, let alone separate encrypted signals from the noise.
Why should we suppose we could detect a civilisation two million years more advanced?
I think the point is that they would have colonised Earth already. It is comparing the age of the Earth (billions of years) to the size of the galaxy (100kly)
Or maybe the value of colonising a planet relative to the cost just does't make it worth it by the time you have the technology to try.
Maybe every advanced civilisation ends up turning inwards to simulations rather than outwards. Maybe there are aliens all around us, but we can't notice them, because we're living in someones phone, in the latest top-hit: SimHumans5000.
There's quite literally an infinite number of possibilities where we're not alone after all, and part of the problem is that we don't just not know the values to plug into the relevant probability calculations, but we don't even know all the variables.
An inwards turning civilization would be much more incentivised to collect all the matter they can since E=mc^2, and for them E is their lifespan.
For a civilization living for trillions upon trilions of years inside a simulation, intergalactic latancy is just a blink of an eye -- they have no reason not to distribute their simulation across the galaxy.
All of the burning stars would be just lost time for such a civilization
Sorry, guess I wasn’t clear. There’s folks who think intelligent life on earth could’ve been caused by alien life splicing genes with pre-existing species on the planet. Still rather loony, just in a different way than the abducted folks.
>If you accept the 3 presuppositions, the only logical conclusion is that there are no other technologically advanced species in the Milky Way or that we're the first(very unlikely).
Or they have a culture/value system that doesn't care about sprawling out to occupy as much space as possible. Why is it presumed that advanced alien civilizations would use their technology to maximize, rather than minimize their ecological footprint?
It stands to reason that environmental degradation could be one of the major filtering mechanisms for advanced life and any civilization that clears the filter will need to have fostered sustainable, symbiotic production systems rather than extractive ones.
Game theory basically. Any civilizations that minimizes would be destroyed by any civ that maximizes, so the survivors must be maximizers. Furthermore any civ that minimizes would never get to an advanced stage in the first place.
>Any civilizations that minimizes would be destroyed by any civ that maximizes
Unless we're talking about hive-minds I don't think this premise holds at all.
Civilizations that culturally value domination and expansion can dominate within their own civilization much more easily than casting out into space to find alien ones. They're more likely to destroy themselves than others.
>the only logical conclusion is that there are no other technologically advanced species in the Milky Way or that we're the first(very unlikely).
Why is this unlikely? The likelihood of intelligent life emerging that gets far enough in their development to try to colonize a galaxy might be so low that it could easily only be once per galaxy or even rarer than that.
Think about how long multicellular life has existed on this planet and yet a very narrow band ended up 'becoming intelligent' and even then it took a long time before industrialization that finally propelled us forward.
More technical advanced meaning they can effectively use all resources on their own planet?
Or even further where they can effectively gather all the energy from their star?
The only reasons to branch out is because your own star is dying or your population (demand for energy) is growing too fast.
Essentially I think that population control is a big hurdle for any advanced species. Which would be the main drive for colonizing space. But at what rate does it grow? After their first star, how long does it take for the next one to be used fully.
Why would a species want to colonize a galaxy? You’re talking about insane travel and communication times after all, and spreading very thin. Sticking to a single system or handful of systems while controlling reproduction rates is a solution that achieves the same ends without spreading your species so far that it amounts to a million different civilizations. At some point over such distances you’re not colonizing anything, just throwing generations into the night with no hope of a return. That same energy could be used to engineer a structure or structured for those same beings to live within a reasonable distance of each other.
The whole Fermi Paradox reasoning is flawed, because the assumptions are flawed.
P(at least one technologically advanced civilization observes itself | there exists at least one technologically advanced civilization) = 1
We will observe ourselves, no matter how it is unlikely for us to exist. We may not be special, but highly improbable and therefore the first and the only.
You're just stating an anthropic-principle argument here, and unfortunately anthropic arguments tend conceal a fallacy of equivocation. You can look up some of William Lane Craig's work for a formal logical analysis showing exactly where it goes wrong, but the traditional plain-language analogy is:
Imagine you're sitting in a prison cell, and you've been told you're going to be executed by a firing squad of a dozen expert marksmen, who will shoot you from a distance of only a few feet away.
You're led out of your cell, tied to a stake, the marksmen line up, they raise their rifles and aim at you, and you hear the mass bang of their simultaneous firing. And... you discover you are unharmed. All twelve expert marksmen completely missed you, at close range!
The anthropic argument says you should be completely unsurprised; after all, the only way for you to be observing the marksmen now is if they all missed, so the probability of them having missed, given that you are observing them, is 1. We should not be surprised by something which has probability 1.
While this is trivially true, it is also utterly unconvincing and utterly pointless. You are perfectly justified in being surprised that all the marksmen missed, and here is the equivocation: the anthropic argument tells you not to be surprised, but the thing it tells you not to be surprised at is not the thing that was actually surprising. Your prior for all the marksmen missing and leaving you completely unharmed should have been very, very low, and no amount of ex-post-facto wrangling can change that.
Or, put more simply, anthropic arguments can be used to launder any past event to a probability of 1; after all, the probability of event X having happened, given that you observed event X, is always trivially 1. But it tells you nothing useful whatsoever.
> the probability of them having missed, given that you are observing them, is 1. We should not be surprised by something which has probability 1. While this is trivially true, it is also utterly unconvincing and utterly pointless.
It's completely convincing to me. If you are unharmed, no single bullet hit you. Even if you've had the "prior" information like that they are "a firing squad of a dozen expert marksmen" if you are "utterly surprised" by the result it probably means that your priors simply miss something, not that it couldn't have happened at all. E.g. maybe that dozen of experts had some additional decision you didn't know about, e.g. they want to influence a regime change and agreed before that it would start with you not being harmed.
Such kind of information turns very improbable event to very probable.
More than that, even improbable events do happen all the time. It is very improbable that I personally will never win a huge prize on the lottery ticket, but the lotteries are intentionally made that somebody from time to time wins it. If we are a "lottery winner" and don't know anything about the rules of the lottery and the number of the participants, we just can't conclude anything else about the lottery probabilities than that we got what we got. As long as more different universes are possible, no matter how we think our is improbable, as long as we have information only about our state we don't have to be able to conclude more than that we have what we have.
So there's no value is empty talk, we know more only by moving the boundaries of what we can measure. We have actually have learned amazingly much about the Universe, but there's no "guaranteed first prize win" ("learning everything") in these endeavors.
Such kind of information turns very improbable event to very probable.
The issue is, as I pointed out, that it works for anything and produces the same result. Anthropic arguments literally consist of stating that "P(A|A) = 1". Which of course is true, but what does that actually tell you?
> The issue is, as I pointed out, that it works for anything
If you find it used "for anything" you can point to such cases where it is improperly used. When it is used for the case where we genuinely don't have enough information to claim any "probability" then it is simply a more colorful statement of that fact: that there is really nothing actually measured and known based on which any other statement than "we know exactly one example that such outcome is possible because we live in this outcome" is honestly impossible.
You may psychologically not like that fact, but when there is no more data, it has to be admitted that there is no more data to claim anything else than that. Maybe the following can help you: Imagine you find yourself in the middle of the desert, in all directions you see only sand. People today know that every desert ends somewhere, but that's based on the fact that enough people don't live in the middle of the desert, so you must imagine that you are the very first person that ever existed and that you find yourself in the middle of the desert. While moving around the place, looking for any signs, you may or may not find the proofs about something else in the distance. But you can genuinely be in the phase where you see only sand and can't conclude anything about the existence of, for example, the rivers or the sees.
The way I see it, the message is not "give up any attempt to learn more about the Universe you are in" the message should be "sitting around and inventing stories doesn't help: you have to move and actively search to find out more." The humanity was for a while good in inventing stories that sounded comforting but were completely unsupported, but we have actually in the last 100 years found the exact and provable signs about much more of our Universe than ever in our history: think about it: 100 years ago nobody knew that there exists anything more than one galaxy: ours. Today we can actually claim that there are one hundred billion galaxies in our Universe. And that most of them move away from ours. And we can calculate very precisely when all of them were on "one place": around 13.8 billion years ago. I'm always so glad that I live in this point of history to be even able to learn about that.(1)
If you want more answers, support the science which looks for the answers -- there have to be serious investments in the real experiments for finding out more because easily reachable information is already found. And the humanity sadly invests more in military equipment than in science. And a lot of what we've learned we have learned from the projects that are less "sexy": you don't have to have many humans walking on the moon to learn more about the moon, for example: using robotic probes is all that we need today.
But you should also not realistically expect the answers that are behind the limits of what's reachable for us. In the example of the first human in the desert, he can reach some land behind the desert, but being just a first human, walking there, he surely won't walk on the moon. He can however look at the moon and the changes in the shades from the sun on different places on the Earth and develop the idea that the Earth is also round. And that's how scientific discovery works: carefully doing enough experiments to be able to conclude something that was before unknown.
1) to reestablish the perspective for the day-to-day events, search for "Everything Is Amazing And Nobody Is Happy"
Anthropic "answers" are used precisely to tell people they should "give up any attempt to learn more". They proclaim trivially that since X happened, the probability of X having happened is 1, therefore there is nothing in need of further explanation, so would you please stop now?
I should mention at this point that my degree is in philosophy, and my thesis was on the argument from design post-Darwin. Much of it consisted of exploring how various people -- not at all religious people! -- have been struck by the seeming improbability of our universe full of complex structured arrangements of matter and energy (including us) and sought explanations beyond "God must have done it". I had to read Barrow and Tipler's door-stopper of a book on the anthropic principle as part of that, along with criticisms of anthropic arguments, and came away convinced that the anthropic principle is at best a tautology possessing no explanatory power.
Anthropic arguments are brought up when someone notices the seeming improbability of the universe we observe around us, and whenever someone suggests that it's worth digging into that to figure out if there's a deeper reason for it. Why, out of all the possible values they could have taken, did the fundamental constants of our universe take the specific values we observe, and which happen to give rise to complex accumulations of matter and energy?
We know of nothing -- yet -- to suggest those particular values were necessary. We know that even tiny changes to those values produce a very different universe in which significant accumulations of matter and energy don't occur.
So the particular universe we observe around us appears to be staggeringly improbable.
The anthropic argument says "well, we observe it, therefore it happened, therefore don't bother digging any more because that explains that". And it's incredibly unsatisfying.
(I should note Barrow and Tipler did go further and proposed a variety of anthropic principles, some of which border on religious, but I'm talking here mainly about the base anthropic argument, as trotted out way back at the start of this comment chain)
There are people who look for deeper theories in which the particular conditions we observe do fall out as necessary. There are people who work on multiverse theories where it's inevitable that at least some universes will have conditions amenable to complex accumulations of matter (and thus to conscious observers). There are even people who've gone way out there proposing theories of natural selection applied at the level of whole universes, where arrangements of the constants that help a universe "reproduce" more (usually through forming singularities) are also coincidentally ones which require complex accumulations of matter.
There's a whole wealth of things out there to think about and explore. But anthropic arguments tell us to just shrug our shoulders, say it's totally unsurprising that a seemingly-so-unlikely thing happened, and stop trying to explore.
(and, of course, they do it in a away that's suspect from both a logical and a probability standpoint)
> the particular universe we observe around us appears to be staggeringly improbable
Says who? Anybody who knows what he really talks about will tell you that there aren’t any real physical data to claim any “probability” at all to the physical constants we observe. Let me repeat: you can’t use “probability” for that in mathematical or physical sense. That’s all.
Also please quote who actually says that doing experiments to learn more should not be done. If somebody just says that some philosophers talk about something they don’t undestand, and that there isn’t enough data from the experiments we’ve already done to say more, that’s what I can imagine.
Ah, you've got me. Literally no-one in the history of humanity has ever suggested that the particular configuration of our universe, suitable for the formation of complex accumulations of matter, is surprising, improbable or unlikely. I completely fabricated that notion, which had never before been expressed by any person, from whole cloth.
Or, you know... https://en.wikipedia.org/wiki/Fine-tuned_Universe for an introduction and some bibliography. If and when you eventually come round to the idea that perhaps this is something people have thought, talked, and argued about, we can continue from there, but I'm having an extremely difficult time assuming good faith on your part right now.
Incidentally, some of your other questions about whether anyone has ever at any point in history said or done something with respect to anthropic arguments will be touched on even in that introductory Wikipedia article.
“Fine tuned” doesn’t mean “highly improbable” in mathematical or physical sense, because probability is defined only as a limit value in case the number of trials is infinite. Once you have only one outcome there is simply no "improbable" probability. Check your sources if they understand that much.
“Fine tuned” claimed by some people is currently only a metaphysical claim: “IF we believe that the physical constants can take any value, then only some exact constants define the Universe with the characteristics like ours.” Note if. There’s no real evidence for that belief, that all the values are/were actually possible. We don’t know if there’s something outside of the results of our current experiments that can support such claim, but given what we exactly know, there’s also no real reason to believe that the other constants were possible.
Note: we also have absolutely zero of exact evidence that multiverses exist. So what we actually have is only this Universe of ours, with only the constants that we observe.
To paraphrase one joke, if you only saw only one sheep, looking black, through you window in your whole life, moving from left to right, the only valid mathematical and physical claim you can make is "there exists at least one sheep there, whose one side is black." Nothing more unless you make more observations. It's basic scientific literacy.
Or to paraphrase another example: you've just went to your search machine and typed "seinfeld license plate picture." You've received the image of the plate on which is "new york assman." Then you go around saying "imagine, from all combinations that I can imagine that the license plates can have, I've got exactly this one. It's so improbable." No, it isn't. Sorry to burst your bubble. Now imagine a relaxed experiment: You live in New York, go outside your door, wait for the first car to pass by. You never saw any license plate in your life, or got any information how they are issued. Then you go out and see the license plate "BGY 3891." You say again "oh look, from all the license plates in New York, if I assume they can have 7 letters and digits, how 'improbable' it is for me to see exactly BGY 3891." Again, in that experiment, there's no mathematical probability involved. You simply observed one license plate. You can invent "fine tuning" story about it, of course: imagine, if the Earth wouldn't exist, there would be no humans, there would be no cars, there would be no license plates, there would be no me making the experiment. Yes. But given all that actually exists, and that you actually went and observed one plate, it's simply the first (the only) license plate you saw.
I've also generated a number of random bytes, hex encoded, which can be observed as one big number:
You obviously don't agree with the argument, but not that long ago you were pretty obviously questioning whether anyone had ever even made the argument. Did you make sure to call and check for buried cables before digging the new holes to move those goalposts into?
I actually asked: ""the particular universe we observe around us appears to be staggeringly improbable"<--Says who?" Don't omit your words I quoted to pretend that I ask something else.
I questioned the use of "improbable" by anybody who knows what "improbable" in mathematical sense means and uses it in that way about the "universe we observe" giving any exact evidence. I still haven't seen such an example. I don't care about these who don't want to even understand what they talk about, or those who make claims with no actual support. My question was with the goal to point to exactly that: those who claim actually don't have any physical evidence for their claims. The wikipedia article mentions "theists," "philosophers," "theologians" etc. Some scientists also write something on the subject but not with the actual scientific support, just in their role of some of the above.
The number of random people claiming something doesn't make the claim more true. The validity of physical claims can be evaluated only based on the physical evidence, and only processed according to the mathematical rules that are valid in our Universe. Anything else is not physics.
Once in a time, there was one constant that physicists talked about as "special" (in that non-physical sense) being at that moment "exactly" 137, according to the best measurements of the moment. It was measured much more precisely since. 137.0359. Then 137.03599. Then 137.035999. Wow, isn't it something special? Maybe it's exactly 137.036? Does it have some special "meaning"? But now we know using much more precise measurements that it's 137.035999139... So it turned out to be quite unspectacular.
What is spectacular is that we have so precise measurements today: for that constant we are uncertain only for "0.23 parts per billion." And by the way, whichever computer or mobile phone you use now, the CPU in it has the basis of 99.9999999% pure silicon. That's also spectacular, it's one grain (i.e. 1 gram) of dirt in 1000 tons of silicon (at least 12 railroad cars needed)! And because it is also a result of a lot of hard work, and available to everybody including "theists" and "philosophers."
In probability theory there's the concept of a filtration.
Rigorously speaking, probabilities are not attributed to singular events, but to sets (possibly, in the discrete case, single-element sets such as {heads}, {tails}).
To be even more precise: a sigma-algebra is a concept of set-of-all-sets that avoids continuous-case paradoxes through technical stuff you can ignore.
A filtration is a decreasing sequence of sigma algebras: for example, if your event space is the six numbers a die might show, you could have
F(0) = { {}, {die number is odd}, {die number is even}}
and
F(1) = { {}, {die number is 1},...{die number is 6}, {die number is 1 or 2},...} (a sigma algebra is closed under intersections)
What this means is that earlier probabilities are defined on coarser spaces of sets of events as compared to later probabilities. This is how you get dynamic stochastic processes that don't look into the future.
The solar system formed, apparently, 4.6 billion years ago. The universe itself is closer to 14 billion years old. Might a solar system like ours have formed, somewhere, 4.7 billion years ago instead? And once it's formed, how long does it take for life to evolve? If the earth had 10% more surface area, would evolution have occurred 10% faster? There were various huge extinction events during the life of the earth, at least one of which appears to be due to a collision with a giant asteroid; if that happened 70 million years ago instead of 65 million, would humans have evolved a few million years earlier?
The point is that, if you rolled the dice on a new universe that had a decent chance of producing something like our Earth, there are lots of sources of variance, such that the "95% confidence interval" during which something like us might have evolved is probably hundreds of millions of years wide. Given that, the chance that two independent civilizations would reach a particular level of technological development during the same million-year period is less than 1%.
Because planets suitable for life could have formed several billion years before the Earth. Our solar system is relatively recent. I don't know when rocky planets first started to form, but I've heard for at least the past 10 billion years. Also, we don't know what evolution always takes as long to result in a technological species. It could be tens of millions of years slower or faster, just depending.
In Fermi's time, the perception was that an advanced civilization would have large powerful radio stations that could be detected from many light years away. Up until the 1930s, radio stations did indeed get bigger and bigger. But after WWII it became clear that many smaller stations are preferable over a few big stations. The trend started with FM radio and today we listen to radio using WLAN systems that barely cover the area of an apartment.
An advanced civilization has antennas that don't waste energy by beaming signals into space.
But the SETI assumption is that an advanced civilization could beam powerful radio signals into space to let other civilizations know they're around. That's part of the premise for Sagan's Contact, with the aliens listening in and then sending a strong signal upon detecting a new fledgling civilization. From the movie, it's apparent there were various listening posts scattered about the galaxy.
> Also, the article seems to use "our galaxy" and "the universe" as interchangeable terms, which of course they are not.
I didn't find that. For example, there is the following paragraph; it clearly distinguishes between our galaxy and the universe:
> In the end, the team’s conclusions do not mean that humanity is alone in the Universe, or that the odds of finding evidence of extra-terrestrial civilizations (both past and present) is unlikely. Instead, it simply means that we can say with greater confidence – based on what we know – that humanity is most likely the only intelligent species in the Milky Way Galaxy at present.
given the distances involved and what we know about the speed of light, an alien civilization would have had to develop a very long time ago in order to send signals that we would be able to detect today.
The Milky Way is only 100,000 light years across, which is a blink of an eye in geological time.
It doesn't matter. The point is that the only local-galaxy civilizations we would fail to detect for being "too far" would be those that were born only yesterday, on the timescales involved. All things being equal this should be an irrelevant fraction of the total.
A different matter is "too short lived", which is taken into account in the Drake equation (the last factor, L).
Seems to me the point is mainly that the Fermi paradox is no paradox at all, since it's entirely plausible that we're alone in the universe.
This doesn't mean we are alone, given the uncertainties. It just means that if we don't see anyone, we don't have to come up with all sorts of crazy scenarios to resolve the paradox, as people often do.
(If you find it implausible that we're the only ones, note we could also be one of just a few, separated by billions of light years, in which case we still wouldn't see anyone.)
If advanced technological civilizations are on average, short-lived, then the separation could be orders of magnitude smaller and your point would still stand. All of human evolution is in the millions of years, modern civilization in the thousands, technology in the hundreds.
This seems to me to be the most likely scenario given what we see here on earth. Our civilization figured out how to remove limits on its growth a lot faster than it is figuring out that we need to self-impose limits on our growth in order to avoid a catastrophic failure of our ecosystem. I see no reason to expect anything different. Evolution does not plan ahead.
> faster than it is figuring out that we need to self-impose limits on our growth in order to avoid a catastrophic failure of our ecosystem
We "figured out" unconstrained growth could be catastrophic a couple of hundred years ago at the very least. And in fact we guessed prematurely as technology extended the date of the catastrophe multiple times.
So we then bounced into a phase of overconfidence whereby some of us decided that technology could overcome all problems.
(Arguably we are still in this phase - or at least some of us hope this turns out to be true as we've lost hope in the possibility of political solutions to problems caused by growth)
Yes, I think you have this exactly right. We've dodged the bullet for a few hundred years, which our brains are predisposed to think of as a long time. Then we (well, not all of us, but enough of us to matter) (mis-)apply inductive reasoning and conclude that because doomsday predictions have always been wrong in the past they will continue to be wrong in the future.
In fact, you can sum up the problem as a pithy slogan: Doomsday predictions are always wrong. Until they aren't.
Given enough civilisations, it's likely that a few will eventually manage to not wipe themselves out (at least for long enough to become fairly redundant).
Exactly, and how many want to be found, or find other species? Of those, how many are so far away that communication is a pointless enterprise? Not to mention that the more difficult and costly communication becomes with distance, the further out in time you are. All in all, using a ton of high tech and energy to say “hi” to us from 10,000 ly away and 10,000 years of time seems... weird. At 100,000ly it starts to be a really exotic proposition, and at a million ly you shouldn’t assume your message will even reach the same species.
Unless your fellow technological life is within a few hundred light years, communication of any kind starts to look like throwing messages in bottles. Of course the closer your target is, the less wistfully academic it is. If Proxima Centauri had a planet with a civilization like ours, you’d have to consider what they could do other than just talk. At that point, the issue switches from insane time scales rendering the whole process moot, to potential forms of dark forest detente.
I originally saw this proposition on reddit - if all humans died where they stood right now, what archaeological evidence of our civilization would exist in 65 million years? 200 million? 350 million?
The discussion at that time assumed vanishingly little would survive that long - potentially just a thin layer of radiation from atmospheric nuclear testing that would be hard to place in time with accuracy more than +/- a couple million years.
I say this more as a thought experiment than a serious suggestion, but if a "technologically advanced" civilization did previously exist on earth, 100+ million years ago, what evidence of it would we expect to find at this point?
This is actually the setting for a sci-fi story I want to write. There's no reason to believe that the dinosaurs didn't have some kind of civilization, because we are so far removed from their extinction that any artifacts would be long gone. Unless they went to space, and the issue there is there's a whole lot of territory.
The rapid spread of animal and plant species around the globe when there wasn’t a supercontinent might be visible in the fossil record. We’d be a crank answer to the “How did potatoes spread from the Andes to the rest of world?” problem.
The arc of technology that would matter or be detectable to a distant civilization is less than 200 years of our history so far. It is yet to be seen how far that arc extends to the future. Even if it does, we’ve already passed the peak of our broadcasting in the EM spectrum. It’s not hard to imagine us coming to terms with the speed of light, and ultimately losing interest in stars that could only be reached by our ancestors tens of thousands of years from now. If we stick within a few tens of light years, and don’t bother to try and contact life we’d have to communicate on a thousand+ year “tape delay” we’d be hard or impossible to detect. That of course assumes that some other form of life is interested in something so far away and in its relative past.
All of that assumes that the arc of technology doesn’t terminate in a cosmic eyeblink after the current era, which it might. Either way, there is a sort of childish assumption that we’re always going to be as exploration-minded as we are today. If ‘c’ really is the cosmic limit, then it makes more sense to stick closer together, because any significant travel will imply either speciation, or a one-way cultural trip.
Before flipping off the site rules and implying that I didn’t read the article, you could think about what I actually said and the comment I responded to. Nothing about a civilization dying off in millions rather than billions of years implies the Great Filter, Dark Forest, or any other exotic explanation. All it requires is for the speed of light to be what it is, and hyper-optimistic “Von Neumann probe the galaxy” hypotheses to be wrong. If we’re technological for longer than many species exist, it would still be an eyeblink out of reach for the vast majority of any other hypothetical life in the universe.
If life is constrained to a reasonable fraction of the speed of light when moving mass, and light when communicating then the universe could be teeming with life out of temporal spatial sync with each other. We might not be alone, but we’d be effectively isolated, and still see ourselves as being alone. It’s only under the assumptions of unlimited growth and expansion that the Fermi Paradox is a paradox.
Mind you, the idea that life which is likely evolve, adapt, and develop significant technology being ultimately self-destructive a la The Filter also seems very likely. The article offers squat in rebuttal to that.
I’m notable for a few talents and abilities. But self-replication and spaceworthiness aren’t among them. I appreciate the baffling non-sequitur though.
> we don't have to come up with all sorts of crazy scenarios to resolve the paradox, as people often do.
Run-away climate catastrophe, nuclear war, or alternatively, the physical impossibility of interstellar travel are not crazy scenarios that explain the paradox. They are real threats that we have put far too little into mitigating. [1][2]
[1] I think we are putting the right amount of effort into mitigating the latter.
[2] A few smart people are playing with mental what-if toys like 'If we could send 100,000 people to a planet which has already suffered a run-away climate catastrophe!' The rest of us are digging our heels in, and finding every excuse to do nothing.
I definitely did not mean to minimize either of your first two threats. However, using them as explanations for the Fermi "paradox" would imply that they're impossible for any civilization to fix. Personally I don't think that's true.
> Thus, the real information is still strongly dominated by beliefs about the width and centering of the distributions, which we still don't know, and as long as enough of the parameters have high uncertainty with some potentially having near-zero values, the result will always tend towards zero.
You state this like it is a problem. I think it's more like a feature?
My take on the main point of the paper: it is commonplace to derive probabilities for being alone in the universe by plugging in point estimates into Drake Equation, but one should take into account that, in fact, we have lots uncertainty in our estimates for the parameters, and without quantifying that uncertainty, the Drake's Equation indeed will produce biased estimates. And if we quantify the uncertainty, we end up with wide distributions (that have some of their mass near zero).
edit. Your point about N = 1 is interesting, though.
edit2. Concerning N = 1: They don't elaborate on it in writing, but if I'm reading the sec 4.0.2 correctly, they do account for our existence in their spatial Poisson update: P(no detection | N, d) = 1 - exp(-(4/3) pi d^3 lambda) = 1 - P("number of civilizations in sphere of radius d when rate is lambda" = 0) = P("number of civilizations in sphere ..." > 0).
edit3. wait, that does not make sense, does it. shouldn't P( no detection | N, d) be P("number of civilizations in sphere when rate is lambda" = 1) = (4/3) pi d^3 lambda exp(-(4/3) pi d^3 lambda), as per spatial Poisson process P(K = 1). now I'm just confused.
have to somehow correct for the fact that we know N is not zero, since we are here
The anthropic principle is the correction. Observing ourselves is a prerequisite of analysis - independently of probabilities which can be any low! So it is not an "independent" observation.
Otherwise we could easily say from armchair, that:
1. Probability of civilization in a certain place is extremely low
4. Expected number of civilizations in universe (without noticing anthropic principle): 1 (at least!)
5. So take Poisson dist. with lambda=1, as a conservative estimate
6. Probability of more than 1 civilization, if we know that there is at least 1 (again without noticing anthropic principle):
P(X>1 | X>0) = P(X>1)/P(X>0) /Bayes/ = 0.718
So with around 72% probability there is a another civilization. ;)
If you assume that a storm put together a Boeing 707 randomly, and there are many storms in the Universe, then yes there will be a good chance for there being Boeings all over.
Not to creatroll the discussion, but the probability of occurrence for each individual independent event that led up to what is life on Earth, in each branch of the tree of life, on each evolutionary path, is so low that number of planets in the Universe multiplied by all the milliseconds in the history of the Universe will not get very far toward reasonable chances. Take, for instance, a protein of 100 aminoacids. 23^100 possibilities, to put it naively, yields a quite low probability. And there a many types of proteins needed for life.
IF one knows of any study that touches seriously on this topic, I'd happily want to know and be corrected!
Well the question is just how physics plays into this.
Proteins are not random because we have laws of physics that govern their structure and even if they were then once you have a protein which is self replicating and it’s better at doing that job in its environment that’s all you need, “natural selection” does not need life it just needs a process that is iterative, repeatable and is dependent on its environment.
Overall life might be an inevitable outcome of our universe as our universe is predisposed to create carbon and life is the most efficient process of hydrating carbon we currently know off if increasing entropy is a natural lawful state of the universe it will be predisposed for creating life to serve that purpose.
That said it’s unclesr if complex and not to mention intelligent life is a predisposition or an anomaly as out of the nearly 4 billion years life existing on Earth and now it seems it started only a few 100Ms of years after the early bombardment period technological intelligence evolved so late into the game that it might have been a fluke.
Other types of intelligence exist we have pretty darn smart animals but despite many of them predating mankind and our ancestors none of them has ever evolved further.
I didn't say anything about proteins being random. But words of hundreds of letters, out of 23 possible letters, come in many, many, many combinations. All governed by laws of physics. Not to mention the many ways a protein could fold, not all ways being useful, but are still possible. The search space, as I said, is huge.
The length of the words doesn't matter much if you have an effective process to filter and make sense of them and those processes are completely natural.
For example if we take self replicating proteins these will self optimize to an exact letter order which would improve their ability to self replicate in the environment and this isn't conscious since the random assortment that would be better would simply over take the other types in a fairly short amount of time.
Beyond that you still have some random process of "mutation" and that is proteins that were not replicated correctly which can happen for multitude of reasons something bumped into them, temperature changed, acidity changed, random high energy cosmic ray hit them and boom one is not like the other.
If that new protein is now better at replicating in that environment whether it's because this new protein can replicate faster or is more resilient to it's environment and it degrades slower it will take over it's environment and the same holds true for the opposite as if that protein replicates slower it will eventually die out because it cannot compete faster proteins for the available raw materials and so it replicates slower until eventually it's gone.
This is the true beauty of natural selection it works on all processes even those we don't consider biological.
Life is not 'self replicating proteins', but rather self-replicating processes. And in those processes, hundreds of chemical components are involved in a dynamical manner, which requires both spacial (chemical formula) and time synchronization.
And as life complexity increases, so does the the complexity of the processes that make it possible. Humans are the result of a huge number of individual, albeit not all independent, events, each with a very low probability. Multiplying those probabilities, as it's required when determining the probability of occurrence of a sequence of independent events, yields numbers that are hard to event write down without special notations.
As I see it, the only way out is to postulate evolution based on other non-probabilistic evidence, and then to conclude that in spite of the low probability those events still happened.
You should not be asking "how likely it is that natural processes could have created life that invented mcdonalds and justin bieber" but "how likely it is that natural processes could have created _some kind_ of intelligent life".
Specific details of a given example (us, the one example we know about) are irrelevant.
In other words: take a coin, throw it 1000 times and record the results. Now calculate the probability of the exact outcome you just got. Pretty low, yet you just did it. Now ask another question -- how likely is that at least 200 of those results are heads? Pretty high.
How unlikely life is is a question of just how big the set of results qualifying as _life_ is. We are (probably) just one possible configuration.
The possible search space and possible combination can be huge even with chemistry however the number of combination that would be the most effective in a given task and environment is fairly limited and natural selection makes this search very easy because anything that isn't suitable for that environment and the task of surviving even if it's a simple protein would die out.
>>The possible search space and possible combination can be huge even with chemistry
Again search/combination space argument suggests there is a process that resets to 0 and moves to the last possible state incrementally, and that process repeats endlessly until a desirable state is reached, avoiding previously reached final states.
The problem with the above argument is words like 'reset', 'desirable state'.
In reality there is no 'search space'. Its really uni directional chemical reactions that go on, each one happening now was due to the reactions before. These reactions go on until you reach a set of chemicals that either just don't react with each other, or the chemical reactions just continue seemingly forever.
The 'desirable state' always exists and that is govern by the laws of physics and the environment.
In this case if we have self replicating proteins in a puddle of amino acids then those which would replicate the best in that environment and that process would be governed by multiple factors such as what is the most "cost effective" combination and what is the most "resilient" combination that will hold up to environmental factors.
For the amino acids themselves it's not that different which ones would appear and at what ratio would also be governed by the laws of physics based on the compounds available, the make up of the solvent existing energy sources and more.
If for example you have amino acids that can survive dehydration and those who cannot then the only ones which will be left after a few cycles of that puddle drying up and filling with water will be those which can survive.
Same will go for the proteins the protein which will be able to survive dehydration the best and or the proteins that would assemble and self replicate the fastest during the wet period would be the ones that are the "desirable state" as these would be the ones that survive this specific environment and process.
No process is ever this random. Its not like the infinite money theorem.
Infinite monkey take 1, Atlas Srugged written? yes : no
Infinite monkey take 2, Atlas Srugged written? yes : no
...
Infinite monkey take n, Atlas Srugged written? yes : no
In reality its more like what happens before effects what comes later. This is different in the case of Boeing because there is a degree of non-reactive stuff coming together there. In case of Amino acids and subsequent formation of life its just chemistry, and that gets progressively complicated.
And this process is happening between the most commonly naturally occurring elements. So these triggers and process would have happened in almost every star system.
I can't cite you a study, but every time I've read about life appearance, the authors believed that the primitive forms were much simpler. Once the current more effective forms evolved, the original ones disappeared or got assimilated.
The Earth is 4.5B years old. Evidence of life appears 260M-1B years later, with the earliest date given by the current earliest equivocal findings and the latest by direct fossil evidence.
A naive 1-in-23e100 calculation is probably the wrong way to go about it. After all, the probability of any poker hand is 1-in-2.6M.
How is the premise of monte carlo analysis flawed because "of course multiplying together..."? Monte carlo is just a way to estimate the total distribution of the resulting drake probability given uncertainties about the other values. The resulting distribution they depict is just the distribution of outcomes according to our uncertainties. It's precisely bayesian using the empirical estimates according to the literature.
The fact that there is at least one intelligent civilization in "the" universe is not necessarily that informative. This depends on the number of universes. If there is only one universe, then it is informative. If there are many universes, the probability of life in one can still be very small but it can still arise, in some universes. The question then is if there is life on our universe.
Talking about the number of universes feels like talking about the number of wheels on a unicycle.
The count is one. Either the "multiple universes" are causally disconnected and therefore irrelevant, or are actually connected and therefore part of the same universe.
It's a argument about how close to 0 the probability of life might be, given that we know of one instance of life.
It is tempting, because we exist, to assume the probability of life arising can't be that low.
But if life is so rare that the odds of one arising in any given (causally disconnected) universe means life almost never arises, and more than one civilisation is unimaginable more rare than that again, if you imagine an infinite number of universes there will still be an infinite number of civilisations.
But being alone might be the most likely scenario if life is present, and nobody would be present in the empty universes to take them into account.
In which case our existence might easily trick us into over-estimating the odds of life arising.
Framed in a less depressing way: It might be we're not seeing the aliens because something is preventing us from observing them rather than because they don't exist.
Yes, these are not causally connected, so we can call the different universes.
I interpreted the original post to be saying that if life is very improbable then it is unlikely we would be here. And If there were only one universe then yes, it would be unlikely we are here. But if there are many universes, this could support a vanishingly small probability of life with most universes being empty, but a small fraction having life.
So these different universes are meaningful only to the extent you are asking philosophical questions like this, but not to most practical considerations.
I agree. Just because there might be a large number of stars without a habitable planet does not decrease the number of alien civilizations; it only reduces the likelihood that a given solar system would have such a civilization. Or to put it in other words, just because the haystack is large doesn't mean there is no needle present.
Possibly, but not the way to bet. Live appeared on Earth very soon after it started to solidify. That would tend to imply that starting life was pretty easy or, life arrived by some kind of panspermia mechanism. There is a next to infinitesimal chance that life started very quickly by some kind of once in billions of years magical circumstance.
To go from life to beings like humanity it's pretty clear that intelligence tends to increase as life goes on. For example some birds so exceptional abilities and they are dinosaurs according to most current understandings. Octopus can be remarkably clever and their lineage split from the "animals with backbones" line long before there were animals with backbones. Nerves and nervous systems seem to have been reinvented by evolution several different times.
Putting this altogether, it seems there's an excellent chance that primitive uni-cellular life evolves pretty readily and multi-cellular animals tend towards increased intelligence soon there after.
There's an assumption that "advanced civilizations" will be communicating with TV and Radio because that's the best we've got. On average most advanced civilizations would have had their technology for much more than 150 years and very likely communicate with methods we are not aware of yet.
So when it's asked "Where are they?" the answer is they are out there, talking up a storm, they're just not still using electromagnetic "smoke signals" to do it.
It took single celled organism 1 billion years to happen on Earth, it then took another 2 billion years for the first multi-celled animal. The jumps you're describing aren't easy at all.
And there's no such thing as evolution towards intelligence. Evolution does not favor certain traits and at this point the most successful organisms on Earth are bacteria and the most successful multi-cellular animals are the insects.
And actually intelligence requires big brains, which requires energy and thus fuel, so in constrained environments it can be a trait that's detrimental for survival. Even if multi-cellular life developed on other planets, to go from that to fairly intelligent dinosaurs or mammals is still very unlikely.
Obviously, the problem with extrapolating what happened on Earth is that we've got a sample of 1.
It's possible that evolution inherently drives life towards intelligence (as a likely outcome, not in the teleological sense). Three counter-points:
* Sure, an extreme minority of organisms are intelligent right now, but considering that intelligence is such a recent invention, we can't assume that this is the terminal state of affairs. Indeed, intelligence is making progress in taking over the biosphere that shows no signs of slowing down. Perhaps thousands or millions of years from now, we will have converted all biomass to sustain our own civilization and perhaps (artificial?) life entirely devoted to intelligence will be dominant.
* "Evolution towards intelligence" can be a "thing" even if this only occurs to a small minority of species. This is not to say that evolution is a linear progression from primitive to intelligent life; rather, if evolution makes life expand like a tree, and the degree of variation and complexity afforded by this tree is sufficiently large, it could be likely that some of the branches eventually reach intelligence (among myriads of other possible adaptations). This would be sufficient for intelligence to arise with high probability.
* Life is inherently concerned with representing and processing useful information about the environment. The chemical feedback loops that probably made up proto-life would have implicitly encoded information about their chemical environment (and the fundamental chemistry of life still echoes this information). Eventually the incorporation of more complex feedback loops combined with evolution led to the development of complex biomolecules, proper replicators, genetic code, multicellular organisms, nervous systems, brains, the mammalian neocortex, and whatever adaptations enabled human-level intelligence. What we see is a gradual progression through great many improvements towards more sophisticated information processing. It's plausible that a similar progression in information processing complexity occurs with high probability whenever life evolves, although the concrete steps could be radically different. Yes, it took billions of years on earth, but the universe has billions of years to spare and it's massively parallel. Again, there doesn't have to be a linear progression towards intelligence for all life; the conditions just have to be favorable for intelligence to occur as a particular adaptive niche in sufficiently complex ecosystems.
Of course, we can't really say much since we don't know what the actual probabilities given that we only have a sample size of n=1 for life and realistic experiments or simulations are far beyond our current capacity.
The question is: What could they possibly be using instead of some sort of electromagnetic signals?
There are only a limited number of fundamental forces in the known universe (unless our current understanding of physics is completely wrong), not all of which lend themselves to being used for long-range communication.
Gravity possibly could be used as a means of communication but I suppose it’d be rather slow and have a low bandwidth.
Some sort of quantum entanglement-based form of communication might be possible though that contradicts our current understanding of how information can be transmitted in the universe.
So, barring some fundamental shortcut such as Star Trek subspace, there aren’t too many options for interstellar communication.
Why not expect new physics we simply don't have the tools to see yet? Maybe something that only shows itself at very high energy or interacts only with neutrinos or antimatter or some exotic matter that isn't produced with fusion or supernovas.
What about beings that operate at very different time scales? What if there is some encoding scheme that blows our information theory out of the water so communication just looks like white noise?
What if evolution converges to a single shared consciousness or fierce independence and communication just stops being interesting after a certain stage? Or if non-interference is the very well enforced law of our corner of the galaxy?
I think it's hubris to think that we know the answers to all of the questions for physics or at least we know what we don't know. My assumption is that there are questions/problems/phenomena that we don't even have the context yet to imagine.
A signal with compressed data does not look like white noise. It looks like a sine or cosine wave that is around a well defined frequency, and the intensity/phase/frequency changes from time to time. The hops may look random, but the carrier signal is easy to spot.
If they use frequency hoping, the changes in the frequency are even more regular. It's more difficult to measure the exact signal, but all the hops are inside an interval.
You are describing FM, which does dump quite a bit of power into the carrier wave.... which is exactly why it's not used if you care about range/efficiency.
Check out things like "Olivia_MFSK" on wikipedia for something more current. It can transmit even with 14dB below the noise floor.
Seems like a any sane communication method for any galaxy spanning civilization would be:
* encoded in such an advanced way that we wouldn't see it, even if using signals within our physics.
* be highly directional
Maybe something like 1% of the asteroid belt would be converted into communication sats that could form a swarm that would over a few ns significant outshine the sun by focusing a signal on whoever you want to communicate with.
Of course typically the cloud of sats would be handling within the solar system communications or nearby stars that didn't require the entire cloud.
I do wonder if you could capture say 0.01% of the solar radiation for 24 hours and with 50% conversion frequency generate a single for 1 ns on the ideal frequency... could you outshine the galaxy on that frequency?
Of course the ideal frequency would be minimally blocked by dust and minimally generated by stars. The "Water hole" as mentioned by Seti (among others) would be a good start.
You’re probably more of an expert but what you’re saying is totally unintuitive to me.
If a signal carries the “this looks like a signal” information in addition to the source information itself... isn’t that a waste of bandwidth? Wouldn’t an ideally encoded stream be indistinguishable from noise?
I can understand for practical reasons leaving some bandwidth on the table to make it easier to detect signals, but when talking about advanced alien life forms we have to assume they are playing closer to the limits of physics.
Compressed data doesn’t looks like noise correctly encrypted data can.
That said however metadata will often be unencrypted and even encrypted data will have discernible patterns for example the transmission of say a 2.4ghz radio network with packets will look like white noise even if all packets are encrypted as you need to somehow modulate the signal for transmission control.
Under _our_ technology, yes. Why would that be so for a civilization even a hundred years more advanced than ours, never mind millions? Why would they be doing planet-scale broadcasts _ever_, instead of focussing the communication as tightly as they could in the right direction?
There's so many assumptions that they'll basically use a version of modern technology but bigger in the "why can't we hear them?" question. Tight-beam, fully encrypted data with sensitive receivers could easily be utterly undetectable by our understanding.
Well the beauty of the universe is that cannot be used as a good reason.
If civilizations are common then there are just as many less advanced civilization than us than they are more.
Heck if technological progression leads to the downfall of a species than by default everyone we should be seeing is around our own level of development if not lower.
If intelligent life is an anomaly then it can be very well be that in the life span of an existing civilization they would be the only intelligent one in the galaxy or even beyond that and they would simply die out even 1M years is a blink of an eye for universal time scale.
And lastly regardless of what the civilization is today if they are on the other side of the galaxy we see a 100,000 year snapshot if they are in say another galaxy even a close one like the Andromeda galaxy we see them as they were 2.5M years ago.
Tight-beam, tachyon transmission, quantum entanglement with retro-causality suspension or w/e they might be using if they are so advanced beyond us that we look like anthills to them is fine but at some point they were just as advanced as we are and less so that's the noise that we should be hearing and we aren't.
That's assuming the exact opposite of this paper; that there's _infinite_ other civilisations, which is as obviously untrue as the idea that there are none.
If the period of time before all high-power communication is encrypted is short, then the likelihood that the period of our ability to pick up unencrypted high-power communications will intersect it is potentially very low.
The universe is _huge_. Most of our communications will not be audible above background radiation after a little way out - this could be no different for any other civilisation's communication technology. You don't use a nuclear reactor to power a cell tower, and if you don't put that much power into it then it's not gonna be detectable ten light years out.
I always wonder, if you draw a sphere around earth, some amount of light years away. How much surface of that outside sphere has our signal gotten in contact with.
(for the sake of making it easier, say that all signals capable or reaching this distance, have done so)
Encoded data can sounds just like noise. Take Olivia for instance. It can transmit digital data at 14dB BELOW the noise floor. It's just a digital mode that hams use to try to transmit between countries for around 1 watt.
Then the answer is: we can only rule out intelligent life as we CAN immagine it. Sure, there could exist intelligent life of sorts that we cannot comprehend, that use physics outside of the standard model; God perhaps?
The fundamental forces are a starting point, but it's not right to assign special significance to EM. The great majority of mass in the universe does not feel electromagnetism. It's dark matter.
Humanity's expertise in electromagnetism is driven by our biologically EM-sensitive eyes, our physical size, the energy levels that permit classical chemistry, and the timeframe in which we live our lives.
Perhaps aliens exist with much larger physical forms, communicating with neutrinos over vast distances and larger time scales. Humanity would be completely blind to these signals.
EM is much easier, any moron wit a HAM radio can broadcast to the whole Earth, and a similar upgraded version that is still small and cheap can be used for interplanetary communication. So I'd guess EM for usual communications, but the idea of something weird like neutrinos for special purposes is nice.
I don't know the size of the current neutrino transmitter/receivers, but IIRC the size is about a big building. Even the initial EM device where smaller.
About dark matter, the current understanding is that it is diffuse and there are no "dark star" or "dark planets", but I always prefer to wait 10-50 (100?) years to call it definitive.
Two other factors that play a big role in not detecting life in the universe is:
1) it's a really big place. It's likely we can't see that far yet.
2) Will we be here a million years from now? Two million? Other civilizations could have come and gone dozens of times in our galaxy alone before we were here. And dozens may come and go after we're gone.
Milky Way is estimated to be what... 13.5 billion years old. Humanity is 200,000 years old. The time span in which we are actually able to detect anything outside our own planet is what.. 100 years old? I suspect it's a lot more than a dozen times. If life is more likely to develop, then humans have only been able to look out into our own galaxy for an entire 0.00000074% existence of our galaxy. Now, let's say a civilization does not last more than 1 million years for whatever reason, that means we exist for a whopping 0.00074% age of the universe to date. I don't know if I did my math all wrong, but if correct there is an enormous chance that 2 civilzation's paths may never cross. We could be living right next to the remains of a civilization before us and never know it.
#1 is a lot crazier, and more like: "we can only look at specific places, our detectors can only detect things which degrade as distance increases, and we might not be listening at the right frequency. So even if there were a civilization in this tiny, tiny region of the universe we are looking at, if they are too far away/obscured by debris/gas clouds, using wavelengths we aren't actively monitoring, we don't see them.
Why can't they see us? All the arguments for life say "the universe is so big there has to be more life, and highly advanced civilizations somewhere." Assuming that, sure, we can't see them. But why hasn't some advanced civilization seen us?
- some have, and have responded, but we aren't looking for them in the right place and in the right way
- humans have only been able to actively send information into space since only about the 1950s (radio waves), and physics dictates that locations 80 light years away are just now starting to see that, provided they are looking in the right location and in the right way (and that they are technologically advanced enough to do so.
- It could be that a sufficiently advanced civilization no longer has the capability to receive and transmit long-wave radio communications. Just like many computers now don't have any way to read a floppy disk, or if someone were trying to communicate via smoke signals, you'd be hard pressed to find anyone who could understand it.
- Space is so large that if a civilization several hundred light years away (an insignificant fraction of the size of our one galaxy, actually) sent us a message that we received and responded to, it's possible they could have collapsed by the time they receive our response. The converse is also true, they could have 'seen' us and sent a message, and we could collapse before we had a chance to receive it and respond.
I always thought that if a civilization was so advanced, why would it spend time interacting with a world like ours?
We may feel special, but that's only because we haven't found anything else. If they found hundreds (or more) of worlds like ours, then maybe we're just another dot on their galaxy map.
Plus if the media is anything to go by, all we do is shoot each other. Might not be too exciting for a civilization who can travel amongst the stars.
Why do humans spend time studying ants? Atoms? To further our understanding of the world. Seems illogical to think there isn't a civilization out there that wouldn't expand and explore other places.
Just seems unlikely that we assume, because the universe is so big, there are advanced civilizations out there - but not ones that can find us, not ones that we can find, for various reasons discussed in these threads.
If our understanding of the universe is true to a certain point, civilizations are far more likely to be essentially alone and "special" than many might think.
That's the subject of about a century of science fiction speculation.
The deal is, if we were one among hundreds or thousands, what are the odds nobody at all in any of those other civilizations would want to interact with us? Because we've received zero interactions. Dead silence.
Why is that? Well, it takes a lot of energy to broadcast, so they'd have to beam us a message. We're not really looking for beams. Don't even know what they'd look like? Laser pulses? Soliton trains? Neutrino bursts?
Someone would have to be very close (30 light years or so) to have noticed us and have time to signal back. Even among those, likely they would just watch us and then decide if they want to chance contacting a race that has thrown nukes at each other.
To put that in perspective there are about 400 stars within 30 light years of earth and among those stars we have found 26 exoplanets. This radius represents only about 0.00014% of just our own galaxy and an infinitesimaly small portion of the entire universe.
And if that is accurate, then essentially we ARE alone in the universe. There simply isn't anything close enough - we'll be enveloped by the sun or some other catastrophe before that happens.
>Putting this altogether, it seems there's an excellent chance that primitive uni-cellular life evolves pretty readily and multi-cellular animals tend towards increased intelligence soon there after.
How much intelligence though? Dinosaurs were around for eons and, as far as we know, never used simple machines or mastered the use of fire. So far, the human branch is the only handful of species that ever has. Even fairly closely related primates haven't figured it out despite the fact that they watch us do it all the time.
The intended meaning was that there's a small chance of something unlikely happening quickly. If it happened "quickly" (in geologic time) once conditions were favorable, that is evidence against it being enormously unlikely once those conditions obtained.
Multicellular life seems to be one of the biggest stumbling blocks. Life appeared very soon after it was possible, but a pretty huge chunk of the time since was limited to single cells.
Why? Once the earth is covered by single cell organisms there's going to be an incentive for even the simplest of cooperation. Quorum sensing can help reproduction, survival, collecting food, surviving drying out, etc. Once cells loosely cooperation that starts providing more specialization. Cells that have thicker cell walls, or better offense/defense, or more expensive pathways for breaking down more complex chemicals, etc. Seems like the natural tendency would be towards more complexity heading towards where you have cells that specialize being outside (and help control PH and water) vs inside (that specialize in other things taking advantage of the ideal conditions).
Keep in mind how many single cell organisms you can have in a single drop, how many drops are in the ocean, and how long a billion years is. Even a small competitive advantage turns into exponential growth for the community of cells. The noise of random mutations is going to eventually lead to a cooperation between cells that's going to cause a feedback loop.
Cell -> multi cellular seems easy and obvious compared to no cells -> cells.
Even asteroids, super nova's, cosmic radiation, bigger but dumber predators (Dinosaurs), etc.. lots of external factors could destroy a planet too, a planet is a fragile place, and keeping life on it is a fragile balance, it's a miracle we've lasted so long, imho.
None of these things is likely to actually be the end of life altogether, just life as it's known at the time. More asteroids won't make the Earth completely uninhabitable (just as it didn't the first time); nor will climate change (just as it hasn't before: https://en.wikipedia.org/wiki/Great_Oxygenation_Event). They'll simply be the end of our current human civilization, along with plenty of other current species. Other forms of life may benefit tremendously; things will evolve to take advantage of the new circumstances.
Quite, but it may turn out to be a great filter that no species, alien or terrestrial, has yet got past. Thus there may no end of life out there but none that has advanced far enough to be noticed.
Perhaps we'll be the first, though I have to say I'm far more pessimistic than say a decade or two ago.
I'm hopeful we can develop immortality tech, if we can do that, then also build generational ships...even if it takes 10,000 years we could colonize every habitable solar system, even without light speed travel and send findings back to earth on discoveries found. Obviously having a working Albucierre (Sp?) drive or similar would make things better, or some sort of wormhole tech. But we can get to alpha centauri in 100 years, sailors used to risk their lives to go on voyages, why is it too much to ask the same of scientists and their families?
Esp if we can figure out cryostasis, we should definitely make a trip to Alpha Centauri with actual people someday, if there's no-place habitable they could just keep moving on till they find a place. In the off chance something bad happens on earth at least they'd be able to reboot/restart mankind on a new planet.
Imagine if 10 generational ships set off in 10 different directions headed to 10 different solar systems looking for a habitable planet to setup camps. Even if it takes 500 years for each to reach their destination, if longevity increases into the 1000's of years, we'd still be able to witness it and it'd be pretty cool.
Those things probably won't escape the atmosphere, however. It would be much harder to bootstrap an energy-hungry civilisation any time within the next few billion years, as we've used most of the easily-accessible sources of fossil fuels. If human civilisation falls, whatever comes afterwards probably wouldn't be able to build a rocket, unless they wait for fossil fuels to regenerate.
A billion years is a very long time if you wait in a railway station. But in milliseconds, for instance, as powers of 10 is a blink of an eye compared to the magnitude of the search space of configurations of a single protein.
The context here is the lifetime of the universe. If that is 13 billion years, 1 billion is a considerable chunk.
If I interpret you right, you state that we might be extremely lucky hitting on a valid configuration so fast. In other words, the distribution might be at average at e.g. 1000 billion years. What shape would the distribution have?
Why is this even still a question... every single time you boil a pot of water convection rolls form. According to equilibrium thermodynamics it is vanishingly improbable that Avogrado's numbers worth of molecules will conspire together to act in a coordinated fashion to form those rolls... yet it happens every time.
The convection rolls win out over diffusion because they dissipate heat faster and there is only so much heat. Life is no different. A structured low entropy process that dissipates energy faster than diffusion - there's no paradox or anything about emergent complexity.
The fundamental concepts here are pretty simple and any discussion concerning P(life) need to start by understanding what life is in the first place and how it can arise.
Point being: life isn't a rare thing or a mystery, anywhere in the universe you find energy gradients you'll find life.
There just isn't a conversation to have here. If we're alone it's because we're first. But we are almost certainly not alone. The universe is just really fucking big.
Just because "a structured low entropy process that dissipates energy faster than diffusion" will win out every-time doesn't answer the question about how a self-replicating structured low entropy process will form in the first place. Once life starts then it will explode, but the problem is how to get life actually started.
I want to understand this argument better. Are you claiming that any "structured low entropy process that dissipates energy" is alive? How are you defining life?
I may be putting words into the parent's mouth here, but I think the argument is just analogical. That there are emergent effects to physical processes that are difficult/nearly impossible to predict from basic physical law. And therefore we should withhold judgement about the difficulty of life forming, given that we so far have a limited understanding of the relevant physics.
Nah I mean it pretty literally. You are very right that the necessary physics isn't really there, this is a really difficult thing to state rigorously.
However I would argue that the underlying reasoning emerges out of fairly fundamental physics.. and that while difficult to state rigorously the core idea is kindof... obvious... to anyone with a strong physics background...
Pretty much. I define life as any self-sustaining structured low entropy process that dissipates energy faster than diffusion. The low entropy part basically means that such a configuration is vanishingly improbable under _equilibrium_ statistical mechanics.
Under this definition convection rolls, viruses and eukaryotes are all alive.
We don't know if there is no life there, we've never been.
Almost every time we've assumed that an environment on Earth was "too harsh" to sustain life we have been proven wrong. We have no model for something that might live on the surface of Venus or below, but there could be something in the clouds.
How would someone living in the dark ages look for intelligent life if they can't imagine the existence of wireless communications beyond a mythical transference of thoughts from human to human. Most of this is theoretical, but all it proves is our lack of understanding is still ever present, and we're all still just a bunch of ignorant fools trying to make sense of it all. I think we should just focus our energy into bigger things, and allow the technological breakthroughs as they will come with time. We just have to put our focus on bigger better & more interesting pursuits and stop blowing people up. Stop destroying civilizations, and start to accept civilizations as they exist. The more we destroy ourselves the harder it becomes for us to overall advance enough people to create the ever necessary brain trust of the planet to build these new sophisticated technologies to maybe one day find a way to decode what was once maybe thought non-existent. Maybe neutrinos are already used to communicate from light-years away, we just don't have enough advanced technology to decode it. We're not there yet, maybe we need fusion technology before other technology makes more sense to us, maybe fusion technology is what brings us to insane speeds in space that we cannot replicate currently since we haven't mastered it yet.
Once again, we argue about the unknown, and only create factions of people hating on others who don't realize they're just ignorant.
While it may be based on what is technically fiction, my views on the Drake Equation were modified to include a “caution in outbound communication factor” of fairly low probability after reading Cixin Liu’s “Dark Forest”.
SPOILER ALERT for prior novel: it convincingly argues that, due to the impossibility of trust across otherwise unrelated civilizations, the rate of technological progress in any given civilization compared to the speed of travel between star systems, and the long-term understanding that resources are limited, and outgoing communication to another intelligent civilization inevitably results in attack and likely elimination of said civilization.
Indeed this feels like the most satisfactory explanation for the Fermi paradox. I’m just starting on the third book, and am starting to think the eye-popping $1 billion that Amazon is rumored to be paying / have paid to allow them to make this into a TV series might not be entirely insane.
Shouldn't the title be '..humanity may be the only detectable advanced civilization..'?
After all, the last two terms of the Drake Equation are all about detectability, and what are the chances that a civilization outside the Milky Way would emit a signal that we could possibly detect? Just slapping a couple of internet calculators together, you'd get about 500dB of loss from Andromeda (our closest galactic neighbor) to us, requiring a 1x10^41 watt omnidirectional emission for us to receive a nanowatt signal. This is within a few orders of magnitude of a supernova. Even if you were to focus the beam, the transmitting civilization would likely scorch planets within a couple hundred light years just to say hi.
The paper seemed to inconsistently pepper 'detectable' into the language and I lost interest before trying to figure out whether or not it included those terms (fc and L) in the final conclusion.
Exactly. The 'L' term at the end (length of time advanced civilization is transmitting detectable signals) may be a very small window for most civilizations.
Energy is precious. If you have the technology to put the signal where you want it to be, why would you spew it all over the universe?
>In the end, the team’s conclusions do not mean that humanity is alone in the Universe, or that the odds of finding evidence of extra-terrestrial civilizations (both past and present) is unlikely. Instead, it simply means that we can say with greater confidence – based on what we know – that humanity is most likely the only intelligent species in the Milky Way Galaxy at present.
I'm going to repost the same comment I've made on this several times already:
This is quite interesting. It certainly sounds like this does dissolve the Fermi paradox, as they say. However, I think the key idea in this paper is actually not what the authors say it is. They say the key idea is taking account of all our uncertainty rather than using point estimates. I think the key idea is actually realizing that the Drake equation and the Fermi observation don't conflict because they're answering different questions.
That is to say: Where does this use of point estimates come from? Well, the Drake equation gives (under the assumption that certain things are uncorrelated) the expected number of civilizations we should expect to detect. Here's the thing -- if we grant the uncorrelatedness assumption (as the authors do), the use of point estimates is entirely valid for that purpose; summarizing one's uncertainty into point estimates will not alter the result.
The thing is that the authors here have realized, it seems to me, that the expected value is fundamentally the wrong calculation for purposes of considering the Fermi observation. Sure, maybe the expected value is high -- but why would that conflict with our seeing nothing? The right question to ask, in terms of the Fermi observation, is not, what is the expected number of civilizations we would see, but rather, what is the probability we would see any number more than zero?
They then note that -- taking into account all our uncertainty, as they say -- while the expected number may be high, this probability is actually quite low, and therefore does not conflict with the Fermi observation. But to my mind the key idea here isn't taking into account all our uncertainty, but asking about P(N>0) rather than E(N) in the first place, realizing that it's really P(N>0) and not E(N) that's the relevant question. It's only that switch from E(N) to P(N>0) that necessitates the taking into account of all our uncertainty, after all!
[Note afterward: Over on Reddit, hxka points out that that should be P(N>1), not P(N>0). Or really it should be P(N>1|N>0)...]
I think this hits the real point. The expected value of the number of advanced civilizations may be high, but that could mean a very small probability of billions of advanced civilizations complemented with a very large probability of zero, one, or just a few.
Let's cheat. Let's say that there's another Earth out there with human beings on it. Let's say they've been producing electrical noise for 200 years like us. Where would planet Earth #2 have to be in order for us to notice it?
I like this approach, but I don't think I like the answer.
If Earth 2 has been producing noise for 200 years, that gives us an upper bound of 200 light years. However – for general terrestrial traffic – our detection range is only about 1 light year [1].
There are also no stars within that distance from Earth 1. So, at current levels of technology, Earth 2 would have to be a sad star-less frozen ball that just happens to drift into our neighbourhood within the appropriate 1 light year window on its eons-long journey.
Given that, I think we got the best deal here on Earth 1.
Exactly. Now remove the restriction that Earth #2 is active with life at the same time we are. Instead intelligent life existed 10 billion years ago. Now we've got a tiny band of space 10 billion light years away where that planet has to be located where we could have possibly detected it. It doesn't seem that unlikely that we haven't found intelligent life.
If I had to bet on the right answer it would be this. The time scales of the universe are mind-fucking and we are just naive if we think we can maneuver around them.
The point of the paradox is that if intelligent civilization is only active for 200 years out of a 10 billion year timespan, there is something seriously wrong with the survival rates of such civilizations.
Or, they became something different, not worse. For example, Victorians would have looked silly to scan the skies for aliens by looking for soot clouds in their atmospheres, because all 'civilizations' would have steam trains.
We're looking for radio waves. Radio waves! Like a million-year-old civilization wouldn't be communicating with nano-soliton-pulse masers across interstellar distances. And to receive one of those, earth's orbit would have to accidentally cross its path, and then only for an instant. The chances are trillions to one, and we wouldn't recognize it if it happened.
> Claiming long-range settlement or communication are impossible requires assuming a surprisingly low technology ceiling. Whatever the answer is, it more or less has to be strange.”
Considering the terrifying energies required for launching anything useful interstellarly, this quote is a bit off.
I'm perfectly capable of believing that there is a tech ceiling preventing useful interstellar travel.
Roughly a century after discovering powered flight, we are pretty much capable of building spaceships that can reach Alpha Centauri in roughly 100 years.
Single digit percentages of the speed of light are not that hard to achieve, but would probably be sufficient to launch some von Neumann probes (or generation ships) and start colonies.
You are the children of a dead planet, EarthDeirdre, and this death we do not comprehend. We shall take you in, but may we ask this question--will we too catch the planetdeath disease?
Yes, but the question is can we launch and protect a payload for 100 years that can self-reproduce where it arrives? Otherwise all our ships can do is explore a tight X-lightyear sphere around Earth.
Can we propel a ship that can sustain 100 years of human (or AI) life and protect it from cosmic rays and provide it with enough equipment to build a self-sustaining colony on arrival?
It's not a limitation in physical laws, but the sheer scale of the undertaking that I wonder about.
Like, is the mass required for a self-sustaining interstellar generation ship, with all the required shielding needed to handle high-velocity interstellar travel (blasting through the interstellar medium an appreciable fraction of c means absorbing a crapload of high-energy particles, and the occasional micrometeorite that can't be detected until it's too late that hits the ship with the energy).
For example, hitting a little 10 gram rock at 0.1C would be a kinetic energy transfer in the gigajoules - like a small (1 kiloton) nuke.
So we've established our ship needs to have armor that's thick as hell. In order to make that economical, it must be huge. Also, in order for it to do useful work there, it needs to contain factories worth of machinery.
The sheer scale of this ship would be staggering. And then that raises the question: is there enough energy available to propel such a thing?
Define "useful". Generation ships could be a thing, we're arguably close to that level of tech right now, and that's with a relative pittance spent on space-related tech for less than a century. Given enough R&D and raw resources it's probably an easier problem than curing cancer, and no one says that's impossible.
There's a strange recurring theme in learning things. Once you reach advanced mediocrity, you tend to believe that you know more or less everything, and all that separates you from complete mastery is some rote work and training. Yet then as you reach early mastery exactly how little you know starts to become clear.
And this seems to happen everywhere in everything. For instance as early late as the early 20th century there were many scientists who felt that physics was basically solved. There were a few inconsistencies to work out and some more measurements to be performed more solidly, but that all of the most fundamental laws of the universe had been discovered and that we knew everything key. Of course quantum mechanics and relativity would be discovered within a couple of decades showing us that we, to this day, know extremely little.
I think the reason for this is that as one reaches advanced mediocrity, it's easy to look back and see how little people who knew less than you truly knew. And so, without knowing what you do not know, it is quite difficult to imagine how you could not genuinely be near the top of the knowledge scale. Yet something that's shown true time and again in our universe is how little we truly know.
---
Beyond this, there's also a technical response to your comment there. Even given our current level of knowledge interstellar travel is not only possible but practical, though communication is not. A common misconception conflates the relativistic effects of fast moving effects between observer and observed. And in particular as objects approach the speed of light relative to an observer, their mass approaches infinity thus sending the amount of energy required to further accelerate them also approaching infinity.
However, this is not true from the perspective of what's being observed. Indeed they'd perceive quite a different outcome. The energy for a vessel in a vacuum to accelerate 1000 km/s does not suddenly change after they've performed such acceleration 300 times. All sorts of strange things would start happening from their perspective including time for everything else seeming to speed up, and physical distances starting to literally contract (such that they themselves would also never perceive themselves as traveling faster than the speed of light), but they could happily keep accelerating to no physical limitation. And the net effect of this is that you can travel millions, billions, and really a practically unlimited number of light years within a single human lifetime.
Here [1] is a handy calculator for how long it would take to travel certain distances given a certain fixed acceleration. For instance 45 years at 1g acceleration would be sufficient for 15 billion light years of travel. The catch of all of this, and the reason communication is impossible, is that for everything 'observing' our traveling ship indeed 15 billion years would pass. So if our ship turned around and made its way back to Earth (I'm ignoring universal expansion for this discussion), though only 90 years would have passed for the ship - 30 billion years of time would have passed for the ship.
This effect even shows up in particle accelerators, in a really hand way. The collisions create unstable particles that though they would decade after let's say 1 millisecond, often can be observed for many times that. It's the same effect in play. Relative to us, time for the particle slows down due to its acceleration. Perhaps this all makes an even better point. Our universe is extremely weird and defies any notion of intuition. And the more we learn, the weirder it becomes. To think that we have even scratched the surface of its secrets shows that though the universe may be vast, it is incomparably small compared to the bounds of our arrogance!
I'm aware of the mechanics of relativistic travel. I'm also aware of the terrifying energies that will be bombarding a ship - the interstellar medium is quite cruel to objects that are approaching C.
The amount of shielding this vehicle would require is massive - solid feet of shielding to protect the occupants. So the economies of scale make it obvious: the vehicle would have to be enormous. More of a hollow asteroid than a spaceship. Is there enough energy available to launch that to another system?
You can write a model for things you can't measure if you like, but how can you know the accuracy of it or if it can predict future (impossible) measurements. It's almost nonsensical to try, unless some alien civilisation comes along with the data I don't see the point.
You can measure the logic of using individual components of the model, and the combination of those components, by exposing them to social scrutiny. That's all measurements are: An agreed upon social standard based on shared subjective experiences.
Plus, if any part of the model can be abstracted, there's a good chance you'll find leverage points to apply to other disciplines or problems.
There's also a socio-political aspect: We are an emotion-driven species. Sharing this kind of stabbing-in-the-dark information may actually drive the advancement of far less speculative technology that could be beneficial to us all.
Models like these only work when considering the order of magnitudes as a feasibility assessment. If you can determine a component reasonably has such a low probability that it pulls the entire thing down to infinitesimal probability then that observation is useful. However if the bounds of accuracy are too wide its not useful, this would be more useful with bounds.
It seems like we have an equation with a bunch of variables that are either extremely close to 1 or extremely close to zero. And really, these variables seem to group together into two essential variables, which are either almost 1 or almost 0. And we're trying to figure out whether, when you multiply them together, you get almost 1 or almost 0.
So, what does (almost_1 * almost_0) equal? It requires more knowledge about "almost" than are able to obtain.
I often joke about the MAYBE logical operator with friends. It's based on the concept of three valued logic [1]. It always came up for me in SQL where you have a cell in a result that can be NULL, NOT NULL or a value. I also use it to refer to spaghetti or complex code where the outcome on first glance is "maybe this or maybe that, or maybe something in the middle". Again, I use it as a joke when it is a totally real concept.
It seems like increasingly we are trying to extract approximate answers from machine learning, statistics, etc. because in so many cases it seems difficult to get to TRUE or FALSE for a given hypothesis. We want "MAYBE" with enough accuracy to believe it is mostly TRUE or 100% TRUE (but often 95% confidence - heck even 85% is still amazingly valuable when you aren't sure at all).
I believe there is a programming language that handles this (I forget which one though - anyone?). I know quantum processors potentially could handle it since they have multiple states for a qubit which could represent different states of "in-between" TRUE and FALSE.
This basically exists in Haskell/Idris as the Maybe type.
It is used slightly differently than what you're saying but in essence it lets you say:
- The thing doesn't exist, return Nothing (unit in a way)
- The inputs produced Just the thing you wanted
You have to cover both "didn't work" and "here you go" but its highly useful to not have to worry about NULL versus NOT NULL.
Your function that decides if something exists or not, could be the thing determining the percentage spiel.
Aka a function returning Maybe Bool could return: a Bool with true/false, or Nothing. Nothing being "its not a bool, your inputs are insane/nonsensical", but if it returns a Just Bool, you just unbox the Bool and voila. Just thought you might want to know this is a real thing and is super useful.
Ah, thank you. I knew it existed in some language, but I couldn't remember where I had seen it. I need to practice Haskell a bit more.. it's not one of the languages I ever seem to get around to.
You’re absolutely correct, f(almost_1,almost_2) should not be “times” in the calculation I described. Or alternatively, it should be a massive number times a tiny number rather than almost 1 and 0.
Let's say we absolutely found another advanced civilization, and in fact somewhat close, reachable in just 500 years. Let's assume we can build a rather large spaceship, say the size of the Empire State Building, and can fill it with enough food and fuel for the journey.
My question, what are the ethics of sending humans on such a trip? It would take 15 to 20 generations of human beings, living their entire lives confined to a prison, to complete this journey. An absolute nightmare of an existence, to be entirely disposable except to procreate in order to produce more prisoners like themselves for a few hundred more years. And when the ship did arrive, the living passengers would not be Earthlings, in the sense they would have little idea about the Earth, none would have ever been there or know anyone who had. We could provide them with every book, movie, etc that ever existed but they would still have no Earth-bound experiences. And it would likely depress them greatly, seeing and hearing things they will never experience. And finally, why would we expect anyone to ever return, to "complete" the mission? Their choice would be to make the best of life on their new planet or to once again volunteer for this bleak, suicidal journey to get back to Earth.
Some may say that future tech, or ultra-advanced VR, or cryogenics, or "downloading" brains and sending inert bodies, may make such a journey possible, but no answers sound very convincing, in my opinion.
That is sort of a plot summary of Kim Stanley Robinon's "Aurora", which has been criticized on the pages of this very website for not being upbeat and optimistic enough, but is IMO a very nice literary treatment of these problems.
I think we'd probably send robotic probes on missions such as this. The only way I can really see a multi-generational mission being done is for some kind of colonization mission.
Nothing personal, but this is exactly what I was mentioning about "future tech" not being sufficient to solve these problems. In-vitro fertilization (test-tube babies) still requires a living uterus in which to gestate. And the idea that a robot could successfully raise a child from birth to even about 8 years old seems VERY far-fetched. I think the physics of velocity, acceleration, etc will be figured out well before the human aspect, as evidenced by our ability to send a rover to Mars but not a person.
The problem with that argument is that a constant acceleration of 1g is an astronomically huge acceleration (for comparison, it corresponds to the surface gravity of a stone-and-iron ball with a whopping 12000km in diameter!). The most extravagant system on https://en.wikipedia.org/wiki/Interstellar_travel#Non-rocket... has a maximal acceleration of 0.3g.
This logic makes any birth unethical too. You are just confining your children to a prison of very large generation ship Earth, with the aim to "procreate in order to produce more prisoners like themselves".
Why is being alone terrifying? If we're alone it would be less "interesting" but it's not really scary. Are you accounting for a "the great filter is in front of us" scenario?
Additionally, the thought that we're not alone and we're not the most advanced species. If that's the case, then why haven't a more advanced species made contact yet?
For the same reason that humans have not "made contact" with ants.
Why do you assume a higher intelligence would even be capable of making contact with us on a level we understand? The discussion would be meaningless, in the same way that an ant would not comprehend our desire to build a freeway right next to their ant hill.
We have "made contact" with pretty much every species that display even a little intelligence though, including ants. Ants just don't notice when they run inside some scientist's experiment. Do you propose that we're so drastically inferior that we don't even know someone is having us find the cheese in a maze?
Yes, I think this is literally true. Messages may be encoded in the arrangement of stars, via PCM-neutrinos, etc., and humanity would be ignorant.
The history of biology is discovering species in places previously thought unlivable: impossibly salty environments, volcanic vents, etc. A natural extension is finding species in places "above" humanity.
An ant cannot "see" not comprehend the purpose of a freeway, no matter how hard we might try to explain it. It would walk right over it and not think it were any different than your average rock.
Sure, but that doesn't change the fact that the ant's ant hill gets destroyed if the freeway is built over it. Or that the ant itself gets crushed if it gets run over while walking across the freeway. The ant will notice those things even if it doesn't understand what caused them.
Exactly. Thus far, all we can say for sure is that we have not yet been carelessly crushed to make room for some intergalactic freeway we do not understand.
That doesn't mean there isn't one right in front of our eyes though.
a) There are only around 500 stars within the 100-or-so-light-year radius that our transmissions have reached. Does intelligent life arise on 1/500 planets? I'd say not even close.
b) Maybe they have no way to reach us. Despite sci-fi speculation, we still have no reason to believe the speed of light is not the ultimate speed limit of the universe.
c) Maybe they don't care. Aliens in movies often come here to invade us, eat us, steal our resources, etc. But, if any species had the level of technology to come all the way here with a force large enough to subjugate us, they wouldn't need to. Far more resources than they could take from this planet are already freely available floating around in space much easier to take.
d) Maybe they're so different from us that they don't even notice we are anything. Maybe we wouldn't notice them either.
a) the kind of ship required to travel interstellar at a reasonable pace, such as a nuclear termal rocket or antimatter rocket or the like, is pretty well understood in theory. It will glow like a galactic beacon and would be visible from vastly further away than 100 light years. Unless we're somehow technologically ahead of all other existing civs in our light window, should see somewhere, anywhere, anywhen, a single ship. But we don't. There just ain't ships out there.
(b) I agree. We have all reasons to believe it is not possible to accelerate something with mass to 'c'. We know this with astronomical levels of certainty. Anything that changes location without acceleration is either pure energy or pure fantasy.
(c) I agree. Even Dyson spheres don't have the raw energy capture to move an Earth of resources out of the sun's gravity well. Perhaps an Earth has rare and valuable resources that are quickly expended, with high value per kilogram. But there's no reason to believe an Earth would be special. The only real reason aliens have to contact us are the same reasons they'd try to immediately genocide us to get us out of their way anyway.
(d) this is an entire field of xenospeculation, but high energy rockets are really bright and very probably the only way to get to a different place in space. You pretty much have to talk microbe scale to go unnoticed in space.
The brighter a craft is, the less efficient it is. Every photon it emits in a random direction is energy that could have been better used to push it toward its destination. So, if any civilization out there really is super advanced, their ships may be extremely hard to detect. Also, a situation like Star Wars or Star Trek where beings in space vessels just cruise around the galaxy for personal reasons is highly unlikely at any technological level. We shouldn't expect to see a lot of ships out there.
The brighter a craft is, the more thrust power (watts) it emits. Both rocket efficiency (exhaust velocity) and thrust (newtons) contribute to thrust power multiplicatively if I remember correctly.
Any craft capable of moving things in our medium size realm or larger will glow. If you want to direct your exhaust along a cohesive line, you're basically talking lasers. And those can't really provide much thrust power.
If the hypothesis is that there are interstellar civs out there that don't exchange beings or materials, it's pretty hard to call them interstellar.
If you want to travel back and forth between inhabitable stars, forget rockets. They're just never going to get you moving fast enough.
I'm basically talking about a propulsion system that we haven't imagined yet. Because nothing we've (realistically) imagined would make an interstellar civilization possible.
I didn't say they wouldn't exchange beings or materials at all. I'm just saying that individuals will not be hopping in their space cars and cruising off to the crab nebula for the weekend.
One theory I've come across (can't recall where though) is that we aren't worth contacting given how un-advanced/different we are relative to whoever else is out there. The analogy goes, when the Spanish Conquistadors arrived in the Caribbean and Americas in the 16th century they walked ashore and didn't even notice the little ants under foot as they claimed the land and killed the indigenous peoples; we're the ants in this scenario, we haven't been contacted because we aren't worth it.
The ants weren't "contacted" in the sense of being asked anything, but that doesn't mean they were undisturbed. The question is not why intelligent aliens way beyond our tech level haven't tried to talk to us; the question is why they haven't just happened to crush us underfoot in the course of going about their normal business.
Part of the explanation of the whole conquistador/ant analogy addressing that point is that maybe other beings travel inter-dimensionally or in some other manner that isn't what we take for granted to be the only logical way to get around the universe, big ass spaceships, those would definitely disturb our ant pile.
Even if there are more advanced species, maybe they can't travel far enough to contact us. Or maybe they contacted other civilisations and we are just too far to bother.
Imagine the various uncontacted tribes that still exist. Considering how many humans there are, why hasn't there been more malicious actors striving to make contact with these people?
Granted this isn't a 100% proper analogy because a lot of these tribes have been contacted prior and have knowledge of the outside world. Yet it could be as something as simple as that; if there are far more advanced species we could be simply left to our own devices on purpose.
1. Space is very, very large.
2. Traveling between stars takes a very long time.
3. Other than a desire for exploration there is no need. Resources and planets are plentiful.
4. It now appears there is only an extremely brief window during which a technological species broadcasts in a detectable way.
#4 is probably the biggest problem. Digital compressed broadcasts would be difficult to detect, and by the time you have that technology you can build much more sensitive receivers and don't waste energy dumping millions of watts into space.
There could be 1000 intelligent species in the Milky Way right now and it is still unlikely any would visit our solar system within the past 10,000 years, nor in the next 10,000 and we would not be able to detect their faint compressed, encrypted broadcasts.
Hopefully they’ll have gone through a similar ethical development that we have and they’re not still going around conquering other civilizations like we were back in the day.
Or maybe they’re their planet’s equivalent of Nazis and they’re bent on wiping out non-purple-eyed humans.
It's like the shortest shorthand for "Where is your civilisation at?"
If you go from left to right, each value determines where you are in the history of galactic civilisations:
R* : you need to understand galaxies and star formation (1930-70s)
fP : you need to have space telescopes (we are here)
ne : we'll probably have this number in the next 10-20 years
fe : might need at least 30 years before we can detect life via atmosphere spectroscopy
fi : 100 years, maybe?
fc : same as above but with more time since we need more than one other data point
L : we're at a Star Trek level civilisation - we know of each civilisation's Shakespeare and Einstein, their history etc. Now we can learn how long technological civilisations last.
The whole basis seems to be picking a range of values for the variables in the Drake equation rather than a single value for each variable. As a result, the variance multiplies significantly meaning that they find a 30% chance we're alone in the galaxy as the lower end of the deviation (and the higher end being the common notion that there should be abundant intelligent life everywhere we look).
I hope they didn't spend too long making this "study" of tossing conservative estimates into a theoretical equation.
On a bit of a tangent from the core content of the article and the discussion here, I take issue with how easily even the smartest of people will take one example of a phenomenon and extrapolate the frequency(or replicability) of said phenomenon. All we really know about life, as we define it, is that it's possible to occur, because we're here, but we don't have any actual information about whether we're a fluke or an unknowing part of a greater community of intergalactic life. Pointing to billions and billions of stars and saying "there must be life out there" is unfounded at this point because we still aren't even close to understanding exactly how life came to be on Earth, as well as why it happened when it did. What the equation of predicting life is missing is the variable of time; without having at least two examples of life occurring independently, there's no way that I'm aware of that we can come up with a useful probability of life occurring anywhere. That is, unless we can create simulations to test theories outside of real time, which will probably be a required next step in science for getting a better understanding of our universe.
Intelligent life on Earth is (effectively) unique. By "intelligent", I mean capable of purposefully leaving our gravity well.
Residing in the goldilocks zone isn't enough. There needs to be external forces causing constant adaptation. Otherwise life will reach an equilibrium and evolve no further.
Our Earth has seasons, plate tectonics, periodic mass extinction events (asteroid strikes). Other less frequent causes are likely nearby supernovae and possibly our galactic arm passing thru a gamma radiation beam.
Even with constant evolution forcing habitats, a galaxy faring intelligence will be rare. As someone else noted, our saurian ancestors didn't stumble upon a tool building transmissible culture. There's just so many prerequisites.
Compare us to the octopus, likely the next most potent intelligent being on Earth, and certainly alien (to us). They are frikkin brilliant. But they unfortunately have no way to transmit, share their knowledge. And their life spans are probably too brief, birth rate too low.
Interstellar 'biomes' would operate on much slower clocks. Human perception time-scale of ~1s is mandated by fast-paced ecosystem of our gravity well. Maybe we just not had enough continuous data in ~50 yrs ~ 10e10s to recognise intersystem channel modulation as such. Economy cycle phase shift across the civilized area cant be much bigger than cycle duration, so for inter-system distances of ~1 ly we are looking at 10^6 years (1 a.u. @ 1 year financial year for humans right now). Any real economy involving Mars will already showcase the need for much slower finance clocks and longer plans than we have in current economy. I guess cryptocurrency clocks will split into temporal tiers according to systems served by that time.
Have we seen any proof of life out there? Well, no... So if there's an alien race out there that's at the same level of technology as we are, could they prove that we exist? Well no...
Let's think about that for a second, because to me, it's too early to tell since we don't possess the necessary technology for interstellar travel or any other technology that could prove it. And no other alien race might not either. Since we simply don't broadcast our location to half the galaxy that we exist we can't expect another race to it.
Finding something that might not want to be found has proven to be tricky and I think drawing the conclusion that there's no other advanced life out there is arrogant at best.
From what I’ve read, we do have the technology, but not the desire to spend so much money on what would be a vanity project with no return on investment possible due to the timescales involved. Especially as we expect our costs to come down and our early ships to be overtaken.
An alien civilisation with our technology and a million year headstart, trivial compared to the age of Earth, would have had time to build a K3 civilisation. (Assuming you can call things with that sort if time lag “a civilisation” rather than “a collection of civilisations the size of a K3”).
My pet theory about Fermi’s paradox is that there was only ever a very brief moment in the history of our civilization where attempted contact was politically and socially viable.
Think about the Voyager project right now. Assembling a rep sample of all of human existence onto a single golden record and including the instructions to decode it in math.
This. Could. Never. Happen. Today.
No one could agree on what to include or exclude or how to execute it. There was an incredibly brief window of opportunity where we had the technology and the social capacity to reach out to interstellar neighbors.
There are tons of other civilizations close by us. They just missed that very tiny window of a chance to say hello.
> Instead, it simply means that we can say with greater confidence – based on what we know – that humanity is most likely the only intelligent species in the Milky Way Galaxy at present.
Didn't read it, but that it's taken us ~4 billion years on this rock to git gud enough to send interstellar transmissions, it's probably taken other civs that long (because we have no evidence to the contrary).
Given that we're looking at systems thousands to millions and if "visible" literally billions of light years away, even if they beat us to interstellar radio, we're still getting messages from them which are hundreds, thousands, millions and billions of years old.
So they might be there just like us, but the Pony Express can only run so fast.
Almost all of these models assume that an "advanced civilization" must have language and radio astronomy and I've always felt that's a rather odd set of parameters. It's completely possible that we haven't found anyone else because we are looking for the wrong things.
We are looking for ourselves in the stars and fail to realize that we aren't "it". We may not even be the most advanced species on Earth.
When all the dolphins rise out of the ocean and into the sky maybe then we'll realize, but it will be too late.
We've had EM comms for 100 years. Let's assume life, and civilisations capable of sending comms beyond their solar system have emerged multiple times during the history of the universe.
If civilisations have a shelf life (for whatever reason) and don't exist indefinitely, it's extremely unlikely the timespan of someone transmitting and us receiving would overlap during those 100 years.
Chances of existence/comms overlap are tiny even for the next 1 or 10 million years - even if Earth origin civilisation is still around then.
The title of this post if completely wrong. The article rather specifically states that:
In the end, the team’s conclusions do not mean that humanity is alone in the Universe, or that the odds of finding evidence of extra-terrestrial civilizations (both past and present) is unlikely. Instead, it simply means that we can say with greater confidence – based on what we know – that humanity is most likely the only intelligent species in the Milky Way Galaxy at present.
Yep. The ugly truth: We are kind of alone out there. There's no one, or there's no one who wants to openly communicate with us, at least not yet. Get over it.
If we started the colonization of the galaxy in the year 2500 with 10 planets I suspect we would just wait and see if we came across a society with the level of technology we have today. After all, why take the risk of meeting a war like society that lobs nukes at each other?
Just leave a few probes nearby to keep tabs on them and see if they advance to a level that's interesting.... or scary.
I've seen that we search for exoplanets with life by using spectral analysis to estimate the temperature, atmospheric composition, etc. of these planets, and seeing whether or not they fall within a range that is similar to Earth's. I wonder if it would also be instructive to search for planets who exhibit signs of a man-made apocalypse, like massive nuclear fallout.
The Drake equation gets too much attention in the popular press . First, the selection of factors is an arbitrary choice of basis that may not relate to what we can measure and second, we don't have good ideas of the values of most of those factors. It's a mental exercise and useful for showing certain concepts, but not a prediction based on observable data.
I have always had a sneaking suspicion that the time scale on which we seek answers is far too narrow for answers to be found. Is it not likely that the period during which advanced civilizations actually exists is so short that they never cross each other's paths? I mean, if this were the case, wouldn't it throw the entire Drake equation out the window?
Amazing how uncomfortable humans are with ignorance. Are we alone or not in the universe? Is there life after death? Is existence infinite? We'll never know. I wish we could be OK with that. "I don't know" is a toxic phrase for leaders to utter, so they prevaricate and confabulate.
Spirituality, to the self, is largely a mechanism to sweep dukkha and its pointlessess, and lack of nuance in messy, real-world, imprecise interpretations under a rug of happy-clappy self-/mutually-reinforced delusions, blissful pseudosimplicity and sentimentality.
The big, unanswerable questions include:
1. Reincarnation: yes/no?
1.5. If yes, are we all one or are there multiple perception entities?
2. Why are you you and me me? (1. asks “am I and you the same?” implicitly)
3. Does cognitive dissonance allow “good”-self-image people to live with their committing unspeakable horrors?
4. If cancers’ semi-“life” existence can be explained by increasing a host’s entropy, why don’t most “pathogens” improve a host’s fitness, enable the host to find, store or expend more energy, reproduce and extend the host’s lifespan? Since when thr host dies, the cancer dies.
5. Is there an absolute limit of entropy?
6. How much neural complexity is needed for self-awareness?
7. Does global intelligence have a maximum useful upper limit?
8. If each piece of a person were systematically-replaced with inorganic parts of close overall function, does the individual retain rights of an organic person?
8.1. Could they be sued for healing or modifying themselves because of intellectual property?
8.2. Could they be legally owned, bought, sold, etc.?
life forms are just collections of atoms from a few elements. these elements exist all over the universe, no reason to think they would not have somehow combined to form life forms else where. I personally don't like the attitude of thinking we are alone in the universe.
We are currently looking for smoke signals. We need to find a loop-hole in quantum theory and exploit it to send and listen for signals that wont be affected by the distance between the sender and receiver.
Instead, should a model not aim to reach 0.5 probability, because we really do not know. Fermi Paradox or Drake equation express rather wishful thinking more than anything else.
Being alone in the galaxy and being alone in the universe seem to be casually equated but are vastly different in every possible way (except perhaps any practical consequences).
I think what this analysis tell us is that if you don't think it's possible to get better, more specific values for the parameters, the Drake Equation is just a bad tool.
Also, the Drake Equation doesn't describe the number of alien civilizations, it describes the likely number of civilizations. Thus, extremely pessimistic estimates have to somehow correct for the fact that we know N is not zero, since we are here. That seems like a highly informative update that is not addressed in an interesting way.