The strongest reason that aliens might be invisible to us is that they are deliberately hiding. In fact I think that this is the only plausible reason.
Why would they be hiding? Well, they might be frightened that they’re in a simulation, and that the simulators have some crude algorithms that search massive tracts of space (many, many hubble volumes in size) by simply looking for inhomogeneity on the galaxy or supercluster level. The advanced aliens don’t want to get caught by the simulator.
This explanation would still work even if we’re not in a simulation: the threat of it is enough. Even a small probability that we’re in a simulation given the intelligence and data that an advanced alien civilization would have might be enough to offset the (perhaps small?) advantages of expansion, especially given recent work on bounded utility functions.
EDIT: This could be combined with the singularity hypothesis: perhaps all superintelligences, convergently decide not to expand.
Superintelligent agents can be expected to evolve out of systems that evolved by random mutations.
The systems that they evolved from can be expected to have goals that result in tracking down and utililising every available source of negentropy.
They will build superintelligent agents to help them attain these goals—and so the resulting systems are likely to be even better at tracking down and utilising negentropy than the original systems were—since they will pursue the same ends with greater competence.
Systems with radically different goals are not logically impossible. I call those “handicapped superintelligences”. If they ever meet any other agents, it seems that they will be at a disadvantage—since nature disapproves of deviations from god’s utility function.
Living systems maximise entropy. If the system dies out, it fails in doing that, and entropy increases more slowly. So: self-perpetuation is pretty much an automatic corollary of long-term entropy-maximisation. The best way to flatten those energy gradients is often to have lots of kids—and to let them help you.
While I’m not in any way an expert in simulation making, wouldn’t it seem just a bit too convenient that, in all the monstrous computing power behind making the universe run, the Overlords couldn’t devise a pretty clever and powerful algorithm that would have found us already? Maybe you can help me see why there would only be a crude algorithm that superintelligences should fear being caught by, and why they wouldn’t have considered themselves caught already.
Apart from this, I’m in agreement with other commenters that a stronger argument is the vastness of space.
Overlords couldn’t devise a pretty clever and powerful algorithm that would have found us already?
Maybe the overlords are very, very, different life forms than us, and so they don’t know what to look for other than large scale statistical anomalies?
But I must admit, this is a weakness of the hypothesis.
Apart from this, I’m in agreement with other commenters that a stronger argument is the vastness of space.
More space = more stars = bigger problem. There are roughly 10^14 stars within 100 million light years of earth (as far as I can tell), which would make for 10^17 within 1 bn ly, and 10^20 within 10bn ly. The universe is a whole 13bn years old.
I think wireheading, religious seclusion or some other memetic effect could keep them from expanding. I get that there is a selection effect in favor of expansionism but I think aliens could be weird enough that the entire range of variability is within isolationist boundaries. Besides, if you think the aliens can all agree to stop expanding to hide from theoretical simulators then I don’t see why some other meme couldn’t do the trick too.
Besides, if you think the aliens can all agree to stop expanding to hide from theoretical simulators then I don’t see why some other meme couldn’t do the trick too.
I can’t think of anything else that would affect every single alien civ out of trillions; in fact, even this explanation seems unlikely to account for no expansion from every single one of trillions of civs.
I think wireheading,
Perhaps we should regard the fermi problem as evidence in favor of wireheading?
religious seclusion
I think that the “religious seclusion” idea is not a good bet. Remember, it has to work with really rather high fidelity.
or some other memetic effect could keep them from expanding
If the simulators are looking for intelligent life, then they may be removing it when they find it, to allow the rest of the universe to continue on producing other original life forms. Then, there would be intelligent life in hiding, and other intelligent life would be selected against.
The strongest reason that aliens might be invisible to us is that they are deliberately hiding.
Couldn’t they just be really, really far away; or have a structure and value system such that their bustling activity looks like natural phenomena; or be trying like hell to get our attention but finding us kind of hard to talk to?
or have a structure and value system such that their bustling activity looks like natural phenomena
No; to see why you need to think about physics, engineering, thermodynamics and information processing. They’d want to capture low entropy sources of energy, and radiate away all that energy as high-entropy infra red radiation. This would be noticeable. Also, it would be an amazing coincidence if alien engineering just happened to look exactly the same as pristine nature.
Or if they entered a region of dust or gas that had a temperature differential to the sphere.
If fact, I just realized that if any such structures like Ringworlds existed, they would be trivially easy to spot if they were at the right aspect to us. It would look like a star with either a band across it, or like the star had a single ring. The ring on the far side of the star would also be tremendously bright (if angled to our solar system slightly) and much easier to spot with a telescope that blocked off the light of the central star.
Stars are too bright and too far away and the ringworld too thin (at least I think there would be stability problems if they were built too wide, but I don’t know enough). Worse, I thought the way we were trying to detect planets these days was by looking at the gravitational effect the planet has on the star. But wouldn’t a ringworld balance out its own gravitation effect on the star?
you need to think about physics, engineering, thermodynamics and information processing
Like this guy did. Short short version: to get the most out of your star, you drain the energy of the emitted photons until they are (almost) at the temperature of the CMB. Infra-red is still harvestable.
Getting rid of all the stars in one part of space and replacing them with near CMB doesn’t seem to produce what we see: it would produce a raised CMB signal in one direction (would it be strong enough to pick out? How close, exactly, to CMB would it be?), a reduced star density in one region, and presumably a bunch of half-eaten galaxies on the boundary of the region.
The point that “it would be an amazing coincidence if alien engineering just happened to look exactly the same as pristine nature” seems reasonable convincing to me.
I think the idea is not all the stars in a region—it’s one star per civilization. The basic idea is to maximally exploit energy transfer between the star and interstellar space, so optimal efficiency makes the artifact appear from the outside to be as close to interstellar space as possible.
Another possibility is that they’ve gone through their own singularity, and no longer have a significant visible (to us) presence in physical space (or at least don’t use radio waves any more), i.e. they’ve transcended.
Naturally we can only speculate about what new laws of physics become identifiable post-singularity, but string theory suggests there’s a lot of extra dimensions around to which we don’t have direct access (yet?). What if civilisations that transcend tend to find 4D spacetime too limiting and stretch out into the wider universe, like a small-town kid making it big?
It look like, if it happens for us, that it will happen within the next hundred years or so. Considering that we’ve only been using radio for a bit over a hundred years, that gives a roughly two hundred year window for any two pre-singularity civilisations to spot each other before one or both of them transcend.
0.4 seconds in the year of the cosmos.
Myself, I suspect the most plausible explanation is that interstellar colonization is simply a lot harder than we think. Nobody yet has managed to build a self-replicating probe that will actually survive through interstellar space intact and manage to set up a colony, and it’s not a given that anyone ever will. Add to this the fact that even if it were possible, it could be horribly expensive, with a return on investment at least hundreds of years away.
Kaj, I’m surprised! I think you should update against this explanation; of all the technologies discussed here (superintelligence, Dyson spheres, megascale engineering, nanotech), space colonization is the only one that is so simple that even contemporary humans have a definite recipe for doing it that would probably work. If humans can do it, to a superintelligence it would almost certainly be trivial.
space colonization is the only one that is so simple that even contemporary humans have a definite recipe for doing it that would probably work
Maybe. The nearest solar system—which would apparently take about a hundred years to reach—is over four light years away, and it isn’t even certain whether it has any planets. The closest planet that is actually known is apparently ten light years away, which would then take 250 years to reach. It’s obviously not impossible that we—or some future superintelligence—might manage to build a craft that survived that whole journey intact enough to continue with a colonization effort, but it doesn’t seem obviously possible either.
It seems fairly clearly possible because I can actually think of ways of getting there, including having all those tedious details filled in. Fleets of nuclear powered ships with 10-20 generations of humans on them could colonize thousands of worlds. The details of the nuclear powered ships and the technology to keep humans alive in space have been hashed out. Colonizing once there is tougher, we haven’t really done the bulk of the work on how to colonize a new solar system, but it doesn’t seem to require magical technology like nanotech to do it, just a large initial ship with a lot of infrasturcture on it and lots of fairly tedious planning and preparation.
Furthermore, I anticipate that technology will not stand still for the next milennium. What we can do now with difficulty, expenditure and much loss of life (colonizing the local stellar neighborhood) will, in 1000 years worth of technological development, be possible without much fuss. Nanotech, good AI systems, near-light-speed Von Neumann probes, and voila.
Since some people have opined that maybe we’re not alone in the universe, I’ll write down the strongest argument in that I can think of in favor of this position. (To win, you must fight not only the creature you encounter; you must fight the most horrible thing that can be constructed from its corpse. )
The strongest reason that aliens might be invisible to us is that they are deliberately hiding. In fact I think that this is the only plausible reason.
Why would they be hiding? Well, they might be frightened that they’re in a simulation, and that the simulators have some crude algorithms that search massive tracts of space (many, many hubble volumes in size) by simply looking for inhomogeneity on the galaxy or supercluster level. The advanced aliens don’t want to get caught by the simulator.
This explanation would still work even if we’re not in a simulation: the threat of it is enough. Even a small probability that we’re in a simulation given the intelligence and data that an advanced alien civilization would have might be enough to offset the (perhaps small?) advantages of expansion, especially given recent work on bounded utility functions.
EDIT: This could be combined with the singularity hypothesis: perhaps all superintelligences, convergently decide not to expand.
It seems like a weak argument to me: living systems typically invade all available niches—if they are here, we would see them.
For a much better argument, consider:
http://en.wikipedia.org/wiki/Fermi_paradox#Intelligent_civilizations_are_too_far_apart_in_space_or_time
Tim: these civilizations would be superintelligences, they would not behave in a way that is typical of dumb life.
But I agree, I find this argument somewhat weak.
Superintelligent agents can be expected to evolve out of systems that evolved by random mutations.
The systems that they evolved from can be expected to have goals that result in tracking down and utililising every available source of negentropy.
They will build superintelligent agents to help them attain these goals—and so the resulting systems are likely to be even better at tracking down and utilising negentropy than the original systems were—since they will pursue the same ends with greater competence.
Systems with radically different goals are not logically impossible. I call those “handicapped superintelligences”. If they ever meet any other agents, it seems that they will be at a disadvantage—since nature disapproves of deviations from god’s utility function.
More important than negentropy is continued existence. If the simulation gets shut down, you’ve already lost all you already have.
Living systems maximise entropy. If the system dies out, it fails in doing that, and entropy increases more slowly. So: self-perpetuation is pretty much an automatic corollary of long-term entropy-maximisation. The best way to flatten those energy gradients is often to have lots of kids—and to let them help you.
While I’m not in any way an expert in simulation making, wouldn’t it seem just a bit too convenient that, in all the monstrous computing power behind making the universe run, the Overlords couldn’t devise a pretty clever and powerful algorithm that would have found us already? Maybe you can help me see why there would only be a crude algorithm that superintelligences should fear being caught by, and why they wouldn’t have considered themselves caught already.
Apart from this, I’m in agreement with other commenters that a stronger argument is the vastness of space.
Maybe the overlords are very, very, different life forms than us, and so they don’t know what to look for other than large scale statistical anomalies?
But I must admit, this is a weakness of the hypothesis.
More space = more stars = bigger problem. There are roughly 10^14 stars within 100 million light years of earth (as far as I can tell), which would make for 10^17 within 1 bn ly, and 10^20 within 10bn ly. The universe is a whole 13bn years old.
I think wireheading, religious seclusion or some other memetic effect could keep them from expanding. I get that there is a selection effect in favor of expansionism but I think aliens could be weird enough that the entire range of variability is within isolationist boundaries. Besides, if you think the aliens can all agree to stop expanding to hide from theoretical simulators then I don’t see why some other meme couldn’t do the trick too.
I can’t think of anything else that would affect every single alien civ out of trillions; in fact, even this explanation seems unlikely to account for no expansion from every single one of trillions of civs.
Perhaps we should regard the fermi problem as evidence in favor of wireheading?
I think that the “religious seclusion” idea is not a good bet. Remember, it has to work with really rather high fidelity.
Some other mimetic effect seems more promising.
If the simulators are looking for intelligent life, then they may be removing it when they find it, to allow the rest of the universe to continue on producing other original life forms. Then, there would be intelligent life in hiding, and other intelligent life would be selected against.
Couldn’t they just be really, really far away; or have a structure and value system such that their bustling activity looks like natural phenomena; or be trying like hell to get our attention but finding us kind of hard to talk to?
Yes, they could lie beyond our cosmic horizon.
No; to see why you need to think about physics, engineering, thermodynamics and information processing. They’d want to capture low entropy sources of energy, and radiate away all that energy as high-entropy infra red radiation. This would be noticeable. Also, it would be an amazing coincidence if alien engineering just happened to look exactly the same as pristine nature.
Could we detect Dyson spheres if they were out there?
Might we be able to see them if they went in front of galaxies or other highly luminous body?
Come to think of it, wouldn’t they look a lot like black holes?
Not really. The black hole might have hawking radiation it should also gravitationally lens the surroundings more than a dyson sphere.
It might also have an accretion disc.
Or if they entered a region of dust or gas that had a temperature differential to the sphere.
If fact, I just realized that if any such structures like Ringworlds existed, they would be trivially easy to spot if they were at the right aspect to us. It would look like a star with either a band across it, or like the star had a single ring. The ring on the far side of the star would also be tremendously bright (if angled to our solar system slightly) and much easier to spot with a telescope that blocked off the light of the central star.
Just an OT thought.
Stars are too bright and too far away and the ringworld too thin (at least I think there would be stability problems if they were built too wide, but I don’t know enough). Worse, I thought the way we were trying to detect planets these days was by looking at the gravitational effect the planet has on the star. But wouldn’t a ringworld balance out its own gravitation effect on the star?
The Kepler Observatory also detects planets by masking the light from the star and then looking for any bright spots (Coronal Masking).
Even more advanced telescopes will use this technique in combination with the gravity wobble. There are also diffraction and coronal masking.
http://www.iop.org/EJ/article/0004-637X/662/1/738/65461.web.pdf
http://spie.org/x24241.xml?ArticleID=x24241
why did I get down voted on that post?
Like this guy did. Short short version: to get the most out of your star, you drain the energy of the emitted photons until they are (almost) at the temperature of the CMB. Infra-red is still harvestable.
Getting rid of all the stars in one part of space and replacing them with near CMB doesn’t seem to produce what we see: it would produce a raised CMB signal in one direction (would it be strong enough to pick out? How close, exactly, to CMB would it be?), a reduced star density in one region, and presumably a bunch of half-eaten galaxies on the boundary of the region.
The point that “it would be an amazing coincidence if alien engineering just happened to look exactly the same as pristine nature” seems reasonable convincing to me.
I think the idea is not all the stars in a region—it’s one star per civilization. The basic idea is to maximally exploit energy transfer between the star and interstellar space, so optimal efficiency makes the artifact appear from the outside to be as close to interstellar space as possible.
Another possibility is that they’ve gone through their own singularity, and no longer have a significant visible (to us) presence in physical space (or at least don’t use radio waves any more), i.e. they’ve transcended.
Naturally we can only speculate about what new laws of physics become identifiable post-singularity, but string theory suggests there’s a lot of extra dimensions around to which we don’t have direct access (yet?). What if civilisations that transcend tend to find 4D spacetime too limiting and stretch out into the wider universe, like a small-town kid making it big?
It look like, if it happens for us, that it will happen within the next hundred years or so. Considering that we’ve only been using radio for a bit over a hundred years, that gives a roughly two hundred year window for any two pre-singularity civilisations to spot each other before one or both of them transcend. 0.4 seconds in the year of the cosmos.
Myself, I suspect the most plausible explanation is that interstellar colonization is simply a lot harder than we think. Nobody yet has managed to build a self-replicating probe that will actually survive through interstellar space intact and manage to set up a colony, and it’s not a given that anyone ever will. Add to this the fact that even if it were possible, it could be horribly expensive, with a return on investment at least hundreds of years away.
Kaj, I’m surprised! I think you should update against this explanation; of all the technologies discussed here (superintelligence, Dyson spheres, megascale engineering, nanotech), space colonization is the only one that is so simple that even contemporary humans have a definite recipe for doing it that would probably work. If humans can do it, to a superintelligence it would almost certainly be trivial.
Maybe. The nearest solar system—which would apparently take about a hundred years to reach—is over four light years away, and it isn’t even certain whether it has any planets. The closest planet that is actually known is apparently ten light years away, which would then take 250 years to reach. It’s obviously not impossible that we—or some future superintelligence—might manage to build a craft that survived that whole journey intact enough to continue with a colonization effort, but it doesn’t seem obviously possible either.
It seems fairly clearly possible because I can actually think of ways of getting there, including having all those tedious details filled in. Fleets of nuclear powered ships with 10-20 generations of humans on them could colonize thousands of worlds. The details of the nuclear powered ships and the technology to keep humans alive in space have been hashed out. Colonizing once there is tougher, we haven’t really done the bulk of the work on how to colonize a new solar system, but it doesn’t seem to require magical technology like nanotech to do it, just a large initial ship with a lot of infrasturcture on it and lots of fairly tedious planning and preparation.
Furthermore, I anticipate that technology will not stand still for the next milennium. What we can do now with difficulty, expenditure and much loss of life (colonizing the local stellar neighborhood) will, in 1000 years worth of technological development, be possible without much fuss. Nanotech, good AI systems, near-light-speed Von Neumann probes, and voila.
Don’t send 1, send 1000.
Granted.