There’s been some attention lately to a contest on designing an interstellar generation ship, a large scale ship that humans live in for generations while it crosses interstellar space to another solar system. As Paul Gilster at Centauri Dreams notes, generation ships are a long time staple in science fiction, albeit with the common trope of the crew forgetting that they’re on a ship, or other things going horribly wrong.
But even before science fiction got into them, the generation ship was explored by early space exploration thinkers like Robert Goddard and Konstantin Tsiolkovsky. Whenever I’m tempted to dismiss current thinking about how interstellar exploration might work, I think about people like Goddard, Tsiolkovsky, Walter Hohmann, and Hermann Oberth, guys working in the early 1900s who were able to predict a lot of the space age, just by carefully thinking through known physics.
Although I find it hard to be too enthusiastic for generation ships. It’s worth thinking about what might have to be true for us to consign a group of people, who would have to be highly skilled, to spending the rest of their lives and that of their descendants in a profoundly isolated environment. It seems like we wouldn’t want to do it unless a number of factors were true.
We would likely want to know that there was a desirable destination worth pursuing. So we probably would have already sent robotic probes to the destination and would have thorough information on the environment. Otherwise the chances of ship’s descendants finding worlds no better than the other planets in our solar system would be too high.
There would also need to be some kind of ideology or religion, some type of manifest destiny involved, something that convinces a society to spend the kind of resources that would be needed for building something like a mobile space colony and accelerating it away at a velocity that allows it to reach its destination in any kind of reasonable time frame. (The contest posits one percent lightspeed, which gets it to Proxima Centauri in four centuries, but would take over a millenia to get to somewhere like Tau Ceti.)
To me, the whole endeavor is easier to imagine, and much less ethically dire, if it isn’t actually a generation ship, but a long duration mission for humans who have achieved immortality, or at least much longer lifespans.
It’s worth noting that the energy to get that large a habitat to even one percent of light (3000 kilometers per second) would be staggering. Although we might imagine it being doable with several gigantic fusion rocket stages. In calculating things like this, we always run up against the tyranny of the rocket equation, which is pitiless in revealing that fuel requirements increase exponentially the heavier our payload and the faster we want to go. (And are even yet more exponentially worse if we need to use the same method to slow down at the destination.)
Earlier this year I did a post asking where the aliens are. At the end, I noted that one possibility to explain why they’re not here, is that maybe interstellar travel is impossible, even for robots. Putting that at the end of the post led a number of people to conclude that was my argument. But I’m actually pretty bullish on the idea of robotic interstellar exploration. (Although I do fear generation, long duration, or sleeper ships might be as good as it gets for sending biological humans.)
Years ago, Paul Gilster made a comment that stuck with me. He noted that the main obstacle to interstellar exploration is energy, but we have all the energy we need in the sun. The trick is to find a way to channel it.
One of the currently most promising options is to use a laser propelled light sail, where a ground based laser, or array of lasers, propel a light sail craft to some substantial percentage of lightspeed. The beauty of approaches like this is they get around the tyranny of the rocket equation by having the energy used for acceleration remain outside of the spacecraft. This is the method envisaged by Breakthrough Starshot.
There are also hybrid approaches involving beaming power to a spacecraft which uses it to accelerate propellant, but the added weight and acceleration times increase the amount of coordination needed and opportunities for things to go wrong.
Breakthrough Starshot is currently aiming for a flyby mission, but to get enough information to support a future human mission, the craft would have to slow down and be able to explore at its destination. Slowing down, which in space takes just as much energy as accelerating, is a non-trivial problem.
A possible solution comes from an old idea. The Bussard ramjet was originally conceived of as a way for a spacecraft to gather its fuel in flight from the interstellar medium using an electromagnetic ram scoop. The problem is that the scoop has been demonstrated to likely produce as much drag as thrust. However, it leads to the idea of using a magnetic sail to break against the interstellar medium, and maybe even switching to an electric sail in the final stages to get down to interplanetary speeds.
Of course, this means a multi-sail design, which adds considerable weight, requiring larger initial light sails and laser arrays. But if we put the lasers on Mercury (as Robert Forward suggested in one of his designs), where solar power would be much more plentiful, such laser arrays start to seem more plausible.
Looking further down the road, the rocket situation could be improved if we can find a way to harness antimatter, aside from black holes, the most dense energy storage mechanism currently known. Manufacturing antimatter is often thought to be the bottleneck here, but again, if the antimatter factories were in close orbit of the sun, utilizing the solar power available there, it might be easier to imagine it happening.
All of which is to say, I don’t think interstellar exploration is impossible. I do doubt it will be practical for humans for a long time. But we seem to have multiple potential approaches for doing it robotically. While some may fizzle along the way, it’s hard to imagine all of them failing.
At least that’s how it looks to me today. But maybe I’m missing something? Are there problems with these approaches I’m overlooking? Or solid reasons to be more optimistic for sending humans?
SelfAwarePatterns 
You might (or being American) might not remember the disaster on Taransay – the BBC documentary about setting up a group of people to learn to live together and survive on a remote Scottish island. Turned out to be quite a horrid year for the guinea pigs. Think Lord of The Flies (not rings!). I suspect the huge ship would end up full of exactly the sort of people we have on earth. The majority pretty decent perhaps, although still prone to disobey the law (think motorists). While other would be swindlers, liars, violent thugs and so forth. I’m not sure immortality would improve the population much? I can’t think what would, really.
LikeLiked by 2 people
I don’t think I’d heard of that one, but I’ve heard of similar experiments. You might well be right. Gilster noted that the winning solution for the generation ship proposed having a population live first in an isolated colony in Antarctica, presumably to weed out problematic genes? But I don’t know that these kinds of problems are amenable to that kind of filtering.
All things equal, this is a lot easier if we can freeze the crew and thaw them at the destination. (Or even better, upload and transmit them to the destination, and print a new body there tailored for the environment.)
LikeLiked by 1 person
None of which would make for nicer people but I suppose at least they would get there without wiping each other out. Although of course their destination planet would fare far better if they never made it!
LikeLiked by 1 person
Sleeper ships wouldn’t make them nicer, but it does give them a lot less time to drive each other mad.
There are people who argue that if we find other biospheres, that we should leave them alone. It’s hard to see them getting their way. Although the sheer cost of moving matter interstellar distances makes it unlikely we’d exploit them for unobtianium or whatever.
LikeLiked by 1 person
The problem is it’s not clear we’re going to find anywhere like the Earth. Even if we’ve messed up the environment, the question becomes whether it’s easier to terraform other planets, or to just fix the mess we made on Earth. And if the thinking is that we should leave to help the Earth, then I’m not sure it’s any more ethical to go messing with other biospheres.
Long term, I suspect the real solution will be to find a balance with our current environment, and alter ourselves for any other biospheres we might want to colonize. That seems a lot cheaper and easier than terraforming. But prediction is hard, particularly about the future. 🙂
LikeLiked by 1 person
“anywhere like the Earth”
Yep. Another problem. We are superbly adapted to Earth, not space, or some other planet. Even a year in relatively friendly near-Earth space degrades human biology in major ways. Gravity (or its lack) and radiation are problems. At the destination, even small differences with Earth in gravity, atmospheric composition, climate, and who knows what else could turn the New Eden into a place where the arriving humans live out their lives effectively in a bunker.
LikeLiked by 1 person
Even if we can make earth a nice place to live for everybody, there will definitely be people who will want to go to new lands. Some will be cults, but some will just be explorers. And if immortality becomes a thing, some of us will just get bored.
But with all due respect [ 🙂 ], I’m not convinced you have the impact of AI correct. As you noted, the galaxy will be explored first by robots. As you didn’t note, these robots will not only gather information, but also generate infrastructure. We might start the ship off with lasers from around earth, but then the ships will be boosted w/ lasers from around Alpha Centauri, and then again with lasers from the next star on the path, and then the braking lasers from the path near the destination, and so forth.
LikeLiked by 1 person
Oh I agree completely about the AI infrastructure. I didn’t get into the idea of self replicating probes and the likely future interstellar internet because they weren’t necessary to make my point.
And I’d say that if mind copying is possible, then versions of us might not be that far behind the robots. If it isn’t possible, I could still see us gradually making our brain into a more rugged substrate, which could then be put by itself in a spacecraft and sent to the stars, growing a new body at the destination, and saving all the extra mass and complications of transporting living bodies over those distances.
My overall point though is we don’t need any of that more speculative stuff to work to do robotic exploration. We do need to develop new technologies, but we’re closer to the minimum than people often assume.
LikeLiked by 1 person
Couldn’t agree more. The human virus
LikeLike
On mind copying, I’m quite confident that won’t be a thing. A perfect copy is (probably) impossible, and a reasonably close copy is (probably) impractical. And then, what would be the point of a “close enough” copy, if the copy isn’t “you”. I liken it to creating a very detailed self portrait, and sending that. Or maybe creating an LLM version. Think the portraits of mages in Harry Potter. Essentially an act of vanity.
And I think the benefit of just sending your brain instead of your whole body may be insignificant.
otherwise, I’m in the choir.
*
LikeLiked by 1 person
At a minimum, an effective copy would be a backup in case the current version of you gets destroyed in an accident. It wouldn’t have to be woken unless you died. Or needed to send it somewhere a biological human can’t go, like to the frontier of the interstellar network.
The brain weighs 2% of the overall human body. Even if we say a rugged version is heavier, say 4%, or even 10%, that’s still saves 90% of the mass of a human body. And mass has an exponential relationship to the energy needed for any kind of space mission, much less an interstellar one. Of course, a copy that’s sent as a stream of photons weighs nothing and reaches 100% lightspeed instantly.
LikeLiked by 1 person
[because I’m in the mood …] what is the percent weight difference relative to the mass of the ship?
And … what makes a copy “effective”?
And … why wait ‘til you die? Why not make 20 copies now and send them on their way?
And … how long does it take to send a stream of bits equal to the number of atoms in the universe?
And … enough …
LikeLiked by 1 person
The necessary mass of the ship has a relationship to the mass of its payload, a multiplied one if that payload needs life support, air, water, food, etc. And the mass of the ship fully fueled (counting disposable stages) has an exponential relationship to the mass of the ship dry. In short, the ultimate payload (data, brains, bodies, etc) has an exponential relationship to the overall cost and energy of the trip.
An effective copy is one whose behavior can’t be reliably distinguished from the original. It can pass a sort of personalized Turing test.
“Why not make 20 copies now and send them on their way?”
As you like to point out, from original-you’s perspective, the 20 copies wouldn’t be you. Although if you’re needed 10 light years away, and copy-you can get there in 10 years but original-you needs centuries or millenia, sending the copy might be the easiest solution.
“how long does it take to send a stream of bits equal to the number of atoms in the universe?”
A long time. I don’t think an effective copy needs anything like that. But if it does, well, we just have to send the brain!
LikeLike
As you say, a generation ship wouldn’t launch without a destination. And even if a near star were a viable destination (and I don’t think any would be for people) and we had the technology you talk about now, then it would be over four centuries before we would even know it could be a destination. Just getting there would be challenge enough, but to be able to live there would require taking vehicles that could land with sufficient supplies and people to begin a colony. That might mean conventional fuel and, if the gravity is a lot stronger than Earth’s, there might be no leaving the planet once the landing is done. The expedition would need to have several hundred people by the time it lands to have a viable breeding population. Obviously severe reproductive control would need to be in place the entire journey to avoid inbreeding and either over or under population. I’m not sure who would sign up for it or why any government or organization would finance it.
Another problem is we don’t know much about interstellar space. Even small particles could do a lot of damage to spacecraft running at 1% light speed. Radiation? Do we need to put heavy shields on the craft to protect from cosmic rays? Are the people who reach the destination going to end up infertile or unable to produce viable offspring from genetic damage? This issue would even affect sleeper ships. The longer the journey the more the problem.
A lot of problems go away if you don’t send humans, but even robot probes would have avoid or be able to absorb collisions and manage radiation damage to computer memory cores.
And what about power supplies? I’m not talking about propellant but power for everything else (like collision avoidance) and just keeping the lights on. No solar or stellar for much of the journey. Nuclear reactors will eventually degrade. And fusion needs a lot of non-fusion power to make it work, doesn’t it?
I’m pessimistic without some breakthrough that finds major exceptions or extensions to known physics.
LikeLiked by 1 person
A robotic probe could be much lighter and so be accelerated via laser array to a much higher percentage of light than an interstellar ark (say 10-30%). It would still take decades, but not centuries.
But most of your other points I agree with. Major issues still to be worked out.
I’ve read that for crewed missions, water, which has to be carried anyway, stored in the outer layer of the ship, can be an effective radiation shield. Would it be enough? No idea. We need a lot more experience with crewed missions in the solar system first.
On power during the journey, if the ship is moving fast enough, the drag against the interstellar medium can actually be a source of energy. Although I’m not sure how much you’d get at only 1%. The Q-drive discussion from a few years ago got power via a magnetic turbine starting at 3-4% c, but that was to power propulsion. Maybe we can run onboard systems with less?
One advantage of a sleeper ship is it could be smaller, and so maybe faster.
But entropy is a system that isolated for that long is a non-trivial problem.
LikeLiked by 1 person
Generation ships are dead ends.
The human portion of an interstellar diaspora is a fraction of what is required to establish an Earthen colony on another planet. Fungi, bacteria, a thousand variety of flora, creatures of a hundred species–an entire ecosystem must be delivered across the light years. Delivered and established, over centuries.
Therefore, roboticized seed ships would be the only way to spread the human virus.
Seed the planet with a starter culture. Let that mature for a century. Seed the seas, the rivers, the lowlands, the deserts with botanical species. See what takes. Maybe genetically modifying species to delay with deficiencies or overabundances.
Then after three or five hundred years, start to grow fauna and release them. Finally, either in the seed ships or in facilities built by robotic agents, germinate human embryos and raise them with humanoid nannies.
VR will be far cheaper than shipping brains or bodies a trillion trillion miles to experience another world. Two different goals, colonization or experientialism.
LikeLiked by 1 person
This reminds me of a book by David Gerrold where a plague followed by weird animals and plants start showing up, before scientists realize they’re experiencing an invasion, just starting with the ecological foundations. And it’s an excellent point that humans, unmodified, would need our own foundations on any planet we hoped to colonize.
Agreed that VR is a lot cheaper for just getting the experience. And as I’ve said many times, for colonization, I think it’s going to be far easier to modify ourselves for a new environment than modifying the environment for us.
LikeLiked by 1 person
Thinking about this further, for colonization purposes, send our current representation of information:DNA, in the form of actual DNA, cells. If we could send only the DNA data, to an atomic assembler, previously sent, that might be the best way to establish colonies. Build the fungi and blue-green algae in-situ and let it rain down on a virgin planet.
Next best thing, send reconstituted swamp and estuary sludge.
Old Man’s War, I believe, used the “change humans’ DNA” motif to adapt to a wider variety of planetary environments. Others no doubt, too.
LikeLiked by 1 person
Ideally we’d be able to send over just enough, maybe self replicating nano-sized probes, to bootstrap things like assemblers, which could then construct everything else over time.
Old Man’s War did have a good take on altered human forms. Although Scalzi then limited its use in the OMV universe to nothing but soldiers, which seems like unfilled potential. At a minimum, people would want to upgrade to new bodies every time their current one gets old. There’s a new OMV book coming out in a couple of weeks. Maybe he’ll go somewhere new with it.
LikeLiked by 1 person
I think there is another way of approaching this, although for many people not as satisfying.
Just send life. What difference does it make if it eventually becomes human or something(s) else?
Humans aren’t going to be around forever anyway. It is part of our makeup to want ourselves to persist. We built monuments. We proclaim thousand year Reich’s. Many believe in life after death. Others want to get off the planet so humans do not vanish, but we will vanish anyway one way or another. We will evolve if we don’t destroy ourselves. Whoever we send will evolve if they make it. The humans on this planet might barely be human 500,000 years in the future whenever the end result of something we sent reports back. It might be as alien to us as we are to them.
LikeLiked by 1 person
The question comes down to what our goal might be. Many people would see our evolved or engineered descendants as still us in some sense. Even our machine progeny can be created in our image. In each case, we’re still leaving a legacy.
But just random life? As in extremophiles or something along those lines? Might be better than nothing, but although it may please some naturalists, I think you’re right that most wouldn’t consider it success.
LikeLiked by 1 person
Maybe the best goal possible is to spread life, not ourselves. Maybe somebody else thought that way about 6 billion years ago.
LikeLiked by 1 person
Just to provide a little more detail. Tell me if you think this could work.
At the plane of Earth, the Milky Way rotates in about 250 million years. If we sent billions (how many would we need?) of pods out from our system backwards in the direction of rotation, much of the Milky Way would eventually rotate around to the pods where some of them would be captured by other systems.
The pods would have mixtures of primitive organisms, certainly extremophiles among them, although future biology might enable some kind of custom life whipped up on the spot and geared to the conditions encountered., They might have some primitive navigation system that allows them to move slowly towards stellar systems and promising planets based upon light, using the light itself as power source. The pods would need to survive landing on a planet, then they would pop open if conditions were right and no other life was already found. They might spent millions of years on some planets waiting for the right conditions.
LikeLiked by 1 person
Interesting idea. Kind of an engineered panspermia.
I’m curious what you’re thinking on sending them backwards against our motion in the galactic orbit. The pods would inherit the motion of the solar system (about 200 km/s relative to the galactic center IIRC). To send them in any direction toward the immediate stars would involve changing their velocity, with similar amounts of energy in any direction. Although if the goal was to have them drift inward toward the galactic center, going anti-spinward would put them in a non-circular galactic orbit with the far side of the orbit closer to the galactic center. (Going spinward would put them in an orbit with the far side farther than the galactic center.)
I think the big issue with the pods is I doubt they’d be able to succeed in other biospheres, at least in the vast majority of cases. And without some guidance, we have to remember that space is big and solar systems, and planets in particular, very small. The probability is that most of them would never land on a planet, and of the ones that do, most of those wouldn’t land on one conducive to life. Although if we’re sending billions, it seems like we can imagine some might.
It’s when we consider whether those few biospheres might be compatible that we could have trouble. A lot depends on just how contingent the biochemistry on Earth is. If evolution usually converges on similar solutions, down to the DNA, RNA, and protein molecular levels, then some of them might have a chance. If not, the chances of them encountering a useful substrate could be very low. Maybe if we only send autotrophs, they could found their own biospheres, if they land somewhere with the right substrate.
LikeLiked by 1 person
I think they would need to be propelled initially sufficiently to get them out of our solar system. The idea is that they would relatively stationary in galactic orbit and the systems in the plane would come to them.
They wouldn’t be designed to succeed in other biospheres. They would be designed to create a biosphere where one didn’t exist. Sure, that won’t work on most planets and moons. However, a volcanic planet with a lot of water (Earth 4 billion years ago) might be hospitable.
LikeLiked by 1 person
With gravity mechanics, they can’t really be stationary. If we sent them out arresting their orbital motion, they’d start falling toward the galactic center. What we could do would be to give them escape velocity to escape the sun’s gravity and put them on random transfer orbits. Although for any chance of success, that would require more than billions I think, more like hundreds of trillions. An easier plan might be to send a million or so to specific stars, but with just enough velocity that they’d get there eventually over the tens of thousands to millions of years (assuming no unforeseen gravitational perturbations).
Seeding new biospheres does make more sense.
That said, I think there’s reason to be optimistic for more. If self replicating exploration probes propagate throughout the galaxy and build an interstellar internet, we would learn which specific stars have the best candidate planets. And depending on how well we’ve mastered bioengineering, we may be able to create new life out of local substrate tailored to the specific environment, as close or far from Earth life as might make sense.
LikeLiked by 1 person
I meant stationary relative to the galactic rotation, like maybe the same speed as the rotation in the opposite direction from the rotation. Maybe that doesn’t work either?
They could certainly be aimed for specific stars or clusters.
LikeLiked by 1 person
Orbiting in the opposite direction would require a lot more energy (~400 km/s delta-v). And you wouldn’t want the pods coming into their destinations much faster than a typical meteor. If we gave them opposite galactic orbits, they’d be coming in way too hot to survive, I think. Which is why I think the transfer orbits are better.
Although that might still have them coming in at the speed of a long term comet, and even that might be too fast. We could give them something to slow down, but now we’re depending on that mechanism working when they arrive, which across tens of thousands to millions of years seems like a tough ask.
LikeLiked by 1 person
3I/ATLAS is coming in at around 61 km/s, according to Google AI.
Speed might not be a problem if the pods were sent in bunches in containers that could launch the pods on a different trajectory once a system was encountered. That would also increase the odds of getting an appropriate environment.
“depending on that mechanism working when they arrive, which across tens of thousands to millions of years seems like a tough ask”
That’s an issue with any kind of exploration, isn’t it?
LikeLiked by 1 person
61 km/s is pretty fast for a natural object. It would almost certainly be completely destroyed if it impacted the Earth (and would likely cause widespread devastation in the process). That’s about the speed we might expect for an object whose galactic orbit just happened to intersect with ours.
“That’s an issue with any kind of exploration, isn’t it?”
It is, but the difficulty skyrockets as the duration increases. Having mechanisms work after a few decades, maybe even a few centuries, may be solvable engineering problems. That’s what we’d be looking at for a “fast” interstellar trip, one where the craft was accelerated to some significant percentage of lightspeed. Asking them to work after hundreds of thousands to millions of years? We have the problem you mentioned above, that it wouldn’t have an energy source to correct degradation from radiation and interstellar dust collisions.
(I noted above that if the craft is moving fast enough, the interstellar medium becomes an energy source. But I think that requires thousands of kilometers per second.)
LikeLike
Barring the discovery of new physics, I think you’re right. Personally, I envision humanity sticking to the Solar System for a good, long while. Terraforming Venus and Mars seem like much easier projects than sending a generation ship to Proxima Centauri (not to say terraforming either of those planets would be easy, just that they’re easier by comparison).
Also, regarding what you said about generation ships knowing in advance if their destinations are habitable, this has become a serious pet peeve of mine in science fiction. We can already figure out what exoplanets are made of and what their atmospheres are like. If it’s the future and you’re on a generation ship, there’s absolutely no reason you wouldn’t know things like that in advance.
LikeLiked by 1 person
I’d love to see us discover some way to do warp drives or wormholes, but yeah, new physics is a good way to put it. We have to go beyond current science to imagine them. And they come with serious causality issues almost never acknowledged in sci-fi. Not that I don’t enjoy stories with them.
In general, sci-fi has always underestimated the degree to which robots would precede humans in space. To some degree, it’s understandable. Stories of someone walking into the unknown are a lot more exciting than seeing them walk in years after the robots have already mapped everything. Doesn’t mean there won’t be things for humans to do, but if there are aliens there, the robots will already have found them.
LikeLiked by 1 person