Although Jane Smiley has made the haunting story of the Viking settlement of Greenland widely known in her novel The Greenlanders (Knopf, 1988), we have few modern accounts that parallel what happened in remote places like Brattahlið and Garðar, where Erik the Red’s settlements, which had lasted for 500 years, eventually fell victim to climate and lack of external supplies. But local extinctions and near-misses are important because, as John Hickman explains in his new book Reopening the Space Frontier (Technology and Society, 2010), they promote the kind of story-telling that Smiley is so skillful at, advancing the case that not just settlements but entire species can fail when conditions turn ugly.
Image: A reproduction of a Norse church in Greenland, with Eriksfjord in the background. Credit: Hamish Laird/Wikimedia Commons.
In this excerpt from the book, Hickman writes about three modern parallels to 15th Century Greenland, the first being the Sable Island mutiny, where provisioning ships to a French penal colony in the north Atlantic stopped arriving in 1602. The demise of the Sadlermiut Inuit is another case in point, the 58 remaining members of this Dorset culture population expiring due to the effect of infectious disease after a Scottish whaling ship reached Hudson Bay. And then there is Clipperton Island, an eastern Pacific atoll that collapsed when provisioning ships stopped bringing supplies during the Mexican Revolution.
Why write about this in the context of space exploration? Because Hickman wants to understand why we pay so little heed to the technologies that could save our planet from extinction from the likes of a rogue asteroid. We see small catastrophes and large around us, from the 100 million who died at the hands of war, genocide and famine in the twentieth century to the hundreds of millions affected each decade by earthquakes, floods and hurricanes. But our own survival is assumed. Writes Hickman:
Recent events such as the 2005 Indian Ocean tsunami, the 2008 tropical cyclone in the Irrawaddy River Delta, and the 2010 earthquake in Haiti which each killed more than 200,000 people provide powerful reminders of the vulnerability of our species to natural disasters. What they have not done is increase awareness of human vulnerability to extinction. Interviews with people who had managed to cheat fate by surviving catastrophes do not incline viewers to consider the possibility of a global catastrophe without human survivors. News audiences around the planet learned about each of these events via television news coverage, which means that the emotional impact of each event was cushioned by the medium’s entertaining and anesthetizing unreality. If these tragedies were rendered to that degree unreal then the possibility of human extinction must seem impossible.
Thus it is that, while we know about the risk of impacts from space and have funded efforts to detect incoming debris, we have yet to devise a well-funded solution to prevent such an impact. Having little experience with truly existential threats, and having seen nothing but the excavations at scattered sites like those above to remind us how a culture can collapse and disappear, we fail to accept the severity of long-term risk. Hickman thinks we’re hard-wired to recognizing only the immediate, a consequence of evolving in environments that did not demand a longer perspective.
Rather than evolve a general perception of risk and a rational calculation of relative risk, our 150 millennia living as hunter-gatherers and the most recent 2 millennia living as farmers prepared us to deal with only some kinds of risk. We respond strongly to those with a high probability of occurring rather than to those with a low or unknown probability of occurring. Warfare, subsistence failure and cooperation failure have produced an animal that is equipped with “specific cognitive adaptations for perceiving cues regarding the likelihood and magnitude of adverse events” and for making rapid decisions based on “risk-risk” trade-off calculations.
And so it is that three million Italians continue to live around the Bay of Naples near Mt. Vesuvius, the consequences of whose historical eruptions are available for all to see in the ruins of Pompeii. People live by the millions along fault lines in California that could spawn catastrophic earthquakes, coping with a threat of extinction that is, in many ways, too abstract to visualize. We get on with daily life. Our perception is deeply human and understandable, and it offers a recipe for playing down security for future generations in favor of proceeding with today and stressing how infrequently catastrophes occur. Short-term means getting through a finite lifespan, and it doesn’t extend to the kind of lapidary effort that builds a safer future for generations beyond.
It’s a difficult truth to acknowledge, but it seems to be part of human nature. Our innate assessment of risk means we have a steep wall to climb to promote the well-being of our distant descendants, and that makes even the most basic attempts to survey the population of near-Earth objects a matter of constant watchfulness to ensure the continuation of funding. Getting into space to prevent an asteroid strike that might not occur for millennia is a hard sell, and so is establishing a space presence to ensure species survival in case anything happens to our planet. Read the excerpt from Hickman’s book and you’ll understand why it may take a near-fatal event (think of the survivable asteroid strike in Clarke’s Rendezvous with Rama) to make long-term danger immediate and reinvigorate our will to master space technologies.
Eniac: OK, I think we have satisfactorily settled the dispute with my and your previous posts.
The way you describe it, those O’Neill colonies could ultimately constitute a kind of Dyson swarm, fully utilizing solar energy.
I still doubt how feasible such colonies could ever be as means of transport to neighboring stars, because of their slowness and continuing dependency on solar system resources.
All in all, I still see those space stations as complementary to planets not as substitutes.
I presume that the eventual (scale of) implementation of these kind of mega-space-structures will depend on two other factors:
1) Our ability to travel to the stars: how fast, how often, how expensive.
2) What we will find (beforehand) near those stars: how common are terrestrial/earthlike planets.
I.e. if we ever manage to make it to the stars in reasonable time and at reasonable cost, and suitable planets are common, I think that planetary colonization will get preference. However, if one of both or both factors appear to be quite disappointing, solar system space colonies will probably get preference.
In other words: if we can get 12 suitable planets, why build 12 O’Neill colonies?
ProtoAvatar:
“Eniac did the math, (…) He proved these 12 O’Neill colonies are far safer than a life supporting planet; far, FAR safer than domes built on Mars (…)”.
No, wishful thinking, he did not prove that, he offered reasonable arguments for those colonies, without proving their superiority to a lifesupporting or colonized/terraformed planet, however.
“You can’t argue with the numbers – credibly, that is, Ronald.”
O yes, I can: once again, one can only multiply *independent* chances. However, in the case of serious internal failure (technical, structural and otherwise), what happens to one colony, will likely happen to others.
Example: the chance of serious failure of the brake system of a car is rather small, the chance of it happening to a large number of cars within the same time period seems infinitesimal. However, Toyota (and other companies?) recently called back many, many cars to their factory for revision because of serious shortcomings in the brake system.
Something similar happened to an aircraft manufacturer (Airbus?) some years ago.
The more complex an artificial system, the more prone to internal failure it generally is. Natural systems are usually quite different in this respect.
Multiplying of chances in statistics works fine only for *independent* chances.
Ronald:
O’Neill colonies are artificial habitats. As such, they have failure modes planets don’t.
But planets have failure modes O’Neill colonies don’t ( changing climate, for example).
The great advantage of O’Neill colonies is that they are autonomous, while planet based settlements are not so. This is valid for many failure modes, ranging from astroid impacts to societal or economic crisis, etc.
It should also be pointed out that, when talking in giga-years, planet wide catastrophes go from highly improbable to near-certain. On the other hand, the chances of all O’Neill colonies being affected by catastrophe at the same time remain very small.
About O’Neill colonies being artificial habitats and needing maintenance in order to function – for thousands of years, humanity has lives in artificial habitats (cities) that need maintenance in order to function. So far, it worked out pretty well.
Ronald: I, too, think we are mostly in agreement. I have a few more quibbles, though:
I think you underestimate the interstellar potential of O’Neill habitats. There will be various reasons for at least some of them to migrate into the outer reaches of the solar system, and rely on nuclear rather than solar energy. Apparently there is Oort cloud material most of the way from here to Alpha Centauri, so raw materials also do not necessarily keep you close to the sun, either. People build houses in the desert, and some might decide to move ever further out into interstellar space.
I disagree that common defects violate independence. If a particular common defect has a 1% chance of causing habitat failure, then two habitats with the same defect will still both fail only in 0.01% of cases. It does not matter if the defect is the same or different, as long as one failure does not _cause_ the other, there will be no correlation. What you are worried about is not interdependence, but unanticipated failure. You are thinking about the case where a large percentage fail, at a rate not anticipated, due to an unrecognized common defect. Although obviously this is a valid problem, it is not interdependence. It is having incorrectly estimated the single habitat failure rate to begin with. In mathematical terms, the power law still applies, but with a larger base.
In fact, there is an anticorrelative effect in common defects: After the first failure occurs, the ensuing investigation and corrective action will _decrease_ the chance of it occurring in the other habitats. We have here a beneficial form of dependence.
This is a very good point. Together with the fact that most planets are not suitable for walking around on without equipment, it shows that the view that we should live on planets because it is natural to us is of very limited validity.
“the view that we should live on planets because it is natural to us is of very limited validity.”
O’Neill described that view as planetary chauvinism.
It can be annoying that no matter how often you remind them, people will forget that even the “most Earth like” planet Mars is a hell of a lot less habitable than the central Antarctic ice cap. They imagine domed cities on Mars, well why not domed cities at Vostok station? Space based settlements on the other hand will have the ability to manipulate the environment in and around them in ways that cannot be duplicated on a planets surface.
Eniac, ProtoAvatar: I think matters are settled well now, very interesting discussion BTW, I am again learning a lot, thanks.
I also have one more quibble left, the economic one: how much living space (area) is produced per invested amount of money and energy, maybe also per time unit.
I do not have the answer ready, but the parameters are approximately as follows: let’s reasonably suppose that terraforming Mars takes about 1000 years and an investment of xxxx billion dollars (net present value). For that amount of money, time and energy invested though you get quite a piece of real estate, about 28% of the earth surface, of which about half would be mainly water and half mainly land.
It is also reasonable to assume that Mars could remain in such a terraformed state for at least 100,000 years without too much corrective effort. Various risks, their damage and mitigating/restorative costs, could be estimated.
My question is: how does that compare to the total cost, including risk mitigation/restoration/replacement, of O’Neill colonies? If we assume the circumference of an O’Neill colony to be 30 km and its lenth also 30 km, its total living area is some 900 km2. Even if we consider only the land area of Mars (approx. half of its total surface area, the other half becoming watery) and only half of this as suitable, the rest being too polar or whatever, i.e. only a quarter of the total area of Mars, this is still some 36 million km2.
In order to acquire the same living area with O’Neill colonies of mentioned size you would need some 40 thousand of those!
Although of course this is to some extent like comparing apples and oranges, it would still be interesting to make a comparative calculation, including the ‘risk insurance’.
What about micrometeroids? My feeling was that a planetary atmosphere would be a highly useful shield against this problem. Mars’ may not be substantial enough, but Venus would be perfect. While O’Neill colonies can defend against radiation with a surrounding layer of regolith (http://www.nss.org/settlement/ColoniesInSpace/colonies_chap12.html), will they be able to deal with micrometeoroids?
Of course, exoplanets are another story. An earthlike exoplanet would be a highly desirable target, O’Neill colonies notwithstanding. The amount and quality of resources and real estate of an ideal (or near ideal) exoplanet would be highly valuable and desired.
O’Neill colonies, of course, will be important in the expansion into space. They can be made to work in many convenient ways. Even if there was an exo-earth 10 light years away, we’d still make use of such structures around this star and the other star. The interstellar possibilities of O’Neill habitats as outlined by Eniac are of great interest. I still see them as being one part of space infrastructure and economy, though – an important part, but not completely dominant.
Yes, a layer of sandbags will work for both.
Unfortunately, the existence of oxygen in the atmosphere is pretty much coupled to the existence of “complex” life. Life is required for oxygen, and oxygen gives rise to “complex” life. So, if there is no life, we will need scuba equipment, at a minimum. If there is, we will likely have to share the planet with other critters, unlikely to be friendly or edible. The latter would not really be so bad, not to mention interesting, but I suspect we will run into the former a lot more often.
An interesting question, but one that I would be loath to attempt to answer, due to the large uncertainties in estimating the economics of O’Neill colonies, and the even larger ones concerning terraforming.
Ronald, as you mention, O’Neill colonies could be built sooner than Mars could be terraformed, also if I’ve only got y investor each with x dollars, I might be able to build an O’Neill colony, whereas terraforming Mars might cost xy x 10^5 dollars, and so would be too big a project for me and my investors.
O/T but this guys got some interesting ideas on terraforming.
http://www.paulbirch.net/#paper
Given planets have weather and sometimes seasons any planet based colonies would have to pay much more for maintaining their infrastructure than what an O’Neil colony would.
ie. On mars you would have all the disadvantages of building in space with all the disadvantages of building on a planet. You would still be operating in what would be to us a vacuum with no breathable atmosphere at all, and yet have to deal with the wind and dust.
Failure of a life-support system would be just as likely to occur on a planet like mars (actually more so due to weather and dust) as an O’Neil colony yet evacuating from a place like mars would be much more difficult than evacuating from and L point near the earth. Meaning the people are more likely to die.
And those planet based colonies would be at a greater threat from asteroids than what an O’Neil colony would be.
1. A planet with it’s gravity field would draw then in.
2. A near miss with an O’Neil colony is a miss, while an impact on a planet can still destroy a colony even if it misses it.
3. An O’Neil colony for the same living area would offer a smaller target to an asteroid.
4. To deflect an asteroid from a planet you would have to pay the extra cost of getting out of the gravity well meaning a slower intercept time, meaning less time to deflect it.
5. An O’Neil colony can try to dodge the larger asteroids.
Of course an O’Neil colony would cost more to build since you would have to build the outer shell for it and that would take a lot of material.
The best thing about O’Neil colonies would be that in a few hundred years when we have matured the technology we need to make them safe they are mobile. An O’Neil colony full of astronomers might decide to take off for the sun’s gravitational focus for example to study some interesting star or galaxy. And when they get tired of that, move to a different spot.
Comparing the earth to O’Neil colonies isn’t valid for such discussions. For one thing there is no other place like the earth in our solar system. And even should we terraform another planet it just isn’t going to be like the earth. It will have a different gravity (which might be very important to human health). It will have a different atmosphere and climate and weather and seasons.
The only valid comparison is whether an “artificial” environment in space or on a planet is better. And in general a planet will have all the draw backs of an O’Neil colony, plus a few more. Other than initial cost.
I see, interesting @ Eniac.
Ronald brings up a good point about economics.
I’d like to wrap this up by saying, great discussion!