aurora

dale skran 400
By Dale Skran
NSS Chief Operating Officer

Category: Fiction
Reviewed by: Dale Skran
Title: Aurora
Author: Kim Stanley Robinson
Format: Hardcover/Paperback/Kindle
Pages: 480
Publisher: Orbit
Date: July 2015
Retail Price: $19.99/$19.99/$5.99
ISBN: 978-0316098106
Find this book

In 2015 Kim Stanley Robinson produced Aurora, his “just-so” tale designed to show that interstellar settlement is not possible. Robinson had long been beloved among space settlement advocates for his Red Mars trilogy, but with Aurora he changed sides and become actively hostile to space settlement. I thought it would be interesting to review Aurora not by a rehash of the plot, but by re-envisioning how a trip to Tau Ceti might actually be successful, and in the process correct some of the potted problems Robinson has stuffed the plot with.

Firstly, Robinson envisions two ships, each with about 2,000 crew, flying to Tau Ceti on a generation ship. One is lost fairly early into the voyage, but the second reaches Tau Ceti, although a bit worse for wear. There is an old saying about redundancy “three is two, two is one, and one is none.” The absolute minimum fleet for such a voyage intuitively is three ships, each staffed at 2/3 of capacity, allowing for the evacuation of a failing ship to the other two.

The ship in Aurora is a double-ringed 0.83G version of the Stanford Torus, split, rather like Biosphere 2, into a large number of biomes. This design seems a bit extravagant, but makes for an entertaining story as a major character tours the various biomes early in the book on a coming of age “wanderjahr.” The ships appear to have been purpose-built for the voyage (i.e. not existing settlements retrofited for the jouney), and although the crew is described as being descended from a carefully selected group of volunteers, over time resentment of the voyage has grown.

Since we are redesigning the voyage to actually work, let’s build three ships with double rings, but a bit larger so that they have a capacity of 10,000 crew each, but sticking to the biome plan, and compensating by making the voyage longer and slower.  To this core fleet of purpose-built ships crewed by selected volunteers, we would add six more ships with double rings, two sets of three, each with a capacity of 10,000 crew. Note that by using a double ring, the torus can be half the size of that in the NASA-Ames 1975 summer study, allowing for more economical construction. But the additional six would be existing space settlements that are joining the voyage, and have already been in operation for at least decades.  Although total fleet capacity is 90,000 crew, everything is staffed at 2/3 capacity to allow for the total loss of three ships of the fleet with full evacuation.  Thus, 60,000 settlers would attempt the voyage.

This number, 60,000, is well beyond the needed genetic minimum, and allows for significantly greater genetic and occupational diversity than two ships of 2,000 each. 60,000 people is a substantial number. You would always be meeting new people, and the possibility exists of exchanges between ships at regular intervals. With nine total ships, the cultural diversity would also be significantly greater. Everything is bigger, more redundant, and more likely to be stable over time. It has been pointed out that the size of the biomes Robinson describes in Aurora are insufficient to support the wilderness species he describes, but with more than 3x the biome space in the re-envisioned fleet, they might be successful as ecologies.

The voyage starts in 2545 C.E., more than 400 years in the future. This may or may not be enough time to get ready for such a voyage. A key assumption I would make is that none of the ships are using untested new technology or ecology, but are relying on systems in widespread usage in 1,000s of orbital settlements in the solar system. Near the end of the novel, it becomes clear that although people have settled many locations in the solar system, it would be an exaggeration to suggest that the eco-tech needed for thriving in space has been achieved in Robinson’s fictional future world. Also, with a larger expedition, the population seems less likely to feel condemned to a voyage they did not choose. I grew up in Michigan, and moved as adult to New Jersey, where I had a family. Did my children feel ‘trapped” in New Jersey? Was I ethically wrong because I exposed them to risks present in New Jersey — e.g. coastal storms — that are not present in Michigan? Few would argue that I did anything wrong in doing this. Of course, New Jersey has a population of millions, not 60K, and they have the option of moving to nearby states such as Pennsylvania, which my son has done. It’s not the same, but there is a point at which the fleet is so large that you aren’t isolating the “crew” to any significant degree. With virtual reality technology, which plays no part in Aurora, an even greater sense of possibility might be created.

During the voyage, the ship suffers greatly from the “island effect” which Robinson seems to have misunderstood and exaggerated. The “island effect” does not mean that all characteristics of life on an island are reduced in quality, i.e. animals become less intelligent, live shorter lives, etc. Of course, it is true that bacteria and viruses evolve faster than large animals, but this is true — and problematic — on the Earth as well, as in Covid-19. All the “island effect”  means is that if resources are limited, the physical size of animals will become smaller. In a well-planned interstellar voyage, there should be no lack of food which might otherwise lead to this effect, and even if there were, genetic engineering technology could be used to mitigate the impact of the effect. Although Aurora technology is from 400 years in the future, genetic engineering technology seems only modestly advanced over our time, and plays little role in the story, which seems unlikely. The crew in Aurora are portrayed as resenting limits on childbearing, and this is certainly a potential issue, although given our current difficulties with falling birthrates in developing countries, the real issue might be inducing the crew to have kids at all.

Robinson just seems to assume that whatever control over breeding is exercised by the “ships councils” it is not effective in overcoming regression toward the mean, with the result that the no-doubt originally super-capable crew becomes less intelligent over time. This is clearly an area where even very modest improvements in our current technology would have made a big impact on the crew.  There is also a serious misunderstanding of what regression toward the mean does. If we always evolved back to some primal mean, evolution would not occur, yet clearly it does. In fact, given the isolation of the crew from the Earth and the no-doubt rigorous selection process, a strong “founder effect” would lead the crew to evolve toward a new mean, not the Earthly mean. Of course, if every crew member was selected to have an IQ of 150, after a few generations, the average IQ would not be 150, but neither would be it be 100. Instead, it would verge toward the actual mean of the crew population and their ancestors, perhaps an IQ of 130. Alternatively, a crew ideologically motivated toward eugenics might impact these issues via old-fashioned approaches to breeding. Sexual selection is a powerful evolutionary force, and only a few generations of ideologically motivated mating would have a profound effect on the crew.

Another odd note in Aurora comes from the role the AI “Ship” plays as it struggles to create a narrative of their voyage. “Ship” seems only modestly advanced over the Large Language Models (LLMs) of 2024, but we can forgive Robinson for not fully predicting how rapid the progress of AI would be. It also seems quite unlikely that only the character Devi would have a deep relationship with “Ship” over the decades of the voyage as Robinson suggests. At one point later in the book, someone regrets that there is no way to know what a deceased character, Devi, might have said about a new issue. Even using current LLM technology, one could readily create a decent simulacrum of what someone might say. With 400 years of progress in AI, and a complete record of everything a person said over their entire life, the avatar would most probably be hard to distinguish from the original. In this regard, Robinson joins a long-standing SF tradition of greatly underestimating progress in computers and AI. This outage is on a par with Heinlein characters using slide rules centuries in the future or the paper tape data input in H. Beam Piper’s The Cosmic Computer.

The crew encounter various other issues — complex failure scenarios, and a lack of sufficient phosphorus — that seem inevitable in such an enterprise. A larger fleet with more redundancy and larger margins seems like a key factor in overcoming such issues.

New problems arise once the crew reach their destination and begin to settle a tidally locked but windswept moon in the Tau Ceti system. As the settlers struggle with various issues, the reader wonders why they are so focused on settling a moon they know little about. O’Neill’s vision is that long before they moved down to a planet or Moon, the settlers would manufacture a significant number of orbital settlements in the Tau Ceti system. On such orbitals, there would be no concern with potential contamination or the vagaries of wind. A small scientific base might be established on a moon or planet, but settlers would move there only over a long period of time after conditions were well understood. Certainly, a major deficiency of the crew’s efforts lies in a failure to start by spending a decade or two exploring the Tau Ceti system, perhaps leaving small bases in several locations.

Now the trouble really starts when a “prion-like” alien life-form wipes out the settlers on the ground and leads to internal conflict within the crew. Robinson makes a simple-minded case that worlds are either sterile and will take a long time to terraform, or “live” and deadly, suggesting this is the answer to the Fermi paradox. He also has the crew failing to understand that an oxygen atmosphere is a sign of life. The ecology of the moon seems too simple — one organism? — to be plausible, but as SF dangers go, this is certainly the kind of thing you might encounter.

The problem lies in the inability of the crew to deal effectively with real challenges. Setting aside Robinson’s misleading and almost certainly false speculations about “co-devolution,” the author is on to something. We don’t have any idea how to govern a generation ship such that the organization at the far end will be able to deal with the challenges of settling a new solar system. The confusion and degraded problem-solving capabilities of the crew are at least plausible. One direction of solution lies in sending not a ship, but a society, that is less likely to degrade over time. But there is another solution, which shows up in the plot a bit later and invalidates Robinson’s notion that interstellar travel by humans is impossible.

The crew is unable to decide to “stay or go” and fall to fighting among themselves. Things are only made right by the “Ship” taking control as a benevolent dictator and enabling the vehicle to be split in two to support one faction returning to Earth while another attempts to survive in the Tau Ceti system. There is an element of socialist fantasy here, with a magically smart AI saving humans from themselves as the ultimate technocratic ruler. Robinson is certainly right that the problem of effective governance cannot be regarded as solved, nor is there a well understood solution for how to govern generation ships. However, let us hope that AI taking over is not the only possible direction, since it may end up being the human future on the Earth as well.

On some level Robinson’s story is logically incoherent. The settlers remaining in the Tau Ceti system can perform large construction tasks in splitting the ship and mining the system for resources to fly back to the Earth, but are unable to build enough biome space to overcome the “co-devolution problem” — assuming this is a real issue. This is just Robinson ignoring the obvious – it would be possible for the settlers to build more orbitals in the Tau Ceti system to the point that the biomes were stable and human life would thrive. There is no need to terraform anything, although that might be an interesting long-term hobby for the settlers.

As the “Backers” return back toward the Earth, the ship starts to break down, literally being consumed by living organisms that have evolved during the voyage. In Robinson’s telling, the ship seems remarkably fragile in design, lacking the ability to, for example, have two circuits to supply power to key components, such that key systems must be turned off to effect repairs. Also, the expected amount of repair for such a long voyage seems to been underestimated by the designers. Technology tends to wear out, and even a modest construct such as my home needs a remarkable amount of on-going maintenance to keep things operating smoothly. Given the radiation environment, there are likely to be issues with new life forms evolving enroute. But we have issues all the time on the Earth with new life forms evolving and attacking humans. Starting in 2020 a million plus died in just the US from Covid-19. Now a new bird flu looms. We are in a continuous state of warfare with everything around us, and everything inside us, not living in some perfect “garden of Eden” that deep-green types seem to imagine. 

Robinson is certainly correct that the on-going evolution of microscopic life on a generation ship will be a continuous challenge, and may doom the voyage. Success can only be achieved by the crew rising to the challenge. This suggests both a larger crew — there is only so much 2,000 people can do — but also that much more of the ship’s resources need to be devoted to science and medicine than Robinson envisions. In fact, much of the crew in Aurora seems to be doing agricultural stoop work or routine maintenance. There are at most a small number of competent technical leaders, and they lack both the numbers and the laboratories to mount a meaningful response to any crisis. With my proposed crew of 60,000, the fleet of 12 ships could include all the resources of a large research university and corresponding research medical center, preferably spread over at least three ships for redundancy. Of course, along with the crew of 60,000, we could send an extensive array of AI-managed laboratory equipment that would enable the fleet to attack new issues with far more firepower than just a larger crew of scientists. For a classic SF example, consider Van Vogt’s Voyage of the Space Beagle with its vast numbers of chemists, physicists, biologists, etc. as they face off with the space creatures that inspired Ridley Scott’s Alien.

In Aurora, a continuous feed of data arrives from the Earth, and apparently reports are sent back, but with one major exception this has little impact on the story. Even with many years to receive a response, a large support team back on the Earth could provide real help to the distant voyagers. This type of exchange would not be personal, but professional and academic, with each side doing research and writing reports that would provide benefit far in the future. Over the course of Aurora, there is exactly one point where information from the Earth suddenly becomes useful. As they return, conditions deteriorate rapidly, leading to mass starvation and ecosystem collapse. In a writers’ trick of deus ex machina, Earth suddenly invents an effective hibernation technology, and the crew applies it to return the remaining 700 or so still living all the way to the Earth, while leaving “Ship” to manage things.  

The big problem with this plot gimmick is that it is a solution that allows at least voyages of centuries to nearby star systems by the original crew, neatly circumventing our inability to manage multi-generational political change as well as potential long-term ecosystem collapse. If the original settlers had developed this technology first — maybe by waiting even 100 years — the voyage to Tau Ceti would have been much more likely to succeed. A larger crew, fresh to the fight, and not at all regressed to the mean, would have been more ready to take on whatever challenges Tau Ceti offered. Hibernation also supports a range of “deep time” projects, including terraforming, by allowing an original, highly-motivated crew to complete an objective over thousands of years.

And hibernation is not some magical technology like faster than light travel — many animals have extensive hibernation abilities, with some, like tardigrades, bordering on the fantastic. In time, we almost certainly will develop technologies to allow humans to sleep away the years between the stars. Even the radiation resistance problem has non-magical solutions — artificial magnetic shields and the usage of biological techniques to increase radiation resistance — which could be well tested on orbital space settlements inside the solar system.

Much of the latter part of Aurora is taken up with the Ship’s efforts to slow down and return the diminishing number of crew members to the Earth. This is accomplished via the usage of the main fusion engines to decelerate, followed by an inadequate laser illumination from the Earth, and lastly by an elaborate series of orbital aerobraking maneuvers. I have seen articles which suggest that the orbital maneuvers would not really work, but this only reinforces Robinson’s point that decelerating from 10% of the speed of light is not easy.

The potted aspect of this is the revelation that the starships have been launched by wealthy inhabitants of the Saturn system who were in the process of terraforming Titan, but sent the starships off without any real reason to believe that their ecosystems could be maintained for centuries. The space dwellers are described as returning to the Earth for “sabbaticals” every ten years that magically prevent the co-devolution from occurring. The main prerequisite for a serious run at interstellar settlement is successful settlement of the solar system, and if people must return to the Earth periodically to survive, this cannot be viewed as characteristic of successful space settlement. Thus, Robinson’s would-be interstellar space settlers are some combination of self-deceptive and incompetent.

As should be expected in a potted “just-so” tale, all ends badly for anyone who wants to settle space. The “stayers” in Tau Ceti go silent after 28 years, and are presumed to have all died due to ecosystem collapse or some unknown disaster. The Ship drops the surviving crew off on the Earth, but is unable to slow down itself, and burns up in the sun. Robinson devotes a lavish amount of text of detailing the crew’s difficulties in re-adjusting to the Earth, with a continuous drip of deaths for one reason or another. The high point, or more accurately the low point, comes when a delegation of the crew is brought to a space colonization conference and briefed on new plans to launch more starships, now equipped with hibernation to overcome some of the issues described above. The leader of the crew, Freya, starts punching the main speaker in a fit of rage, shouting that interstellar settlement will never succeed. This is a bit like having Greta Thunberg attend the NSS International Space Development Conference, and run up on stage to punch Gerard K. O’Neill or Elon Musk. That it might be a plausible human reaction in the context of the story does not change its propagandistic nature. To really turn the knife, Robinson describes the pro-starship speakers as being “white men,” suggesting that no people of color support space settlement or interstellar travel. I guess Mae Jemison, the Principal of the 100 Year Starship organization (https://100yss.org/mission/team) is a fictional character.

Robinson clearly loves researching and writing about technology, often to the detriment of telling an engaging story. Even his concluding description of the surviving crew members joining groups rebuilding post-climate change beaches on the Earth is an ode to human ambition and technology more than an example of ecological balance. But Robinson’s notion that humans are forever restricted to the Earth by some de-evolution phenomenon that seems more magical than scientific in Aurora must be regarded on the same level as clerics saying we could not sail around the world due to the presence of dragons in the ocean. This does not make interstellar travel easy, just not impossible. Although I strongly advocate for the immediate settlement of the solar system, I do not support premature efforts at interstellar settlement and colonization. First, we need to really understand and implement on an engineering and a human level rotating space settlements in the solar system. When generations have lived happily in orbital settlements, with virtually no one making routine trips back to the Earth, and 99% plus of all goods used in space are being sourced off the Earth, then we can start building starships. If you want to learn more, a good place to start is Milestone 30 of the NSS Roadmap to Space Settlement. To expand your knowledge of interstellar flight, check out NSS President Isaac Arthur’s Science and Futurism YouTube channel’s playlist on the subject of generation ships.

See also these other critiques of Aurora:

© 2025 Dale Skran

NSS index of over 500 book reviews

Share:

Facebook
Twitter
Pinterest
LinkedIn
Picture of National Space Society

National Space Society

1 thought on “Aurora Re-envisioned: An Essay/Book Review”

  1. (Humor mode) Ah! So producing a space faring society _is_ rocket science.(Exit humor) Good to know people smarter than me are working on it. Maybe Robinson was writing from the perspective that a guy like me would really mess up the works, if put in charge of such an undertaking.

    Reply

Leave a Comment

Search
Categories
future 1

Don't Miss a Beat!

Be the first to know when new articles are posted!

Follow Us On Social Media

JOIN THE
GREATEST ADVENTURE

Give The Gift Of Space: Membership For Friends and Family

Book Review

Archives

ISDC 2025:
Together Beyond!

In Orlando at the Rosen Centre Hotel.
June 19 - 22, 2025

FEATURED BLOG

Image of Kalpana One space settlement courtesy Bryan Versteeg, spacehabs.com $32,000 in Cash Awards Given for Best Space-Related Business Plans — Deadline March 1, 2024

Category: Nonfiction Reviewed by: John J. Vester Title: Nuclear Rockets: To the Moon and Mars Author: Manfred “Dutch” von Ehrenfried Format: Paperback/Kindle Pages: 270 Publisher:

Partially Successful Flight Reached Space and Demonstrated New “Hot Staging” System The National Space Society congratulates SpaceX on the second test of its Starship/Super Heavy

Ad Astra, the NSS quarterly print, digital, and audio magazine, has won a 2023 MARCOM Gold Award. The awards are given yearly for “Excellence in

By Jennifer Muntz, NSS Member Coordinator On October 10th, an inspiring breakfast event took flight at the Center for Space Education at the Kennedy Space

By Grant Henriksen NSS Policy Committee Benefit sharing is a concept that refers to the distribution of benefits derived from the exploration and use of

People residing and working in space, space settlements, or on long-duration space flights will need to produce infrastructures and food to maintain healthy lifestyles. The

Image: Artist’s concept of the Blue Moon lander. Credit: Blue Origin. Second Human Landing System Contract Encourages Competition and Innovation The National Space Society congratulates

Your Doorway to New Worlds