The sun is high, slightly after midday, as a group of men look southwest over the savannah from a ridge of stone in Tanzania, with their backs to the edge of the Great Rift Valley. They are just west of salty Lake Eyasi, which is getting larger, along with the valley itself, at a rate of one inch per year. The valley will become a flooded extension of the Red Sea sometime within the next 10 million years, due to the slow, inevitable action of plate tectonics that constantly shifts the geography of Earth. The Somali Plate will continue to distance itself from the Nubian Plate, effectively tearing this southeastern portion of land away from the continent of Africa.
In the meantime, these men are here to hunt game with the longbows they carry with them, and the quivers of arrows slung over their backs. They have climbed this hill to survey for signs of life below. The hunters are members of the Hadza, an indigenous people who have lived in this part of the world for tens of thousands of years. A little less than one-third of the 1,200 or so remaining Hadza still live an entirely traditional “hunter-gatherer” lifestyle that may be as old as the early human ancestor fossils found in Olduvai Gorge. Just thirty miles away, on the other side of the Lake Eyasi, the gorge has held the fossilized bones of Homo habilis for nearly 2 million years.
The photo of the men on the ridge, taken by anthropologist Brian Wood in 2005, was the opening image in a slideshow talk given by George Sowers for the Space Studies Institute in 2019, at a conference held at the Museum of Flight, outside of Seattle.1 The institute is one of the heirs to the L5 Society, the organization that was formed to help advance and promote the space settlement ideas of Gerard O’Neill. Sowers—former vice president and chief scientist at United Launch Alliance (ULA)—became a professor of practice at the Colorado School of Mines in 2017, helping to launch their new degree program in space resources, where students are being prepared to imagine how humans might one day mine asteroids and exploit the resources of the Moon.
He showed this image, of some of the last hunter-gatherers on Earth, to make a point about the next era of life in outer space. In this presentation, and in an earlier version of it that he gave to a House of Representatives Committee on Science, Space and Technology,2 Sowers wanted to advocate on behalf of a worldview that linked human prosperity to resource and energy access, and thereby to argue in favor of access to resources in space. Along these lines, he used “energy capture” as an indicator for “economic wealth.”
Human existence, for about the first 100,000 years or so, he argued, had depended on the kind of lifestyle still practiced by the Hadza: hunting game like the men on the ridge were doing, and foraging through a loosely bounded territorial region. This practice provided them with access to a certain amount of calories per person, per day. That amount roughly doubled during the “Agricultural Revolution,” when people arrived at the understanding that plants and animals could be purposely cultivated alongside human settlements, allowing for different kinds of societies with division of labor inside them. And then energy access roughly doubled again, Sowers said, during the “Industrial Revolution,” as humans figured out that some of the material they were mining underneath of the planet could be burned for fuel, unleashing unused energy—which had lain dormant for hundreds of millions of years—to power a meteoric rise in both population and potential energy access. For both of these metrics, the graphs Sowers showed are going up and to the right.
Sowers is making a point about successive revolutions in general, and he tells Congress, and the Space Studies Institute conference, explicitly that access to space resources will facilitate the “third major economic revolution in human history.” In an echo of O’Neill’s testimony forty-five years previously, he presents mining the Moon as, once again, the first step in a path that will lead to permanent human inhabitation of space, and to a permanent expansion of the human capacity to access and use ever more resources and energy.
Sowers, as a former leader of a legacy space company, is a person with roots in what some might call Old Space. ULA formed in 2005; it was a joint venture of the space launch divisions of Boeing and Lockheed Martin—the two largest and longest-standing providers of orbital launch vehicles in the world. In the aughts, the two companies realized that there were not enough launch customers for them to compete over, so they sought, and got, approval for a merger.
What they didn’t know at the time was that a then small company called SpaceX, whose lawsuit over the merger was dismissed almost immediately, was about to demonstrate capabilities that could open up space access to all kinds of new players. Over the next decade, SpaceX’s reusable rocket technology drastically lowered launch costs. The price to send an astronaut to Earth orbit via SpaceX’s Dragon 2 capsule, atop their Falcon 9 rocket in 2020, is one-seventh as expensive as it was with the Apollo capsule and the Saturn V. The Dragon 2 price is one-third that of the Space Shuttle orbiter, and just over half the cost of a ride in ULA member Boeing’s Starliner capsule. As of this writing, the Starliner is still in testing phases, and has not yet launched with people onboard; Dragon 2 did it successfully twice in 2020.
Launch costs are going down, while rockets are going up, and even more of that will happen in the near future, when Blue Origin’s New Glenn rocket comes online later this decade. These two companies, and the whole ecosystem of secondary providers and startups that have grown up around them, are collectively forming a new generation of private spaceflight models. Nevertheless, many of the worldviews they draw on as they’re doing that are grounded in the past.
Space Origin and BlueX
These are maybe the only true things you can say about both of the two most well-known “NewSpace” companies—SpaceX and Blue Origin—and their founders, Elon Musk and Jeff Bezos, respectively: They are both large private spaceflight companies, well funded by two billionaires who both made their fortunes in online commerce. Both founders created their rocket companies to pursue lifelong dreams. Their near-term objective is to make reusable space vehicles and to lower launch costs, and they have a track record of success where this is concerned. But their long-term goal is to open up access to space for the future of humanity. They were also both very lucky to have been around when Space Shuttle launches stopped in 2011. During the COVID-19 pandemic, both billionaire founders got even wealthier, and at this writing, Elon Musk and Jeff Bezos are the two richest individuals on Earth. SpaceX and Blue Origin seem to have a lot in common, and they are real and imagined rivals with one another for both customers, capabilities, and cosmic Common Tasks.
That’s where the similarities end, though. The origins of the specific planetary imaginations of their founders, and the methodologies the two companies use to pursue those imaginations, could not be more different.
Bezos was a science fiction fan growing up. As a kid, he spent time every summer at his grandfather’s ranch in West Texas, where the local library had an extensive collection of science and science fiction books. This is probably where he encountered O’Neill’s work about the future of technology, business, and humans in space. Bezos’s high school graduation speech in 1982 was basically about the O’Neill paradigm. The Miami Herald published excerpts from valedictorian speeches that year, including Bezos’s, noting his desire to build “space hotels, amusement parks, yachts and colonies for two or three million people orbiting around the earth.”3
The Herald also highlighted Bezos’s hope that this scenario might help turn Earth into something like a national park, a concept straight out of O’Neill’s books, and one reflected in talks Jesse Strickland would have been giving in the Midwest around the same time. Today, nearly forty years later, Blue Origin’s vision statement, and their unofficial motto, reads like it could be an excerpt from that speech:
Blue’s vision is a future where millions of people are living and working in space. In order to preserve Earth, our home, for our grandchildren’s grandchildren, we must go to space to tap its unlimited resources and energy. If we can lower the cost of access to space with reusable launch vehicles, we can all enable this dynamic future for humanity.4
Blue Origin started in a kind of stealth mode, funded entirely by the fortune Bezos amassed as the founder, president, and CEO of online retailer Amazon. Under Bezos’s direction, Blue Origin spent several of its early years brainstorming and researching alternative methods for launching payloads and people off of Earth, with science fiction author and long-time Bezos advisor Neal Stephenson as the company’s first employee. They studied non-rocket systems, including variations on O’Neill’s mass driver, and more baroque proposals like a giant whip made of chain-links the size of railroad cars, and a spinning scooper that could swoop down from orbit to pick up and boost a payload from a plane.
All of this took place before the company’s existence or purpose was even acknowledged publicly. In the end, they arrived back where they started. Coming to the same conclusion Konstantin Tsiolkovsky had reached more than a century earlier, they decided that a multistage chemical rocket—which uses the combustion energy released when a propellant meets an oxidizer to create a directed, controlled explosion—was the simplest place to start. Bezos returned to West Texas and secretly bought up hundreds of acres of property so that he and his growing company could safely, and secretly, refine their ability to blow things up and launch them.
Musk, like Bezos, was suffering from disillusionment in NASA’s ability to keep advancing the long-term goals in human spaceflight that he thought were important. For Bezos, this void meant more steps needed to be taken to fill out the plan laid out by O’Neill, “a future where millions of people are living and working in space.” Musk, however, had a different overarching goal, as he put it in the title of a 2016 talk he gave to the International Astronomical Congress, “Making Humans a Multiplanetary Species.”5
Bezos, following the O’Neill paradigm, doesn’t seem to care about living on new planets; they are, after all, not the best surfaces for the expansion of industrial civilizations. In this worldview, we humans need more room to grow and live, so we need more surfaces. And so, we mine material to make them. But as we do this, we realize we don’t want to mine where we live, and we don’t want to live where we mine. Musk has a different set of things he’s worried about, mostly relating to human extinction. The origins of Musk’s “multiplanetary” ambitions are about getting away from existential threats to human futures. Eventually, Musk told the International Astronautical Congress (IAC), some kind of “doomsday event”—plague, super-volcano, asteroid, war, or runaway technological singularity—will occur on Earth, so some humans should be elsewhere when that event occurs. Mars, he says, is the best option for being that elsewhere.
Musk traces the development of his own ambitions in private spaceflight to a certain moment in which he saw something in his own planetary imagination: a vision of plants growing on Mars. Musk hoped that if he could independently launch a small greenhouse to Mars, and if that greenhouse contained systems that allowed it to broadcast footage of life from Earth extending out to the furthest point it had yet been, that it would inspire more investment in space exploration—thereby advancing the plan to make Mars a backup planet for human life. After spending some time researching launch costs, and even trying to buy his own rockets in Russia during the early 2000s, he condensed his own origin story (also rooted in online commerce) and ambitions into an elevator pitch, which he used when calling people in the aerospace industry out of the blue. One friend of Musk’s still claims to remember it verbatim:
I’m Elon Musk, I’m an internet billionaire, I founded PayPal and X.com. I sold X.com to Compaq for 165 million dollars in cash and I could spend the rest of my life on a beach drinking Mai-Tais, but I decided that humanity needs to become a multi-planetary species to survive and I want to do something with my money to show that humanity can do that and I need Russian rockets and that’s why I’m calling you.6
Musk was working in the growing space between NASA’s signaled capabilities and its actual technical capacity in the post-Apollo era. He wanted to help them along with his own signal and image, the television broadcast of plants growing on Mars. This is typical of his approach: to get something done, start with the image and work backward. To make the image happen, certain technologies and gadgets are needed, in this case a rocket. First, attempt to apply money directly to the problem—if the gadget exists, try to buy one. If that turns out to be unfeasible, as it was in his efforts to buy a Russian rocket, then apply expertise and make the gadget yourself.
And the company he founded, in order to make his own rockets, is SpaceX. Over the next several decades, he made many successful images with his rockets, and he made them in public, usually via livestreamed webcasts. In 2008, SpaceX streamed its first successful flight to orbit, showing images of the launch, and images from the upper stage itself as it reached space. It was the first privately funded and privately designed rocket to do so, literally from the ground up.
In 2015 SpaceX and Musk made history again, successfully landing an orbital rocket’s first stage, upright, like a spaceship that Chesley Bonestell might have painted for Wernher von Braun in the 1950s. Only a month earlier, Blue Origin had achieved similar success, launching their New Shepard rocket on a simpler ballistic up-and-down flight to the edge of space, and landing the first stage vertically. SpaceX had achieved their landing after the much more difficult and dangerous task of reaching orbit, and, whether it succeeded or failed, viewers of the livestream would know the results of the landing test immediately. The footage that SpaceX streamed was dramatic and risky, like a giant robotic gymnast, the automated systems on the rocket stuck the landing perfectly, restarting its engines at seemingly the last second after a long free fall. Characteristically, Blue Origin had done their test in secret, away from public view, only announcing it on the internet a day later, in a slickly produced video with a soundtrack.7
Stacks and Gadgets
Bezos’s way of working—quietly, sometimes secretly, constructing whole systems from first principles, step by step—is similarly visible in the development of Amazon’s structure. Amazon builds or absorbs platforms; then, after integrating them, it either rents them out to others (if it sees an advantage in doing so), or keeps them to itself (if that is more worth its while). One example of this process can be seen in Amazon Web Services, their cloud computing subsidiary. AWS descended from a suite of technologies developed in-house to support the growing needs of their online marketplace to coordinate information while remaining distributed across several continents.
An internal paper from some company engineers suggested that this suite could be reorganized so that it could be, among other things, leasable to third parties who also needed these same kinds of capabilities. The result has been a paradigm shift in which customers of Amazon Web Services include not only clients, but competitors such as Netflix, and government agencies, including NASA and the CIA. This is the same structure that Amazon Marketplace uses, in which their existing e-commerce platform is leased to others who can create their own online store-within-the-store, for a price and a cut.
This is a variant on the “full stack” approach to information technology. Development is full stack if it encompasses not only the design and ownership of the “front end” customer and client facing interfaces, but also the “back end” set of software and servers that exist in the infrastructure, and every other layer in between. Amazon started with books and expanded from there, reaching out and absorbing, controlling, or recreating every step of the supply chain that a book, or by extension any item, would go through on its way from producer to consumer. And then, once their stack was built out, they realized further profit by leasing some layers out to others.8
If I were to order a copy of this book from Amazon.com, I would visit a website hosted in an Amazon server, running AWS cloud computing software. Given the need to optimize the transaction for speed and smoothness, this would be connected to hardware as near as possible to my location in Northwest Baltimore. Once I clicked “Buy Now,” a worker in a local Amazon warehouse, probably the one located just outside Southeast Baltimore, on the site of a former steel mill, would get a notification. A robot, built by Amazon Robotics, would also mobilize, picking up a shelf of items that includes the book and carrying it to a picking station staffed by a human worker. Before 2012 this robot would have been made by Kiva Systems; after that year Amazon bought the company and brought all of its operations in-house. The reason was simple: Kiva worked so well that they didn’t want anyone else using it. The book would then go to another worker, who would receive a printout of the relevant scanning code and instructions about which materials to pack it in. Then the book, in a box or envelope, would sail through the facility on a vast network of conveyor belts, until it arrived at a delivery truck. Until the late 2010s, this might have been a truck from FedEx, UPS, or even the US Post Office. But lately, Amazon has been expanding its logistics program as well.
Now, in Baltimore, almost all packages are delivered in deep blue Amazon vans. But it is not technically an Amazon employee who is placing the package with the smiley face logo on my porch; it is a member of a private contractor team who pays a license fee to use Amazon’s routing software and other systems. Meanwhile, they have used these fleets of branded delivery trucks to expand their mapping and routing capabilities. Some trucks have lidar and cameras on top, using every delivery as an opportunity to make a map of the city that they hope will rival parallel systems built by Apple and Google. If this is the last local copy of the book, stock in the warehouse might be replenished on the next visit from Amazon Air.
Musk’s approach, as visible in the way his other companies operate, is completely different. Where Bezos and Amazon tend to think and act in terms of whole systems, which remain abstract and recede from visibility, Musk thinks about technology in terms of the concrete gadget or artifact. The rocket is part of a family of objects that include the car, the truck, the house, the solar panel, the battery, the satellite, and so on. And under Musk’s influence, his companies tend to make newer, and arguably better, versions of all of these things. Where Bezos takes apart whole industries and whole worlds, putting them back together in ways that suit his company’s needs—and sometimes with disruptive results in the world outside—Musk will simply make a version of the thing that is more efficient, more cost-effective, and cleaner—without requiring any massive change to lifestyles or the status quo. With Tesla, Musk’s automotive and electricity company, we just swap our gasoline-powered cars for electric ones, and change out our asphalt shingles for solar panels. They even mimic the forms of the objects they replace; what homeowner’s association could possibly object? Starlink, his company that provides satellite internet, is just a higher-speed, more prolific option for broadband.
One other point of contrast between his approach to technological world-building and Bezos’s is the way that each figure uses intellectual property. Where Amazon has bought a robotics company in order to acquire their successful IP and keep it from competitors, Tesla has created standards for things like charging stations and then opened up their patents for anyone to use, betting that their standards will become more useful if they are allowed to proliferate. Similarly, its mass transit project, the hyperloop, was written up in several white papers from his companies, and explicitly “open sourced” so that other companies can make their own versions of it. But in Musk’s implementation, run by his Boring Company, the idea of high-speed trains with vacuum and magnets slowly morphed into a scenario where Tesla owners can get to drive their cars in specially designated tunnels, made just for them. The Boring Company’s first test tunnels will connect SpaceX’s factory, the Hawthorne Municipal Airport, and a family home owned by the company. This is a mundane solution to a mundane problem. Musk founded the Boring Company in 2016 after announcing on Twitter: “Traffic is driving me nuts. Am going to build a tunnel boring machine and just start digging.”
Musk’s 2016 talk on “making humans a multiplanetary species” did not specify what the new residents of Mars would do once they got there; his sequence of renderings ended with their landing. There was little mention of what their lifestyle might be like. But if this planetary imagination is sounding a little suburban, with solar-powered single-family homes and private vehicles transplanted through space, that vibe was underscored by Musk’s target price tag for the trip. He had based it, he told the audience, on a very specific benchmark:
If we can get the cost of moving to Mars to be roughly equivalent to a median house price in the US, which is around $200,000, then I think the probability of establishing a self-sustaining civilization is very high. I think it would almost certainly occur. Not everyone wants to go—in fact, I think a relatively small number of people from Earth want to go—but enough would want to go, and who could afford the trip, that it would happen. You keep looking at sponsorship and I think it gets to the point where almost anyone if they saved up, and this was their goal, they could ultimately save enough money to buy a ticket and move to Mars. And Mars would have a labor shortage for a long time, so jobs would not be in short supply.9
In this equation, the capacity to save enough to buy a home is immediately correlated with civilization, which is further linked to the ability to access a new job market.
In a follow-up talk about his Mars plans to the IAC, in September of 2017, Musk elaborated on what life might be like. This Mars, with its underground vacuum tubes for travel, self-driving electric cars, solar power, and even the hint that the leader of this new world might have a very specific and unusual biblical name, matches up well with von Braun’s The Mars Project from 1948. Musk showed a slide with an aerial view of a city, laid out as a rough grid of medium-sized domes, all unalike, connected by tubes to each other, and to a space-port with several of his Starship landers parked there. This rocket is part of SpaceX’s ITS, or Interplanetary Transportation System, with an eventual plan, at a scale worthy of von Braun, to launch fleets of these rockets every two years, when the planets’ alignment makes travel time efficient. Tellingly, the original name for the rocket was the MCT—the Mars Colonial Transporter.
In images from the 1950s that show human settlements on Mars, like those made by Bonestell for Collier’s magazine, the main focus is usually a single large, all-encompassing dome, enclosing space for growing food and sharing public life, in between smaller buildings where people might live and work with more privacy. In these older domed cities, there is a shared risk: one puncture in the dome would require collective effort to fix. But it would also be safer overall than the smaller outbuildings: the larger the size of the air mass inside, the longer it would take before a leak became dangerous. A team would therefore have plenty of opportunity and capacity to fix things problems as they arose.
In Musk’s city, however, there is no center, no implication that it includes public space or any shared community. The individual modules proliferate across the Martian landscape, sprawling in every direction. Musk’s city suggests an every-dome-for-yourself scenario, and it’s easy to imagine that afflicted modules would have to be evacuated and sealed off from the network before they could be fixed.
Musk’s planetary imagination seems to take for granted this tendency toward sprawling individualism, he’s already launched his own Tesla to Mars orbit. We do not need trains or public transportation, his business plans signal, if everyone can be inside their own private self-driving electric car, in a tunnel. Or, even better, Musk’s future customers can be inside the heavily armored Cybertruck. A design object that’s pulled straight out of dystopian science fiction, the recently unveiled truck is designed to be bulletproof. Its hard angles aggressively reject the world outside it, while its medical-grade air filter protects the occupants against pandemic infection or wildfire smoke. This is an object designed for the kind of endof-world scenarios that Musk wants to hedge against with his Mars plans.10
Conflicted, too, is his tendency to technologically normalize existing conditions of sprawl and suburbanism. Musk’s intention to bring down launch costs by adding more launches is a classic example of something economists call “induced demand”: sometimes people don’t know they want something until they have the capability to get it affordably. Electric cars and solar panels make suburban life easier and more convenient, and if traffic gets bad, just build another tunnel. Musk’s companies are already sprawling in space, thanks to the expanded capacity SpaceX enables.
The Starlink satellite internet system that SpaceX is launching will raise connectivity and bandwidth on Earth, especially in rural regions and suburbs inducing more demand for the internet, but it may also start to crowd low Earth orbit. As of this writing, there are over 1,200 Starlink satellites in space, each about the size of a mini-fridge. But SpaceX already has approval, and the demonstrated ability, to launch up to 12,000 of them—a situation that makes those who track collisions in space, like the residents of the International Space Station, a bit nervous. In the future, a cascading chain reaction of satellite collisions might make this region of space unusable. And the small satellites are already interfering with ground-based astronomy, showing up in photographs as bright streaks in the sky.
Musk has a complicated relationship with the concept of induced demand. Even though his plans with SpaceX depend on it, he rejects induced demand when it is used to criticize his ideas about transportation on Earth. When transit experts questioned his car-tunnel plans on this basis, he tweeted back: “Induced demand is one of the most irrational theories I’ve ever heard. Correlation is not causation. If the transport system exceeds public travel needs, there will be very little traffic. I support anything that improves traffic, as this negatively affects almost everyone.”11
This kind of cognitive dissonance is bound up in a world-view that values expansion above all else, and one that seeks salvation in the proliferation of gadgets and gadget-based solutions, hedging bets against a catastrophe that it may also be enabling. Like the attempt to reduce traffic by adding more highway lanes or more tunnels, lowering the opportunity costs for the end of the world might be a strategy that backfires.
If Musk is following von Braun’s paradigm, right down to the imagined existential threat of something like World War III, and the dramatic launch of a fleet to Mars, Bezos is still following steadily in O’Neill’s footsteps. Bezos will, like O’Neill, start with the Moon. At a 2019 press conference in Washington, DC, Bezos revealed another project that Blue Origin has been working on away from public view: a full-scale mockup of its design for a robotic lunar lander cargo vehicle, named Blue Moon.12
Subsequently, NASA announced that it would be working with Blue Origin and Blue Moon to consider the possibility of using the lander as part of its Artemis program to return to the Moon, with a landing date set for some time in 2024. Bezos and NASA both hope to investigate the possibility of “in-situ resource utilization” (ISRU)—that is, the mining of the Moon—to support further developments in human spaceflight. There is oxygen and aluminum bound up in the Lunar regolith, but the most promising molecule here is H2O.
In 2008, an Indian space probe confirmed what had been long suspected, that large deposits of water ice exist on the surface of the Moon. This ice is mostly concentrated in craters in and around the Moon’s south pole, where the Moon’s rotation has left some areas in shadow, hidden from the harsh sunlight, for billions of years. This water is valuable for life, of course, but it is maybe even more valuable, from a certain point of view, for another purpose. With easily obtainable electricity from solar power—some of the mountain peaks in this region are permanently in sun—H2O can be “cracked” to its component molecules of hydrogen and oxygen, which can then be recombined as a propellant and oxidizer to make the controlled combustion that Tsiolkovsky proposed. Water on the Moon can be used for rocket fuel.
Bezos wants to use these lunar resources—rocket fuel and all—to get to the next steps in the O’Neill paradigm, and in Blue Origin’s long-term vision statement, “millions of people living and working in space.” In his 2019 talk, he all but directly addressed Musk’s proposals delivered to the IAC in the previous two years. The title of Bezos’s talk was “Going to Space to Benefit Earth,” and he went out of his way, without naming him, to score points against Musk’s idea that being multiplanetary is a viable backup plan for Earthly catastrophe. “Earth is the best planet. It is not even close,” he said, “We have sent robotic probes to every planet in the solar system—this is the good one.”13
In addition to echoing his frequent disparagement of “plan B” scenarios like Musk’s, Bezos also showed a clip from the 1975 television show in which O’Neill spoke with Isaac Asimov about his ideas. Asimov describes the knowledge gap that he and other science fiction writers had encountered around imagining alternatives to living on planetary surfaces, calling it “planetary chauvinism.”14 The capacity for large-scale heavy industry and the production of new living space was available already, without the need to use other planets—or planetoids like the asteroids and Earth’s Moon—as anything more than a resource base. Bezos also directly echoed O’Neill’s language about stasis and change:
The good news is that if we move out into the solar system, for all practical purposes we have unlimited resources. We get to choose. Do we want stasis and rationing or do we want dynamism and growth? This is an easy choice. We know what we want. We just have to get busy. If we’re out in the solar system, we can have a trillion humans in the solar system, which means we’d have a thousand Mozarts and a thousand Einsteins. This would be an incredible civilization.15
He showed a series of renderings of the interiors of Stanford tori and O’Neill cylinders, contemporary updates of the views that Don Davis and Rick Guidice created for Gerard O’Neill and NASA’s 1975 Summer Study. Unlike those original paintings, though, these images were produced with literal cut-and-paste techniques like those used in Adobe Photoshop editing software. One shows a Brutalist college library next to a red clapboard family farm, with the crops getting watered by a drone.
In the distance, there is a city that looks like Seattle, where the headquarters of Amazon and Blue Origin are located. In another we can see the Duomo from medieval Florence, with Beijing’s Forbidden City visible behind it. Another, with a denser city covered in greenery inside a Stanford torus, mashes up Norman Foster’s Gherkin building from London with Moshe Safdie’s Marina Bay Sands luxury hotel in Singapore. All of this was from past futures, and it somehow looked less optimistic and futuristic than the images made almost half a century ago.16
The behavior and nature of the companies that Jeff Bezos owns and operates offer hints about his worldview and the nature of the planetary imagination that might inform the creation of these new worlds. In the past ten years, Bezos has acquired everything from national media outlets to grocery store chains. Amazon itself was founded with the aspiration to eventually become a store that sells everything.
In 2017, Bezos and Amazon issued a kind of request for qualifications to all of the cities in North America, asking them to demonstrate why they would be the best place for a new second headquarters. The winner would receive all of the attendant jobs and other potential economic benefits that would come along with the presence of one of the country’s largest technology companies. This so-called HQ2 initiative mobilized municipalities to offer Amazon unprecedented tax incentives for the chance to host them. In the end, the winner was in Northern Virginia, not far from one of Bezos’s primary residences and the offices of the Washington Post, which Bezos bought in 2013. Meanwhile, back at HQ1 in Seattle, Bezos built “the Spheres,” a set of spaces for Amazon employees to work in, take meetings at, and relax with. The Spheres are filled with exotic plants, and double as a conservatory and research complex, preserving and studying ecosystems and their interaction. This is a project not entirely unlike Biosphere 2, an attempt, at the end of the Cold War, to recreate as many of Earth’s biomes as possible, under a sealed skin. Biosphere 2 was financed by another billionaire interested in space settlement, Texas oil fortune heir Ed Bass.
The full-stack approach visible in Amazon and elsewhere might only be a rehearsal. In the suite of companies Bezos operates, and in the many layers of platforms that form the strata of Amazon itself, we can see experiments in world-building. These forays into logistics, software, information and journalism, urbanism, and ecology seem to signal his intentions regarding the pieces, and the ways in which they will be put together. Unlike Musk’s collection of disparate, individualistic domes sprawling across Mars, Bezos’s Stanford tori and O’Neill cylinders will each be a unified environment, all run from one central guiding authority at every level, from infrastructure to superstructure. The copy/pasted collage cities in the renderings that Bezos showed in 2019 would be company towns. HQ3 and HQ4, and on and on, in space. These places would have all of the services, amenities, and products and platforms provided by Amazon, or by some future descendant and heir to its most potent capability: the capacity to absorb everything that it touches and to turn it all to its own purposes.
The Spheres sit in the shadow of Amazon’s flagship office building, a thirty-seven-story slab tower named “Day 1.” The name, which Bezos has applied to every Amazon building in which he’s had a personal office, is a reference to a frequent touchstone of his, reaching back to Amazon’s days as a tech startup. It captures the ambition and motivation that drove the company then and now. In a 2017 letter to Amazon shareholders, Bezos wrote, “Day 2 is stasis. Followed by irrelevance. Followed by excruciating, painful decline. Followed by death. And that is why it is always Day 1.”17
It’s easy to imagine that Jeff Bezos found a kindred spirit in, and was subsequently influenced by, Gerard O’Neill, who also feared stasis more than even threats to personal safety. This spirit of change and development is also captured in Blue Origin’s official motto: “Gradatim Ferociter,” translated as “Step by Step, Ferociously.” But this equation, where the process of building new worlds is linked to a fear of death, while also connected to the fierce developmental experimentation of startup culture, is dangerous. What would it mean to live in a world that was put together, at every level, on the principle that each day was Day One?
Mining the Sky
Musk’s and Bezos’s e-commerce fortunes served as the private investments that enabled their spaceflight ventures, sometimes supplemented by other funding from private capital, and sometimes reinforced by contracts and grants from NASA. These projects represented a limited amount of personal risk; after all, both founders still owned many other companies that might continue to be profitable if their ambitions to create orbital launch vehicles had failed to materialize. So despite the cosmic scale of their overall plans, they would probably still be comfortable, even if they had never made a rocket that flew, much less one that landed.
But Blue Origin and SpaceX were not the only NewSpace companies out there, and their subsequent success enabled a whole new constellation of other startups to take further risks. Like the railroads from almost two centuries ago, these two companies have built relatively affordable infrastructure that can then be used by others to perform all kinds of potentially profitable activities.
One of those activities, some entrepreneurs hoped in the mid 2010s, would be to mine the asteroids. The logic of this kind of in-situ resource utilization had been obvious even to Tsiolkovsky. The protagonists of his 1920 novel Beyond the Planet Earth see the benefits of building new habitats from material already found in space, on the Moon and in the asteroids. And asteroid mining was an important phase in the speculative timeline developed by Gerard O’Neill and NASA Ames, and illustrated by Rick Guidice. The large rotating habitats that formed the most charismatic component of their scenarios all depended on asteroid mining for their existence, and they were meant to be the places where the future asteroid miners lived, when they were not out in the field working. Just as the railroads made access to the North American interior catch the eye of new speculators and prospectors, interest in asteroid mining had another revival when launch costs began to fall in the NewSpace era.
In the early 2010s, two startups, Deep Space Industries and Planetary Resources, publicly announced business plans based on asteroid mining. They sought, and received, funding from venture capital funds and private investors, including Ross Perot and Google heads Larry Page and Eric Schmidt. Even with this backing, the would-be asteroid miners faced an uphill battle. Their effort encountered technical obstacles, but it also ran into challenges in international law.
The “Treaty on Principles Governing the Activities of States in the Exploration and Use of Outer Space, including the Moon and Other Celestial Bodies,” more commonly known as the Outer Space Treaty, took effect, via the United Nations, in 1967. This treaty, signed by 134 countries, including the United States, forbids any nation from making a sovereign claim on territory in outer space. “The exploration and use of outer space,” the treaty affirms, “shall be the province of all mankind.”18 It further stipulates that nations are responsible for the behavior of their own private companies acting in space. Thus, the legal status, under international law, of a “claim” to an asteroid or a mining site on the Moon is ambiguous, making asteroid mining not only complicated, but difficult to sell to investors.
One of the first and largest efforts these companies undertook, with the bulk of the work done by Planetary Resources president Chris Lewicki, was to lobby the US government for a legal interpretation of the treaty in their favor. In 2015, they succeeded. President Obama signed the Spurring Private Aerospace Competitiveness and Entrepreneurship (SPACE) Act into law. It contained language that, while not allowing claims of sovereign ownership, upheld the right of the US government and American citizens and companies to “engage in the commercial exploration and exploitation of space resources.”19 No one could claim territory, but they could claim materials. The water on the Moon and the metal in the asteroids were there, the law says, and anyone with the capability to make use of them could do so.
At the same 2019 Space Studies Institute conference that space-mining professor George Sowers spoke at, another speaker made the connections clear. “Sticking to property rights for a second, I think everyone here should actually applaud Chris for all the work he did to get that agreement done in 2015, because it was an incredible amount of progress,” said Eva-Jane Lark, a space settlement advocate and investor. “When you look at history and precedents for how things were structured, how you determine claims, it’s a crucial issue … When you look at history it was determined by the miners themselves,” Lark said. “Nobody really wants The Hague telling potential miners what to do.”20
In 2020, President Trump issued an executive order that went even further than the SPACE Act. After affirming the rights to “recover and use space resources,” the order goes on to write into official policy that, “outer space is a legally and physically unique domain of human activity, and the United States does not view it as a global commons.”21 The order then empowers the US secretary of state to act on behalf of the country to pressure the leaders of other nations to affirm all of these things as well. Not only was it an explicit rejection of any interpretation of the Outer Space Treaty as disallowing space mining, but it also sought to expand of this rejection into other signatory nations.
In denying the status of outer space as a “global commons,” Trump’s order was a direct rebuke to another UN document. In 1967, when the Outer Space Treaty went into effect, no human had yet landed on the Moon. The United Nations repeatedly revisited the ideas that drove the treaty, though, and in 1979 they presented the “Agreement Governing the Activities of States on the Moon and Other Celestial Bodies,” known more widely as the Moon Agreement.22 The Moon Agreement, calls for international regulation of the use and exploitation of extraterrestrial resources, thus making outer space an official category of international law.
The language about “global commons” which Trump rejects, was first deployed at The Hague to protect cultural property during wartime, which was recognized as “the common heritage of humanity.” It was further extended to cover the ocean floor, its surface, and the air above it, and it is broadly considered to extend outward into space beyond the atmosphere. Some legal experts argue that the affirmation of the right of satellite overflight, a principle that helped define the early space age, was also a de facto acceptance of outer space into the common heritage. This is the principle rejected by Trump, Obama, and Lewicki. The United States is not a signatory to the Moon Agreement; in fact, no nation with space launch capabilities is. Its signatories are mostly smaller, less wealthy countries in the global South and East. In 1979, it was the L5 Society, convened by advocates of Gerard O’Neill’s work, that helped successfully lobby against its adoption.
Even these latest developments stop short of allowing any nation or private actor to claim territory off of Earth as their own, but certain technical problems unique to places in outer space might make that issue moot in the first place. Mining on the Moon, even landing a spacecraft on the Moon, might create hazards that necessitate the creation of “safe zones” around these activities. Anything that disturbs the lunar regolith, like a drill or rocket exhaust, inevitably launches bits of grit, dust, and gravel upward. Since there is no atmosphere on the Moon, these particles fly away just as fast as the rocket explosions that push them, and they don’t slow down. Since there is comparatively little gravity on the Moon, these particles can travel very far, sometimes even into orbit, endangering other spacecraft or stations above. For sites like mines or spaceports, the private company or government operating them might need to claim territory anyway, potentially to a radius of several miles, in order to make sure that they would not endanger others working on the Moon. It may turn out to be the case that any nation who wanted to make a de facto sovereign claim to part of the Moon or the asteroids just has to set up some kind of scientific installation there, and demand that others stay out for safety’s sake. Or it may be the case that the kind of international regulatory structure called for in the Moon Agreement might be necessary after all.
George Sowers, in his 2019 presentation for the Space Studies Institute, framed the question of access to space and its resources in terms of revolutions. Each revolution, he argued, allows more humans to access more and more calories and territory. But the Hadza, as they hunt along the Great Rift Valley, have plenty of access to energy and space. One of their customs, strict sharing of surplus food among themselves, ensures that everyone has what they need in terms of calories. They have plenty of room, too; the territory they live in and move through breaks down into about one square mile per person. If everyone on Earth lived with as much access to surface area as the Hadza do, they would need to fill about 136 more planets like ours. Much of the land that the Hadza occupy is in national and international preservation parks, some of them recognized by the UN as part of the common heritage of humanity. The government of Tanzania exempts the Hadza from the hunting bans in these parks, because they have the demonstrated ability to regulate themselves without taking too much, and because the presence of human predators helps the other animals do the same. Since they do not produce surplus to sell or trade, they are also exempt from Tanzanian taxes.
Sowers’s point about revolution is really about the same kind of induced demand that Musk has a complicated relationship with. Just as adding lanes to highways leads to more cars on the road, launching more rockets into space creates more demand for more rocket launches. And access to space resources will create more need for mining on the Moon and the asteroids. The metric that Sowers relied on only counts calories, and it fails to capture the uneven ways that those calories are distributed. By itself, the existence of calories does not end human suffering; it just makes more humans. Bezos’s hope for “a thousand Mozarts and a thousand Einsteins” in space will not be a part of an “incredible civilization” of a trillion humans unless it can avoid the corresponding scalar effects. Presuming no other variables change, in such a world those effects would generate over 3 million deaths per day due to hunger.
Following up from Sowers, Chris Lewicki, the former president of Planetary Resources, presented his own case for mining in space at the 2019 conference. He also invoked the Industrial Revolution, and its attendant change in humans’ relationship to the world. But then in a subsequent slide, he pivoted about life in space. “It’s actually not going to be anything different,” he said, pointing out that humans would still be predictable in terms of their activities and needs. “There’s not some bright new different, never before seen future there,” Lewicki said. “It’s just us continuing to do the things we’ve always done.” There will be people who go to space because “it’s a fun hobby, or an interesting pastime,” but also, he says, “there will be people who do it because their life depends on it, or because they are fleeing something.”23
The future, therefore, will be more of the same. The aims of NewSpace, and the expression of those aims, is caught between the need to call for a new and better world, and the need to call for a better engagement with worlds in general. The new generation of private spaceflight and space mining companies is trying to maintain a complicated relationship with the different audiences to whom they are signaling. They need to build support by telling idealists and visionaries that everything will change and get better, but they also need to reassure investors and government that everything will stay basically the way it already is.
A revolution rearranges the terms of the human engagement with the world they are in. That’s what a revolution, like the agricultural one, or the industrial one, is, and that’s what the space mining revolution will be. Unless, as Ursula K. Le Guin suggests, humans widen the scope of their categories—the category of common heritage, but also the category of technology—the status quo will not change.
Lewicki, after speculating about refugees in space, went on to reflect on lessons learned from his time at Planetary Resources. After successfully lobbying Congress and the Obama administration to issue a legal interpretation of the Outer Space Treaty that recognized resource exploitation rights, both Planetary Resources and Deep Space Industries ran out of funding. Finding themselves unable raise more money within the usual ten-year period that startup companies need to get to the next level, both companies were bought for their assets by 2020. Lewicki now works for ConsenSys, a company that makes blockchain and cryptocurrency technology, which now has a space division. He lamented the fact that conventional funding structures seemed inadequate to deal with the long term research needs necessary to bootstrap space capitalism. Regarding venture capital funding and bond markets, Lewicki seems to echo Le Guin. “We created these things,” he said. “Much more than any technology problem … this particular aspect of financial engineering, is something where we need to see a new invention.”24
The 1967 Outer Space Treaty is a unique document. It sets out priorities and goals that are explicitly utopian, in a way that Le Guin probably approved of. As she writes in her “Rant about ‘Technology,’ ” “technology is how a society copes with its physical reality.”25 The Outer Space Treaty is a device for coping with the harsh and dangerous realities of space. One provision of the Outer Space Treaty in particular, Article V, writes about the consequences of the human presence in such a deadly environment—the requirement for mutual aid:
States Parties to the Treaty shall regard astronauts as envoys of mankind in outer space and shall render to them all possible assistance in the event of accident, distress, or emergency landing on the territory of another State Party or on the high seas. When astronauts make such a landing, they shall be safely and promptly returned to the State of registry of their space vehicle.
In carrying on activities in outer space and on celestial bodies, the astronauts of one State Party shall render all possible assistance to the astronauts of other States Parties.
States Parties to the Treaty shall immediately inform the other States Parties to the Treaty or the Secretary-General of the United Nations of any phenomena they discover in outer space, including the moon and other celestial bodies, which could constitute a danger to the life or health of astronauts.26
If you are in space, and you learn about a potential or present condition of danger or harm, you are required to warn and render aid to anyone else who needs it. The legal interpretations of the Outer Space Treaty, in the SPACE Act and in Trump’s executive order, guarantee the right to access water ice for rocket fuel, or iron for a rocket engine. Article V of the treaty itself seems to guarantee access to water for drinking, and to calories for food, if a lack of access to those things constitutes a danger to life and health.
Alexander MacDonald, the NASA economist and author of The Long Space Age, has noted that some people in space science refer to the Outer Space Treaty as “our constitution,”27 that is, as a set of foundational shared values, captured in a piece of political technology that can be used to accomplish shared goals. A space future that deploys the potential that is latent in this technology might look more like the life of the Hadza after all. If this is true, companies like Planetary Resources, Blue Origin, SpaceX, or any that come along next in NewSpace should adapt their technology to the existing work, instead of attempting to reorganize that document to their own ends. To do otherwise is only to end up reinventing the rocket engine as a missile, with all of its expansive and destructive power unregulated.
Elon Musk called his early failures and explosions on the launch pad RUDs—“rapid unscheduled disassemblies.” But once the explosion was done and the fires were out, his SpaceX engineers got to work figuring out how to put the pieces back together in a better way. Anything made by humans can be unmade, or remade. The pessimistic and even cynical assignment of poverty, displacement, hunger, and lack to a category of inevitable side effects of human nature will only create more of the same.
It is a question of induced demand, and the breakdowns themselves are signals that the engineers should well heed. Meanwhile, all of the legal and political technology that is needed to make a world without those things already exists: common heritage, human rights, and mutual safety. They’re simply not put together in the right way.