Contents | May 2003

More on science and technology from The Atlantic Monthly.

From the archives:

"Stairways to Heaven" (March 1992)
Since rockets will never be a cheap or efficient way to lift payloads into space, alternative—and highly speculative—technologies are being studied. By Gregory Feeley

The Atlantic Monthly | May 2003
Long Shot

Defying the odds, even before the recent loss of the space shuttle Columbia, an eccentric company called Sea Launch has become the first private enterprise to send large rockets into space—from an enormous floating launch pad that sails to the equator for blast-off. Has the era of private space travel begun?
by Gregg Easterbrook
verybody knows about the Kennedy Space Center, in Florida, from which the space shuttle flies. Many people know about Vandenberg Air Force Base, in central California, from which the military launches spy satellites aboard big Titan rockets originally designed to hurl nuclear bombs at the Soviet Union. But hardly anybody knows about a third spaceport in the United States, which can be found adjoining a bird sanctuary at one end of the isthmus on which the Port of Long Beach sits, just outside Los Angeles.

No government agency or military organization owns this new facility. Its heart is an enormous, otherworldly-looking vessel called the Odyssey—a floating launch pad, twenty stories high, built atop the same kind of floodable pillars as an oil platform. When viewed from a distance, the ship suggests the alien-designed teleportation tower in the sci-fi movie Contact. Intended to sail to the Equator and send into orbit an unmanned rocket that can carry as much as the space shuttle, the Odyssey is the operational part of the first entirely private effort to put into space entirely private large rockets carrying entirely private payloads. The ship is so large that when it came through the Suez Canal, in 1998, its owners, a company called Sea Launch, had to rent both lanes. Moored near the Odyssey at Long Beach is a large companion ship, the Sea Launch Commander. Built in Glasgow, the Sea Launch Commander scraped the sides of the Panama Canal on its passage from Scotland to California.

Sea Launch's core idea is a novel one; neither the National Aeronautics and Space Administration nor anybody else has tried the ocean approach to space. And it works: Sea Launch can send rockets into space more cheaply per payload pound than anything the government can offer, and those rockets place satellites exactly where they are supposed to go. Since its debut, in 1999, the Odyssey has launched seven large satellites, including the orbital broadcast towers for the new XM Satellite Radio subscription network—two multi-ton techno-marvels dubbed Rock and Roll.

Is Sea Launch an oddity or a harbinger of a new space age? With the loss of the space shuttle Columbia, the nation has been reminded that space flight is hazardous; and even when all goes well, NASA's current systems are absurdly expensive. Each space-shuttle launch costs some $640 million—meaning, among other things, that just the bottled water that astronauts drink on the new International Space Station costs around $400,000 a day. And good luck explaining what purpose the space station serves, other than as a jobs program for aerospace contractors and a destination to justify space-shuttle funding. Space also seems a hopelessly impractical destination, except for scientific exploration, military use, and telecommunications satellites. Most commercial ideas for space—microgravity manufacturing, for example—haven't really panned out. And history's first space tourists, two rich men who bought rides on old Soviet rockets launched from Kazakhstan in 2001 and 2002, paid about $20 million each to be crammed into a tiny capsule and subjected to agonizing G-forces at blast-off, to eat freeze-dried food and bump into floating Russians in orbit, and, finally, to come home motion sick. There's a limit to that market.

From the archives:

"Freedom of the Skies" (June 2001)
Inventors, entrepreneurs, and government visionaries have teamed up to create new kinds of small planes that can take off from and land almost anywhere. By James Fallows

From Atlantic Unbound:

Flashbacks: "The Soul of a New Flying Machine" (May 25, 2001)
James Fallows, the author of Free Flight, argues that the next generation of small planes could usher in a new age of travel.
Yet in the early days of aviation, airplanes were absurdly expensive and impractical too. Budget-busting government programs dominated, and flight applications were too specialized for the typical person to care about. In 1935, when Pan Am's first Clipper took off to chase the sun across the Pacific, air travel seemed destined always to be an experience exclusively for the super-rich; in 2001 U.S. airlines alone carried 622 million passengers, including tens of millions of the working class, and even the poor. When Federal Express proposed in 1973 to move packages anywhere overnight, the idea seemed a costly extravagance best suited to big business; now average Americans routinely get shirts or CDs delivered overnight. And in 1910, when crowds gathered throughout the United States to watch the touring Blériot monoplane struggle to gain altitude, anyone who said that someday thousands of Americans would own their own airplanes for personal transportation would have been considered certifiable.

Could space flight undergo a similar progression, morphing from a rarefied, expensive, and government-dominated activity into a mundane, affordable, and mostly private one? Some entrepreneurs think so. Sea Launch is not alone; start-up firms have spent considerable money in the past decade trying to develop cheap rockets, reusable rockets, space planes that are towed or catapulted into the sky, and even a rocket-powered helicopter that rotors up into orbit. And a few mavericks within government have been urging NASA and the Air Force (which has the world's second largest space budget) to adopt revolutionary, low-cost means of reaching orbit. So far every new idea except Sea Launch's has failed. But on the day the Wright Flyer skimmed up into the Carolina breeze, all previous attempts at flying heavier-than-air vehicles had flopped, and the whole idea of entrepreneurs doggedly pursuing the skies struck most sensible people as hilarious.

he genesis of Sea Launch came roughly a decade ago, when Boeing, which then had no rocketry division, was casting about for a way to get into the satellite-launch market. At the time, satellite customers—mainly telecommunications firms—were dissatisfied with every option available to them. American launch services relied strictly on Atlas, Delta, and Titan rockets, all of which used engineering from the late 1950s, when the rockets had been designed as intercontinental ballistic missiles. (By convention, "missile" denotes a military device, and "rocket" a civilian one; missiles are aimed at targets, and rockets are a means of flight.) Though they got the job done, the aging rockets were expensive, and their designs reflected little of what had been learned about electronics and materials in more recent years. To top it off, the manufacturers of the three rockets had fallen behind in production. The result was that in the early 1990s one had to wait a year or more to have a telecommunications satellite put into orbit.

The main international alternative to these rockets, the Ariane, built by the French company Arianespace, was experiencing its own problems. Arianes have a nasty tendency to veer off course. This means they have to be blown up, to avoid a crash, instantly reducing to glowing slag painstakingly crafted satellites that could be worth as much as $250 million apiece. The shortage of reliable launch systems in the early 1990s, coupled with the eagerness of telecom firms to put hardware into orbit, prompted the first President Bush and then President Bill Clinton to sign waivers allowing American commercial payloads to fly on Chinese rockets. The moves were ill-fated: in 1998 it was alleged that Loral Space and Communications, a U.S. defense contractor, had given China classified technology that could have been used to improve its rocket accuracy. All in all, at around the time Sea Launch was being dreamed up, anyone who could offer satellite customers a reasonably priced launch using a reliable modern rocket seemed to have a promising future.

The Soviet Union was breaking up in those days, and its engineering bureaus were winning permission to market technology for commercial use. The old Soviet system may not have been able to produce blue jeans, but it was good at rocketry. From the end of the Moon race, in the early 1970s, until the end of the USSR, the Soviet space program outlaunched America's by a wide margin, establishing a solid safety record in the process. Whereas NASA went "cost-no-object" with the shuttle, Soviet engineers concentrated on perfecting simple, affordable approaches to space flight. Soviet rocket engines became sufficiently high in quality and low in cost that today the latest version of the Atlas—a rocket originally designed to obliterate Russian cities—uses a Russian main engine that is extremely powerful, but has far fewer moving parts than the space shuttle.

The last rocket the Soviets designed before the disintegration of the USSR was called the Zenit. Though in appearance the Zenit is nothing more than a long tube, it represents the distillation of everything Moscow's engineers had learned about rockets. Its engines are enormously powerful, yet spare compared with American designs. Supersensitive handling is not required: Zenit components can be laid on their sides and moved around by hoist. Most rockets must be treated gingerly, which adds to the handling expense. The skin of the Titan is so frail, for example, that a mechanic once caused one to explode just by dropping a tool on it from a platform. The Zenit was also designed for simplified flight: its entire launch sequence can be automated, and executed in just a few hours—no weeks of fine-tuning in the clean room, no swarms of lab-coated technicians on the gantry. (Most rockets require lengthy preparation cycles. The latest Delta, the Delta IV, needs six to eight days of prep time on the pad before flight.) Affordable, relatively simple to use: the Zenit seemed just what commercial satellite customers were looking for.

When the secrecy surrounding the Soviet space program ended, in the early 1990s, a rocket-loving faction within Boeing learned the details of the Zenit and wanted to buy some of the rockets for commercial use. But rather than bring them to the United States and contract with the Air Force for launch at Cape Canaveral (all other rockets in the world fly from government facilities), Boeing's rocket lovers wanted a purely private venture. That raised the question of where to launch the Zenits. At some point Valery Aliev, an engineer who had helped to design the Zenit, said, "You should launch it at the Equator, off a ship."

Rocket buffs knew what Aliev meant. Any projectile launched to the east from the Equator derives maximum benefit from Earth's rotation; the closer a rocket is to the Equator at blast-off, the more momentum it has and, therefore, the more it can carry. The Kennedy Space Center is located where it is because, of all points in the continental United States, Cape Canaveral offers one of the best combinations of low latitude and ocean to the east. Ocean to the east means "range safety"—that is, no inhabited area on which a malfunctioning rocket might fall. Arianespace built its launch center in France's quasi-territorial possession French Guiana, which has ocean to the east and is only five degrees north of the Equator. There are places exactly on the Equator where NASA or Arianespace or anyone else might build a spaceport, but doing so would entail complex negotiations with a host government. Launching from a ship would mean that one could just pick a spot in the sea along the Equator where there is no land to the east, sail there, and have an instant spaceport—no treaties required, no fees to anyone, and rocket performance boosted by nature.

No government had ever attempted a sea launch to orbit; the notion was considered too risky and technically daunting. (Ballistic nuclear missiles launched from submarines do not go into orbit but fly less challenging suborbital paths.) The people who founded Sea Launch became determined to prove that this approach would be not only practical but also cost-cutting. The company began, in 1995, as a joint venture among Boeing and the Russian and Ukrainian companies that manufacture the Zenit, along with a fourth partner, Norway's Kvaerner Group, which built and operated the two ships. (Many of the world's high-end commercial vessels have Norwegian bridge crews.) Preparations began, and Sea Launch eventually spent about a billion dollars on its ships, its home port, its testing, and its staff. Not a dime came from government.

Then, in 1997, the ground shifted: Boeing took over McDonnell Douglas and in the process acquired the Delta rocket program. The program is expensive and has insider status with the aerospace old-boy network—both NASA and the Air Force use Deltas. After the merger Boeing's top management is said to have had second thoughts about its private rocket. Executives worried that Sea Launch—a lean, unconventional operation with a focus on price-cutting—not only would steal business from the more expensive Delta rocket but also might offend NASA and the Pentagon, both important Boeing customers, by showing up the government's top-heavy space programs.

Following the merger, Sea Launch developed an uneasy relationship with its parent. Boeing still owns its original 40 percent of the project, but Sea Launch is pretty much on its own, even though all the world's other launch systems enjoy some form of government support and government guarantees. So although Sea Launch started with a borrowed rocket design and Boeing capital, by the time it was ready to fly, it had become an outsider organization, preparing for frontal assault on NASA and the aerospace giants. This is just what advocates of private space flight had been dreaming would happen.

ocket parts arrive by ship at the Sea Launch home port, where they are transferred to an assembly hall aboard the Sea Launch Commander. The Commander was designed to do rocket processing anywhere, and the fact that it floats offshore—though, at home port, only inches away from California soil—eases export and union rules.

In the assembly hall of the Commander, Russians work on the Russian part of the system, Ukrainians work on the Ukrainian part, and only Americans are allowed to work on the satellites themselves, where the sensitive technology is found. The China scandal ultimately turned out to be a boon for Sea Launch, because President Clinton ended up banning the use of Chinese rockets by American satellite customers—eliminating one government-subsidized competitor to whom Sea Launch might have lost business. But initially, as newspapers expressed outrage over the possibility that U.S. space technology had been leaked, the State Department ordered Sea Launch shut down and the Russians and the Ukrainians left Long Beach.

The project eventually resumed, with elaborate conditions in place. Only Americans are now allowed to work with, or even to see, the payload; satellites must be sealed and rendered inaccessible before being transferred to Russian technicians, who mount them on the upper stage. Americans are supposed to keep their hands off the Russian technology. At home port Sea Launch's Eastern bloc has a separate office building, known as "the Russian embassy" (although the Ukrainians use it too). Aboard the Commander are Russians-only and Ukrainians-only work areas that even top American officials don't enter. After Sea Launch suffered its sole failure to date (a 2000 flight that went haywire attempting to put into orbit a communications satellite called ICO-F1), the company had to obtain a special federal license for American and Russian experts to exchange technical data while figuring out what went wrong.

Once a rocket is assembled on the Commander, it is transferred to the Odyssey, which then departs for a long sail to a point on the Equator about 1,400 miles south of Hawaii. The Odyssey's destination is the Pacific doldrums, where the waters are usually flat, the winds are usually light, and lightning is rare. It's a place with no shipping lanes and nothing but open ocean for hundreds of miles to the east. At the Kennedy Space Center thrill seekers in boats and private planes often populate the downrange area, where a malfunctioning rocket can become a menace; Coast Guard crews and other range-safety personnel number in the many dozens during launches. But Sea Launch need not worry about some idiot in a Cessna 172 buzzing the liftoff path. An hour before launch the Commander's helicopter sweeps the downrange zone. Pilots have never seen anything but ocean.

Three days after the Odyssey sets sail, most of the company's personnel board the Commander and catch up with its slower sister ship. Crew members on the two ships combined number 310—most of the staff of the Sea Launch enterprise, except for the security guards, business managers, and secretaries minding the home port. This is a tiny staff, relatively speaking: Ariane, Atlas, and Delta launches each require about 1,000 people. The Russians and the Ukrainians on the Sea Launch staff earn much less than the Americans, and so tend to view their time on the ship as luxurious—the food is free, the chefs boast of preparing restaurant-quality meals, and drinks cost a dollar. Most of the crew members are male. Typically, one or two executives from the company whose satellite is being launched come along, and for them there are VIP staterooms. A State Department official also travels with the crew. One of the official's tasks is to monitor compliance with each launch's export license, because as far as federal law is concerned, the technology on anything bound for outer space is being exported.

When the launch destination is reached, after a journey of approximately eleven days, the Odyssey takes on seawater in order to lower itself in the waves and become stable. A temporary bridge links the vessels, and personnel shuttle back and forth; both ships have underwater "thrusters" that allow them to stay almost motionless in fair seas. The Commander sports two complete control rooms side by side: one has computers that operate in English and is laid out to resemble NASA's Mission Control, in Houston; the other has computers that operate in Russian and is laid out to resemble the old Soviet space-program stronghold, at Baikonur. The Russian control room manages the rocket, the American room the spacecraft. The Americans, being the lead investors, have final say.

Much of the time at the Equator the rocket is still lying on its side in a protective building atop the launch ship. The day before blast-off, in the quick, hands-free prep of which the Zenit's designers are so proud, an automated system rolls the rocket out and sets it upright. Five hours before the launch everyone leaves the Odyssey, the Commander pulls away to a three-mile distance, and the automated system fuels the tanks. At ignition this very powerful machine—it has 1.6 million pounds of liftoff thrust, far more than earlier rockets of the same size—climbs up from the pad so quickly that the Odyssey bears few scorch marks.

Most commercial satellites are bound for geostationary orbit, 22,300 miles above the Equator. In this orbit an object hangs the same distance from Earth as Earth is around; the effect is that the satellite moves at a pace matching Earth's rotation, and therefore remains over the same point. (An alternative is "sun synchronous" positioning, which Sea Launch doesn't offer, in which a satellite mimics the apparent movement of the sun.) Sea Launch's fabulous accuracy when "injecting" to high orbits is a selling point, because a satellite in exactly the correct orbit will perform better and last longer. All Sea Launch flights have deposited the payload within a few meters of the ideal point; the Ariane and other rockets have sometimes missed by hefty margins.

When a high orbit is the destination, several short burns in a rocket's final stage are more accurate than one long firing, so the Zenit uses a stop-start motor, the most complicated kind of rocket engine, in its final stage. Control of the final stage is necessary for hours, because high orbit is reached in phases rather than all at once. The Commander's huge antenna, a seaborne tracking station, is employed for this task. When the Commander is moored at home port, Russian personnel have been known to use the antenna to chatter with orbiting cosmonauts. The subject of discussion has sometimes been whether Sea Launch has job openings.

As the satellite ascends to its final orbit, usually over several weeks, the vessels sail back toward Long Beach, with the customer executives sometimes departing as soon as the ships are within helicopter range of the nearest place to catch a flight to Hawaii. The Commander is designed to carry a second Zenit, which could be transferred to the Odyssey in mid-ocean for another launch before returning to port; system planners wanted the ability to put a lot of satellites into orbit fast. But in the current slack market it's back to Long Beach and months of downtime until the next payload. Most of the Russians and Ukrainians are sent home between missions, because paying their passage back to the motherland costs less than putting them up in California.

ea Launch was the only wholly private rocket venture to become operational during the 1990s, but this wasn't for want of trying by others; the decade witnessed a private rocket-research boom. Until the loss in 1986 of the shuttle Challenger, NASA had essentially held a legal monopoly on commercial satellite launches within the United States. In the aftermath of the Challenger calamity, however, President Ronald Reagan signed orders banning commercial payloads aboard the shuttle and appealed to industry to compete against NASA. Space-launch research, stagnant since the end of the Moon race (the United States had not flown a new kind of rocket since the Saturn V moon rocket, which first went aloft in 1967), was revved up.

Then, in the 1990s, came forecasts of fantastic growth in the satellite industry. The burgeoning demand for portable phones was going to be addressed by "constellations" of hundreds of low-altitude repeater satellites, operated by companies with names like Iridium, Teledesic, and Globalstar. Venture capitalists spoke of private launch companies that would fire a rocket to orbit every day and of "space planes" that would travel back and forth in such numbers that space tourism might become, if not common, frequent enough to be unremarkable.

Believing that private space launching would be the next great technology growth sector, numerous start-up firms opened their doors, each promising a different idea for launch engineering—from supercheap throwaway boosters to the rocket-accelerated helicopter. A company called Orbital Sciences designed a mini-rocket, dropped from beneath the belly of a plane, that can carry mini-satellites into orbit; in 1990 that mini-rocket, the Pegasus, became the first privately developed object to fly to space. (Because the Pegasus can handle only little payloads, purists don't put Orbital Sciences in a class with Sea Launch.) A wealthy Dallas banker named Andrew Beal sank more than $100 million of his own money into a private attempt to build a simple but powerful throwaway rocket that would cost half as much per pound launched as the least expensive existing system. Beal Aerospace hired about 200 people, built an office complex, and got as far as the "static" firing, on a test stand, of the largest privately designed rocket motor ever to ignite. The rocket motor worked fine, but Beal Aerospace did not. The company closed its doors in late 2000.

Another important start-up was Kistler Aerospace. Based near Seattle's tech wonderland and backed by the cell-phone baron John McCaw, Kistler is said to have raised about $500 million by 2001, more capital than any of the other pure start-ups. Kistler's plan was to build the world's first fully reusable rocket, powered by inexpensive, reliable Russian engines. The stages of the rocket, after dropping off, would float back to Earth on a parachute and, at the last second, deploy a giant cushion for landing. The rocket's ascent path would be manipulated so that the stages would float back close to the launch point, not hundreds of miles downrange. Kistler hired lots of engineers, burned through lots of money, and broke ground on a spaceport in Woomera, Australia, a location near the Equator. But no launch was ever attempted. A remnant of the company still exists, and continues to promise the first blast-off of a totally reusable rocket.

During the 1990s many entrepreneurs also became enamored of the prospect of a space plane—that is, a winged flying machine that would ascend to orbit in level flight, gradually building up speed, rather than roaring straight skyward. The idea wasn't new: the United States had developed a prototype, called the X-15, at the beginning of the space race. It was a small, one-person winged aircraft that was carried aloft by a B-52. Released at the bomber's maximum altitude, the X-15 would fire a rocket engine and rise as high as sixty-seven miles (commercial aircraft cruise at around five miles), which is the boundary of the atmosphere, before returning to land at an airfield under the pilot's control. From its debut, in 1959, to 1968 the X-15 flew to the edge of outer space almost 200 times, with only one crash—a remarkable feat considering that the entire project was undertaken before pocket calculators were in stores.

The X-15 and related space-plane projects were canceled in 1967, and during the Nixon Administration both NASA and the Air Force committed themselves to using only the space shuttle for future launches. To this day space buffs are driven to distraction by the thought that the United States had a functioning space plane almost half a century ago—albeit one not powerful enough to achieve orbit—yet abandoned research into this comparatively low-cost idea in order to focus on the very expensive space shuttle. A space plane is an attractive idea for other reasons besides cost: it might take off from anywhere and return to anywhere, and its relatively gradual acceleration could mean a smooth ride that a healthy person could bear without months of astronaut training. Such characteristics made space planes sound like airplanes, at least to entrepreneurs—and since building and flying airplanes is a successful private industry, perhaps this was the model that would make space flight profitable.

Start-up firms researched space planes that would be towed into the sky by a 747 for air launch; that would take off with regular jet engines and then receive in-flight fueling with rocket propellants from a tanker aircraft; that would rise off runways on the backs of mammoth carrier aircraft, in a grand update on the X-15; and that would have lasers beam them energy as they flew, to heat special chemical fuels. The most ambitious design came from a start-up called Andrews Space & Technology, which proposed a space plane that would take off on the back of a giant carrier aircraft, without any liquid oxygen for its rocket engines. As the tandem gained altitude, a system aboard the carrier aircraft would manufacture liquid oxygen out of the air. Thus the space plane would not reach full takeoff mass until it was already 30,000 feet up and moving at 500 miles per hour. Andrews has no prototype yet, but it does have $3 million in NASA funding to study the idea. Private-rocketry advocates who speak of a revolutionary overthrow of the space status quo have, like the big aerospace "primes," found that it's a lot easier to stay in business with government contracts.

A watershed moment for the independent space start-ups came in 1999, when a space helicopter called the Roton flew. One would be hard put to come up with a goofier idea than a helicopter bound for orbit, but that's what the Roton (as much as it sounds like the star of a Japanese monster movie) was to be. The original idea was this: A gumdrop-shaped spacecraft would hang underneath four rotor blades. At the end of each blade would be a rocket motor. The rockets would fire sideways, accelerating the blades to unheard-of revolutions per minute; the spinning motion itself would pump out the rocket fuel, avoiding the expense and weight of the complicated turbo-compressors that are key items on the space shuttle's main engines. Up toward space the Roton would flutter, churning through the air. As the atmosphere thinned, the rockets would swivel downward for regular firing. (Above the atmosphere a little thrust goes a long way.) On the return from space the rotors would provide braking power, so that the Roton could soft-land exactly where it took off, allowing immediate refitting for the next flight.

The helicopter approach turned out not to be practical for launching the Roton, but it was designed to be used for landing. True space believers loved the Roton, because it was technology the aerospace establishment laughed at (They laughed at the Wright brothers! They laughed at Montgolfier!) and because the company that made it, Rotary Rocket, was run by Gary Hudson, one of the first public advocates of private rocket development. Around 2,000 people attended the Roton rollout, in 1999, at an airfield in Mojave, California, from entrepreneurs to space groupies to the science-fiction writer Jerry Pournelle, a NASA critic who somehow got the ear of Vice President Dan Quayle during the first Bush Administration. Out of a hangar came the weird rotor-topped spaceship, built at a cost of about $5 million. Later, at the first launch, gases hissed, up into the air the Roton went, and then it landed as a helicopter. They laughed at Montgolfier!

The Roton prototype never got above a few hundred feet, but Hudson beams with pride about what he accomplished. "Five million dollars is what the primes spend on paint for their prototypes," he says. "For that amount I did the basic design of an all-new concept, put a working test vehicle into the air with a pilot, and brought him back safely. It's true that private space entrepreneurs haven't done much compared to the big boys, but we haven't had access to even a tiny fraction of their capital. Give me ten percent of what the primes have and I will clean their clocks."

That said, Roton, Beal, Kistler, and the rest have all either folded or gone into hibernation. Why? Part of the answer is cell phones. Iridium and other space-based telecommunications companies that were going to put triple-digit numbers of satellites into orbit instead fizzled out—bested by the humdrum earthbound technology of repeaters mounted on poles and arranged to broadcast in cells. When the cheap cell phone appeared, demand for rocket-launch services collapsed. Today the average number of annual space launches is little different from what it was in the 1960s; the huge market in commercial space launches simply hasn't emerged. Most investors have withdrawn their capital from the start-ups. Space hope springs eternal, though: the entrepreneur Elon Musk, who became wealthy when he sold his creation PayPal to eBay last fall, just announced a new company called SpaceX, whose goal is an entirely private rocket much cheaper than any in existence today.

And although the start-ups had ambition and dreams, none ever had enough capital in the first place. Andy Beal's conviction that he could get a large rocket, even a simplified one, flying for $100 million was an illusion; the development of auto models has cost more. It took $2 billion to $3 billion to develop the 777 airliner—a plane that flies between airports using well-established technology. Given that, how can private space-plane developers seriously expect to perfect a machine, for less than many billions of dollars, that reaches orbit using all-new technology? As long as demand for space launches is relatively modest, capitalists will have little reason to put up the huge sums required for a revolutionary rocket or space plane—even if the result might ultimately be cheaper to operate and better than anything existing today.

ea Launch, at least, is flying, and its hope of survival is price. Right now the company is offering to put about six tons into geostationary orbit for about $65-70 million. This is loss-leader pricing; the company wants to charge more but can't in the current market. But consider that a space-shuttle mission costs about ten times as much as a Sea Launch flight, yet carries a comparable payload. This is why the space establishment, with thousands of jobs and more than $4 billion in annual spending tied to the shuttle program, is not happy about Sea Launch.

According to Marco Caceres, an analyst for the space consulting organization Teal Group, Sea Launch's competitors are asking significantly more per payload pound to high orbit than Sea Launch. And despite the fate of its fellow start-ups, Sea Launch has competitors in spades. Lockheed Martin offers three versions of the Atlas rocket, including the Atlas V. Lockheed also sells launches aboard the Proton, a workhorse of the old Soviet space program. Boeing offers three Delta models, including the new Delta IV, sporting the first all-new U.S. rocket-engine design in some thirty years. Arianespace offers two models of the Ariane. A Russian firm also sells launches on the Soyuz, a rocket barely changed from the one that made Yuri Gagarin the first person in space, four decades ago. Japan's rocket program, though troubled by launch failures, continues to solicit customers. China hopes to regain position in the commercial launch business; India recently entered it, and Brazil hopes to enter. There are even companies offering surplus U.S. or Soviet ICBMs converted to launch small payloads into space.

If Sea Launch charges less than many of these companies, how can they stay in business? The explanation is that everybody else has government contracts, and governments don't necessarily drive hard bargains. Space launching for NASA and the military is "a very stable, good-margin business," Albert Smith, the executive vice-president of the Lockheed Martin division that operates the Atlas rocket, told the industry publication Aviation Week and Space Technology in 2001. Boeing's Delta IV and Lockheed's Atlas V received a total of about $1 billion in Air Force development funding, and each company was guaranteed Air Force use. Arianespace is guaranteed all French government launches and conducts most of the launches of the European Space Agency. The Chinese, Japanese, and Indian rocket programs are anchored in government money. Sea Launch's lower overhead means it can hang in, but as a true free-market project competing against projects with extensive government backing, it has a tenuous existence.

Many in the aerospace business are quietly rooting for Sea Launch to fail, so that its innovations and affordability will stop providing uncomfortable comparisons with "good margin" government projects. The situation is especially awkward for Boeing, which now offers customers both the privately developed, low-cost, and ready-to-fly Sea Launch system and the tax-subsidized, higher-cost, years-behind-schedule Delta IV. The old boys worry that Sea Launch, or any start-up that succeeds, will kill the goose that lays the golden eggs. The last thing big aerospace contractors want is to set in motion price cuts in government work.

each low orbit, and you're halfway to anywhere in the solar system," the writer Robert Heinlein once supposed, and from the standpoint of propulsion he was correct: the power required to go the first 200 miles up from the ground to orbit is about the same, per pound moved, as that required to traverse the next 50 million miles to Mars. But the price of that initial launch remains so high that the rest of the solar system might as well be light-years distant. Talk of returning to the Moon, or of people's exploring Mars, or of protecting Earth against extinction-causing asteroid strikes, will remain whimsy until someone figures out a much less expensive means of putting pounds into space.

Rocket lovers continue dreaming nonetheless, sure that the breakthrough must someday come. They foresee going back to the Moon in numbers, to establish laboratories or asteroid-watch stations or even defensive fortifications against alien attack. Space tourism as a commonplace event is viewed in these circles as a question of when, not if. Mars is a particular obsession. For years a group of astrophysicists, rocket lovers, sci-fi writers, and space buffs held a Case for Mars conference, advocating a grand adventure to the Red Planet. (I once attended a Case for Mars conference and can report that they didn't talk much about launch cost.) Space-buff Web sites tout, among other things, fusion-driven scramjets, antimatter power, and "zero point energy" propulsion based on quantum fluctuations in subatomic particles. Hey, and what about a skyhook? In theory a heavy counterweight could be put in orbit and then used to lower a hook to Earth's surface. Giant solar collectors floating in space could produce electricity to run winches that would pull up the skyhook, allowing access to orbit by means of a very spooky but low-cost elevator ride. Once we built a skyhook, we'd be halfway to any destination in the solar system.

Meanwhile, Sea Launch readies its private rockets to make the long journey to the Pacific doldrums. Out on those flat seas there are few natural sounds. People standing on the deck of the Sea Launch Commander can hear the clanking noises of the automated rocket-erection devices on the launch ship miles away, and the hissing of the propellants as they load. Then the sound changes from distant, muffled reverberations to the earsplitting roar of blast-off. In a single instant more energy is unleashed above the Odyssey launch pad than the entire medieval world used in a year. The rocket arcs upward on blinding flame, and in a moment the light vanishes from sight. A pop-music satellite goes into orbit, a small accomplishment in the scheme of things, and two ships sail back to port, their crews hoping that more business awaits them.

Will such private access to space someday be a part of everyday life? Or will space always be inaccessibly far off and expensive for most people? The private rocketeers of Sea Launch can only wonder as they watch their rocket disappear into the infinite darkness of space.

What do you think? Discuss this article in Post & Riposte.

Gregg Easterbrook is a contributing editor of The Atlantic, a senior editor of The New Republic, and a visiting fellow at the Brookings Institution. His most recent book, a novel, is The Here and Now (2002).
Copyright © 2003 by The Atlantic Monthly Group. All rights reserved.
The Atlantic Monthly; May 2003; Long Shot; Volume 291, No. 4; 64-75.