Return to the Table of Contents.
N O V E M B E R 1 9 9 7
by Freeman J. Dyson
BOUT twelve years ago I visited the Johnson Space Center, in Houston, and climbed around in the space shuttle that is kept there for visitors to examine. That was before the Challenger disaster, when the shuttle was advertised as a safe ride for congressmen and schoolteachers. What impressed me about the shuttle was the immense quantity of stuff on board for the care and comfort of human passengers. It felt more like a hotel or a hospital than a rocket ship. I made rough calculations of how many tons of material were needed to keep seven passengers alive and well for a couple of weeks. I was thinking, Why don't we rip all this out and fly the thing from the ground by remote control?
Discuss this article in the Science & Technology forum of Post &
From the archives:
A look back at Atlantic articles tracing our long preoccupation with the extraterrestrial.
At that time most of the shuttle missions were carrying unmanned satellites
into orbit for various purposes -- some scientific, some commercial, and some
military. These launching jobs could just as well have been done automatically.
Only a few of the shuttle missions really need people on board, to do
experiments or to repair the Hubble Space Telescope, for example. It would have
made sense to reserve two shuttle ships with all their hotel equipment for
missions in which people were essential and to use the other two for
satellite-launching jobs. A freight-only version of the shuttle could carry
bigger payloads for less money than the passenger version, without risking any
lives. Unfortunately, when I suggested this to people at Houston, they did not
think it was a good idea. Their whole existence is centered on the training of
astronauts and the operation of manned missions.
After failing to eviscerate the shuttle, I wandered into the museum of the Johnson Space Center, where there is a collection of rocks that astronauts brought back from the Moon. Many of the Moon rocks have been lent to scientists in other places, but a large number remain in Houston. Scientists who are interested in Moon rocks are usually also interested in meteorites: their tools for analyzing meteorites work on Moon rocks as well. The space-center museum has a fine collection of meteorites, too, some of which were sitting in glass cases next to the Moon rocks. Among them were two from Mars.
It seemed like a miracle. Here I was, in the museum in Houston, twelve inches away from a piece of Mars, with only a thin pane of glass to stop me from grabbing hold of it. In those days the National Aeronautics and Space Administration was talking seriously about grandiose missions to Mars, costing many billions of dollars. One of the reasons for going to Mars was to bring back samples of rock for scientists to analyze. And here were samples of Mars rock already in Houston, provided by nature free of charge. I found it odd that nobody seemed to be studying them. As far as I could tell, I was the only person in Houston who was excited about the Mars rocks. I stood and gazed at them for a long time. Nobody else came to look at them. I remarked to the NASA people that they might usefully spend some time studying the Mars rocks they already had, instead of planning billion-dollar missions to collect more. At that time the administrators in Houston seemed little interested in anything that did not cost billions of dollars.
Things have changed since then. Now NASA is interested in cheap missions, and many more scientists are interested in Mars rocks. Last year some of the rocks were examined more thoroughly than ever before. Two contain chemical traces that might be interpreted as evidence of ancient life on Mars, and scientists have also found microscopic structures that might be relics of ancient microbes. The evidence that these traces have anything to do with biology is highly dubious; we cannot say on the basis of it that life must have existed on Mars. These traces are important for two other reasons. First, if we are seriously interested in finding evidence of life on Mars, we now know that Mars rocks on Earth are the most convenient place to look for it. Instead of waiting for many years for an expensive sampling mission to land on Mars and return a few small chips of rock to Earth, we can find a supply of bigger chips lying in Antarctica, where meteorites accumulate on the ice and are freely available. Second, these rocks show that if life was established on Mars at any time in the past, it could have been transported to Earth intact. In the first billion years after the solar system was formed, when Mars had a warm climate and abundant water, asteroid impacts were much more frequent than they are now. Mars rocks fell on Earth in great numbers, and many Earth rocks must also have fallen on Mars. We should not be surprised if we find that life, wherever it originated, spread rapidly from one planet to another. Whatever creatures we may find on Mars will probably be either our ancestors or our cousins.
NOTHER place where life might now be flourishing is in a deep ocean on Jupiter's satellite Europa. Jupiter has four large satellites, discovered almost 400 years ago by Galileo: Io, Europa, Ganymede, and Callisto, in order of their increasing distance from Jupiter. The Galileo spacecraft now orbiting Jupiter is sending back splendid pictures of the satellites. The new pictures of Europa show a smooth, icy surface with many large cracks but very few craters. It looks as if the ice is floating on a liquid ocean and being fractured from time to time by movements of the water underneath. The pictures are strikingly similar to some pictures of the ice that floats on the Arctic Ocean; it would not be surprising if Europa had a warm ocean under the ice. Io is blazing hot, with active volcanoes on its surface; Ganymede's surface is icy like Europa's but not so smooth; and Callisto looks like a solid ball of ice covered with ancient craters. All four satellites are heated internally by the tidal effects of the huge mass of Jupiter, but the internal heating falls off rapidly with distance from Jupiter. We should expect that below the surface Europa is much cooler than Io and much warmer than Ganymede and Callisto. Since Io is hot enough to boil away all its water, and Callisto is cold enough to freeze solid, Europa might well have a warm liquid ocean. Ganymede might also have a liquid ocean, but it would be covered by a much thicker layer of ice. Of all the worlds that we have explored beyond Earth, Mars and Europa are the most promising places to look for life.
To land a spacecraft on Europa, with the heavy equipment needed to penetrate the ice and explore the ocean directly, would be a formidable undertaking. A direct search for life in Europa's ocean would today be prohibitively expensive. But just as asteroid and comet impacts on Mars have given us an easier way to look for evidence of life on that planet, impacts on Europa give us an easier way to look for evidence of life there. Every time a major impact occurs on Europa, a vast quantity of water is splashed from the ocean into the space around Jupiter. Some of the water evaporates, and some condenses into snow. Creatures living in the water far enough from the impact have a chance of being splashed intact into space and quickly freeze-dried. Therefore, an easy way to look for evidence of life in Europa's ocean is to look for freeze-dried fish in the ring of space debris orbiting Jupiter. Sending a spacecraft to visit and survey Jupiter's ring would be far less expensive than sending a submarine to visit and survey Europa's ocean. Even if we did not find freeze-dried fish in Jupiter's ring, we might find other surprises -- freeze-dried seaweed, or a freeze-dried sea monster.
Freeze-dried fish orbiting Jupiter is a fanciful notion, but nature in the biological realm has a tendency to be fanciful. Nature is usually more imaginative than we are. Nobody in Europe ever imagined a bird of paradise or a duck-billed platypus before it was discovered by explorers. Even after the platypus was discovered and a specimen brought to London, several learned experts declared it to be a fake. Many of nature's most beautiful creations might be dismissed as wildly improbable if they were not known to exist. When we are exploring the universe and looking for evidence of life, either we may look for things that are probable but hard to detect or we may look for things that are improbable but easy to detect. In deciding what to look for, detectability is at least as useful a criterion as probability. Primitive organisms such as bacteria and algae hidden underground may be more probable, but freeze-dried fish in orbit are more detectable. To have the best chance of success, we should keep our eyes open for all possibilities.
A similar logic suggests warm-blooded plants as a reasonable target in the search for life on the surface of Mars. By "warm-blooded'' I do not mean that the plant will have a circulatory system or a precise temperature control. I mean only that the plant will be able to keep its internal temperature within the normal range of a cool greenhouse, roughly freezing to 80° Fahrenheit. Any form of life that survived on Mars from the early, warm and wet era to the present, cold and dry era had two alternatives: either it adopted an entirely subterranean lifestyle, retreating deep underground to places where liquid water could be found, or it remained on the surface and learned to protect itself against cold and dryness by growing around itself an insulating greenhouse to maintain a warm and moist environment. The first alternative is more likely but would be much more difficult to detect. Organisms living deep underground, without sunlight, would probably be microscopic, like the bacteria that live deep in the earth. To find such organisms would require deep drilling and heavy machinery. The second alternative, though less likely, would be easier to detect. The two missions that arrived at Mars this year -- the Pathfinder,landing on the surface, and the Global Surveyor, remaining in orbit -- are not intended to detect living greenhouses or other possible forms of life. Their purpose is to explore the planet in a general way and to raise questions that later missions could answer.
Many species of terrestrial plants, including the skunk cabbage that sprouts in February in the woods of Princeton, New Jersey, where I live, are warm-blooded to a limited extent. For about two weeks the skunk cabbage maintains a warm temperature by rapidly metabolizing starch stored inside the part of its anatomy known as the spadix, which contains the hidden flowers with their male and female structures. According to folklore, the spadix is warm enough to melt snow around it. The evolutionary advantage of warm-bloodedness to the plant is probably that it attracts small beetles or other insects that linger in the spadix and pollinate the flowers. The spadix is not a greenhouse, and the supply of starch is not sufficient to maintain a warm temperature year-round. No terrestrial plants are able to stay warm through an Arctic winter. On Earth polar bears can flourish in colder climates than trees can. It seems to be an accident of history that warm-blooded animals evolved on Earth to colonize cold climates, whereas warm-blooded plants did not. On Mars plants might have been pushed to yet more drastic adaptations.
Plants could grow greenhouses (so far the idea remains a theory) just as turtles grow shells and polar bears grow fur and polyps build coral reefs in tropical seas. These plants could keep warm by the light from a distant Sun and conserve the oxygen that they produce by photosynthesis. The greenhouse would consist of a thick skin providing thermal insulation, with small transparent windows to admit sunlight. Outside the skin would be an array of simple lenses, focusing sunlight through the windows into the interior. The windows would have to be small, to limit the loss of heat from outward radiation. The plant would also need deep roots, to tap water and nutrients from warmer layers underground. Inside the greenhouse the plant could grow leaves and flowers in an oxygen-containing habitat where aerobic microbes and animals might also live. Groups of greenhouses could grow together to form extended habitats for other species of plants and animals. An attendant community of microbes and fungi might help the plants to extract nutrients from the local ice or soil. Pores in the outer skin of the greenhouse might open to admit carbon dioxide from the atmosphere outside, with miniature airlocks and cold traps to keep losses of oxygen and water to a minimum.
If warm-blooded plants exist on Mars, they may or may not be easy to see. We cannot predict whether they would stand out from their surroundings in a visual or photographic survey. Two clues to their presence would almost certainly be detectable: leakage of heat and leakage of oxygen. Neither thermal insulation nor atmospheric containment is likely to be perfect. If we looked for heat radiation from anomalously warm patches on the Martian surface at night, or for anomalous local traces of oxygen in the atmosphere in daytime, we might find places where warm-blooded plants are hiding.
That we might find warm-blooded plants living wild on Mars or elsewhere in the solar system, it must be admitted, is only a remote possibility. It is much more likely that we will find Mars to be sterile, or inhabited only by subterranean microbes. In that case warm-blooded plants could be important in a different way -- as a tool for human settlement. I now leave the subject of science to talk about human space travel and colonization. Human colonization of the solar system would not be primarily a scientific enterprise; it would be driven by motivations that go far beyond science.
HE space-shuttle program was in trouble even before the Challenger accident, because it was based on a confusion of aims. It was trying both to open the way to human adventure in space and to serve as a practical launch system for scientific, commercial, and military missions. As I saw when I climbed into the shuttle at Houston, the two aims were never compatible. The shuttle was the result of a political compromise between people who wanted a reliable freight service into space and people who wanted to keep alive the tradition of the manned Apollo missions to the Moon. No single vehicle could do both jobs well. The shuttle tried, but it was too expensive for the first and too limited in its performance for the second.
In the future the two aims of the space program will be pursued separately. The twenty years since the birth of the shuttle have seen spectacular progress in the technologies of data processing, remote sensing, and autonomous navigation. With these technologies almost all the practical needs of science, commerce, and national security are now better served by unmanned missions than by the shuttle.
The future shape of a manned program pursuing idealistic aims is the great unknown. The shuttle is inadequate as a vehicle for human adventure. It resembles a Greyhound bus rather than a Land Rover. Another spending spree like the one for Apollowould be inadequate, even if it were politically possible. Does the manned space program have a future?
The confusion of aims afflicting the space program from the beginning was in essence a confusion of time frames. The practical aims of scientific and military activities in space made sense in a time frame of ten years; the basic technology for unmanned space missions took only ten years to develop. The aim of opening the skies to human exploration and adventure makes sense in a time frame of a hundred years -- what it will probably take to develop the technologies needed for significant numbers of human explorers to roam space at a price that earthbound citizens will consider reasonable. The Apollo missions, tied to a ten-year time frame, gave a false start to human exploration. They were far too costly to be sustained, and at the end of the ten-year program they had reached a dead end. If it had been made clear from the beginning that manned exploration would be a hundred-year program -- one with a stable and affordable budget -- we might by now have a light two-passenger spacecraft instead of the shuttle. We might already have a few people learning how to live permanently on the Moon, using only local resources.
We are now at the beginning of a revolution in space technology, when for the first time cheapness will be mandatory. Missions that are not cheap will not fly. This is bad news for space explorers in the short run and good news in the long run. Finally cheapness has a chance. Missions to the planets have been few and far between in the past ten years because they became inordinately expensive; they were expensive because of an imbalance in funding between ground-based and space-based science. For thirty years it was easier politically to obtain ten dollars for a space-science mission than to obtain one dollar for astronomy on the ground. The unfair competition injured both parties, starving ground-based astronomy and spoiling space science. The injury to space science was greater: ground-based astronomy flourished in spite of starvation, while planetary missions almost came to a halt in spite of big budgets. The rules are now changing, in the direction of fair competition between ground and space. This means that in the future space missions will be cheap. Once the barrier of high cost is broken, missions will be more frequent and the pace of discovery will be faster.
The essential first step in making either unmanned or manned operations cheap is to eliminate the standing army of people at Mission Control who take care of communication with spacecraft day after day. Spacecraft and the instruments they carry must become completely autonomous. The second step is to develop new technologies for launching payloads into space cheaply. The current Mars missions are making only small steps toward these goals. The coming era of cheap space operations will begin with unmanned missions, which will exercise the new technologies of propulsion and operation. Cheap manned missions will come later. Cheap unmanned missions require only new engineering; cheap manned missions will require new biotechnology. The chief problem for a manned mission is not getting there but learning how to survive after arrival. Surviving and making a home away from Earth are problems of biology rather than of engineering.
No law of physics or biology forbids cheap travel and settlement all over the solar system and beyond. But it is impossible to predict how long this will take. Predictions of the dates of future achievements are notoriously fallible. My guess is that the era of cheap unmanned missions will be the next fifty years, and the era of cheap manned missions will start sometime late in the twenty-first century. The time these things will take depends on unforeseeable accidents of history and politics. My date for the beginning of cheap manned exploration and settlement is based on a historical analogy: from Columbus's first voyage across the Atlantic to the settlement of the Pilgrims in Massachusetts was 128 years. So I am guessing that in 2085, 128 years after the launch of the first Sputnik, the private settlement of pilgrims all over the solar system will begin.
HE main lesson I draw from the history of space activity in this century is that we must clearly separate short-term from long-term aims. The dream of expanding the domain of life from Earth into the universe makes sense only as a long-term goal. Any affordable program of manned exploration must be centered in biology, and its time frame tied to the time frame of biotechnology; a hundred years, roughly the time it will take us to learn to grow warm-blooded plants, is probably reasonable. The people who decide to go to Mars or Europa will know whether or not indigenous life exists there. If it does exist, they will know how to nurture and protect it when they come to build their own habitats. If it does not, they will bring new life to make good nature's lack.
The most important part of their baggage will be the seeds of plants and animals genetically engineered to survive in an alien climate. On a world that has only a thin atmosphere, like Mars, or no atmosphere at all, like Europa, the most useful seeds will be the seeds of warm-blooded plants. After a hundred years of development of genetic engineering, we will know how to write the DNA to make plants grow greenhouses. Plants as large as trees could grow greenhouses big enough for human beings to live in. If the human settlers are wise, they will arrive to move into homes already prepared for them by an ecology of warm-blooded plants and animals introduced by earlier, unmanned missions.
Warm-blooded plants will not by themselves solve all our problems. The essential requirement for a successful human colony will be a deep understanding of the local ecology, so that human beings can become a part of it without destroying it. The Biosphere 2 experiment in Arizona, in which eight people tried unsuccessfully to live in a closed ecology for two years, was not a failure but a valuable object lesson. It taught us how human beings without sufficient understanding of their habitat could unexpectedly run out of oxygen.
Why should anybody wish to live on Mars or Europa? The only answer we can give to this question is the answer that George Mallory gave to the question why he wanted to climb Mount Everest: "Because it is there." There may be economic, scientific, or sentimental reasons attracting people to remote places; people always have a variety of reasons for moving from one place to another. One of the few constant factors in human history is migration, often over huge distances for reasons that are difficult to discern. I have little doubt that as soon as emigration from Earth becomes cheap enough for ordinary people to afford, people will emigrate. To make human space travel cheap, we will need advanced biotechnology in addition to advanced propulsion systems. And we will need a large number of travelers, to bring down the cost of a ticket. These are the reasons human space travel will not be cheap until fifty or a hundred years have gone by.
The most important fact about the geography of the solar system is that the habitable surface area is almost all on small objects -- asteroids and comets -- rather than on planets. Planets have most of the mass but very little of the surface area. Asteroids are usually rock, and orbit in the inner part of the solar system, inside the orbit of Jupiter. Comets are usually ice, and orbit in the outer part of the solar system, farther from the Sun than Neptune. Comets of average size are visible from Earth only on the rare occasions when gravitational perturbations cause them to fall close to the Sun and their volatile surfaces boil off to form bright tails in the sky.
Comets are more significant than asteroids in the ecology of the solar system, and a huge swarm of them can be found in a ring-shaped region called the Kuiper Belt, outside the orbit of Neptune. Only in the past few years have some of the largest Kuiper Belt objects been seen, first with ground telescopes in Hawaii and more recently with the Hubble Space Telescope. According to my rough estimate, the total surface area of the trillions of objects in the Kuiper Belt is about a thousand times the area of Earth.
Why are comets of greater ecological interest than asteroids? First, they are vastly more numerous. Second, ice is better than rock as a basis for life, and comets contain not only ice but also most of the other chemical elements that are essential for biology. Third, the orbital speeds of comets are much slower than the speeds of asteroids. The Kuiper Belt may seem to us today to be a cold and inhospitable place, but it is probably less inhospitable to life than Mars. It has the advantage of being an archipelago -- a collection of small, habitable islands not too far apart from one another. Because their relative speeds are slow, communication and travel between islands would be easy. If you were living on a mile-wide comet in the Kuiper Belt, another mile-wide object would pass by within a million miles about once a month, on average. Objects a hundred yards wide would pass by within this distance every day. It would take only a few days, using a small spacecraft with a modest propulsion system, to hop over and visit neighbors or replenish supplies. If you were bored by the scenery or unhappy with your family, you could move permanently and try your luck on another comet, just as colonists moved to Providence and places west.
If a community occupying a Kuiper Belt object outgrew its habitat and wished to expand, it could increase its living space by attaching tethers to neighboring objects as they floated by. A metropolis could grow in the twenty-second century by the accretion of objects as rapidly as Chicago or San Francisco grew in the nineteenth by the accretion of real estate. A Kuiper Belt metropolis would probably be a flat, disk-shaped collection of cometary objects, linked by long tethers and revolving slowly around the center to keep the tethers taut. To continue the accretion of desirable properties while avoiding destructive impacts, a metropolitan border patrol would engage in an interesting game of celestial billiards, tracking approaching objects with telescopes, nudging them gently with space tugs, and hooking them with tethers.
Recently the inhabitants of Earth have become aware that our planet is exposed to occasional impacts of asteroids and comets that may cause worldwide devastation. The most famous such impact occurred 65 million years ago, in Mexico, and may have been responsible for the demise of the dinosaurs. During the next hundred years, as the technologies of astronomical surveillance and space propulsion move forward, it is likely that active intervention to protect Earth from future impacts will become feasible. We may see a mutually profitable merger of the space-science enterprise with the business of planet protection. The cost of protection would be modest, provided that the warning time before an impact was as long as a hundred years. To deflect an orbiting object enough to cause it to miss Earth, a slow, steady push applied by a solar-powered engine would be much more effective than a nuclear explosion. With a hundred-year warning time the power required for the steady push would be only about two kilowatts for an average-size comet, with a mass of a billion tons. Two kilowatts is power on a human, not an astronomical, scale. Even as far from the Sun as the Kuiper Belt there is enough power in sunlight to supply two kilowatts with a solar collector of reasonable size. Once human communities were established in the Kuiper Belt, their border patrol would be in a position to offer its services to Earth, to detect objects that threatened to collide with Earth and deflect them in timely fashion at minimal cost.
Another service that Kuiper Belt communities might provide for human beings on Earth is scientific exploration. The belt contains enough unknown objects to keep explorers busy for thousands of years. The comets are cold and ancient enough to preserve detailed records of the formation and early history of the solar system. It is likely that we would find objects there that are older than the Sun. It should be possible to trace the history of our system back into the pre-solar era.
It could well happen that within a few hundred years most of the inhabitants of the solar system will be living in the Kuiper Belt. Accustomed as we are to living on a high-gravity planet close to the Sun, it is difficult for us to imagine what it would be like to live with low gravity far away. One of the first steps a human colony would take to establish itself in the Kuiper Belt would be to surround its cometary habitat with an extended efflorescence of mirrors in space to collect sunlight. An array of mirrors sixty miles in diameter could collect a steady thousand megawatts of energy anywhere in the Kuiper Belt, out to three times the distance of Neptune from the Sun. That is enough energy to sustain a considerable population of plants, animals, and human beings with all modern conveniences. The mirrors would not have to be optically perfect. The material out of which to construct them, a few thousand tons of metal or plastic, would probably be available on any Kuiper Belt object. After a century of progress in biotechnology we would not need to manufacture the mirrors. We would teach our plants to grow them.
Life in the Kuiper Belt would be different from life on Earth, but not necessarily less beautiful or more confined. After a century or two there would be metropolitan centers, cultural monuments, urban sprawl -- all the glories and discontents of a high civilization. Soon restless spirits would find the Kuiper Belt too crowded. But there would be an open frontier and a vast wilderness beyond. Beyond the Kuiper Belt lies a more extended swarm of comets -- the Oort Cloud, farther away from the Sun and still untamed.
Freeman J. Dyson is the president of the Space Studies Institute, and was until his retirement a professor of physics at the Institute for Advanced Study, in Princeton, New Jersey.
Illustrations by Giacomo Marchesi
Copyright © 1997 by The Atlantic Monthly Company. All rights reserved.
The Atlantic Monthly; November 1997; Warm-Blooded Plants and Freeze-Dried Fish; Volume 280, No. 5; pages 71-80.