After the widely publicized 1994 impact of Comet Shoemaker-Levy 9 with Jupiter, Congress asked NASA to work with the US Air Force Space Command to develop a program plan to accelerate the timetable for NEO search and discovery, and we participated in the joint NASA/USAF study team led by Gene Shoemaker. In the same year the Air Force launched the first of its Clementine series of military spacecraft to near-Earth asteroids, while NASA moved toward a 1996 launch of its own Near Earth Asteroid Rendezvous mission. In this chapter we discuss the outcome of the Shoemaker team report and the continuing relationship between NASA and the Air Force, as both potential partners sought to define their role in asteroid searches and planetary defense. We describe the various asteroid spacecraft missions and the new proposals for less expensive ways to carry out the Spaceguard Survey. Meanwhile the US military dropped all reference to nuclear solutions, presumably in deference to public concerns about nuclear weapons in space, and proposed a formal planetary defense mission for the US Air Force. This chapter brings us up to date on where each of these programs stands, as we prepare for a new round of congressional testimony in 1997.
CHAPTER 15: THE PUBLIC POLICY DEBATE If a big impact happens, it would be an unprecedented human catastrophe. There's something like one chance in two thousand that such a collision will happen in the lifetime of a newborn baby. Most of us would not fly in an airplane if the chance of crashing were one in two thousand. (In fact, for commercial flights the chance is one in two million. Even so, many people consider this large enough to worry about, or even to take our insurance for). When our lives are at stake, we often change our behavior to arrange more favorable odds. Those who don't tend to no longer be with us. -- Carl Sagan, in Pale Blue Dot (1994) Recognition of the NEO hazard has influenced far more than academic scientists, as we have seen in the previous chapters of this book. Compared with other natural hazards, impacts are quite remarkable. They are the only known natural hazard that can threaten the lives of billions of people and the survival of civilization. They are also the only major hazard that can, at last in principle, be avoided by the judicious application of technology. Many issues of public policy are associated with our new awareness of impacts. Is there a government responsibility to protect us against impacts? How much should be spent on this effort? Should NEO defenses be a national or an international responsibility, and should they be in the hands of civilian or military agencies? To address these questions, we must also inquire into public reactions to this newly-identified danger, and to how well we are communicating these issues to the citizens of the United States and the world.
Most of the scientific debates we have been discussing took place behind closed doors, involving only a few dozen technical experts. However, as the impact hazard attracted press interest, rumors of the conflicts between the astronomers and the weapons scientists inevitably leaked. In some cases, it was these conflicts, more than the impact hazard itself, that became the object of press coverage. The bitter 1992 Los Alamos confrontation between NASA astronomers and star warriors attracted a good share of press interest, especially after reporters were banned from the meeting itself. A week after Los Alamos, The New York Times printed an op-ed piece by physicist Robert Parks called "Star Warriors on Sky Patrol", which supported the astronomers but criticized the motives of Edward Teller and Lowell Wood, asking "who will protect us from the 'Nukes Forever!' mentality?". Then a front-page story in The Wall Street Journal titled "Never mind the peace dividend, the killer asteroids are coming!" falsely claimed that NASA "is about to recommend that the Earth start planning to assemble an arsenal of nuclear missiles to head off an attack by asteroids". These and similar press stories contributed to NASA's decision to dissociate itself from the Los Alamos meeting and focus on non-defense aspects of NEOs.
Some commentators saw a different issue involving turf conflicts between government agencies. In a long article published in March 1992, reporter Fran Smith of the San Jose Mercury News (serving the high-tech community of California's Silicon Valley) suggested that the asteroid astronomers had "inadvertently walked into a battle over NASA's fate" when they "seized on the public fascination with dinosaurs to stir interest in the lowly field of asteroid astronomy". She speculated that the debate between the astronomers and the star warriors reflected a broader struggle in Washington, over the survival of NASA's civil space program in the face of challenges from the Department of Defense to militarize space. She wrote that some asteroid astronomers were already backpeddaling, expressing their concern that proliferation of space weapons was more dangerous than the asteroid danger these weapons were supposed to counter. Clark Chapman told her that the projects proposed at Los Alamos were "Outrageous. Not outrageous because they're technically wrong. Outrageous because they're so radically more expensive and dangerous than the modest threat they would address."
On the whole, the American news media in the 1990s have treated the NEO hazard issue fairly and given it serious if inconsistent attention. The few really negative comments came primarily from the political right. Both the Washington Times and the Wall Street Journal editorialized that the impact "threat" was a ploy invented by astronomers to obtain more research funds from the government. In Europe, however, press coverage of impact issues tended toward wild exaggeration, alienating public support and frustrating astronomers who wanted to have their message heard.
After Shoemaker-Levy, the danger from NEO impacts was no longer treated as an unproven hypothesis. In fact, public interest in the comet crash stimulated production on several major television shows, both factual and fictional, that were broadcast in 1996 and 1997. The first documentary (name) was produced by the Public Television series NOVA, followed by a production (On a Collision Course with Earth) broadcast on the new Science Fiction cable channel. Much more widely viewed were a 2-hour production (Three Minutes to Midnight) broadcast on the Discovery Channel and a 1-hour National Geographic Special on NBC, both appearing in February 1997. Each of these shows presented balanced and technically accurate pictures of the role of impacts in Earth history, past and present, much of it described by scientists who were directly involved in researching these issues.
Probably even more widely seen, however, was the 4-hour NBC miniseries Asteroid, which depicted the destruction of Dallas by the impact of an asteroid several hundred meters in diameter. In spite of advertizing hype, this show turned out to be technically weak and nearly devoid of drama or character. In apparent response, the TNT network resurrected the 20-year-old film Fire from the Sky, in which a comet wipes out Phoenix. Both plots deal with the local effects of Tunguska-like impacts, not with a global environmental disaster. Even if these dramatizations were widely criticized, it is unlikely that many people remain in the US in 1997 who have never heard of the possibility of an impact catastrophe. Whether they consider such a risk credible is, of course, a different issue. The boundaries between fact and fiction are frequently blurred on American television.
Even among people who were aware that science recognized a threat from the sky, however, the actual risk seemed extremely small. Indeed, of all the hazards we face, an impact is the least likely to happen in any given year. We are concerned, not because a big impact is likely, but because each large impact can produce extremely severe consequences, in which hundreds of millions of people could die. It is the combination of low probability with high mortality that sets this risk apart from others.
There is a discipline of social science that deals with risk perception. In the late 20th century, American society has become preoccupied with risk evaluation, and many political controversies are based on perceived risk, in issues as diverse as the health effects of breathing second-hand tobacco smoke, the value of requiring motorcyclists to wear crash helmets, or the hazard of constructing nuclear power plants. Most issues involving product testing, drug screening, and environmental impact statements deal with risk and risk-perception. We have become a risk-averse society, and the study of how we deal with hazards is a field with tremendous practical as well as academic implications.
One of the pioneers in academic risk perception is Paul Slovic, a professor at the University of Oregon. We first met him in 1992, when he was speaking at the annual meeting of the American Association for the Advancement of Science. This organization, which publishes the widely quoted journal Science, is one of the few that brings together practitioners of the vast panoply of disciplines and subdisciplines that make up the contemporary scientific enterprise. Its annual meetings span the social, biological, and physical sciences, with talks presented so that the average educated layperson or scientists from another discipline can understand.
Following the planetary science presentations (organized to honor the explorations of the Americas begun 500 years earlier by Columbus), we wandered into a symposium on public perception of risk in the nuclear power industry. One soft-spoken man in the front row seemed to be the authority to whom other participants frequently turned. This was Paul Slovic, whose name was vaguely familiar to us from an article of his we had seen in Science. In that paper, Slovic had asserted that how people react to various hazards depends as much on their perception of the immediacy or dread of the danger as on its actual calculated risk. At that time Slovic had not heard of the cosmic impact risk, but we struck up an acquaintance that led to our collaboration in jointly studying this aspect of the impact issue.
For Slovic and other social scientists, the impact risk provided a fascinating new area for research. Discovering a new risk was analogous, for them, to the discovery by astronomers of a new celestial object or physical phenomenon. Starting to work now, before many people were aware of this risk, they had a unique opportunity to track this issue and gauge what influenced public opinion and political action as the impact hazard became better known and more widely discussed.
In a paper he wrote with us in 1993, Slovic asked how the public might respond to the statistical threat of impact, absent a specific prediction. He noted reasons from his previous studies to predict lack of concern about this issue, on the one hand, and -- for different reasons -- the possibility of a high degree of concern. There were several reasons Slovic expected lack of concern and possible opposition to large expenditures. First, natural hazards (perhaps including impacts) tend to be less frightening than mysterious Dr. Strangelovian hazards of modern technology; people perceive nature as fairly benign, and they react rather apathetically to the threat from natural hazards. For example, despite the very real and rather obvious hazards from earthquakes and floods, people continue to build homes along the San Andreas fault and in the floodplains of streams and rivers.
Moreover, the probabilities of an impact may be too low and the risk apparently too remote in time to trigger concern, in spite of their high consequences. Chances of "one in a million" (and that's roughly the annual danger from major impacts) mean that the chances are essentially zero, in the popular meaning of the phrase. Since such chances of impact are often phrased in terms like "every million years" or "every hundred thousand years," people mistakenly think that it means that it won't happen until a hundred thousand (or a million) years have elapsed, not realizing that an impact could as well happen tomorrow as on any particular day in the distant future.
Also, people are often insensitive to very large losses of life. Maybe it is because we are overwhelmed and numbed by stories of thousands (or millions) dying in a war or famine. Or maybe we simply can't relate to such large numbers. But we will anguish over the death of a single child or expend great effort to find an organ donor to save an individual life, but in a context of impersonal numbers or statistics, the lives of individuals lose meaning. Thus many people may respond to our words about the prospects of a billion people dying pretty much the same as if we had said that "only" a million might die.
Finally, Slovic considered that people might be unimpressed by the prospects that we could reduce the chances of an impact. We have noted that the Spaceguard Survey, and even all that the Defense Department might muster, could do nothing to save us from the 20% chance of a big impactor being a comet rushing at us from the edge of the solar system. While experts might value the protection from the 80% of doomsday rocks that we could find long ahead of time, most people tend to expect -- however unrealistically -- 100% insurance against a threat: if impact defense systems cannot provide complete protection, they may be undervalued.
On the other hand, there are also reasons Slovic expected the public to be concerned enough about impact hazards to support action. First, the risk is demonstrable (it happened to the dinosaurs), the news media showed comparable events happening on Jupiter a few years ago, and the reality of the hazard is endorsed by credible scientists. Moreover, the potential consequences of large impacts are uniquely catastrophic and are qualitatively different from other natural hazards. Catastrophic potential, whether from a nuclear melt-down or an Act of God, is an element of hazard that increases public concern. Indeed, the hazard is new in the public consciousness, it strikes (unless we do more) without warning from the skies, and it is (at present, at least) uncontrollable. Lack of control, dread, and catastrophic potential are all qualities of other risks studied by Slovic that are associated with public perception of high risk and a resulting strong desire for action to reduce risk.
Finally, there is the simple, rational fact that the probabilities of catastrophic impacts, while small, are not trivial. Considerable public funds are already being spent to deal with hazards with even lower annualized rates of deaths, such as death or injury from tornadoes or terrorist attacks.
At the end of 1992 (before the discovery of Comet Shoemaker-Levy 9), Slovic carried out a preliminary survey of attitudes and perceptions using a sample of 200 college students, 75% of whom had not heard about the impact hazard prior to participating in the study. He found that after reading about the impact risk, his group ranked impacts near the middle of a range of possible dangers faced by the American public. The impact risk was judged higher than risks from prescription drugs, medical x rays, bacteria in food, floods, and air travel, but lower than risks from earthquakes and hurricanes. Impact risks were rated as extreme with regard to being unknown to scientists and the public, distant in time (non-immediate), uncontrollable, and catastrophic. There was modest support for detection efforts but considerable opposition to use of weapons in space, even to deflect a threatening asteroid. Slovic also concluded that the scientists who are expressing concern about this threat were considered credible by his survey group, suggesting that the media have treated these activities in a positive light and have not interpreted these public statements as particularly ill-founded or self-serving. This feeling applied to the astronomers who had dominated media coverage to that point, but it might not be true of the nuclear scientists who want to build asteroid defense systems.
Perhaps unfortunately, asteroid defense issues inevitably become entangled with public concerns about nuclear energy. The simple fact is that if we were faced with the prediction of an impact, nuclear energy provides us with the most effective, widely applicable tools for self-protection. There are no other options to deflect or destroy a typical incoming NEO, at least with present or easily projected technology. (Small objects, especially with long lead times, could be dealt with using more benign approaches. But it is difficult to get around the fact that nuclear weapons possess about a million times the power of equivalent chemical devices, per pound, and we are limited in how much sheer weight we can launch into space to meet an intruder.) And, through all of the other hazard discussions, Edward Teller has held fast to his conviction that we need to pursue a program of nuclear testing in space to learn how to deflect or destroy asteroids and comets. Repeatedly, the Father of the H Bomb argued that science depends on experimentation (meaning that we ought to blow up an asteroid to see what happens), and that we must not trust our future survival to an untried technology. Teller, with his great prestige, would not let the nuclear issue be swept under the rug.
One astronomer who was deeply disturbed by these proposals was Carl Sagan, who had written about the hazard of comet impacts in his 1985 book Comet. Sagan, probably the best-known scientist in the U.S., began to participate in the hazard debate in 1993 and remained active until his untimely death in 1996. Unlike Teller, Sagan had never known political power, but he compensated with a commanding public presence. Teller tells a story about having breakfast with Sagan in an airport, where three people approached to ask for Sagan's autograph, but no one asked for one from Teller. Teller remarked "Sagan won." In addition to his own distinguished career as a planetary researcher, Sagan had taught a generation of planetary scientists, and both of us had been his students in the 1960s, when he was a young Assistant Professor at Harvard.
Many participants in the impact hazard discussions have expressed concern about the potential danger of maintaining fleets of nuclear-tipped missiles as part of the space defense system. It seems obvious that the risks inherent in such a defense system -- risks from either accident or misuse -- might be greater than the danger of an asteroid impact, unless effective international controls are developed. But Sagan went a step farther. He questioned whether knowledge of the deflection technology itself might raise the possibility of someone, sometime, using it against the Earth. After all, he argued, the same technology that could deflect an asteroid from a collision course and save the Earth could also deflect an asteroid that would have missed the Earth to hit the planet. And there are many more "near misses" that could be deflected toward us than there are asteroids that must be deflected away to save the planet.
Sagan was not the first person to suggest that asteroids might be used as weapons. About a decade earlier an anonymous paper had circulated within the U.S. defense establishment called "Ivan's Hammer", suggesting that the Soviet military might develop a technology for aiming asteroids at the United States, raining far more megatonage upon us than could be achieved by a simultaneous attack with all of their missiles and warheads. This paper had also speculated that the Soviets might engineer such an attack secretly, blaming the resulting catastrophe on natural causes. At the time, no one took this "Ivan's hammer" threat seriously. Sagan, however, was quite serious about possible misuse of asteroid defection technology.
Sagan presented his arguments against developing deflection technology to a variety of audiences, including a mid-1994 article in Parade Magazine and as a chapter in his book Pale Blue Dot (published late in 1994). In collaboration with radar astronomer Steven Ostro of JPL, he also published these arguments as technical papers in Nature and the journal Issues in Science and Technology. Through these efforts, Sagan's arguments against asteroid defenses reached many people who had not previously been aware of the impact hazard issue in any form.
To make his point, Sagan turned to a story from ancient history. In about 560 BC the Sicilian city-state of Camarina was suffering from pestilence, which was attributed to the presence of marshlands that surrounded most of the city. At great expense, the rulers of Camarina undertook a public works project to drain the marsh and thereby improve the health of the citizens, surely a worthy and laudable project. The water was drained, and everyone was happy -- until the neighboring city of Syracusa recognized that the swamp, which had protected Camarina from invasion, was no longer present. When the attack came, the citizens of Camarina found themselves unable to defend the enlarged perimeter of the city; they were defeated, the city destroyed, and its citizens all killed. Thus for Sagan, the marsh of Camarina became a metaphor for the well-meaning misuse of technology, in which an act that seems to serve a worthy purpose backfires to produce disaster.
Sagan's basic issue is that the techniques for altering asteroid or comet orbits can work both ways: to deflect them away from the Earth or, in the wrong hands, to direct an otherwise benign projectile toward us. His focus in this concern was on objects large enough to cause a global environmental catastrophe. As we have noted, such impacts take place naturally only once every few hundred thousand years. But suppose, Sagan asked, that we completed the proposed Spaceguard Survey and identified all of these objects. For every one that can hit the Earth, there will be about 4000 others that come as close as the Moon, or about one of 1.7-km diameter every century. About once per decade an asteroid larger than 500 m in diameter comes as close as the Moon.
Sagan then asked us to consider a world in which the technology has been developed to modify asteroid orbits, perhaps by "herding" them through a series of small nuclear explosions. What are the chances that there will someday be a madman somewhere with access to such technology and a desire to destroy civilization? We have already had Stalin, Hitler, Pol Pot, and Idi Amin this century. Given the potency of ethnic and religious conflicts, it is surely possible that someone, sometime, should conclude that it was time for the Day of Judgement and be willing to help God achieve His apocalyptic objectives. If such a madman existed, he could find an inoffensive asteroid and turn it into a weapon with thousands of times the destructive power of all the world's nuclear arsenals. Presumably there will be few such opportunities and, we may hope, no such despots with the capability of taking such action. But can we be sure that this could never happen? Like the ancient Camarinans, we might create a greater hazard than the one we are protecting against. The cure might be worse than the disease. Sagan called this the "deflection dilemma".
Sagan's dreadful scenario was widely regarded as unlikely. Is the technology to deflect an incoming asteroid really sufficient to redirect an innocent asteroid into a collision course? The methods suggested so far to deflect asteroids by stand-off nuclear explosions are quite crude. We do not know with any confidence how large the explosion should be or how close to the asteroid; that is why we have considered the possible need for a series of explosions to accomplish the orbit change. Furthermore, the asteroids themselves are irregular in shape and often rapidly rotating, and it is impossible to predict with assurance just how one would respond to a nearby nuclear blast. We would be lucky if it went even approximately in the direction we wanted. Fortunately, it is not necessary to change the orbit in any particular way to protect the Earth. We really don't much care how much the orbit is changed; the body can pass to the east or the west, the north or the south, just so it doesn't hit us. We all know that it is easier to get out of a parking place than to get in. In this case the difference is much greater, and the contrast between missing the Earth and being directed toward any particular direction or target is huge.
Furthermore, the orbital change required to bring an asteroid from lunar distance into an impact trajectory is about a hundred times greater than that required to miss the Earth. If 5 bombs would be enough to accomplish the defensive deflection, it would require 500 to change an orbit from lunar distance to intersection. In this case defending the Earth from impact is a great deal easier than launching an offensive strike against the Earth, and the technology of deflection is probably not capable of steering a benign asteroid into a threatening orbit. Similarly, if a madman did manage by multiple small nudges to get an asteroid into a collision course, it would be relatively easy for a defender to give it a bigger, less precise impulse and get it out of the way.
So what was Sagan imagining? First, he suggested that someone will develop the technology to fine-tune an asteroidal orbit. Assuming that such a thing is possible using nuclear explosives, it would require much more sophisticated technology than the sort of defense system contemplated today. Second, he imagined that a madman gains control of this technology and convinces a large cadre of technical experts to join in this suicidal scheme. Third, he required that this group of conspirators launch dozens (or more) of warheads and carefully redirect an asteroid over a time scale of several years in order to shift it into an impact trajectory. Finally, he imagined that no other nation or organization exists on Earth to detect this activity and counter with a defensive countermeasure.
While it is difficult for us to take such a scenario seriously, it certainly cannot be proved to be impossible, or that another analogous scenario, which nobody has thought of yet but the madman might, could lead to the awful consequences that Sagan worried about. And that was Sagan's larger point. Unless we can be better than 99.9% sure that there isn't a viable madman scenario that could be used to turn asteroid deflection technology against us, then the dangers from the madman exceed the very remote possibilities of an asteroid striking an unprotected planet Earth. And how often are we 99.9% sure of anything, let alone about an esoteric possibility built of cutting-edge technology and the psychology of madmen?
Teller had a much shorter response when asked in the summer of 1994 about the possible misuse of deflection technology. Bristling, his voice rose as he responded: "Who could suggest such an idea? Only what's-his-name -- Sagan! I do not understand him. I think he must know many evil people to have such ideas!"
Sagan had his own opinions on Teller. Writing in his 1995 book The Demon-Haunted World, Sagan concluded that Teller has done a great deal of harm to society, and he interprets Teller's life-long advocacy of nuclear explosives as an appeal that "somehow, somewhere, he wants to believe [that] thermonuclear weapons, and he, will be acknowledged by the human species as its savior and not its destroyer".
In spite of his concern with the deflection dilemma, Sagan recognized the reality of the danger posed by impacts, and he struggled in his articles and lectures to find a suitable compromise. He wrote in Pale Blue Dot: "If we develop and deploy this technology, it may do us in. If we don't, some asteroid or comet may do us in. . . If we are too quick in developing the technology to move worlds around, we may destroy ourselves; if we are too cautious, we will surely destroy ourselves." He suggested, therefore, that we proceed at once with the Spaceguard Survey, to obtain a complete census of the larger near-Earth objects, while deferring the development of deflection technology to a time when either an immediate threat is identified or else maturing international safeguards allow us to invest in such technology safely. Optimistically, Sagan assumed that political institutions will improve and the risk of powerful madmen will be reduced to negligible levels within the next few centuries.
If the NEO issue is primarily seen as one of defending the Earth from catastrophic impacts, an obvious case can be made for the military to play a leading role. In 1994 a this idea received official recognition in a futuristic report from the US Air Force Air University called Spacecast 2000, prepared with the advice of industry and academic figures (including Carl Sagan). In their analysis of global power in the 21st century, the Air Force planners wrote that the military mission is to conduct "counterforce operations ... aimed at opposing or defending against threatening force anywhere on the planet or in space". The Spacecast 2000 document asserts that "although not a traditional enemy, the asteroids are nonetheless a threat that the DoD should evaluate and defend against. The role of the military has traditionally been to operate in and expand the frontier of space. ... Provisions for defense of the planet, as far away from the planet as possible, need to begin."
The Spacecast 2000 report was written at the same time the USAF was participating in the Shoemaker committee to study a new, streamlined Spaceguard Survey. Pete Worden and others wanted the Air Force to assume responsibility for NEO searches. This is an appealing idea, since the military have both the telescopes and the manpower to operate them in their existing GEODSS surveillance system. But there are concerns as well, primarily associated with issues of secrecy. Astronomers are by nature open in their communications with each other and the public, but the same cannot be said about the military, whose mission requires discipline and (on occasion) secrecy. It is interesting to speculate what the military would do if they found an NEO headed for the Earth.
In 1996, when Hollywood film-maker Paul Almond was researching a script on defense against asteroids, he asked a number of public officials, including former cabinet officers, how the government would react to the discovery of an impact threat. Several told him that they would favor keeping the discovery secret from the public while establishing a top secret, military-type organization to defend against the NEO. Authors of several earlier novels and filmscripts have made such a government cover-up a central part of their dramatization of an impact scenario. Yet we, and most of our fellow scientists, consider the idea of keeping the risk secret to be absurd, both morally and practically. It is impossible to imagine scientists such as Tom Gehrels, Gene Shoemaker, Steve Ostro, or Brian Marsden -- the people today most likely to discover a new NEO and calculate its orbit -- keeping this information to themselves. The skies are an open book, there for anyone to read. Nor can we imagine a civilian government agency such as NASA keeping such a secret for more than a few days, even if they wanted to. Dramatizations that depend on such a conspiracy for their plot elements are pure friction.
On the other hand, it is possible to imagine that a tightly-disciplined military organization like the USAF Space Command might be able to keep such a secret for quite a long time. We have no way of knowing whether they would want to, but some highly placed officials did tell Paul Almond that they would favor such a course. Perhaps this is a good reason to keep the Spaceguard Survey in civilian hands.
CONCLUSION STILL TO COME
5200 words (5/7/97)
CHAPTER 15: WHAT WE SHOULD DO: THE PUBLIC POLICY DEBATE
Most of the scientific discussions we have been describing took place out of the public eye. However, rumors of the conflicts between the astronomers and the weapons scientists leaked out, stimulating widespread press interest, which has grown steadily since 1992. Additional events have accelerated the public debate: the widely publicized impact of SL 9 with Jupiter, publication of nearly a dozen popular books, and several TV accounts (both fictional and factual) of impacts that were broadcast early in 1997. In this chapter we analyze the public and press reaction, including studies of the psychology of risk perception. We then turn to current policy issues. Carl Sagan originally raised the question of the unexpected consequences of a planetary defense system, in which the risks of the defenses themselves might be greater than the impact risk they are intended to mitigate. Astronomers and weapons scientists have argued over the relative importance of the very large impacts, compared with more common but smaller impacts like Tunguska. We also ask whether the defense of the planet should be in the hands of military or civilian agencies, and to what extent the effort should be international. When we look closely, we see that comets pose a much more difficult problem than asteroids, one that current technology may not be able to handle. There are difficult questions, but ultimately these issues will be faced. On the long run, we must learn to protect ourselves against comets and asteroids. Our very survival as a species depends on it.