The Atomic Origins of Climate Science

How arguments about nuclear weapons shaped the debate over global warming.
An imprint of the nuclear symbol in snow
“Nuclear winter” skeptics created institutions that later challenged global warming.Illustration by Paul Sahre

A nuclear weapon is a certain thing—atomic or hydrogen, fission or fusion, bomb or missile, so many megatons—but nothing could be more uncertain than the consequences of using one. Nine nations have nuclear weapons; only the United States has ever used one, and that was in 1945. Our nuclear-weapons policy rests on a seven-decade-long history of events that have never happened: acts of aggression that were not committed, wars that were not waged, an apocalypse that has not come to pass. Strategists attribute the non-occurrence—the deterrence—of these events to the weapons themselves, to bombs on airplanes, missiles in silos, launchers on submarines. The power of deterrence, however, is a claim that cannot be proved. If, while a police car is parked in front of your house, your house is not robbed, you might suspect that a robbery would have taken place had the police not been there, but you can’t know that for sure. Nuclear-weapons policy is a body of speculation that relies on fearful acts of faith. Doctrinally, it has something in common with a belief in Hell.

This belief is about to be tested. The United States and its only nuclear rival have been reducing their arsenals since the end of the Cold War. In 1985, the United States and the Soviet Union held a combined stockpile of more than sixty thousand warheads; today, the U.S. and Russia have fewer than fifteen thousand between them. Dangers remain: a computer error, a malfunction in a silo, a rogue state, nuclear terrorism. In 2007, the former Secretaries of State Henry Kissinger and George P. Shultz, the former Defense Secretary William J. Perry, and the former senator Sam Nunn warned in an essay published in the Wall Street Journal that “the world is now on the precipice of a new and dangerous nuclear era . . . that will be more precarious, psychologically disorienting, and economically even more costly than was Cold War deterrence.” The time had come, they argued, for the eradication of nuclear weapons. Global Zero, an international nuclear-abolition organization, was formed the next year. In 2009, in a speech in Prague, Barack Obama pledged “America’s commitment to seek the peace and security of a world without nuclear weapons.” That speech helped earn him the Nobel Peace Prize. The reality did not match the rhetoric. Congress exempted nuclear weapons from mandatory cuts in the military budget, and, in exchange for support for the New START treaty, which reduced deployed warheads by two-thirds, the President pledged eighty-five billion dollars to modernize the nation’s aging nuclear arsenal. “I think we can safely say the President’s Prague vision is dead,” the Alabama congressman Mike Rogers said in 2015, at a weapons conference sponsored by Lockheed Martin, General Dynamics, and Northrop Grumman. “And I’ll leave it to the Nobel committee to ask for its prize back.”

The new President’s vision is unclear. “We have to be extremely vigilant and extremely careful when it comes to nuclear,” Donald Trump said during the campaign. But he also refused to rule out conducting a first strike, even on Europe (“Europe is a big place”); suggested that it might be a good thing for more countries to acquire nuclear weapons; and argued that it was pointless to manufacture weapons that could never be used, asking, “Then why are we making them?” In December, Vladimir Putin told military leaders in Moscow that he intended to bolster Russia’s nuclear arsenal. “Let it be an arms race,” Trump said in response. “We will outmatch them at every pass and outlast them all.”

Trump has often contradicted himself on the subject of nuclear weapons, but one of the more interesting things he’s said about them is that they are far more dangerous to the planet than global warming is. It’s a revealing comparison. The damage from a nuclear explosion does not respect national boundaries, and this adjustment in scale, from the national to the global, was the key argument put forward by advocates of disarmament. That argument has been won: Trump’s tweets aside, there is a bipartisan consensus in favor of significant arms reductions. Bipartisan agreement about the future of the planet falls apart not over the bomb but over the climate. Historically, though, they’re inseparable: the weapons and the weather are twisted together, a wire across time, the long fuse to an ongoing debate about the credibility of science, the fate of the Earth, and the nature of uncertainty.

In 1981, when Jimmy Carter delivered his farewell address, part of it was written by Carl Sagan. The Senate had proved unwilling to ratify a treaty that had come out of a second round of Strategic Arms Limitation Talks; Carter wanted to take a moment to reckon with that loss, for the sake of the planet. He turned to Sagan, whose thirteen-part documentary, “Cosmos,” first broadcast in 1980, was the most-watched PBS series ever. “Nuclear weapons are an expression of one side of our human character,” Carter said, in words written by Sagan. “But there’s another side. The same rocket technology that delivers nuclear warheads has also taken us peacefully into space. From that perspective, we see our Earth as it really is—a small and fragile and beautiful blue globe, the only home we have.”

Sagan was an astronomer, but early in his career he’d worked on a classified nuclear-weapons project. This was not unusual. Since the Second World War, the military has funded the preponderance of research in the field of physics, and, as historians have now established, a close second was its funding of the earth sciences. Although the environmental movement may not have started until the nineteen-sixties, the research that lies behind it began in the fifties, in the U.S. military. Indeed, the very term “environmental science” was coined in the fifties by military scientists; it was another decade before civilian scientists used the term.

Beginning on the day black rain fell on Hiroshima, nuclear weapons shaped environmental science. In 1949, the U.S. Weather Bureau launched Project Gabriel, a classified meteorological study of weapons and weather. The next year, the Department of Defense, in a study titled “The Effects of Atomic Weapons,” coined the word “fallout.” Researchers considered making the quantity, spread, and duration of fallout the standard measure of the force of a nuclear explosion, but found that approach to be too dependent on the weather. (Instead, they chose blast radius.) They measured and modelled the best weather conditions for explosions and the effects of those explosions on the natural world; they invented and refined tools to detect atmospheric weapons tests conducted by the Soviets; and they investigated the possibility of using nuclear weapons to alter the weather and even the climate of adversaries. Sagan, after finishing his Ph.D. at the University of Chicago, in 1960, worked on a secret military project code-named A119, which had begun in 1958, a year after Sputnik. Sagan was charged with calculating “the expansion of an exploding gas/dust cloud rarifying into the space around the Moon.” The idea was to assess whether a mushroom cloud would be visible from Earth, and therefore able to serve as an illustration of the United States’ military might.

Government-funded environmental scientists began noticing something curious: nuclear explosions deplete the ozone layer, which protects the Earth’s atmosphere. This finding related to observations made by scientists who were not working for the military. In the wake of Rachel Carson’s “Silent Spring,” published in 1962, the U.S. government formed a number of advisory and oversight organizations, including the Environmental Pollution Panel of the President’s Science Advisory Committee. The panel’s 1965 report, “Restoring the Quality of Our Environment,” included an appendix on “Atmospheric Carbon Dioxide,” laying out, with much alarm, the consequences of “the invisible pollutant” for the planet as a whole. In 1968, S. Fred Singer, an atmospheric physicist who had worked on satellites and was now a Deputy Assistant Secretary of the Interior, organized a symposium on “Global Effects of Environmental Pollution.” Four papers were presented at a panel on “Effects of Atmospheric Pollution on Climate.”

Changing weapons policy opened new avenues of research. In 1963, the U.S. and the U.S.S.R. signed the Limited Test Ban Treaty, an agreement to stop testing nuclear weapons in the atmosphere. Using longitudinal data to study the ozone both before and after the test ban, the Berkeley chemist Harold Johnston found that stopping the testing had slowed the depletion. Research into the environmental consequences of nuclear explosions and of other kinds of pollution shared a planetary perspective, a vantage greatly enhanced by the space program; gradually, the meaning of the word “environment” changed from “habitat” to “planet.” The first photograph of the whole Earth was taken in 1972, by the crew of Apollo 17. It became an icon of the environmental movement. It also shaped arguments about arms control.

Nuclear-weapons research was usually classified; other environmental research was not. During the nineteen-seventies, military-funded environmental scientists continued their top-secret research into the environmental effects of nuclear weapons. Given the test ban, these studies relied less on experiments on Earth than on computer models and on empirical findings involving dust on other planets, most notably Mars. Meanwhile, some environmental scientists pursued—and published—research on how chlorofluorocarbons, the exhaust from jet engines, and fossil-fuel consumption affected the ozone layer; this research demonstrated, crucially, that even tiny amounts of certain chemicals could catalyze dramatic changes, with planetary consequences. In 1974, the director of the U.S. Arms Control and Disarmament Agency asked the National Academy of Sciences to prepare a report on the effects of nuclear explosions on the ozone. That report, “Long-Term Worldwide Effects of Multiple Nuclear Weapons Detonations,” married the logic of nuclear deterrence to the logic of environmental protection, or what might have been called pollution deterrence. Stephen Schneider, a climatologist at the National Center for Atmospheric Research, speculated that the fallout from a nuclear war might make the world colder by blocking sunlight, and that the diminished industrial production in a postwar world could change the climate, too.

“Look, if I have to explain the meaning of existence, then it isn’t funny.”

Sagan had by this time become an advocate of nuclear disarmament, a cause that gained a great deal of momentum early in 1982, when The New Yorker published a four-part series by Jonathan Schell called “The Fate of the Earth,” which did for nuclear weapons what Carson had done for chemical pollution: freaked everyone out. That fall, Representative Al Gore, the chair of the Subcommittee on Investigations and Oversight of the House Committee on Science and Technology, convened hearings into “The Consequences of Nuclear War on the Global Environment.” The consequences of nuclear war on the environment, like its consequences on the balance of power, were difficult to prove; most data came from computer models, and from research on other planets. A battle began between those who were willing to place their faith in the speculations of military strategists and those who were willing to place their faith in the speculations of environmental scientists.

At the center of that battle was a plan to build a defensive missile shield: weapons that would orbit the planet. On March 23, 1983, President Ronald Reagan announced the Strategic Defense Initiative, in what came to be called his “Star Wars” speech: “I call upon the scientific community in our country, those who gave us nuclear weapons, to turn their great talents now to the cause of mankind and world peace: to give us the means of rendering these nuclear weapons impotent and obsolete.” To its many critics, S.D.I. undermined the nuclear paralysis that had set in with the idea of mutual assured destruction: neither the U.S. nor the U.S.S.R. would launch a missile, the theory went, since everyone would end up dead. But, if the U.S. could defend itself against a missile attack, M.A.D. no longer applied. The Union of Concerned Scientists prepared a hundred-and-six-page report opposing the project. Sagan, who was in intensive care after a ten-hour esophageal procedure, dictated a letter of objection from his hospital bed.

What Sagan did next is recounted in a new book by Paul Rubinson, “Redefining Science,” a history of science in a national-security state. The story of Sagan’s campaign against S.D.I., though little remembered, has been told before, in Lawrence Badash’s 2009 book, “A Nuclear Winter’s Tale”; in “Merchants of Doubt,” by Naomi Oreskes and Erik M. Conway, in 2009; and, most richly, in a terrific 2011 journal article by Matthias Dorries. Sagan launched a campaign to warn the world about “nuclear winter”; the very term, as Dorries points out, brought together the weapons and the weather.

Sagan collaborated with four scientists. James Pollack, his first graduate student, had written a dissertation, in 1965, on the greenhouse effect on Venus. Thomas Ackerman had attended Calvin College, a Christian school, as an undergraduate, earned a Ph.D. at the University of Washington, and then worked with Pollack at NASA. (Later, he declined an invitation to work on S.D.I., citing his moral objections as a Christian.) Richard Turco, an atmospheric scientist, was the co-author of a study called “Possible Ozone Depletions Following Nuclear Explosions,” which was published in Nature in 1975, the year that Owen Brian Toon, a graduate student at Cornell, submitted to Sagan a dissertation called “Climatic Change on Mars and Earth.” Sagan decided to use his celebrity to bring the research of these scientists to the broadest possible public audience, as fast as the scientific method allowed—or maybe faster.

“To the village square we must carry the facts of atomic energy,” Albert Einstein said in 1946. “From there must come America’s voice.” Sagan, in his understanding of the role of science in a democracy, had Einstein behind him, but, more, he had John Dewey, along with a generation of Progressive engineers, New Deal reformers, and Manhattan Project-era atomic scientists. In 1946, the Federation of Atomic Scientists, which had been founded to advocate for international, civilian control of atomic energy, had established the National Committee on Atomic Information. Atomic scientists organized a speakers’ bureau: they spoke at Kiwanis clubs, at churches and synagogues, at schools and libraries. In Kansas alone, eight Atomic Age Conferences were held. The F.B.I. launched an investigation. In 1948, the head of the National Committee on Atomic Information, suspected of being a Communist, was fired. His entire staff resigned in protest, and the committee disbanded. The next year, the Soviet Union tested its first nuclear weapon. Public-spirited science yielded to the demands of a national-security state.

Sagan received his training in that world. But that world did not survive Vietnam, or the Love Canal disaster. This much Sagan understood. But what he could not have fully understood were two forces that had gained strength in the nineteen-seventies, both of which were at odds with his neo-Deweyism: a postmodern critique of objectivity, fashionable among intellectuals, artists, and writers; and a conservative movement determined to expose the liberal bent of the academy and of the press. Sagan waded into these waters early in 1983, with a paper he prepared with Turco, Toon, Ackerman, and Pollack. The paper, known by its authorial acronym, TTAPS, used meteorological models derived from the study of volcanoes to calculate the effects on light and temperature of different kinds and numbers of nuclear explosions, factoring in the dust, smoke, and soot produced by the burning of cities; some of the data came from Mars. Moving beyond Schell, whose essays had predicted the end of humanity, TTAPS forecast a nuclear winter that might result in the end of all life on the planet.

Sagan circulated a draft to fifty scientists, then convened a conference at the American Academy of Arts and Sciences in April, 1983, which was attended by twice as many. Meanwhile, he sent George F. Kennan a draft of an essay he’d written for Foreign Affairs. “I cannot tell you what a great thing I feel you have done,” Kennan wrote in reply, thanking him for providing a “clear and irrefutable demonstration of the enormity of the danger presented by these vast nuclear arsenals.” Next, Sagan and an assortment of colleagues submitted two papers to the peer-reviewed journal Science and planned another conference, to be held at a Sheraton in Washington, with five hundred participants, a hundred members of the press, and a live “Moscow Link.” The day before the conference, Sagan published an article about nuclear winter in Parade. Using only the worst-case numbers, he admitted no room for doubt about what was, after all, a theory, presenting nuclear winter as the consensus of more than a hundred scientists from around the world.

Two charges were levelled at Sagan: that he shouldn’t be writing for a Sunday newspaper supplement, and that he’d exaggerated the certainty of an untested theory. “In the scientific community you don’t publish first results in Parade magazine,” George Rathjens, of M.I.T., wrote. (Unknown to Sagan’s critics, the two Science papers had already been accepted for publication when the Parade article appeared.) In the Wall Street Journal, S. Fred Singer, at that time a fellow at the Heritage Foundation and later a consultant for the tobacco industry, argued that the theory relied too heavily on predictive models. The physicist and S.D.I. enthusiast Edward Teller wrote Sagan, privately, “My concern is that many uncertainties remain and that these uncertainties are sufficiently large as to cast doubt on whether the nuclear winter will actually occur.” He added, “I can compliment you on being, indeed, an excellent propagandist—remembering that a propagandist is the better the less he appears to be one.” In Nature, Teller attacked the theory (“A severe climatic change must be considered dubious rather than robust”), and offered this sermon: “Highly speculative theories of worldwide destruction—even the end of life on Earth—used as a call for a particular kind of political action serve neither the good reputation of science nor dispassionate political thought.” That S.D.I. rested on highly speculative theories of worldwide destruction was not lost on Teller’s many critics.

Sagan’s grandiosity helped him gain a vast popular audience; it also hurt his cause. So did his partisanship: he declined an invitation to dine with the Reagans at the White House. His celebrity knew no bounds. The Pope asked him for an audience. Talking about warheads seemed like a fabulous way to be famous. “This morning, Trump has a new idea,” Lois Romano wrote in a Washington Post profile of Donald Trump in November, 1984, the week after Reagan defeated Mondale. “He wants to talk about the threat of nuclear war. He wants to talk about how the United States should negotiate with the Soviets. He wants to be the negotiator.” He knew just how to do it. “It would take an hour-and-a-half to learn everything there is to learn about missiles,” Trump told Romano. “I think I know most of it anyway.”

Sagan was widely resented, and he made some poor decisions, but he was a serious scientist. Despite a number of adjustments—Stephen Schneider ran his own numbers and determined that the likeliest consequence of nuclear war was something more like a nuclear autumn—the theory gained widespread scientific acceptance. Declassified documents demonstrate that Navy scientists, for instance, were persuaded by TTAPS. In May, 1984, William Cohen, a Republican senator from Maine, wrote to Reagan about the paper and suggested that the Administration conduct a study. The House Republicans Jim Leach and Newt Gingrich joined their Democratic colleagues Tim Wirth, Buddy Roemer, and Al Gore in proposing a budget amendment mandating a “comprehensive study of the atmospheric, climatic, environmental, and biological consequences of nuclear war and the implications that such consequences have for the nuclear weapons strategy and policy, the arms control policy, and the civil defense policy of the United States.”

Hearings on nuclear winter were held that summer, before a House subcommittee that, Paul Rubinson says, “essentially put deterrence on trial.” If a misfire or an accident would mean the end of all life on the planet, could there really be any strategic argument in favor of a nuclear stockpile? During another round of congressional hearings, Sagan said that he’d give a recent and inconclusive Department of Defense report on nuclear winter a D or “maybe a C-minus if I was in a friendly mood.” Assistant Secretary of Defense Richard Perle said that he’d give Sagan an F. “I didn’t hear a word of science this morning,” Perle declared. “I heard a shallow, demagogic, rambling policy pronouncement.” William Buckley observed, “Carl Sagan gave a half-hour’s performance so arrogant he might have been confused with, well, me.”

In December, 1984, Reagan’s National Security Council presented the President with a summary of a report on nuclear winter prepared by the National Academy of Sciences. While allowing that the model’s quantitative risk assessment involved uncertainties, the report argued that the model’s calculations suggested “temperature changes of a size that could have devastating consequences” and, with urgency, called on all available agencies and scientists to conduct investigations to narrow the range of uncertainty. Nuclear winter could be debated, but it couldn’t be dismissed.

“You were always my favorite to manipulate.”

Nuclear winter did not end the Cold War, but it did weaken the logic of deterrence, and not merely by undermining the idea of a winnable nuclear war. Nuclear winter relied on computer models and projections; its predictions were uncertain. Deterrence relies on computer models and projections; its predictions are uncertain. At one point, Richard Perle said that he wished Sagan would go back to his laboratory and stop “playing political scientist.” And that, inadvertently, got to the heart of the matter. However much Sagan might have overreached, his intellectual extravagance was nothing compared with the entirely hypothetical musings and game-theory models of the political scientists and strategists on the basis of whose speculations the United States government spent more than five trillion dollars between 1940 and 1996.

Reagan was himself persuaded by nuclear winter; a nuclear war, he said, “could just end up in no victory for anyone because we would wipe out the earth as we know it.” In the U.S.S.R., nuclear winter energized dissidents. In 1985, when the Soviet physician Vladimir Brodsky was arrested, one of the charges was “transmitting a letter to the Soviet Academy of Sciences requesting greater publicity about the nuclear winter.” Protesters in Moscow’s Gagarin Square chanted, “Tell the truth about the nuclear-winter phenomenon to our people.” Eduard Shevardnadze, the Soviet foreign minister, talked about nuclear winter in a speech at the U.N., and Mikhail Gorbachev alluded to it on another occasion. In 1985, the Federation of American Scientists presented Sagan with an award honoring him as the “Most Visible Member of the Scientific Community on the Planet Earth.” In 1986, Turco won a MacArthur prize. After that year, the number of nuclear weapons in the world began to decline.

Since the end of the Cold War and the collapse of the Soviet Union, most nuclear-weapons talk has been about non-proliferation and coercion. Under the terms of the Non-Proliferation Treaty, signed in 1968, non-nuclear powers agreed to forgo nuclear weapons in exchange for the assurance that they could develop nuclear energy, and for a promise from nuclear powers to pursue disarmament in good faith. Since then, no nuclear nation except South Africa has dismantled its arsenal, which is why non-nuclear states continue to press nuclear states to make good on the promise they made in 1968.

In a new book titled “Nuclear Politics: The Strategic Causes of Proliferation,” the Yale political-science professors Alexandre Debs and Nuno P. Monteiro struggle with a very small data set. Of the eight nations other than the U.S. that have nuclear weapons, three (the U.K., France, and Israel) are American allies; two (India and Pakistan) are friendly; and three (China, North Korea, and Russia) are adversaries. Two of these countries (North Korea and Pakistan) acquired nuclear weapons since the eighties, which is very worrying, but both acquired them at great cost. (Zulfikar Ali Bhutto, then the Pakistani foreign-affairs minister, said his people could “eat grass” if that’s what it took.) The U.S. has stopped several states from developing nuclear weapons, either by threatening to abandon an alliance (in the cases of Taiwan and West Germany), or by threatening, indirectly, to use military force (Libya), or by using it, a perilous course (Iraq). Under what circumstances do states develop nuclear weapons? Debs and Monteiro argue that most states are too weak to do so; most weak states aren’t interested; some weak states aren’t especially threatened; and most weak states that are threatened are protected by stronger states. These findings question conventional wisdom, which has it that the bomb is a tool of weak states. “No doubt, the atomic bomb would enable a weak state to stand up to more powerful adversaries,” they write. “So far, however, no weak unprotected state has ever managed to obtain it.”

In “Nuclear Weapons and Coercive Diplomacy,” just published, another pair of political scientists, Todd S. Sechser and Matthew Fuhrmann, investigate nuclear coercion, an idea that has all but replaced deterrence in some policy circles. Deterrence involves stopping your enemy from doing what you don’t want; coercion involves getting your enemy to do what you want. The theory of deterrence rests on an analysis of the balance between two roughly equal superpowers. Those conditions no longer apply. Coercion is a theory for a single superpower: a new game, requiring a new game theory. Does it work? Not really. As Sechser and Fuhrmann demonstrate, nuclear powers have not generally been able to coerce other nuclear powers: in the sixties, the Soviets’ nuclear superiority didn’t help solve territorial disputes with China; and, more recently, the United States hasn’t been able to coerce North Korea into abandoning its nuclear-weapons development. Nor have nuclear powers been able to alter the behavior of non-nuclear powers, the authors argue, and their list is long: “The shadow of America’s nuclear arsenal did not convince Afghan leaders to hand over al Qaeda operatives after the group conducted terrorist attacks against American targets in 1998 or 2001. Great Britain could not coerce Argentine forces to withdraw from the Falkland Islands without a fight in 1982, despite deploying nuclear forces to the South Atlantic. The Soviet Union could not force Iran or Turkey to hand over disputed territory in the early 1950s, after Moscow acquired the bomb. China has similarly been unable to make relatively weak states—including Brunei, Malaysia, Philippines, Taiwan, and Vietnam—abandon their claims to the disputed Spratly Islands in the South China Sea.”

These arguments are very interesting, but they are based on the shaky science of very small numbers. An _n_th case could unravel any of them. The most contested number in this debate is zero. In 2009, Thomas Schelling, an economist, national-security expert, and Cold War deterrence theorist, who had won a Nobel Prize for his game-theory analysis of conflict, issued a dire warning:

A “world without nuclear weapons” would be a world in which the United States, Russia, Israel, China, and half a dozen or a dozen other countries would have hair-trigger mobilization plans to rebuild nuclear weapons and mobilize or commandeer delivery systems, and would have prepared targets to preempt other nations’ nuclear facilities, all in a high-alert status, with practice drills and secure emergency communications. Every crisis would be a nuclear crisis, any war could become a nuclear war. The urge to preempt would dominate; whoever gets the first few weapons will coerce or preempt. It would be a nervous world.

Schelling’s nervous world is the setting for “The Case for U.S. Nuclear Weapons in the 21st Century,” a careful and balanced study by Brad Roberts, the director of the Center for Global Security Research at Lawrence Livermore National Laboratory. Lamenting the hardened lines between advocates and abolitionists, Roberts calls for a fresh and broad-minded debate: “Whether nuclear weapons will continue to be effective in preventing limited wars among major powers is an open question.” The case for deterrence began to fall apart in the nineteen-nineties and two-thousands, he argues, owing to a lack of leadership. This isn’t entirely correct. Presidents in that period did fail to make the case for deterrence, but deterrence had already been dismantled by the spectre of nuclear winter.

The biggest consequence of the nuclear-winter debate, though, has had to do not with nuclear-weapons policy but with the environmental movement. In the short term, the idea of a nuclear winter defeated the idea of deterrence. In the long term, Sagan’s haste and exuberance undermined environmental science. More important, the political campaign waged against nuclear winter—against science, and against the press—included erecting a set of structures, arguments, and institutions that have since been repurposed to challenge the science of global warming.

In 1984, in an effort to counter Sagan and to defend the Strategic Defense Initiative, the George C. Marshall Institute was founded by Robert Jastrow, a NASA physicist; Frederick Seitz, a former president of the National Academy of Sciences; and William Nierenberg, a past director of the Scripps Institute of Oceanography. Jastrow argued that “the Nuclear Winter scenario could not serve the needs of Soviet leaders better if it had been designed for that purpose.” One of the Marshall Institute’s first projects was to try to persuade PBS affiliates not to air a documentary critical of S.D.I.; citing the Fairness Doctrine, the institute argued that equal time ought to be given to its own report, in favor of S.D.I. The report represented the views of three scientists—Jastrow, Seitz, and Nierenberg—while a statement expressing concerns about the science behind S.D.I. had been signed by sixty-five hundred scientists. Nevertheless, most PBS stations decided not to broadcast the documentary. With funding from the Marshall Institute, Seitz’s cousin Russell Seitz, a physicist at Harvard’s Center for International Affairs, published an essay in The National Interest, in the fall of 1986, dismissing the nuclear-winter paper as nothing but “a long series of conjectures” and declaring nuclear winter dead: “Cause of death: notorious lack of scientific integrity.” In 1988, funded, in part, by ExxonMobil, the Marshall Institute turned its attention to the science behind global warming.

Another of Sagan’s most vociferous critics, S. Fred Singer, had repeatedly challenged nuclear winter on the grounds of its uncertainty. “Sagan’s scenario may well be correct,” Singer wrote in 1983, “but the range of uncertainty is so great that the prediction is not particularly useful.” A longtime consultant to ARCO, Exxon, Shell Oil, and Sun Oil, Singer is currently the director of the Science and Environmental Policy Project at the Heartland Institute, founded in 1984. Its position on global warming: “Most scientists do not believe human greenhouse gas emissions are a proven threat to the environment or to human well-being, despite a barrage of propaganda insisting otherwise coming from the environmental movement and echoed by its sycophants in the mainstream media.”

The nuclear-winter debate has long since been forgotten, but you can still spy it behind every cloud and confusion. It holds a lesson or two. A public understanding of science is not well served by shackling science to a national-security state. The public may not naturally have much tolerance for uncertainty, but uncertainty is the best that many scientific arguments can produce. Critics of climate-change science who ground their argument on uncertainty have either got to apply that same standard of evidence to nuclear-weapons strategy or else find a better argument. Because, as Sagan once put it, theories that involve the end of the world are not amenable to experimental verification—at least, not more than once. ♦

An earlier version of this article erroneously stated that Sagan had begun his career working on a classified nuclear-weapons project. It also misidentified the operation Sagan was recovering from when he dictated his letter of objection.