Doubling Down on Denial

Despite the best efforts of Jastrow, Seitz, Nierenberg, and Singer to create doubt, the scientific debate over the detection of global warming was reaching closure. By 1992, Hansen's 1988 claim that warming was detectable no longer seemed bold. It seemed prescient. The only remaining issue really was whether we could prove that the warming was caused by human activities. As scientists had acknowledged many times, there are many causes of climate change, so the key question was how to sort out these various causes. Now that warming had been detected, could it be definitively attributed to humans?

"Detection and attribution studies" work by considering how warming caused by greenhouse gases might be different from warming caused by the Sun—or other natural forces. They use statistical tests to compare climate model output with real-life data. These studies were the most threatening to the so-called skeptics because they spoke directly to the issue of causality: to the social question of whether or not humans were to blame, and to the regulatory question of whether or not greenhouse gases need to be controlled. As these studies began to appear in the peer-reviewed literature, if s not surprising that Singer and his colleagues tried to undermine them. Having taken on the patriarch of climate change research, they went after one of its rising young stars: Benjamin Santer of the Program for Climate Model Diagnosis and Intercomparison at the Lawrence Livermore National Laboratory.

Santer had done his Ph.D. work in the 1980s at the University of East An-glia, England, where he had compared climate •model results to observational data, using so-called Monte Carlo methods to make a rigorous statistical analysis.109 Until this point, model comparisons had been mostly done qualitatively. Scientists looked at maps of model output and compared them to maps of real-life observations to identify similarities and differences. Santer and his Ph.D. supervisor, Tom Wigley (the director of the Climatic Research Unit at East Anglia, U.K.), thought statistical analysis offered more to climate science than such qualitative comparisons. Besides, other parameters—detailed patterns of surface pressure, precipitation, and humidity—might actually provide better tests of the models than global mean temperature. Depending on the driving force—greenhouse gases, volcanic dust, or the Sun—you'd expect different changes in some of these parameters.110

After finishing his thesis with Wigley, Santer was invited to the Max Planck Institute for Meteorology in Hamburg. One of the Institute directors was Klaus Hasselmann, a physicist who spent much of his spare time working on unification theory: the effort to merge the four known fundamental forces in the universe into a single field at extremely high energies like those that theoretically existed at the universe's first few moments of existence. This was pretty far from climate science, but Hasselmann had also made a number of major scientific contributions to climate questions. One of those was a paper in 1979 that proposed a new detection and attribution technique called "optimal fingerprinting."111 The idea was derived from signal processing theory, and fhe paper was so technical, so elegant, and so laden with dense tensor field mathematics, that Santer at first didn't get it. Santer recalls it as "a thing of beauty. It was many years ahead of its time. I was just too dumb to understand it."112

Hasselmann’s key insight was that climate scientists faced the same basic problem as communications engineers: how to detect a weak signal—the thing you're interested in—amid lots of noise that you don't care about. In climate science, the noise is caused by phenomena that are internal to the climate system, such as El Nino. The "signal" is something caused by things that are external to the Earths natural climate system: the Sun, volcanic dust, or man-made greenhouse gases. Engineers had worked for a century to develop mathematical techniques to sort out signals from noise, but they were largely unknown to climate scientists. They also aren't simple to master.

Santer got started, but progress was slow. The results of his Ph.D. thesis hadn't been all that encouraging, either. He and Wigley had shown that some of the models used for the IPCC's first assessment had large errors in surface pressures; it was what scientists call a "negative result." Still, it was important to point out such errors, and based on this and his preliminary work with Hasselmann, he was offered a position at the Climate Model Intercomparison project at the Lawrence Livermore National Laboratory, in California. The program's founder, Lawrence Gates, believed that if models were to be used for policy purposes—and they obviously would be if climate policy were to be based partly on model forecasts—it was important to evaluate them to see whether they were reliable or not. Gates pioneered the idea of "benchmark experiments"—getting climate model centers around the world to perform exactly the same calculation with their models—to permit scientists to rule out differences in model design as an explanation for the differences in model performance. (Model benchmarking was a radical idea at the time; now it is standard procedure.) Gates also argued for making the results of these experiments widely available, fo that model diagnosis became an activity of the entire climate science community, not just the responsibility of the modellers themselves—who might not be entirely objective. The lab, in other words, was trying to make modelling more rigorous, more objective, and more transparent.

Santer had the good fortune to arrive at the lab not only in the middle of one of the first major model intercomparison projects, but also at a time when Livermore colleagues Karl Taylor and Joyce Penner were performing an innovative set of climate model experiments that considered not only greenhouses gases, which cause warming, but also sulfate aerosol particles, which generally cause cooling. The Taylor and Penner experiments clearly showed that human influences on climate Were complex: changes in C02 and sulfate aerosols had distinctly different climate fingerprints.

Fingerprinting proved to be a powerful tool for studying cause-and effect relationships. Up to that point, much of the scientific argument about the causes of climate change had gone like this: if greenhouse gases increased, then you would expect temperatures to increase, too. They had.

So the prediction had come true—textbook scientific method. The problem with the textbook method, however, is that if s logically fallacious. Just because a prediction comes true doesn't mean the hypothesis that generated it is correct. Other causes could produce the same effect. To prove that greenhouse gases had caused climate change, you'd have to find some aspect of it that was different than if the cause were the Sun or volcanoes. You needed a pattern that was unique.

We saw in chapter 4 that V. Ramanathan, a prominent atmospheric scientist, had suggested one: the vertical structure of temperature.113 If warming were caused by the Sun, then you'd expect the whole atmosphere to warm up. If warming were caused by greenhouse gases, however, the effect on the atmosphere would be different, and distinctive. Greenhouse gases trap heat in the lower atmosphere (so it warms up), while the reduced heat flow into the upper atmosphere causes it to cool. Collaborating with colleagues at the Max Planck Institute, and six other research institutions around the world, Santer started to look at the vertical variation of temperature.114 Before they'd finished the work, Santer was asked to become the convening lead author for "Detection of Climate Change and Attribution of Causes," chapter -8 of the second IPCC assessment.

Nowadays, there's a lot of prestige associated with the IPCC, since they shared the 2007 Nobel Peace Prize, but back in 1994 most scientists considered it a distraction from their "real" work—doing basic research—and Bert Bolin was having a difficult time finding someone to take the lead on the detection and attribution chapter. In the spring of 1994, after some of the other chapters were already started, Tim Barnett of the Scripps Institution of Oceanography called Santer to ask if he'd be willing to do it. Barnett had been one of two lead authors of the equivalent chapter in the First Assessment, and he convinced Santer that it would be a feather in his cap. Santer signed on.

The job of the convening lead author (this position is now called the coordinating lead author) is to produce an assessment of some aspect of climate science based on "the best scientific and technical information available."115 This involves working together with other "lead authors" and "contributing authors" to agree on the structure and scope of the future chapter. Individual scientists are then assigned the task of drafting different sections of the chapter. Once all sections are drafted, the convening lead author and the lead authors attempt to hammer out a complete draft that’s acceptable to the entire group. Santer's chapter ultimately had four lead authors, including his old mentor at East Anglia, Tom Wigley, Tim Barnett, and thirty-two additional contributing authors—in other words, thirty-six of the world's top climate scientists.116

The chapter ft author group met in Livermore, California, in August 1994 to identify the key scientific areas that needed to be addressed. There were a total of twenty participants (from the United States, Canada, the United Kingdom, Germany, and Kenya). After this initial meeting, most of the author group's discussion took place by e-mail. Then, in October through November, Santer attended the first of three so-called drafting sessions, involving the lead authors and the convening lead authors of all chapters of the IPCC Working Group I Report.

The first drafting session convened in Sigtuna, Sweden, and Santer encountered his first challenge: a disagreement over whether the chapter should include a discussion of model and observational uncertainties. Since the topic was covered in other chapters, some authors thought it would be redundant to do it here, but Santer didn't think readers would search other chapters to find it, and in any case his panel would have no control over what was said in those chapters. Santer prevailed, and the published version contained about six pages of discussion of model and observational uncertainties.

Shortly after the Sigtuna meeting, chapter 8 went through an initial round of peer review. The "zeroth" draft was sent out to roughly twenty scientific expert! in detection and attribution work, to all scientific contributors to the chapter, and to the lead authors of all other chapters of the report. After updating their chapters in response to the peer review comments, the IPCC lead authors met for a second drafting session in March 1955 in the British seaside resort of Brighton. In May, a complete draft of the entire IPCC Working Group I Report, as well the Summary for Policymakers, was submitted for full "country review" by the governments participating in the IPCC. The governments chose reviewers—a mixture of scientists and laypeople—who were supposed to provide comments to the lead authors prior to the third drafting session, in Asheville, North Carolina, in July, but because Santer had been chosen so late as the convening lead author, this schedule didn't quite work out for his group. Santer arrived in Asheville having yet to receive the government reviewers' comments.

At the Asheville meeting, Santer presented the results of his fingerprint study of changes in the vertical structure of atmospheric temperatures; which by this point had been submitted to Nature.117 One scientist present at the meeting reported that Santer's presentation electrified the audience; it was "mind-boggling to a lot of the scientists there.""8 It looked like Santer and his colleagues might just have proved the human impact on climate.

After Asheville, all chapters were revised in response to the country review—all, that is, except chapter 8, because Santer was still awaiting the comments. The final stage in the process was the IPCC plenary meeting, scheduled to start in Madrid on November 27. In October, drafts of the Working Group I Report and the Summary for Policymakers had been sent to all the government delegates to the Madrid meeting. When Santer arrived at the Madrid meeting, he was handed a sheaf of comments— including comments from the U.S. government—that he had never seen before.

Meanwhile, sometime in September, a draft of the entire Working Group 1 Report was leaked. The central message of chapter 8, that the anthropogenic fingerprint had been found, drew widespread attention."9 "In an important shift of scientific judgment, experts advising the world's governments on climate change are saying for the first time that human activity is a likely cause of the warming of the global atmosphere," the New York Times declared on its front page. This, of course, wasn't quite right. Scientists had been saying for a long time that human activity was a likely cause of warming. They were now saying that it was demonstrated. The New York Times didn't get it. But the skeptics did, and they went on the attack.

Two weeks before the plenary session in Madrid, the Republican majority in the U.S. Congress launched a preemptive strike. In a set of hearings held in November, they repeatedly questioned the scientific basis for concern. The star witness was another well-known contrarian, Patrick J. Michaels, who had completed his Ph.D. at the University of Wisconsin-Madison in 1979, building models relating climate change to crop yields. In 1980, he was appointed state climatologist of Virginia by Republican governor John Dalton (although many years later Michaels was forced to forego that title when it was shown that Dalton had acted without legal authority).120 In the 1980s, Michaels had published scientific work on the climate sensitivity of various crops and ecosystems, but by the early 1990s, he was mainly known not for mainstream science, but his contrarian views.121 He had joined Fred Singer in publicly attacking the mainstream view of ozone depletion in a series of columns in the Washington Times122 He produced a quarterly newsletter called the World Climate Review, funded at least in part by fossil fuel interests, and used it as a platform to attack mainstream climate science. The Review was circulated free to members of the Society for Environmental Journalism, ensuring that its claims got wide attention.123 In the early 1990s, he had worked as a consultant to the Western Fuels Association—a coal mining industry group—to promote the idea that burning fossil fuels was good, because it would lead to higher crop yields as increased atmospheric CO2 led to increased photosynthesis and therefore increased agricultural productivity.124

In the Republican hearings, Michaels was presented as an expert who somehow knew more than all the scientists working within the IPCC umbrella. His own personal analysis of the difference between a model prediction of greenhouse gas-induced warming and atmospheric temperatures derived from NOAA's weather satellites showed, he claimed, that the IPCC climate models had heavily over-predicted global warming and could not be trusted. He complained in the hearing that while he'd made many critical comments on the various chapters of the IPCC report, his comments had been ignored, resulting "in not one discernable change in the text of the IPCC drafts."125

Congressman George E. Brown Jr. of California asked Jerry Mahlman, director of NOAA's Geophysical Fluid Dynamics Laboratory (GFDL), to respond to Michaels's claims. The particular model study that Michaels attacked was the work of GFDL scientist Syukuro Manabe—probably the world's most respected climate modeller—the man who, along with Jim Hansen, had presented his work to the Charney committee back in 1979. Mahlman explained that Michaels's analysis contained an elementary flaw. Manabe's study was designed to investigate the impact of CO2 on climate, and had deliberately omitted other factors—including volcanic dust. However, there had been a set of large volcanic eruptions in the early 1990s, most famously Mt Pinatubo in 1992. The satellite measurements obviously did incorporate these other real world phenomena, so naturally, they'd be different from the model results.

"The bottom line," Mahlman concluded, "is that there is no logical basis for a direct comparison of this GFDL model experiment with that of [satellite] data sets or any other data set."126 A legitimate comparison between models and observations could only be carried out when the models and observations examined the same things. It was obvious why the IPCC had ignored Michaels's complaints.

The hearing wasn't very successful at getting press attention, receiving no notice from the New York Times, the Washington Post, or even the Washington Times. Among major newspapers, only the Boston Globe seems to have bothered covering it It wasn't exactly news by late 1995 that the Republican congressional leadership opposed environmental protection; there had been discussion that year of repealing the Clean Water Act, one of the cornerstones of American environmental improvement. But the lack of press attention didn't matter; the hearing had the desired effect of reinforcing the Republican majority's do-nothing attitude. Writing to Fred Seitz after the hearing, Nierenberg said, "I doubt that Congress will do anything foolish. I can also tell you that at least one high-level corporate advisor is advising boards that the issue is politically dead. Happy holiday."127

Santer presented the findings in chapter 8 on November 27,1995, the first day of the plenary session (and the same day Nierenberg proclaimed the issue politically dead in his letter to Seitz). The chapter was immediately opposed by the Saudi Arabian and Kuwaiti delegates. In the words of the New York Times's reporter, these oil-rich states "made common cause with American industry lobbyists to try to weaken the-conclusions emerging from Chapter 8."128 The lone Kenyan delegate, Santer remembers, "thought there should not be a detection and attribution chapter at all."129 Then the chairman of a fossil fuel industry group, the Global Climate Coalition, and automobile industry representatives monopolized the rest of the afternoon.130 Finally the IPCC chairman, Britain's Sir John Houghton, closed the discussion and appointed an ad hoc drafting group to work out the disagreements and to address all of the late government comments. The working group included the lead authors, and delegates from the United States, Britain, Australia, Canada, New Zealand, the Netherlands, Saudi Arabia, Kuwait, and the lone Kenyan.

A portion of the ad hoc group hammered out an acceptable language. Steve Schneider convinced the Kenyan that there really was a scientific basis for the chapter's central conclusion that anthropogenic climate change had been detected.131 But the Saudis never sent a representative to the ad hoc sessions, and when Santer presented the revised draft, the Saudi head delegate protested all over again. A bit of a shouting match ensued, and Houghton had to intervene, effectively tabling the issue while the working group finished negotiating the Summary for Policymakers. There the entire issue boiled down to a single sentence, in fact a single adjective, drawn from Santer’s chapter "The balance of evidence suggests that there is a [blank] human influence on global climate."132

What should the adjective be? Santer and Wigley wanted "appreciable." This was unacceptable to the Saudi delegate, but it was too strong for Bert Bolin, too. One participant recalls the group trying about twenty-eight different words before Bolin suggested "discernible." That clicked, and the outcome of the Madrid meeting was this sentence: "The balance of evidence suggests that there is a discernible human influence on global climate."133 This the would be quoted repeatedly in the years to come.

With the Summary for Policymakers settled, the individual chapters had to be revised in the light of all the late review comments, and Houghton instructed the lead authors to make the necessary changes after the meeting.134 Santer went from Madrid to the Hadley Center in Bracknell, England, where he made the changes in long-distance collaboration with Wigley and Barnett. The most significant of these changes was structural. The draft chapter 8 had summary statements at both beginning and end of the chapter, but none of the other chapters did. They only had summaries at the beginning. Therefore, Santer had been instructed to remove the summary statement at the end of the chapter so that it would have the same structure as the rest of the chapters. That, Santer remembered years later, was a fateful decision, as critics would later attack him for "removing material."135

Then Fred Singer launched an attack. In a letter to Science on February 2, 1996, four months before formal release of the Working Group I Report, Singer presented a litany of complaints. The Summary for Policymakers, he claimed, ignored satellite data that showed "no warming at all, but actually a slight cooling." On this basis he claimed that the climate models, which all showed warming, were wrong. The IPCC had violated one of its "major rules" by including the fingerprinting work, because "the research had not yet, to my knowledge, appeared in the peer-reviewed literature." The panel had also ignored an "authoritative U.S. government report” that had found the twenty-first-century warming might be as little as 0.5°C, making global warming a non-problem. (Singer didn't cite the report.)

Finally, he concluded, "The mystery is why some insist in niaking it into a problem, a crisis, or a catastrophe—'the greatest global challenge facing mankind.'"136

Tom Wigley responded to Singer's criticisms in Mafch. Rejecting the "no warming" claim entirely, he simply stated, "This is not supported by the data; the trend from 1946 to 1995 is .3 C. As shown in chapter 8 of the full report (figure 8.4) there is no inconsistency between the observed temperature record and model simulations." There were some differences between measurements made with satellites and measurements made with "radiosondes"—instruments on balloons, with radios attached to transmit the results—but climate scientists didn't expect them to perfectly track. each other; the reasons were explained in both chapters 3 and 8. "There are good physical reasons to expect differences between these two climate indicators," Wigley noted, because they were in different places measuring somewhat different things.

The claim that the pattern recognition studies violated the IPCC's rules was wrong on two counts. First, Wigley explained, the IPCC allowed use of material from outside the peer-reviewed journals as long as it was accessible to reviewers. This was to ensure the report was "up to date" when published. Moreover, the specific work Singer referred to "on the increasing correlation between the expected greenhouse-aerosol pattern and observed temperature changes, is in the peer-reviewed literature/'137

Moreover, Singer was again creating a straw man. "Singer refers to the [Summary for Policymakers] as saying that global warming is 'the greatest global challenge facing mankind,'" Wigley and his coauthors wrote. "We do not know the origin of this statement—it does not appear in any of the IPCC documents. Further, it is the sort of extreme statement that most involved with the IPCC would not support."138

Wigley was right. The IPCC had not described global warming as the "greatest global challenge facing mankind." The words Singer attributed to the IPCC don't appear in either the Working Group I Report or in its Summary for Policymakers. Singer was putting words into other people's mouths—and then using those words to discredit them.

The IPCC had in fact bent over backward not to use alarmist terms. Bert Bolin had deliberately imposed a policy of extreme conservatism of language; witness his rejection of "appreciable" in favor of "discernible." The opposition of the Saudi and Kuwaiti delegations had ensured only least common denominator statements. Everyone involved had seen how the process led to a conservative estimation of the threat. What was Singer's response to this refutation of his allegations? He provided the missing citation for his claim that there would be only a 0.5°C warming in the twenty-first century.139

The IPCC had contracted with Cambridge University Press to publish the Working Group 1 Report, scheduled to appear in the United States in June 1996. In May, Santer and Wigley presented their chapter at a briefing in the Rayburn House Office Building on Capitol Hill, organized by the American Meteorological Society and the U.S. Global Change Research Program. The two scientists were now challenged by William O'Keefe of the American Petroleum Institute and by Donald Pearlman, an industry lobbyist and registered foreign agent of several oil-producing nations.140 O'Keefe and Pearlman accused them of "secretly altering the IPCC report, suppressing dissent by other scientists, and eliminating references to scientific uncertainties."141

"Who made these changes to the chapter? Who authorized these changes? Why were they made?" Pearlman demanded. "Pearlman got up and in my face, turned beet red and [started] screaming at me," Santer recalls. AMS officer Anthony Socci "finally separated us, but Pearlman kept following me around."142 Santer explained that he'd been required by IPCC procedures to make the changes in response to the government comments and discussions at Madrid, and the chapter had never been out of his control, but the truth did not satisfy the opposition.143

The Global Climate Coalition meanwhile had circulated a report entitled "The IPCC: Institutionalized Scientific Cleansing" to reporters, members of Congress, and some scientists. By chance, anthropologist Myanna Lahsert interviewed Nierenberg about his "skepticism" about global warming two weeks before the Working Group I Report was published, and found that he had a copy of the coalition report. He had evidently accepted its veracity, even though there was no way to compare its claims against the real chapter 8 (since the latter had not yet been released). He quoted its claims to Lahsen, telling her that the revisions had "just altered the whole meaning of the document. Without permission of the authors." Moreover, he claimed, "Anything that would imply the current status of knowledge is so poor that you can't do anything is struck out."144 That was hardly true; Santetis pane| had included six pages of discussion of uncertainty in the final text. But Bill Nierenberg knew all about altering scientific reports for political reasons, so perhaps he followed the adage that the best defense is offense. Or perhaps he was guilty of "mirror imaging" as Team B had accused the CIA of in 1976: assuming that his opponents thought and operated the way he did.

Then Fred Seitz took the attack to the national media. In a letter published in the Wall Street Journal on June 12,1996, he accused Ben Santer of fraud. "In my more than 60 years as a member of the American scientific community, including my services as president of the National Academy of Sciences and the American Physical Society, I have never witnessed a more disturbing corruption of the peer-review process than the events that led to this IPCC report." Seitz repeated the Global Climate Coalition's charges that unauthorized changes to chapter 8 had been made after its acceptance in Madrid. "Few of these changes were merely cosmetic; nearly all worked to remove hints of the skepticism with which many scientists regard claims that human activities are having a major impact on climate in general and on global warming in particular," Seitz claimed. If the IPCC couldn't follow its own procedures, he concluded, it should be abandoned and governments should look for "more reliable sources of advice to governments on this important question."145 Presumably, he meant the George C. Marshall Institute, of which he was still chairman of the board.

Santer immediately drafted a letter to the Journal, which forty of the other IPCC lead authors signed. Santer explained what had happened, how he had been instructed by Houghton to make tiae changes, and why the changes were late in coming. At first the Journal wouldn't publish it. After three tries, Santer finally got a call from the Jpurnal's letters editor, and the letter was finally published on June 25. Saucer's reply had been heavily edited, and the names of the forty other cosigners deleted.

What the Journal allowed Santer to say was that he had been required to make the changes "in response to written review comments received in October and November 1995 from governments, individual scientists, and non-government organizations during plenary sessions of the Madrid meeting." This was peer review—the very process that Seitz, as a research scientist, had been a part of all his life. Only it was extended to include comments and queries from governments and NGOs as well as scientific experts. But the changes didn't affect the bottom line conclusion.

Santer also pointed Out that Seitz wasn't a climate scientist, hadn't been involved in creating the IPCC report, hadn't attended the Madrid meeting, and hadn't seen the hundreds of review comments to which Santer had to respond. In other words, his claims were just hearsay.146

Bert Bolin and Sir John Houghton also responded with a long letter de-fending Santer and the IPCC process. "Frederick Seitz's article is completely without foundation," they replied unequivocally. "It makes serious allegations about the Intergovernmental Panel on Climate Change and about the scientists who have contributed to its work which have no basis in fact. Mr. Seitz does not state the source of his material, and we note for the record that he did not check his facts either with the IPCC officers or with any of the scientists involved."147

Well, that’s what they'd wanted it to say, but the Journal edited that statement out, too, along with three more paragraphs explaining the drafting process in some detail. The journal allowed them to say only that:

... [in] accordance with IPCC Procedures, the changes to the draft of Chapter 8 were under the full scientific control of its conveningLead Author, Benjamin Santer. No one could have been more thorough and honest in undertaking that task. As the responsible officers of the IPCC, we are completely satisfied that the changes incorporated in the revised version were made With the sole purpose of producing the best possible and most clearly explained assessment of the science and were not in any way motivated by any political or other considerations.148

We know how the journal edited the letters because Seitz's attack and the journal's weakening of the response so offended the officials of the American Meteorological Society and of the University Corporation for Atmospheric Research that their boards agreed to publish an "Open Letter to Ben Santer" in the Bulletin of the American Meteorological Society, where they republished the letters in their entirety, showing how the journal had edited them. They voiced their support of Santer and the effort it had taken all the authors to put the report together, and categorically rejected Seitz's attack as having "no place in the scientific debate about issues related to global change."149 They began, finally, to realize what they were up against.

[There] appeared] to be a concerted and systematic effort by some individuals to undermine and discredit the scientific process that has led many scientists working on understanding climate to include that there is a very real possibility that humans are modifying Earth's climate on a global scale. Rather than carrying out:a legitimate scientific debate including the peer-reviewed literature, they are waging in the public media a Vocal campaign against scientific results with which they disagree.150

But the attack was far from over. On July n, the Wall Street Journal published three more letters reprising the charges, one from Fred Seitz, one from Fred Singer, and one from Hugh Ellsaesser. (Ellsaesser was a retired geophysicist at the Lawrence Livermore National Laboratory who previously had questioned the evidence of the ozone hole. He served in the mid-1990s on the Marshall Institute's Scientific Advisory Board, and in 1995 wrote a report for the Heartland Institute on The Misuse of Science in Environmental Management.) Singer and Seitz simply repeated the charges they'd already made, but Singer also took the opportunity to turn the IPCC's caution against it. The IPCC had bent over backward to be judicious, arguing at length to choose just the right, reasonable adjective--"discernible." Singer dismissed the IPCC conclusion as "feeble," at the same time insisting illogically that it was being used to frighten politicians into believing that a climate catastrophe is about to happen.151

Santer and Bolin responded a second time to the attacks in letters the Journal published July 23—prompting another attack by Singer.152 This time, the Journal wouldn't publish it, and Singer circulated it by e-mail instead. Santer responded by e-mail, too. There was, Singer maintained, no "evidence for a current warming trend." According to Singer, chapter 8 had been based primarily on Santeris "unpublished work," and the panel should have included as a lead author "Professor Patrick J. Michaels, who, at the time, had published the only refereed paper on the subject" of climate fingerprinting. And he repeated the charge of "scientific cleansing." Santer rejected all of Singer's charges. Chapter 8 was based on more than 130 references, not just Santer's two papers. The claim that Michaels had published the only "refereed paper on the subjecf of pattern-based recognition before mid-1995 was incorrect: Hasselmanris theoretical paper on the subject was published in 1979, and Tim Barnett and Mike Schlesinger had published a "real-world" fingerprint study as early as 1987. Michaels had been invited to be a contributing author to chapter 8 but had refused. Finally, Santer noted, chapter 8 contained several paragraphs discussing Michaels's paper, but when Wigley had approached Michaels for comments, "Prof. Michaels did not respond."153

Singer's claims were not only false, but had been shown to be false. Still, he wasn't finished repeating them. Now he would claim that Fred Seitz was the real victim of the whole affair.

In November, Singer penned an article for the tied "Global Warming Disinformation?'' By this time, the IPCC report had been published and available for months, so Singer could have seen for himself that chapter 8 contained six pages of discussions of model and observational uncertainties, as Santer had insisted it should all along. Still, Singer repeated the claim that chapter 8 had been edited to remove uncertainties, and then asserted that "Seitzfone of the nation's most respected scientists, was attacked for factually reporting the revisions made by the IPCC leadership, which clearly affected the sense of the report!"154

Joined by Bill Nierenberg, Patrick Michaels, and a new ally—MIT meteorologist Richard Lindzen—Singer then attacked the AMS/UCAR Open Letter. After repeating the refuted charges of "substantial and substantive" deletions of uncertainty, Singer cast the deletions as a conspiracy that Santer was now trying to cover up. "Santer ... has not been forthcoming in revealing who instructed him to make such revisions and who approved them after they were made. He has, however, told others privately that he was asked [prevailed upon?] to do so by IPCC co-chairman John Houghton." To Singer and his co-authers on the letter, this was evidence of political meddling in the chapter. He continued, "You may not have seen the 15 November [1995] letter from the State Department instructing Dr. Houghton to 'prevail upon' chapter authors 'to modify their texts in an appropriate manner following discussion in Madrid.' "155 Singer's presentation of it as some sort of clandestine conspiracy was absurd: Bolin and Houghton had already identified themselves months before as the source of Santefs instructions.

In her 1999 analysis, Myanna Lahsen pinned Singer's efforts to "envelop the IPCC in an aura of secrecy and unaccountability" to a common American conservative rhetoric of political suppression.156 As we have seen in previous chapters, if anyone was meddling in the scientific assessment and peer review process, it was the political right wing, not the left. It wasn't the Sierra Club that tried to pressure the National Academy of Sciences over the 1983 Carbon Dioxide Assessment; it was officials from the Department of Energy under Ronald Reagan. It wasn't Environmental Defense that worked with Bill Nierenberg to alter the Executive Summary of the 1983 Acid Rain Peer Review Panel; it was the White House Office of Science and Technology Policy. And it was the Wall Street Journal spreading the attack on Santer and the IPCC, not Mother Jones.

The over-fie-top attacks on Santer began to have consequences for Nierenberg. In April, Nierenberg had invited Tom Wigley to a conference he wanted to hold at Scripps that November on the costs and benefits of global warming, but Wigley smelled a rat. "I have decided to withdraw from your November meeting," he wrote. "The reason for this is the letter you co-signed which appeared in BAMS [Bulletin of the American Meteorological Society]. I have no desire to cooperate with anyone who endorses such an unmitigated collection of distortions and misinformation."157

Nierenberg tried flattery to keep Wigley on board. "The personally difficult part for me is that your work, Klaus [Hasselmanrfs] work, and [Bill] Nordhaus' work have had the most influence (and still do) on my thinking." He lamented the rift that was developing in the climate science community over the ongoing public attacks, but then followed Singef’s lead in imputing conspiracy, this time in the scientific journals. "I remind you in this instance of something that touched you personally about which I only had the slightest information from the gossip columns and some hallway talk. I was told that you faced great opposition in getting your Nature paper published. That great pressure was put on you."158

Wigley evidently had no idea what Nature paper Nierenberg was talking about. "It seems that you have not only NOT been influenced by, but actually disagree with (or are unaware of) the vast bulk of my scientific work: in particular the work on detection, which the BAMS letter you co-signed has unfairly, unjustifiably, unscientifically and incorrectly criticized." Wigley also rejected the imputation that Nature had pressured him. 'To which paper are you referring? I have published 22 papers in that particular journal. No matter which, you shouldn't take any notice of what you hear in 'gossip columns and hallway talk.'" He concluded, "So Bill, what I said in my previous email stands. Your 17 April 'response' gives me no reason to change my mind—just the opposite. The BAMS letter makes it quite clear that you think my IPCC detection work with Ben Santer was distorted for political motives. I am surprised, therefore, that you would even want a person like me to attend a meeting of yours. I still think you are being duplicitous, and I still suspect your motives."159

Wigley wasn't the only one to begin to understand what Nierenberg was really up to. Klaus Hasselmann also wrote to Nierenberg: "I have followed the attacks on Ben Santer during the last year and found them to be grossly unfair and clearly politically motivated. In a letter I wrote to the Wall Street Journal (which was not published, with many other similar letters) I pointed out that it was ridiculous to imply that the conclusions of Chapter 8 had been willfully or unintentionally altered against the will of the Madrid delegates."160 Hasselmann was still willing to come to a meeting about the costs and benefits of global warming—a subject that interested him greatly—but he wouldn't come to a meeting with a polmcnl agenda, Tniieivaf the pronounced political colouring of the BAMS letter I am not couviimed at this point that the concerns of Tom Wigley are not justified."161

Perhaps after so many years as Svengali, Bill Nierenberg did not realize that this time he had gone too far. Nierenberg, despite his intellect, really didn't seem to understand that by participating in this assault on Ben Santer, he was attacking the entire community of climate modellers. By signing on to Singer's letter, he marked himself in their eyes as a political actor, not a scientific one. Nierenberg's comment that he feared the polarization of the community was both perceptive and blinkered; the climate science community was most definitely becoming polarized, but it was due to his own actions, and those of a small network of doubt-mongers.

We might dismiss this whole story as just infighting within the scientific community, except that the Marshall Institute claims were taken seriously in the Bush White House and published in the Wall Street Journal, where they would have been read by millions of educated people. Members of Congress also took them seriously. Proposing a bill to reduce climate research funding by more than a third in 1995, Congressman Dana Rohrabacher called it "trendy science that is propped up by liberal/left politics rather than good science."162 In July 2003, Senator James Inhofe called global warming "the greatest hoax ever perpetrated on the American people."163 As late as 2007, Vice President Richard Cheney commented in a television interview, "Where there does not appear to be a consensus, where it begins to break down, is the extent to which that's part of a normal cycle versus the extent to which if s caused by man, greenhouse gases, et cetera"—exactly the question Santer had answered a decade before.164 How did such a small group come to have such a powerful voice?

We take it for granted that great individuals—Gandhi, Kennedy, Martin Luther King—can have great positive impacts on the world. But we are loath to believe the same about negative impacts—unless the individuals are obvious monsters like Hitler or Stalin. But small numbers of people can have large, negative impacts, especially if they are organized, determined, and have access to power.

Seitz, Jastrow, Nierenberg, and Singer had access to power—all the way to the Whit| House by virtue of their positions as physicists who had won the Cold War. They used this power to support their political agenda, even though it meant attacking science and their fellow scientists, evidently believing that their larger end justified their means. Perhaps this, too, was part of their professional legacy. During the Manhattan Project, and throughout the Cold War, for security reasons many scientists had to hide the true nature of their work. All weapons projects were secret, but so were many other projects that dealt with rocketry, missile launching and targeting, navigation, underwater acoustics, marine geology, bathymetry, seismology, weather modification; the list goes on and on.165 These secret projects frequently had "cover stories" that scientists could share with colleagues, friends, and families, and sometimes the cover stories were true in part. But they weren't the whole truth, and sometimes they weren't true at all. After the Cold War, most scientists were relieved to be freed of the burdens of secrecy and misrepresentation, but Seitz, Singer, and Nierenberg continued to act as if the Cold War had not ended.

Whatever the reasons and justifications of our protagonists, there's another crucial element to our story. Its how the mass media became complied, as a wide spectrum of the media—not just obviously right-wing newspapers like the Washington Times, but mainstream outlets, too—-felt obligated to treat these issues as scientific controversies. Journalists were constantly pressured to grant the professional deniers equal status—and equal time and newsprint space—and they did. Eugene Linden, once an environment reporter for Time magazine, commented in his book Winds of Change that "members of the media found themselves hounded by experts who conflated scientific diffidence with scientific uncertainty, and who wrote outraged letters to the editor when a report included their dissent."166 Editors evidently succumbed to this pressure, and reporting on climate in the United States became biased toward the skeptics and deniers because of it.

We've noted how the notion of balance was enshrined in the Fairness Doctrine, and it may make sense for political news in a two-party system (although not in a multiparty system). But it doesn't reflect the way science works. In an active scientific debate, there can be many sides. But once a scientific issue is closed, there's only one "side." Imagine providing "balance" to the issue of whether the Earth orbits the Sun, whether continents move, or whether DNA carries genetic information. These matters were long ago settled in scientists' minds. Nobody can publish an article in a scientific journal claiming the Sun orbits the Earth, and for the same reason, you can't publish an article in a peer-reviewed journal claiming there's no global warming. Probably well-informed professional science journalists wouldn't publish it either. But ordinary journalists repeatedly did.

In 2004, one of us showed that scientists had a consensus about the reality of global warming and its human causes—and had since the mid-1990s. Yet throughout this time period, the mass media presented global warming and its cause as a major’s debate. By coincidence, another study also published in 2004 analyzed media stories about global warming from 1988 to 2002. Max and Jules Boykoff found that "balanced" articles—ones that gave equal time to the majority view among climate scientists as well as to deniers of global warming—represented nearly 53 percent of media stories. Another 35 percent of articles presented the correct majority position among climate scientists, while still giving space to the deniers.167 The authors conclude that this "balanced" coverage is a form of "informational bias," that the ideal of balance leads journalists to give minority views more credence than they deserve.

This divergence between the state of the science and how it was presented in the major media helped make it easy for our government to do nothing about global warming. Gus Speth had thought in 1988 that there was real momentum toward taking action. By the mid-1990s, that policy momentum had not just fizzled; it had evaporated. In July 1997, three months before the Kyoto Protocol was finalized, U.S. senators Robert Byrd and Charles Hagel introduced a resolution blocking its adoption.168 Byrd-Hagel passed the Senate by a vote of 97-0. Scientifically, global warming was an established fact. Politically, global warming was dead.


End  of  chapter