Of course, Crusoe was a fictional character (though Alexander Selkirk, the castaway whose story apparently inspired Defoe, was not). And the challenge facing the next generation is vastly larger than Crusoe’s challenge. But is it so unlikely that our species, a congeries of changelings, would be able to do exactly as Crusoe did—transform our lives to meet new challenges—before we round that fateful curve of the second inflection point and nature does it for us? I can imagine Margulis’s response: You’re imagining our species as some sort of big-brained, hyper-rational, cost-benefit-calculating computer! A better analogy is the bacteria at our feet! Still, Margulis would be the first to agree that removing the shackles from women and slaves has begun to unleash the suppressed talents of two-thirds of the human race. Drastically reducing violence has prevented the waste of countless lives and a staggering amount of resources. Is it really impossible to believe that we wouldn’t use those talents and those resources to draw back before the abyss?
Our record of success is not that long. In any case, past successes are no guarantee of the future. But it is terrible to suppose that we could get so many other things right and get this one wrong. To have the imagination to see our potential end, but not have the cultural resources to avoid it. To send humankind to the moon but fail to pay attention to Earth. To have the potential but to be unable to use it—to be, in the end, no different from the protozoa in the petri dish. It would be evidence that Lynn Margulis’s most dismissive beliefs had been right after all. For all our speed and voraciousness, our changeable sparkle and flash, we would be, at last count, not an especially interesting species.
* * *
*1 Darwin argued that the fossil record was then too incomplete to show transitions, and that later discoveries would fill in gaps. Almost all scientists believe he was correct. Since then, paleontologists (dinosaur specialists) have uncovered many “missing link” species. One example: Kulindadromeus zabaikalicus, discovered in 2014. A small, bipedal dinosaur, it has both bird-like feathers and dinosaur-like scales, exemplifying how dinosaurs evolved into birds.
Appendix A: Why Believe? (Part One)
Years ago I sat in on an introductory course given by Lynn Margulis. At a certain point she referred to evolution as the cornerstone of modern biology. A student raised her hand and said she didn’t believe in evolution. “I don’t care whether you believe it,” Margulis replied. “I just want you to understand why scientists believe it.” After that, she said, the student could decide whatever she wanted.
Not long after, I spoke to John Maynard Smith, a distinguished British evolutionary biologist. In our conversation I recounted Margulis’s injunction to her student. Maynard Smith laughed heartily. The reason for his amusement, he explained, was that Margulis was a prominent critic of mainstream evolutionary theory. She didn’t dispute the existence of evolution by natural selection. But Margulis thought that natural selection was just part of the picture—in the long run, symbiosis and chance were more important sources of evolutionary innovation. Maynard Smith thought Margulis was off base. “But I’ll grant her one thing,” he said, as I remember it. “She’s a remarkably good skeptic. Even if she’s wrong, she’s fruitfully wrong.”
In the main text, I tried to follow Margulis’s example in my discussion of climate change—lay out why the great majority of scientists believe that it is occurring and caused by human activity. This belief is the culmination of a century and a half of investigation into atmospheric chemistry and physics. But I also argued that this belief, by itself, doesn’t compel specific action—it’s hard to know what duties we owe to faraway descendants. People have to decide what to think on their own.
Now I would like to follow Margulis again, by arguing for the role of skepticism. By skepticism I don’t mean the ad hominem claim that climate science is a “hoax.” About four thousand scientists from eighty countries and countless representatives from those countries were involved in the last report of the Intergovernmental Panel on Climate Change. To imagine that these thousands of researchers and government officials were all part of a devilish plot—and that they all kept silent about it, as did their predecessors in previous IPCC reports—is foolish. It is particularly hard to figure out why oil states like Iran and Saudi Arabia, which participate in IPCC reports, would enlist in a fraudulent endeavor to rid the world of fossil fuels.
At the same time, though, the “hoax” rhetoric reflects something real. It is a way, however inexact or distorted, of expressing the fear that the risks of rising carbon-dioxide levels are being systematically exaggerated by environmentalists. The activists, in this view, are using climate change as a tool to win social changes that they cannot get in other ways. In 2007 the activist-journalist Naomi Klein published The Shock Doctrine, which argued that right-wing elites were using—or even manufacturing—economic crises as a pretext to force societies to adopt corporations’ preferred policies (slashing social-welfare programs, reducing taxes, cutting regulations, and so on). The policies were presented as solutions to the crisis but in fact were designed to enrich the already-wealthy people who proposed them. Some global-warming opponents see climate change in the same way—as an overstated or even fictitious “crisis” that left-wingers like Klein use as an excuse to force other people to do what they want (reducing consumption, overturning capitalism). When activists retort (accurately) that the great majority of researchers agree with them—well, that proves the researchers, too, are in on the plot.
To avoid this, decouple for a moment the science (rising carbon dioxide levels leads to a warmer planet) from the proposed remedies (getting rid of fossil fuels).
If the basic physical understanding of atmospheric carbon dioxide were proven to be incorrect, it would be a remarkable event in the history of science. As a rule, entire disciplines don’t get big things wrong if they are part of the arena of study. True, physicists wrongly believed for several centuries that outer space was filled with a mysterious substance called the ether, but the longevity of that error was due mainly to researchers’ inability to test for the ether’s existence. In the nineteenth century appropriate testing methods were developed, and the belief was quickly exploded. Climate change has been studied off and on since Tyndall, systematically since about 1960, and intensively since about 1990. Given that long effort, it would be highly unusual for the general consensus—pouring lots of carbon dioxide into the air increases average global temperatures—to be wrong.*1
It would be especially unusual given that the models have made many successful quantitative predictions. An early example was the 1967 prediction by Syukuro Manabe and Richard T. Wetherald—two researchers at the National Oceanic and Atmospheric Administration, in Washington, D.C.—that the lower atmosphere and the stratosphere would act in opposite ways: warming in the former would be accompanied by cooling in the latter. Because the stratosphere is hard to observe, confirmation did not occur until 2011. But it did occur. Twelve years after that initial prediction, Manabe and two other scientists predicted something else: land areas would heat up faster than ocean areas, with the slowest warming being around Antarctica. That, too, has turned out to be the case. There are many other examples.
Yet despite these successes, we still do not understand either the rate at which climate change is occurring or its precise effects. (I discussed this when I looked at climate sensitivity.) Here there is room for skepticism. It may be that the atmosphere won’t respond to carbon dioxide nearly as fast as the doomsters fear. Or that the impact will be distributed in ways that we don’t yet understand. As I wrote this book, I asked half a dozen climate scientists to identify what they see as the biggest uncertainties—the reasons that their fears could be wrong. Here are some of their answers.
The first stems directly from the fact that no computer today can handle calculations that cover the entire surface of the earth and its atmosphere. In consequence, researchers simplify their models by treating the atmosphere and surface as an array of cubes, each perhap
s fifteen or twenty miles on a side (different models have different sizes). The cubes are treated as if they were uniform, but in the real world, of course, a cube of air many miles on a side can hold many different types of clouds. Or the supposedly uniform ground in a cube could be partly covered by a lake and partly by a mountain. To handle such variables, scientists write equations intended to approximate what is going on. Inevitably, small errors build up. To account for them, the models must be “tuned.” This means manually tweaking the parameters (the change in temperature as altitude changes, say, or the way heat moves in the ocean). The tweaked models are then compared to twentieth-century weather records to see how well they reproduce them. Unfortunately, the models are also supposed to explain those same weather records. The unavoidable risk is that scientists will fool themselves, inadvertently using the tweaks to mask the inaccuracies of a model. It also means that other scientists using the model won’t know whether a specific prediction arises directly from the underlying science (good) or is largely due to tuning (bad). None of this is shady in any way—all large physical models must be tuned. But doing it correctly is a constant worry.
All the models have difficulties with clouds, especially clouds over the oceans (because the oceans cover most of Earth’s surface, this means most clouds). As the ocean grows warmer, the air above it becomes more humid, which encourages clouds to form. The clouds, constantly changing and turbulent, interact in a highly complex way with the constantly changing winds and currents in the atmosphere. Turbulence, convection, airflow—all are notoriously intractable physics questions. Indeed, the reason that the military has built so many costly wind tunnels is that aeronautical engineers have wanted to watch what happened when a plane was exposed to wind turbulence because they couldn’t predict the consequences well with mathematics. This problem has only gotten worse—the need now is to account for entire global systems of clouds. Low-altitude clouds tend to reflect sunlight, cooling the air around them. High-altitude clouds tend to trap infrared radiation, heating the air around them. Will warmer oceans end up driving clouds higher into the atmosphere? Will they create more low-altitude thunderstorms? Will air and water currents in a warmer planet push cloud formations north or south of where they would have been? Which effects will dominate? More than a few scientists think this will long be a source of uncertainty.
In these descriptions of climate models, one word has not appeared: biology. Climate models are constructed by physical scientists. Yet one lesson that biology has taught is that living creatures profoundly shape the world. Consider the role of methane hydrates in the Arctic. The land sheds organic molecules into the water like a ditchdigger taking a shower. Sewage plants, fertilizer-rich farms, dandruffy swimmers—all make their contribution. Plankton and other minute sea beings flourish where the drift is heaviest, at the continental margins. When these creatures die, their bodies drizzle slowly to the seafloor. Microorganisms feed upon the remains. In a process familiar to anyone who has seen bubbles coming to the surface of a pond, the microbes emit methane gas as they eat and grow. (Methane, one recalls, is a potent agent of climate change.) Under the high pressure of these cold depths, water and methane react to each other: water molecules link into crystalline lattices—“methane hydrates,” in the jargon—that trap methane molecules. Such vast amounts of methane are stored in methane hydrates that researchers fear that their release into the air could set off catastrophic changes in climate.
In 2017 scientists from the Woods Hole Oceanographic Institute tried to measure the methane seeping up from Norway’s Arctic coast, where the oceans are warmer than they used to be. They discovered that the warming that was pushing up the methane was also pumping nutrients from the seafloor to the surface. The nutrients were feeding huge blooms of phytoplankton. To the scientists’ astonishment, the plankton took in so much carbon dioxide via photosynthesis that they more than canceled out the effects of the methane. This is a small example of a general problem: our continuing ignorance about the impacts of life.
The issue is not restricted to microorganisms—no one has an accurate figure for how much carbon dioxide is absorbed in forests around the world. All in all, raising carbon-dioxide levels tends to increase plant growth—a negative feedback, because it tends to reduce those same carbon-dioxide levels. But increasing the area covered by plants (“leaf-area index,” in the jargon) also makes some bare areas darker—a positive feedback, because dark areas tend to reflect less sunlight into space. The effects are complicated and not uniform. In northern areas, thicker forests seem to change global wind patterns. In southern areas, higher leaf-area indexes affect groundwater levels, and in turn are affected by them. In 2017 an international research team estimated in Nature Climate Change that “the greening of the earth” had “mitigated” surface warming by about one-eighth in the previous thirty years. No one knows if this will continue.
To repeat, none of this means that climate change is not occurring, or that the science underpinning it is incorrect. It means instead that the basic physical mechanisms elucidated by Callendar and his successors are modulated, in ways that we do not yet understand, by other factors (clouds, plant growth), which could reduce—or, just as likely, amplify—their impacts. Because of these uncertainties, climate systems are full of “natural variation”—changes with causes that scientists can’t yet identify. In Chapter 7, I described my friend Rob’s model, published in 2016, for how rising carbon dioxide could cause the Antarctic to melt quickly. Temperatures in the West Antarctic—roughly speaking, the portion of Antarctica that fronts onto the southern Pacific Ocean—have risen in recent years. It was easy to couple the rising temperatures and Rob’s theoretical calculation and produce a picture: the Antarctic is responding to our carbon dioxide, and that could flood coastal cities from Miami to Mumbai. Quite naturally, the result was newspaper headlines around the world.
But a year later two other research teams looked at the historical data and concluded that the rising temperatures in the West Antarctic were within the range of natural variation (with the possible exception of the temperatures in the Antarctic Peninsula, the tongue of land that extends almost to the tip of South America). These two teams did not contradict my friend’s work. What they did was look at different data, and come up with a different picture: the Antarctic is warming, but not in a way that is different from the past.
Everyone involved believed that human-caused warming is occurring—the question is how fast, how long, and what it will do. Rob was saying: the speed and effects look like this. The two other teams were saying: no, we think they look like that.
Here, to my mind, is the great value of skepticism. No matter whether Wizards or Prophets prevail, whether the future is full of thousands of nuclear plants or millions of networked solar installations, reducing the risks of climate change will be pricey and politically difficult. If the costs of fighting climate change are offset by the benefits of (say) avoiding rapid sea-level rise, they will be well worth paying. But if large sums are spent on a chimera, the effects will be negative—spending too much can be as bad as spending too little.
People sometimes try to wish this problem away by saying that some preferred measure could be easily paid for by gutting this or that pointless or corrupt government program. But much of the supposed “abuse” actually reflects different beliefs about what government should be doing—one person’s “wasteful spending” is another person’s “vital government function.” For people of one disposition, spending too much on reducing carbon-dioxide levels could involve the peril of leaving the nation open to hostile incursions or failing to improve its physical infrastructure. For people of another disposition, the risk involves cheating the poor today of education, sanitation, and social investments in the name of a purported benefit to their descendants. This worry can be put more bluntly. If, as most economists believe, people tomorrow will be more affluent than people today, the hazard is that we end up valuing tomorrow’s rich more than today’
s poor.
The great value of fruitful skeptics—the reason that, as Maynard Smith suggested, they should be celebrated even if they prove to be wrong—is that they force advocates to think through these issues. And, of course, sometimes the skeptics are right. Maynard Smith didn’t think that Lynn Margulis’s skeptical ideas about natural selection’s relative import to evolution were correct. But he had been equally dismissive about Margulis’s skeptical ideas about the lack of importance of plants and animals in the evolutionary tree and—well, that was spot on.
* * *
*1 A possible counterexample is the theory that dietary fat causes obesity. This idea held sway from about 1970 to about 2000, but has since been strongly challenged. If the dietary-fat hypothesis truly proves to be incorrect, it will be an outlier in scientific history—and it will have been believed for much less time than the carbon-dioxide theory.
Appendix B: Why Believe? (Part Two)
The similarities between disbelief in the scientific evidence for the risk of human-caused climate change and the scientific evidence for the safety of genetically modified food plants and animals are inexact but striking. In both cases skeptics argue that large bodies of scientific evidence are untrustworthy because the scientists themselves have bad motives—they are, variously, paid shills of greedy corporations or witting tools of anti-human agitators. And in both cases the skeptics often claim that the other side is acting in bad faith—that it has drummed up a phony crisis (a heating world, a risk of famine) to which it can present its preferred ideas (anti-industrialism, corporate capitalism) as the only possible solution.
There is another parallel. In the previous appendix, I argued that recognizing a problem doesn’t automatically imply accepting either its seriousness or any particular solution to that problem. In useful discussions, this leads to seeking to manage risks: the risk that the problem is as serious as activists think, the risk of not achieving other goals if too much time, money, and attention is devoted to solving a problem that turns out not to be very serious.