Issue Number Three: Sentinel Devices

The scientist as sentinel

Climate scientists have a frightening message, but the public doesn’t seem worried enough. Naomi Oreskes argues that the dispassionate ideal of science might be getting in the way.

Scientists have been warning the world for a long time about the risks of “dangerous anthropo­genic interference in the climate system,” but they also struggle with how to explain what that really means. People don’t experience the climate system, people experience weather events. So it was perhaps inevitable that a day or two after “Super­storm” Sandy hit New York City, a journalist contacted me to ask if I thought Sandy would be a tipping point. Would this be the crucial event that convinced an otherwise skeptical American public that climate change is underway, and is, indeed, dangerous?

My immediate thought was that this event should be a tipping point, but whether it would be is another matter. Virtually as soon as the floodwaters stabilized in the subways, the usual cadre of self-proclaimed “skeptics” were spinning their usual arguments: that no one weather event proves systemic climate change.

These folks are formally correct, but then so was the tobacco industry when it insisted that no one lung cancer death proved smoking caused cancer. No one event proves climate change, because by definition climate is a pattern, and patterns can only be proved by, well, patterns. But every pattern is made up of individual contributions, and for some time now scientists have been seeing an emerging pattern of weather events consistent with what they have predicted. And while scientists in the 1960s could not explain why smoking caused the particular pattern of disease that it did, scientists today can explain why climate change is causing the observed pattern of extreme weather events. Carbon dioxide traps energy in the earth’s atmosphere. That energy has to go somewhere, and one place it goes is into weather. Among other things, more energy in the system permits the development of more powerful storms. Storms are very dangerous. They kill people. They do billions of dollars in damage. They destroy cultural heritage and disrupt communities, sometimes permanently. Even when people re-build, their sense of safety and security is diminished. We saw this after Hurricane Katrina and we will be seeing it again in the weeks, months, and years to come as New York rebuilds after Superstorm Sandy.

So why aren’t more people more afraid? Public opinion polls have consistently shown that only a small slice of American society is very worried about climate change, and less than half are even somewhat worried. Maybe we don’t see the pattern. On my favorite radio station the commentator informs us each morning of the record high and low temperatures for that day. The record high is usually fairly recent—usually within the last thirty years and often within the last twenty—but the record lows are typically long ago, often fifty or even a hundred years. This is consistent with a warming trend. Yet the announcer never seems to notice… Fair enough. Most people are not poring over temperature records, much less analyzing them statistically to see if record highs are being broken more often these days than record lows. (They are.) But there seems to be a deeper problem.

If we take 1992—the year the United Nations Framework Convention on Climate Change was adopted—as a reasonable starting point when ordinary people paying a modicum of attention to public affairs would have learned that climate change posed a potential threat to human health and well-being as well as to other species—then we might expect that since then many people would have been at least somewhat worried. Two years before that, President George H.W. Bush—remembered by some as the President who did not know what a supermarket scanner was—told the press that “We all know that human activities are changing the atmosphere in unexpected and in unprecedented ways.” Yet, despite knowing about the threat, and despite seeing ravaging floods, droughts, wildfires and now superstorms, only 20-30% of Americans over the past two decades have indicated to pollsters they were “quite concerned” about climate change. As of September 2012, only 36% thought that global warming is hurting people in the United States.[1]

Maybe this will change now, but maybe not. Certainly the harms are evident, but will people connect the harms to climate change? Will they see this as the “dangerous anthropogenic interference in the climate system” that our first President Bush promised to prevent? Perhaps, but perhaps not. Many thought Americans’ attitudes towards climate change would change after Katrina, which remains the costliest event in U.S. history ($62 billion in insured damages, $1.25 trillion in overall losses, and 1,322 lives lost). They did, temporarily, but then seemed to settle back into the prior pattern of inattention. Reactions and responses to risk are obviously complex, and entail diverse and interacting social, political, cultural, economic and epistemic factors. But one element of climate change stands out as different from at least some other forms of risk addressed in this volume: the matter of who is trying to communicate the danger.

Fires are announced by fire alarms, a familiar presence in every school and public building. Earthquakes are generally ‘announced’ by the earthquakes themselves, when we feel our houses shake and see our windows rattle and even roll. Emerging epidemics get communicated largely by public health officials whose job it is to protect us from infectious diseases and other environmental pathogens. These officials take this aspect of their job seriously, and dedicate considerable attention to doing it effectively. (Whether they are successful is another matter.) Patients expect their doctors to alert them to less immediate health threats, and to try to persuade them to do something about them, whether it is quitting smoking, losing weight, or getting a flu shot. But communicating the risk of climate change falls not to familiar technological devices, immediate experience, doctors or public health officials, but to research scientists, mostly physical scientists. And this is a group singularly ill-equipped to communicate effectively to ordinary publics, particularly about issues that trigger alarm or fear.[2]

Consider this. Two years ago, I was on a panel at the American Geophysical Union with several extremely distinguished climate scientists. Everyone on the panel had the same message: Climate change is real, it is underway, and it is dangerous. During the question period, a woman stood up and said: “You are telling us that we have a very serious problem, but you don’t sound at all worried. You don’t even sound upset!”

She was right. The scientists in the room didn’t sound worried. They certainly did not sound upset. And they almost never do. Because scientists take great pains in their work and demeanor to be rational, and scientists link rationality to dispassion. In my experience, scientists working on climate change consider it to be very important—indeed, crucial—to stay calm, to remain unemotional, and never, ever, get hysterical. In scientific circles, if you are emotional, it is assumed that you have lost your capacity to assess data calmly and therefore your conclusions become suspect. Robert Merton famously claimed that the norms of science were universalism, communism, organized skepticism, and disinterestedness, but he left out an important additional one: dispassion. Scientists can be counted on to stay calm and carry on. Or at least to try to.

In a recently published article my colleagues and I have shown that scientists have systematically underestimated the threat of climate change. We suggest that they have done so for normative reasons: The scientific values of rationality, dispassion, and self-restraint lead them to demand greater levels of evidence in support of surprising, dramatic, or alarming conclusions than in support of less alarming conclusions. We call this tendency “erring on the side of least drama.”

Climate change is very dramatic, and it is very worrisome. Superstorms, raving floods, devastating wildfires, not to mention ocean acidification and the threat it represents to the base of the food chain – are alarming. And scientists have terrible difficulty talking about them. It’s not only that they err on the side of least drama in their conclusions, it’s also that they speak without drama, even when their conclusions are dramatic. They speak without the emotional cadence that normal people expect to hear when someone is genuinely worried. So even when they are worried—and most climate scientists will tell you in private that they are – they just don’t sound it.

How can you communicate danger without drama? How can you tell someone he or she should be worried when you don’t sound worried yourself? How can you be a sentinel if you don’t have a trumpet to blow, and wouldn’t feel comfortable blowing it even if you did?

[1] See (Leiserowitz et al. 2012). Ironically a recent report by Munich Re suggests that North America has been “particularly hard hit by weather catastrophes in recent years: Hurricane Katrina, tornadoes, floods, wildfires, searing heat and drought. The intensities of certain weather events in North America are among the highest in the world, and the risks associated with them are changing faster than anywhere else. They estimate over $1 trillion dollars in weather related damages (2011 dollars) and 30,000 lost lives during the period 1980-2011. They attribute the greater impact in North America to a combination of geographic and social factors. Geographically, “The North American continent is exposed to every type of hazardous weather peril—tropical cyclone, thunderstorm, winter storm, tornado, wildfire, drought and flood. One reason for this is that there is no mountain range running east to west that separates hot from cold air.” Socially, North America is characterized by large population, urban sprawl, and high wealth, which makes damages relatively greater than they would be if the affected areas were poor (Re 2012).
[2] There is a now a voluminous literature on communicating climate risk, and scientific societies have recently dedicated many conference panels to discussing the issue. Many of these efforts follow the deficit model of public understanding, suggesting that if only scientists explained the scientific evidence clearly, on a level that ordinary people could understand, then they accept and act upon it. This approach fails to acknowledge the deep social, cultural and economic interests that mitigate against action on the scale required, as well as the psychological reasons why people react to fear with anger, denial, and even violence. For entry into this literature, and a critique of why the deficit model is insufficient, see (Moser and Dilling 2004, Moser and Dilling 2007, and Boykoff 2010).


Boykoff, Maxwell T. 2011. Who Speaks for the Climate?: Making Sense of Media Reporting on Climate Change. Cambridge: Cambridge University Press.

Brysse, Keynyn, Naomi Oreskes, Jessica O’Reilly and Michael Oppenheimer, 2012. “Climate Change Prediction: Erring on the Side of Least Drama?” Global Environmental Change, 23(1):327-337.

Bush, George H.W. Speech at Georgetown University, 1990.

Leiserowitz, Anthony et al. (Yale Project on Climate Communication). 2012. Climate Change in the American Mind: Americans’ Global Warming Beliefs and Attitudes in September 2012. Accessed at:

Meehl, Gerald A., Claudia Tibaldi, Guy Walton, David Easterling et al. 2009. Relative increase of record high maximum temperatures compared to record low minimum temperatures in the U.S. Geophysical Research Letters 36(23): L23701.

Moser, Susanne C. and Lisa Dilling. 2004. “Making climate hot: Communicating the urgency and challenge of global climate change.” Environment 46(10):32-46 .

Moser, Susanne C. and Lisa Dilling. eds. 2007. Creating a Climate for Change: Communicating Climate Change and Facilitating Social Change. Cambridge; Cambridge University Press.

Munich Re. 2012. Severe Weather in North America. Accessed at:, see also,