John Kelly was right to kill EPA’s 'red team' climate exercise

John Kelly was right to kill EPA’s 'red team' climate exercise
© Getty Images

EPA Administrator Scott PruittEdward (Scott) Scott PruittTrump's relocation of the Bureau of Land Management was part of a familiar Republican playbook Understanding the barriers between scientists, the public and the truth Overnight Energy & Environment — Biden makes return to pre-Trump national monument boundaries official MORE’s plan to organize a “red team-blue team” exercise about climate change, consisting of a public debate over whether and to what degree human activities are causing global warming, was reportedly halted by Gen. John KellyJohn Francis KellyMORE, the White House chief of staff. Kelly should be applauded for putting an end to a terrible idea that would have undermined, rather than increased, public understanding of a critical issue.

The “red team-blue team” approach has long been used in military, security, and business settings. It typically features a “red team” tasked with testing assumptions, identifying vulnerabilities, and role-playing from adversarial perspectives, to see whether a “blue team” can adequately defend prevailing ideas. Such exercises can be extremely useful in many contexts, but using this approach to debate peer-reviewed science is a misapplication that would only sow confusion.


Pruitt has already demonstrated strong prejudice against the evidence of global warming. He has staffed the EPA with climate change skeptics and dissented from the idea that human activity is a primary contributor of climate change. His very close ties with the fossil fuel industry are well documented, and there is ample reason to distrust the objectivity of any Pruitt-administered exercise about climate science.


But even in its least biased, best-faith rendition, the use of a red team to test scientific consensus about climate change represents a category mistake — science is fundamentally different from military, security, and business strategy, and the merits of a red-team exercise in those contexts don’t apply to science.

The contexts in which red teams have traditionally been used involve information that is not publicly available and questions that cannot easily receive broad, diverse input. Businesses, for example, don’t broadcast their strategies or operating assumptions, and military information is classified. Absent the benefit of public scrutiny and debate, empowering a fully briefed red team to test contrarian views is a sensible practice for guarding against faulty insular thinking.

Climate science, on the other hand, has developed in a system of public information and open debate. Scientists publish their data and conclusions, which are scoured for errors by panels of experts through the peer-review process. Not only are findings and data public, but there are incentives for scientists to reach counter-conventional conclusions: scientific publishing rewards originality, and journals and scientists around the globe aspire to discoveries that defy widely held conceptions within the scientific community. Whereas a red team exercise in military or business contexts can fill the void of the marketplace of ideas, the scientific marketplace in open societies is alive and well.

Furthermore, the red team-blue team approach suggests an adversarial structure that exists in military and business contexts, but not in science. When the objective is to outmaneuver the enemy, or beat out competitors for market-share, assigning a team to behave like the enemy makes sense because the success of a strategy depends on the adversaries’ reaction. But the validity of a scientific hypothesis is determined by collecting and examining data, not through a simulated contest with a competitor.

The scientific value of red team-blue team climate exercise is nil, but its effect on public knowledge could actually be negative. Pruitt’s idea originated with an op-ed by NYU physicist and climate change-skeptic Steven Koonin, in which Koonin states that “the public is largely unaware of the intense debates within climate science.” Koonin claims that a red-team exercise would offer the benefit of “publicly demonstrating scientific reasoning and argument.”

But there is not intense debate within climate science about the general conclusion that human activity contributes to climate change – 97 percent of actively publishing climate scientists agree that climate-warming trends over the past century are extremely likely due to human activities. (And a recent paper found that the 3 percent of publications that reach contrary conclusions are rife with flaws.)

Pitting two teams of scientists against each other would validate the view of a minuscule minority, creating the misleading appearance of parity between two “sides” and ignoring the demonstrated scientific consensus about climate change. Consider the likely scenario in which the two sides fail to reach agreement. Would the arbiter, installed by Pruitt, then declare a winner? Would she declare a tie? Either outcome would obfuscate the true state of science about the issue. Pruitt had indicated that the process might be televised, and a broadcast that suggested deep disputes about the fundamentals of climate change would have been an irresponsible distortion, akin to spotlighting the utterly unfounded and dangerous claim that vaccines cause autism.

Pruitt’s plan revealed a fundamental misunderstanding of the scientific peer-review process, and it follows the same playbook used for decades by both the tobacco and fossil-fuel industries. These groups have worked to cast doubt on scientific evidence in order to delay policy solutions that would benefit society but hurt their companies’ profits. A red-team exercise is poorly matched to climate science, but it would have allowed Pruitt to give airtime to the witches’ brew of climate science denial. We’re all better off without more distracting noise on this critical issue.

Richard L. Revesz is dean emeritus and Lawrence King Professor of Law at New York University School of Law and director of the Institute for Policy Integrity.