The Politicization of the Scientific Method

On January 26, 2026, the Attorneys General of some 26 states, under the leadership of West Virginia Attorney General John McCusky sent a remarkable letter to Judge Robin L. Rosenberg, the Director of the Federal Judicial Center (FJC), asking the FJC to withdraw the chapter on climate change from its Fourth Edition Reference Manual for Scientific Evidence (Manual) . Their claim was that that chapter was plagued by an incurable conflict of interest, because it was prepared by two authors, Jessica Wenty and Radley Horton, both of whom are closely allied with Columbia University’s Sabin Center for Climate Change Law, which makes no bones that “its mission is to combat the climate crisis and advance climate justice.” 

The FJC published the Manual in cooperation with the National Academies of Sciences, Engineering, and Medicine (NASEM). Their joint manuals have long been regarded as the basic guide toward explaining how the basic principles and practices of the scientific method should guide the Supreme Court and lower courts in evaluating scientific evidence in courts and before administrative agencies on such vital matters as global warming, COVID vaccines, patents, and mental illness. The Third Edition of that Manual had received, for example, uniformly favorable reviews.  

Jessica Weinkle, writing for Civitas Outlook, also condemned the Manual for its chapter called “How Science Works”” by pointing out how even this chapter took substantive positions that supported one side. In addition, the Wall Street Journal wrote in an editorial entitled “a Judicial Climate Scandal,” again, taking dead aim at the Sabin Center. It was therefore no surprise that the chapter was withdrawn by the FJC after its  review, even though NAREM kept the report in place on its website, which is itself strong evidence of a deep left-right divide within the overall legal community.    

At the outset, I should say that I have been and continue to be active on the defendant side of these global warming cases, including Suncor Energy v County Commissioners of Boulder, Inc., in which my Civitas colleague John Yoo and I will participate with an amicus brief, in part because the purported evidence on global warming does not, in my view, begin to support the massive claims for restitution that have been put forward as part of a vigorous denunciation of capitalism. More concretely, there is strong evidence that three of the major claims for global warming, dealing with the alleged decline in polar bears, the Great Coral Reef, and the loss of small Pacific islands, are suspect if not incorrect. At this point, a dose of public caution is needed in making determinations of possible liability that could transfer huge sums of money that could, for example, bankrupt a significant segment of the fossil fuel industry or lead to wasted investments in solar energy that is too often produced when it is not needed, and impossible to store until it is needed. 

It should come as no surprise that the attack on the first chapter of the reference work has carried over to its chapter on “How Science Works,” whose lead author is Michael Weisberg of the University of Pennsylvania, who is accompanied by Anastasia Thanokos of the University of California, Berkeley. Recently, there has been an active effort to show that this work, despite its lofty title, again seeks to tilt the balance in favor of being regarded as an implicit, if evasive, endorsement of the conventional wisdom on topics such as global warming and pandemics like COVID-19. The tone of this chapter is that of a learned treatise that is generally satisfied with the current set of practices dealing with the use of the scientific method in litigation. Unfortunately, that impression is totally displaced by the contrast between its first two citations. The first is Pew Research Center, What Americans Know About Science (2019), which notes that limited public knowledge on this topic is good reason to be careful about the field. At this point, guidance is needed for juries in these cases. Then the chapter offers two disconcerting notes of caution: “Yet in a trial, a jury may have to weigh evidence regarding advanced scientific knowledge that is still emerging and about which lines of evidence may conflict or be inconclusive. Complicating matters, public relations campaigns have misled the public about the true state of scientific consensus regarding certain scientific issues.” Footnote 2 then gives away the game, for the apparent go-to source is a highly influential, but deeply polemical, book by Naomi Oreskes & Erik M. Conway, Merchants of Doubt: How a Handful of Scientists Obscured the Truth on Issues from Tobacco Smoke to Global Warming (2010). If there is any doubt from the title, just one sentence from the book’s abstract clinches the picture: “Merchants of Doubt tells the story of how a loose-knit group of high-level scientists and scientific advisers, with deep connections in politics and industry, ran effective campaigns to mislead the public and deny well-established scientific knowledge over four decades.” So it is always the adversaries' bad faith that poses an obstacle to the widespread propagation of truth.  

But why accept that claim on faith? The writings of three eminent scientists, Richard Lindzen, Will Happer, and Steven Koonin, have offered powerful criticisms of the received wisdom. But their work is not mentioned even once in this report, which is an instructive indication of the one-sided substantive analysis. 

The contrast to the parallel passage is the shorter (by far) Third Edition of this chapter prepared by David Goodstein, whose lodestar on the scientific method was the great English thinker Francis Bacon, followed by a discussion of Karl Popper’s notion of falsifiability as a way to get at the truth or falsity of any scientific claim. In the Fourth Edition, Bacon is ignored, and Popper is criticized. But what is most striking is that the third edition makes only one equivocal reference to “consensus,” whereas in the fourth edition that term plays a central role, appearing some 70 times. 

That shift in usage reflects a failure to acknowledge the profound difference between scientific and legal inquiry. For science, the use of the term “consensus” is a harbinger of danger. The consensus thinking can be either safe or lazy. That approach may lead to incremental advantages. But its big downside is that it prevents major paradigm shifts that can reshape how any specific area is understood. It may well have been that some social (or theological) consensus challenged the Darwinian principle of natural selection. But as new evidence from the finches came in, the basic paradigm shifted to natural selection. When it was clear that there was no mechanism to drive that change, the theory of Mendelian inheritance pushed the change. But when difficulties with an individualist theory began to break down the earlier consensus, the critical conception of inclusive fitness, introduced by W.D. Hamilton, emerged, showing how natural selection is driven by shared family genes. All that matters is whether the theory and the evidence match. 

Legal proceedings are wholly different, but not by some paradigm shift. There is a practical dispute that requires the application of standard principles and a dispute as to how these should be determined. The great case of Daubert v. Merrell Dow Pharmaceuticals (1993) imposes limitations on the qualifications that must be demonstrated before an expert may testify. The need for tighter standards was driven by a strong sense that too many earlier cases used “junk science” to impose large liabilities on various firms. And the Frye standard allowed junk science in, so long as it was accepted within “the relevant scientific community”. Those two standards in practice did not differ much, compared to the far laxer standard found in Oxendine v. Merrell Dow Pharmaceuticals (1986). But the relevant scientific community could be, and often was, a group of quacks with nothing but a collective interest in their junk science being accepted as valid so they could continue receiving research grants and handsome compensation for their “expert” testimony. Daubert helped correct that problem, but as it dealt only with admissibility of evidence it left lots of room for bad science to make its way into legal proceeding, 

Unfortunately, Weisberg and Thanukos in the Fourth Edition do not see the fundamental difference in the two kinds of inquiry, so they place great stock in Daubert, but it hardly tells a tale that science operates well in a legal setting when the case itself ended up with a devastating denunciation of then-current legal practices in the follow-on decision by Judge Alex Kozinski who eviscerated the plaintiffs evidence when the case was returned to the Ninth Circuit. 

Nor is there any reason to think that all forms of judicial inquiry produce the right result. To be sure, there are apparent success stories for the current practices, such as the lengthy reliance that our authors show on the idea that compliance with appropriate research norms will lead to sound outcomes. Weisberg and Thanukos’ heavy reliance on the work of Mario Molina and F.S. Rowland on the role of CFCs in the depletion of the Ozone layer may have earned them a Nobel Prize in 1995. Even so, they do not mention some skeptical voices that cast some doubt on the politics of how and why the standard was implemented.  

Even if, however, their account were 100 percent accurate, few, if any, inferences can be drawn from that story to the modern legal system. As far as I know, their work has never been used in litigation to resolve any judicial review. The work, moreover, was simple compared to many lawsuits that make their way to litigation spanning decades and continents. The questions of global warming are far more complicated than the science that involves the effect of a single chemical on another single chemical. What our authors needed to do was to see whether the practices that have arisen with both COVID and global warming are amenable to that same kind of confident conclusion.  

The answer to that question is surely not. The establishment on science has gone out of its way to attack any scientist who takes a dissenting position on either of those two topics so as to affirm the orthodox answer on such questions as to whether the transmission of COVID from one person to another requires the extensive quarantines put into place, or whether the rules on quarantine and the restrictions on travel were for better or worse, or whether the release of any vaccine is likely to do more harm than good. So too with global warming: there is a question of whether carbon dioxide should be regarded as a positive agent that increases photosynthesis, or a negative one that traps greenhouse gases. The literature on both points is enormous and highly disputed. It is, though, our authors live in a time warp if they think they can offer a coherent defense of their version of the scientific method by ignoring all the controversial issues of our time, in which the use of science, as noted earlier, has been highly contested. 

Richard Epstein is a senior research fellow at the Civitas Institute at the University of Texas at Austin.