‘I’m not a scientist’: Why judges need the climate chapter in the Reference Manual on Scientific Evidence

For over 30 years, the Federal Judicial Center’s Reference Manual on Scientific Evidence has served as a go-to document judges can use when they deal with issues involving scientific testimony. The third edition, produced jointly with the National Research Council, came out in 2011 and covered fields such as forensics, mental health, neuroscience, and toxicology. The fourth edition, published in 2025, expanded its coverage to include chapters on new topics such as computer science, artificial intelligence, and, until a few weeks ago, a chapter on climate science. This chapter, authored by two experts, was removed unilaterally by the Federal Judicial Center after 27 Republican attorneys general sent a letter demanding that it be deleted.

The full document, including the climate science chapter, is still available on the National Research Council website, but this too is at risk, as a subset of these attorneys general have sent a letter demanding its removal and implicitly threatening consequences if this is not done.

The Reference Manual is a crucial, unbiased tool to provide judges with background information about science. Lawyers and judges are not generally known for their proficiency in math and science; law students generally come from the liberal arts and humanities.[1]

Those students become lawyers and some of them become judges. Along the way, their science proficiency does not improve. The Reference Manual is an essential tool to help them do their jobs.

The need for judicial education in science is obvious. A telling example of this comes from Massachusetts v. EPA, a 2007 Supreme Court case involving EPA’s power to regulate greenhouse gases. At oral argument, Justice Antonin Scalia mistakenly referred to the stratosphere instead of the troposphere.[2] In terms of climate change, whether one is dealing with the stratosphere or the troposphere really matters. A lawyer for Massachusetts gently corrected Scalia. “Troposphere, whatever,” Scalia responded. “I told you before I’m not a scientist. That’s why I don’t want to have to deal with global warming, to tell you the truth.”

Yet in the 2024 Loper Bright Enterprises v. Raimondo decision,[3] the Court overturned the existing rule set forth in Chevron v. NRDC,[4] which instructed courts to defer to agency fact finding. Loper Bright asks judges to do exactly what Justice Scalia sought to avoid and for which he felt ill-equipped.[5]

Without that rule, which allowed judges to defer to scientists and other experts within federal agencies, the Reference Manual is one of the few tools available to help judges to do what Loper Bright requires.

Judges as gatekeepers. The first edition of the Reference Manual was published in response to the Supreme Court’s 1993 decision in Daubert v. Merrell Dow Pharmaceuticals, Inc., which set the standard for admitting expert testimony in federal courts.[6] The Court described judges as gatekeepers and held that they must ensure that proffered expert testimony is based on good grounds. For scientific expertise, this meant to be admissible as evidence, the basis for the opinion had been adequately tested, had acceptable error rates, was peer reviewed, and was generally accepted in the pertinent field. The Daubert standard was later codified in an amendment to the Federal Rule of Evidence 702, which requires, among other things, that the court determine that “the testimony is based on sufficient facts or data,” that “the testimony is the product of reliable principles and methods,” and that the “opinion reflects a reliable application of the principles and methods to the facts of the case.”[7]

The Reference Manual gives judges the information they need to carry out their gatekeeping responsibilities under Rule 702. Without this resource, a judge with little background in science will have considerable difficulty evaluating the reliability of the foundation for proffered expertise. Although the parties before the Court will duly assert their respective scientific views, the adversarial system tends to lead to polarized presentations. The Reference Manual provides an independent introduction and overview of the general science of the respective fields covered. This provides a framework by which judges can adjudicate complex scientific cases.

Understanding the research basis for scientific evidence requires knowing fundamental concepts of hypothesis testing, general research design, and statistics. This is the language of science, and the Reference Manual is a subject-by-subject guide regarding the methods and general findings of the fields included.

The Reference Manual does not instruct judges on how to decide individual cases. The science covered in the Reference Manual is general. Issues concerning science are constantly litigated. For example, juries might be asked to decide whether benzene caused the plaintiff’s leukemia, whether the DNA profile of the defendant supports guilt, whether the bitemark on the victim matches the defendant’s teeth, or whether the severity of a particular storm event was enhanced by human influence. However, before the jury ever hears testimony on benzene’s effects, DNA random match probabilities, bitemark matching, or the impacts of a superstorm, the judge must determine whether this testimony is admissible, drawing on his or her knowledge of, respectively, epidemiology, population statistics, metrology, and climate science. The Reference Manual helps them make those decisions.

A flawed attack. The authors of the climate science chapter have posted a defense of its contents here, in which they systematically dismantle each of the arguments used to exclude it from the handbook. Many of the authors of other chapters in the Reference Manual have publicly decried this political attack.

Some background to this scientific “debate” is illuminating. In 2025, Energy Secretary Chris Wright commissioned a “team of independent experts to critically review the current state of climate science.” Secretary Wright concluded that this review contradicted the mainstream narrative, but failed to note that the Energy Department report misrepresents the scientific understanding of human impacts on temperature at different altitudes in the atmosphere. This fundamental error is just one of many egregious errors in the Energy Department report identified in a comprehensive analysis by 85 experts in climate science.

Earlier this year, a federal judge ruled that the Energy Department violated the law by handpicking a secret group to write this climate report, but did not strike the report from the public record. This means that in spite of its serious scientific flaws and the illegal manner in which the report was produced, the report is still available, and is still being presented as a credible scientific assessment. For example, one of the complaints levied by the attorneys general about the climate science chapter of the Reference Manual is that the authors “do not cite, for instance, any of the leading experts from the Department of Energy’s recent report on climate change.”

Science dies in darkness? Over the last 30 years, the Reference Manual has proven to be an essential tool for judges and, perhaps equally, for lawyers. Without that resource, judges would need to resort to lesser alternatives, including AI-aided internet searches. Although not without its uses, AI can produce uncurated, possibly hallucinating internet content. This is a poor substitute for the deeply researched and peer reviewed content in the Reference Manual.

Another benefit of a comprehensive resource like the Reference Manual is that it is transparent. Critics of the climate science chapter can make their objections known generally and to courts in specific cases. The Reference Manual, like any resource, is not perfect or fully comprehensive on the subjects it covers. A party that disagrees with any part of the Reference Manual will have an adequate opportunity to make its disagreement known. In contrast, a litigant whose judge consulted an AI chatbot for background information on the subject is at a greater disadvantage than a litigant who knows that the source of the judge’s knowledge base is the Reference Manual. It is easier to battle a known enemy than one imagined.

Knowledge is rarely, if ever, advanced through censorship. The Reference Manual’s chapter on climate science is an invaluable resource and a major contribution to the literature. It does not need to be, and is not, the last work on the subject. Both science and the law advance by open debate and a willingness to hear alternative perspectives. The complaints of the attorneys general, therefore, should be addressed by the publication of alternative views on the state of the science. These alternative views can then be subjected to rigorous scientific scrutiny—as in the case of the Energy Department’s climate report.

Removing the climate science chapter advances neither science nor the law. It only deprives judges of a resource needed to do their jobs.

ENDNOTES
[1] See Karen Sloan, College students with these majors crush the LSAT. End of logic games may change that., REUTERS (June 4, 2024), https://www.reuters.com/legal/legalindustry/collegestudents-with-these-majors-crush-lsat-end-logic-games-may-change-that-2024-06-04/[https://perma.cc/4BSK-J7PW] (“Fewer than 5,000 STEM majors applied to law school in 2023— just 6.6% of the national applicant pool, according to data from the Law School Admission Council. More than 50,000 arts and humanities or social sciences majors applied.”).

[2] Transcript of Oral Argument at 22, Massachusetts v. E.P.A., 549 U.S. 497 (2007) (No. 05-1120).

[3] Loper Bright Enters. v. Raimondo, 144 S. Ct. 2244 (2024).

[4] 467 U.S. 837 (1984).

[5] David L. Faigman, Overcoming Judicial Innumeracy: A Proposal to Bring the Venerable Process of Peer Review to the Enduring Problem of Courts’ Scientific Illiteracy, 26 Colum. Sci. & Tech L. Rev. 76 (2025).

[6] 509 U.S. 579 (1993).

[7] Fed. R. Evid. 702.