THE SHIELD PART THREE: Monsanto
Trust the Regulators
The Supreme Court’s ruling rests entirely on one EPA determination. Here is what that determination is actually worth.
In Part Two of this series, we examined what the Supreme Court actually held in Monsanto v. Durnell and what the dissent documented that the majority chose not to address. The ruling closed the courthouse door on more than 100,000 plaintiffs. It did not declare glyphosate safe. It rested entirely on a single factual predicate: the EPA has determined that glyphosate is not likely to be carcinogenic to humans, and therefore has never required a cancer warning on Roundup’s label.
When a new study published in the Annals of Global Health found this past week that glyphosate causes DNA damage in human cells at the lowest exposure levels tested, Bayer’s response was immediate. The study, the company told The New Lede, was not conducted with the same rigor as studies submitted to regulatory agencies. Assessments by regulatory authorities, Bayer said, do not support the conclusions of this study.
In other words: trust the regulators.
That is a reasonable ask. It is also an invitation to look closely at what the regulators actually did. When you do, the picture is not reassuring.
What the EPA’s Own Paper Said
The EPA’s determination that glyphosate is not likely to be carcinogenic to humans is not a simple finding. It is the headline conclusion of a Cancer Paper the agency produced as part of its registration review of glyphosate. That paper reviewed the available epidemiological studies, animal carcinogenicity data, and mechanistic evidence, and arrived at a bottom-line hazard descriptor: not likely to be carcinogenic to humans.
What the paper said along the way to that conclusion is a different matter.
The epidemiological data the EPA reviewed did not point clearly in one direction. Effect estimates across the case-control studies and meta-analyses the agency examined were greater than 1, meaning most of the studies indicated that glyphosate exposure was associated with an increased risk of non-Hodgkin’s lymphoma. The meta-analyses showed effect estimates ranging from 1.3 to 1.5, indicating that people exposed to glyphosate were between 30 and 50 percent more likely to develop NHL than those who were not.
The EPA reviewed that data and discounted it, attributing the elevated estimates to chance, bias, or confounding factors. That is a legitimate scientific judgment call. What is not legitimate is what the Cancer Paper said next.
The paper explicitly stated that a conclusion regarding the association between glyphosate exposure and risk of NHL cannot be determined based on the available evidence. That sentence, written by the EPA’s own reviewers, means the agency could not reach a conclusion on the central question. The evidence was, in the agency’s own words, insufficient to decide.
The EPA then selected not likely to be carcinogenic to humans as its final hazard descriptor.
Under the EPA’s own Cancer Guidelines, that descriptor is appropriate only when the available data are considered robust for deciding that there is no basis for human hazard concern. The EPA said in one paragraph that it could not determine the cancer risk. It said in the next, in effect, that the data were robust enough to conclude there was no meaningful risk. Those two statements cannot both be true.
That internal contradiction sat inside EPA’s Cancer Paper for years. The scientific debate continued. The lawsuits multiplied. The agency did not revisit its conclusion.
What a Federal Court Found
In 2022, the Ninth Circuit Court of Appeals examined EPA’s glyphosate cancer determination in Natural Resources Defense Council v. EPA. What the court found was not a close scientific disagreement between regulators and critics. It was a documented series of methodological failures that rendered the agency’s conclusion legally unsupportable.
The first problem was the one already visible in the Cancer Paper itself. The court found that EPA could not reasonably treat its inability to reach a conclusion about NHL risk as consistent with a conclusion that glyphosate is not likely to cause cancer. The agency said it could not determine the risk and then issued a determination. The court said that is not how science or law works.
The second problem was the way the EPA handled its animal carcinogenicity data. The agency used historical control data selectively, invoking it only when it undermined tumor results, never when it might support them. EPA’s own Scientific Advisory Panel flagged the practice during the review process, warning that to subjectively choose to use historical control data only in situations where it undermines tumor results is to potentially introduce bias. The agency proceeded anyway.
The third problem is the most striking. The EPA discarded tumor results occurring at or above what it called a 1,000 milligrams per kilogram per day limit dose, treating those results as uninformative for cancer assessment purposes. The Ninth Circuit found that EPA’s own Health Effects Test Guidelines establish no such limit. The guidelines state only that the highest tested dose need not exceed that figure. They do not say results above it should be discarded. The EPA invented a threshold that does not exist in its own methodology and used it to eliminate unfavorable data. The Scientific Advisory Panel called it an ad hoc decision that is not well-justified and is not justified on the basis of the Cancer Guidelines.
The court’s verdict was unambiguous. Inconsistent reasoning is, absent explanation, the hallmark of arbitrary action. EPA’s glyphosate cancer determination did not survive substantial evidence review. The court vacated the relevant portion of EPA’s 2020 interim registration decision and sent it back to the agency for a methodologically sound explanation.
The court acknowledged it was possible EPA could reach the same conclusion on remand. But it would need to get there honestly, through reasoning that did not contradict itself and did not rely on thresholds the agency invented for the occasion.
What EPA Did About It
The Ninth Circuit issued its ruling in June 2022. What happened next is the accountability story at the centre of this series.
EPA withdrew its interim registration decision in response to the ruling. The withdrawal notice, filed by the agency itself, stated explicitly: EPA’s underlying scientific findings regarding glyphosate, including its finding that glyphosate is not likely to be carcinogenic to humans, remain the same.
Read that again. The agency whose methodology a federal appeals court found internally contradictory and legally unsupportable announced, in the very document withdrawing the vacated decision, that its underlying finding was not changing. Not: we are reviewing our methodology. Not: we are examining the specific failures the court identified. The finding remains the same.
Four years passed. As of 2026, EPA had not finalized a revised cancer determination. Glyphosate products remained on the market under existing labels. The agency described its final registration review as anticipated to be completed in 2026, a timeline that had been anticipated and missed before.
In March 2026, with the Supreme Court case already accepted and oral arguments months away, EPA made a public commitment. The agency announced it would undertake a comprehensive, transparent, and rigorous scientific review of glyphosate, including analysis of independent peer-reviewed research on potential cancer and developmental risks and expanded public access to underlying data. The statement was detailed and specific. It named coordination with USDA, HHS, and state agencies. It mentioned strengthened monitoring of residues in food and water.
Three months later, the Supreme Court ruled seven to two in Bayer’s favour, using EPA’s existing not likely to be carcinogenic determination as the load-bearing foundation of the entire decision. The promised rigorous review had not been completed. The vacated methodology had not been corrected. The finding the Ninth Circuit found legally unsupportable was treated by the Supreme Court as controlling federal science.
More than 100,000 plaintiffs lost their clearest path to court on the basis of a determination that a federal appeals court had already found did not hold up to scrutiny.
What the Science Outside the Agency Found
While EPA ran out the clock on its registration review, the independent scientific literature was not standing still.
In 2015, the International Agency for Research on Cancer, the cancer research arm of the World Health Organization, classified glyphosate as a probable human carcinogen. That classification was based on a systematic review of the published scientific literature, conducted by an independent panel of scientists with no industry funding. It has not been retracted.
In March 2026, a group of experts gathered in Seattle to evaluate the scientific evidence on glyphosate health impacts. Luoping Zhang, adjunct professor of toxicology at the University of California Berkeley, told The New Lede that the science is pretty clear on glyphosate’s ability to cause genetic damage in exposed humans and in human cells.
This past week, the Annals of Global Health published the study Bayer dismissed in its statement. Researchers tested ten pesticides and found eight caused DNA damage in human cells. Glyphosate produced cell damage at the lowest exposure levels tested. The researchers concluded there may be almost no safe threshold of human exposure. It was the first study to confirm that pesticide mixtures were also toxic to cells, a finding with implications beyond glyphosate alone.
None of this proves definitively that glyphosate causes cancer in humans at real-world exposure levels. Scientific consensus on complex risk questions rarely arrives in a single study. But the accumulating independent literature points in a consistent direction, and that direction is not toward the EPA’s not likely conclusion.
Bayer’s instruction to trust the regulators over independent peer-reviewed science is therefore not a neutral appeal to authority. It is an appeal to the specific regulatory determination that shields Bayer from liability, produced through a methodology a federal court found arbitrary, restated unchanged after that court’s ruling, and now enshrined as controlling federal law by the Supreme Court.
The Question Now
EPA promised in March 2026 a comprehensive, transparent, and rigorous scientific review of glyphosate. That promise was made before the Supreme Court ruling that eliminated the legal pressure to deliver it. The question of whether that review now happens, and whether it honestly addresses the specific methodological failures the Ninth Circuit identified, is the live accountability thread this series will continue to monitor.
There is a second question, less comfortable but necessary. The EPA does not conduct its pesticide registration reviews in isolation. The process involves industry-submitted studies, agency reviewers, scientific advisory panels, and decades of accumulated regulatory decisions that create institutional inertia. The Ninth Circuit identified not one methodological failure but a pattern of them, each working in the same direction, each discounting data that pointed toward cancer risk.
When a regulatory process produces a consistent pattern of errors that consistently benefit one party, the question that follows is not simply: did the agency make mistakes? It is: why did the mistakes always point the same way?
That question does not have a simple answer, and this series will not pretend it does. What can be said, on the basis of primary documents alone, is this: a federal appeals court found EPA’s glyphosate cancer methodology internally contradictory and legally unsupportable. The agency restated its conclusion without fixing the methodology. The Supreme Court used that unrevised conclusion to close the courthouse door on 100,000 plaintiffs. And Bayer responded to new independent science by telling the public to trust the regulators.
In Part Four of this series, we map the full architecture of how that outcome was built, track by track, and ask what accountability looks like when the courtroom is no longer an option.
Editor’s Notes
Primary sources for this piece: Natural Resources Defense Council v. U.S. Environmental Protection Agency, 38 F.4th 34 (9th Cir. 2022); EPA withdrawal notice following Ninth Circuit vacatur, 2022; EPA public statement on glyphosate registration review, March 2026; Annals of Global Health, pesticide genotoxicity study, July 2026 (as reported by The New Lede); International Agency for Research on Cancer, glyphosate classification as probable human carcinogen, 2015; EPA Cancer Paper, glyphosate registration review (internal language quoted via Ninth Circuit opinion); Bayer AG statement to The New Lede, July 2026.
Confidence tier: All findings attributed to the Ninth Circuit opinion are drawn from the published decision at 38 F.4th 34. EPA’s withdrawal notice language is quoted as reported in court filings and agency records. The Annals of Global Health study is peer-reviewed; the finding that glyphosate showed cell damage at the lowest tested exposure level is reported directly from The New Lede’s coverage of the study. No claims in this piece assert definitively that glyphosate causes cancer in humans; the piece reports what the scientific literature, the EPA’s own Cancer Paper, and the Ninth Circuit found, accurately and at the correct epistemic level.
The Old Guardian | theoldguardian.ca