Click here to subscribe to our Patreon, where you'll get free merch and receive exclusive access to The Pebble, which features Ethan's full monthly column and snippets of The Sweaty Penguin's latest climate solutions stories.
With birthright citizenship, transgender athletes, and the independence of the Federal Reserve dominating the end of June’s wall-to-wall Supreme Court coverage, you’d be forgiven for missing the big environmental decision that snuck out with them.
I’ll spoil it. The chemical company won, 7-2. In dissent was Justice Ketanji Brown Jackson and… wait for it… Justice Neil Gorsuch.
The company? Monsanto, the infamous manufacturer of the herbicide Roundup.
Their argument? That the Environmental Protection Agency should have more authority over how Roundup is regulated.
What?!
Regular news viewers might find this bizarre right out the gate. There’s liberals and conservatives in the majority, and the dissent. Liberal Justices Sonia Sotomayor and Elena Kagan sided with Monsanto. Conservative Justices Samuel Alito and Clarence Thomas sided with the EPA. And that whole group is on the same side.
The case itself, called Monsanto v. Durnell, was not asking whether Monsanto is the heart of all evil. Rather, it concerned the question of whether a state court in Missouri is allowed to find Monsanto liable for not putting a cancer warning label on Roundup if the federal EPA approved a label without one and did not find Roundup to be carcinogenic. Seven said it is not. Two said it is.
Both views are completely understandable, in theory and in practice. If the federal EPA approves a warning label, it makes sense that state courts should not fine or punish a company for using that approved label, as the seven felt. It would be quite burdensome to abide by contradictory state and federal regulations, and federal beats state. On the flip side, one could argue — actually, two could argue — Congress made it illegal for companies to misbrand the dangers of their pesticides, and the EPA’s safety judgment is not the be-all and end-all. If a state jury hears a case and rules differently, that’s their prerogative. Either interpretation of the law is fair, and neither is “pro-environment” or “anti-environment.”
But the court’s strange split is far from the juiciest part of the story. In 1985, the EPA classified glyphosate, the active ingredient in Roundup, as a possible carcinogen; but reversed course in 1991, classifying it as having “evidence of non-carcinogenicity for humans.” Many of the studies the EPA relied on were conducted by Monsanto and not in peer-reviewed journals or available to the public, and multiple EPA scientists refused to sign the 1991 report. That decision has been shrouded in decades of scandals, from a laboratory contributing studies for Monsanto getting exposed for fraud with multiple executives criminally convicted to a major 2000 scientific paper getting retracted last year due to Monsanto’s undisclosed involvement.
In 2015, the International Agency for Research on Cancer (IARC) classified glyphosate as “probably carcinogenic to humans.” Monsanto led a multi-million dollar campaign to discredit IARC.
The list goes on.
In 2018, a San Francisco jury ordered Monsanto to pay $289 million to a former school groundskeeper who was dying of cancer and argued Roundup contributed to his disease. His attorney? Robert F. Kennedy Jr. Just in case the scrambled political coalitions weren’t confusing enough.
Periodically, the EPA is required to reevaluate the pesticides they regulate under the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA), including glyphosate. Their next deadline is October 1, 2026. And because of this new Supreme Court ruling, if the EPA continues to defy the findings of countless toxicologists and deem glyphosate non-carcinogenic, states won’t be able to require Monsanto put a warning label anyway, making the specific EPA classification more consequential than before. That’s not a critique of the court. It’s a demand for the EPA to take this reevaluation seriously.
The history
In 1974, the newly-created EPA registered glyphosate as a pesticide. Monsanto, who owned the patent, launched their glyphosate-based herbicide Roundup that same year. Roundup specifically inhibits an enzyme in the shikimate pathway of plants — a metabolic route that helps plants produce the amino acids that are required for synthesizing proteins and lignin in the cells. Losing those amino acids takes away plants’ ability to grow and develop, causing them to die. Since humans and animals do not have this shikimate pathway, the herbicide was assumed to pose no risk to humans.
As scientists would learn over the coming decades, that logic isn’t quite right. Yes, humans do not have this pathway. But bacteria do, and humans depend on bacteria in our gastrointestinal tract to synthesize amino acids. Disrupting the formation of these critical amino acids can create a host of health problems in humans and animals, even without the shikimate pathway that Roundup targets.
When Roundup was launched, Monsanto marketed it as effective, biodegradable, and safe for humans. Two years later, their scientific accolades were called into serious question. Federal regulators began uncovering serious irregularities at Industrial Bio-Test Laboratories, then one of the country’s largest private toxicology labs in Illinois. Investigators found that some experiments were conducted under appalling conditions with loose wild rats in the labs, records were missing or falsified, animals were replaced without documentation, and data were sometimes simply invented. The level of scientific fraud was unprecedented.
Monsanto, their biggest customer, not only knew of these problems, but participated. When early toxicity studies on polychlorinated biphenyls (PCBs) produced results Monsanto found concerning, the company had IBT repeat some of the experiments using "cleaned up" PCB samples in hopes of obtaining more favorable safety findings. After a later IBT draft report described one type of PCB as "slightly tumorigenic," Monsanto requested that the conclusion be rewritten to say it "does not appear to be carcinogenic," and IBT did it. A Monsanto employee Paul Wright even switched jobs in 1970 to be the director of IBT’s toxicology lab overseeing Monsanto’s studies, where he was caught fabricating data. His then-lab assistant testified that he’d watched Wright “make up numbers out of his head.” After cooking the books at IBT, Wright then moved back to Monsanto in 1972 as Monsanto’s toxicology manager. The scandal eventually led to criminal indictments and convictions for Wright and three IBT officials. Even though IBT’s data was found fraudulent, Monsanto has continued citing it, even as recently as 2018.
Through the rest of the decade, as Monsanto submitted testing data to the EPA in support of glyphosate’s original registration, EPA staff grew concerned about the data’s integrity. In one case, the lab apparently stated that they took specimens from the uteri of male rabbits, despite male rabbits not having uteri. Eventually, in 1983, Monsanto submitted a mouse study to the EPA which was not peer-reviewed or made publicly available. The study used 200 mice, with 50 in a control group, and three other groups of 50 exposed to varying doses of glyphosate. One mouse at the middle dose and three at the highest dose developed kidney tumors called adenomas, which can be a precursor to cancer. In 1985, EPA scientists declared glyphosate a Group C substance, which describes a “possible human carcinogen.”
Unhappy with the finding, Monsanto hired four outside scientists to review the original tissue slides from the mouse study, issuing a memo that implied a favorable outcome in the review. Sure enough, one of the scientists saw a tumor in one of the mice in the control group, shattering the previously determined correlation between glyphosate and the tumors. When the EPA re-reviewed the kidney slides and found no such tumor, Monsanto had five different experts write opinions and testify to the EPA’s FIFRA Scientific Advisory Panel (SAP) insisting they believe the control group had a tumor. The SAP included a panelist who had worked for ChemAgro Corporation and another who headed biochemical toxicology at the Chemical Industry Institute of Toxicology, an institute founded and funded by pesticide manufacturers and industry groups. Despite EPA scientists’ objections to the Monsanto experts and the panel’s intrusion (including one writing that the odds of the kidney tumors being random were about 156 to 1), the SAP proposed reclassifying glyphosate to Group D (a non-classification requiring further study). On that advice, the EPA tossed the mouse study and requested Monsanto repeat it.
Monsanto refused. After years of back and forth, it was the EPA who eventually caved in 1989, concluding that a repeat study was not required at that time. To this day, Monsanto has never repeated that study.
In 1991, the EPA downgraded glyphosate to Group E, which covers substances that “show evidence of non-carcinogenicity for humans based on the lack of convincing evidence in adequate studies.”
Monsanto’s next move saw Roundup sales skyrocket. In 1996, Monsanto launched the first “Roundup ready crops,” which were genetically engineered to withstand being sprayed with glyphosate. Now, a farmer could spray glyphosate on an entire field, killing every weed and leaving their crop unharmed. U.S. applicators went from using 12.7 million pounds of glyphosate in 1990 to 40 million in 1995 to 98.5 million in 2000 to 180 million in 2005 to 261 million in 2010. The surge saw glyphosate become the most widely used herbicide in the world.
The science
Meanwhile, scientists conducting actual peer-reviewed research started to discover cause for alarm. A 2002 Leukemia & Lymphomapaper from Sweden found that farmers exposed to glyphosate were more likely to develop non-Hodgkin's lymphoma than unexposed farmers, though with a relatively small sample of glyphosate users. In 2003, an Occupational and Environmental Medicine study found a similar correlation among a larger sample size of farmers in the midwestern U.S. Human studies continued to trickle in, many with concerning albeit limited findings. Meanwhile, several long-term animal studies reported increases in uncommon tumors, and scientists were even finding the actual DNA damage and oxidative stress in the liver, kidneys, and other parts of rodents, as well as in cultured human cells. These sorts of studies help scientists go from finding correlation between glyphosate and cancer to actually building evidence for a causal relationship.
In addition to Monsanto’s obvious meddling, there were some major problems with applying the 1980s research to the real world. First, people were not exposed to high quantities of glyphosate in a lab, but smaller quantities over several decades. Second, people were not exposed to just the chemical glyphosate, but commercial pesticide formulas such as Roundup. While scientists must research the individual chemical, scientists did find that the surfactants and other co-formulants that Roundup contained to help its performance also produced greater human health risk than glyphosate by itself.
All of this research came to a head in 2015 when the World Health Organization’s International Agency for Research on Cancer (IARC) reviewed glyphosate. A working group of 17 experts from 11 countries convened to review all publicly available published scientific evidence, ultimately reviewing about 1,000 studies. Based on the combination of limited evidence of cancer in humans from real-world exposures and sufficient evidence of cancer in animals from lab experiments with pure glyphosate, IARC classified glyphosate as “probably carcinogenic to humans.”
Several months before IARC actually met to evaluate glyphosate in 2015, Monsanto was gearing up for the fight of their lives. They were so confident that the agency would classify glyphosate as a carcinogen that they’d begun preparing a coordinated response before IARC had issued any conclusion. Internal documents called for a strategy to “orchestrate outcry,” attacking the classification and the credibility of IARC as an institution through aggressive media and social media outreach, op-eds and commentary from outside experts, pressure on regulators from growers and industry associations, and support from organizations presented as independent voices.
Monsanto’s allies went so low as to attack individual IARC scientists of corruption, conspiracy, lying, secrecy, and even that they were motivated by “personal profit and ideological vanity,” despite all IARC scientists being volunteers. They even fed a false narrative to a Reuters journalist suggesting that the IARC scientist leading the review had withheld key data. Her news story later won the 2017 Science Story of the Year Award from the Foreign Press Association (with whom Monsanto’s parent company happens to have a sponsorship deal) and became a key source used by the U.S. House of Representatives in calling for a hearing to investigate IARC scientists’ misdeeds and possibly cut U.S. funding to the group. And the letter to call for the hearing in Congress? Literally drafted by Monsanto consultants.
That’s not all the secret writing Monsanto did. Internal emails discussed having Monsanto employees ghostwrite review papers, and outside scientists ”would just edit & sign their names so to speak.” Some of these papers were published as an entire supplemental issue of the journal Critical Reviews in Toxicology, falsely stating that no Monsanto employee reviewed the manuscript. The journal’s publisher later released an “Expression of Concern,” but declined to retract the papers.
Over five years, seven front groups spent a whopping $76 million supporting Monsanto’s glyphosate defense. And the EPA’s verdict? Still unchanged since 1991.
The case
Despite the EPA’s stubbornness, the IARC finding spurred thousands of lawsuits across the U.S. with plaintiffs who had used Roundup and developed cancer. Several plaintiffs won multi-million dollar payouts from Monsanto. But while the body of scientific research would surely back up the plaintiff’s claims, the lawsuits clashed with the federal EPA regulations that Monsanto, of course, had not violated. That confusion is where the Supreme Court was asked to weigh in.
In 2019, John Durnell sued Monsanto in Missouri state court, alleging that he had used Monsanto’s Roundup products for about 20 years and that they had caused his non-Hodgkin’s lymphoma. Durnell brought a failure-to-warn tort claim, which essentially says that if a company fails to provide adequate safety warnings on a product, they are false marketing and can be held liable under state laws. His claim argued that Monsanto should have included a cancer warning on Roundup’s label. A Missouri jury agreed and awarded Durnell more than $1 million, which Missouri’s Court of Appeals upheld.
Monsanto’s argument to the Court of Appeals, and then the Supreme Court, was that Missouri could not find them liable under state law due to something called preemption. Preemption is the legal theory whereby a higher level of authority (e.g. the federal government) overrules a lower level of authority (e.g. a state government) in the case of a clash between two laws. This can be “express preemption” where a statute explicitly says lower levels of government cannot make laws contradicting it, or “implied preemption” where judges determine the higher authority intended to fully control the issue or it would be impossible to comply with both laws simultaneously. Preemption has the drawback of denying states or municipalities the ability to set their own regulations on a particular issue, but it often does make more sense for the rules to be the same nationwide.
In this case, Monsanto pointed to FIFRA, the federal statute that governs the registration, distribution, sale, and use of pesticides in the United States. FIFRA, they argued, has an express preemption clause entitled “Uniformity,” stating that a “State shall not impose or continue in effect any requirements for labeling or packaging in addition to or different from those required under this subchapter.” Since the EPA had registered Roundup, classified glyphosate as non-carcinogenic, and approved the label, Missouri could not require a different label. Monsanto further advanced the impossibility argument, claiming that once the EPA approves a label, they are required to use it exactly as written and could face civil or criminal penalties for deviating. Were they to follow Missouri’s law, they’d be violating federal law.
Durnell’s attorney disagreed. He pointed to language in FIFRA stating that EPA registration of a pesticide should not be “construed as a defense for the commission of any offense under [FIFRA],” but that registration is “prima facie” evidence of compliance with the registration provisions. Prima facie evidence, in English, refers to evidence that has merit and can be taken “at face value,” but is not irrefutable. In other words, a jury should take the EPA’s approval of Roundup into consideration, but is welcome to consider other evidence too. FIFRA, he argued, requires that companies not misbrand pesticides and properly warn of their dangers, and Missouri’s requirement is not “in addition to or different from” that. It’s the same. The statute expressly says that an EPA registration is not indisputable proof of compliance, and the Missouri court was free to find otherwise.
Monsanto won. Writing for the majority, Justice Brett Kavanaugh declared that “FIFRA preempts a state-law labeling requirement that differs from the federal labeling requirements imposed under FIFRA. ‘Uniformity’ in labeling — the textually stated objective of FIFRA’s preemption clause — would otherwise be impossible to achieve.” Unless and until the EPA changes Roundup’s approved label, Missouri cannot demand a different one. The decision makes good legal sense, and would feel far more satisfying if the EPA’s Roundup classification weren’t so scientifically disputed and fraught with scandals. One could imagine a reverse scenario where the EPA demanded a warning label, a state keen on supporting industrial agriculture demanded the warning be removed, and this exact same ruling upholds environmental protection. It is far more efficient to have companies follow one uniform regulation than to have to navigate fifty regulations in fifty states, hence FIFRA’s uniformity clause.
Justice Jackson’s dissent, joined by Justice Gorsuch, insisted that the most convenient reading of FIFRA is not the correct reading. “Durnell’s failure-to-warn claim is not ‘in addition to or different from’ FIFRA’s mandates; it is equivalent to FIFRA’s key labeling requirement — the misbranding prohibition. And Durnell’s claim does not conflict with any other FIFRA ‘requirement’ because the EPA’s registration of a pesticide and approval of its label does not create a labeling requirement under FIFRA.” She concluded with a refute of Monsanto’s impossibility argument, pointing out that Monsanto could easily follow contradictory labeling requirements from Missouri and the EPA by (a) asking the EPA to let them put the cancer warning which they’d be unlikely to say no to or (b) simply not selling Roundup. Both laws, she contends, require Monsanto not to sell a misbranded product, and leaving off this warning is misbranding according to the Missouri courts. It’s certainly messier than the majority opinion, but all very good points.
The court
I recently finished listening to Last Branch Standing, a new book on the Supreme Court by Sarah Isgur. I’ve long loved calling out the failures of mainstream media outlets and environmental advocates to communicate environmental Supreme Court decisions accurately. Though I haven’t read many cases and am no legal expert, every environmental case I have read concerned separation of powers and/or ambiguous phrasing in a law. Decisions are almost always along the lines of “Congress, this is what we think you meant, but please override us if we’re wrong.” Headlines cover them as “Supreme Court guts climate agenda,” and only the 6-3 or 5-4 conservative-majority cases make the news at all.
So I was relieved that Isgur, an actual expert, not only verified my frustration, but elaborated on it significantly. This court is not a 6-3 court, but a 3-3-3 court, she argues. Along an ideological axis, there are the three liberals on the left and six conservatives on the right. But there’s a second “institutional” axis, measuring how much justices defer to past precedents, value building consensus on decisions, consider on-the-ground consequences of a decision, and otherwise care about upholding the Supreme Court as an institution. On this axis, Justice Kagan scores highest among the liberals, whereas the conservatives have a clear split: Chief Justice John Roberts, Justice Amy Coney Barrett, and Justice Kavanaugh as high-institutionalists and Justices Gorsuch, Thomas, and Alito as low-institutionalists.
That’s not to say these high-institutionalist justices have never overturned precedent or disrupted people’s lives with their decisions. But it does offer an explanation for why the court so infrequently decides cases along ideological lines. In the 2024-2025 term, the most common outcome for the court was a unanimous one, representing 42% of all decisions. That’s quite a feat considering the Supreme Court tackles only the absolute hardest legal questions vexing America’s lower courts. Only 15% of decisions had all three liberal justices in dissent, and quite symmetrically, 15% had three or more conservative justices in dissent. In that term, Justices Kavanaugh and Barrett agreed with Justices Kagan and Sotomayor more frequently than they agreed with their fellow conservatives Justices Gorsuch, Thomas, and Alito.
The 2025-2026 term saw more cases fall along the ideological axis, surely to complicate Isgur’s book tour. But not dramatically so. 44% of decisions were unanimous, 24% saw the three liberals in dissent, and 15% again saw three or more conservatives in dissent. Notably, in the three most flagrant cases of President Trump issuing disastrous executive actions well beyond the president’s power — tariffs, birthright citizenship, and firing a member of the Federal Reserve’s Board of Governors — Chief Justice Roberts did not blink. He authored all three opinions himself, built cross-ideological coalitions of himself, the three liberals, and one or more of Trump’s appointees each time, and shot down the president’s demand. On the Mississippi mail-in ballot case, another Trump priority, Justice Barrett took the lead alongside the three liberals and the Chief. Often, critics of the 3-3-3 theory posit that “well, the most important and most political cases always fall on ideological lines.” Certainly not so in the latest cycle.
And although one of the most memorable moments of the Chief’s tenure will no doubt be the overturning of Roe v. Wade against his wishes, the Roberts court is actually overturning past precedents at a lower rate than any court in the last century.
Politicians, pundits, and activists on the left and right alike have derided the Roberts court as corrupt, partisan, and every other insult in the book. But directing anger at the court in most cases makes little sense for those concerned about the on-the-ground impact of a ruling. No ruling is etched in stone; the Supreme Court merely interprets what the law is currently, whether it can be changed by a simple majority in Congress, or whether it must be changed via a supermajority through a Constitutional amendment. Attacking the legitimacy of the Supreme Court may score political points in some circles, but the justices are the one branch actually doing their job. It’s on Congress to hold up their end and come to work.
So after listening to Isgur’s book, I was not too surprised to see a case fall 7-2 with two low-institutionalist justices in dissent, or see that case get very little news coverage when it didn’t fit the preferred ideological narrative. In fact, during Justice Jackson’s first term on the court, there were four cases where she and Justice Gorsuch joined an opinion together, just the two of them. In each case — whether to maintain an emergency COVID-19 immigration policy post-pandemic, how big a fine the government could issue a defendant for non-willfully failing to report foreign bank accounts, whether the IRS could summon bank records of a delinquent taxpayer without notice, and whether a county could keep the profits from a house they seized and sold from another delinquent taxpayer — Justices Jackson and Gorsuch converged on a distrust of administrative overreach and a desire to protect legal remedies for the little guy. Pundits even started coining the term “Jackson-Gorsuch axis” to ponder their surprising alliance.
In part due to the relative rarity of solely two-person opinions, it took until this term for the Jackson-Gorsuch axis to reemerge, marking their first joint dissent on a merits docket case (the others were an interim relief docket dissent and three concurrences) and their first time as “the two” in a 7-2. It fits the pattern. They set practical and institutional concerns aside to express skepticism of an unclear administrative power and preserve the little guy’s right to seek relief. Justice Jackson may get there via concerns of civil rights, procedural fairness, and statutory structure while Justice Gorsuch may get there via ardent textualism, separation of powers, and his libertarian streak. But given their low-institutionalist posture, it’s not particularly shocking that a case would emerge where the two of them split from their seven peers. This one had the exact ingredients for it: a little guy seeking relief from a big company, an administrative agency exercising broad power, and a statute where when you really squint at the text and set aside practical considerations, you can rule for the little guy and reign in administrative power.
The deadline
Every 15 years, the EPA is required to reevaluate every registered pesticide under FIFRA. Their next deadline is October 1, 2026.
As with any Supreme Court decision, the court does not have the final word. With this one, the solution is even simpler than most. The EPA must get this reevaluation right.
It may feel easier to blame the Supreme Court than to ask Trump’s EPA to do their job, especially after the Trump administration invoked the Defense Production Act to classify glyphosate as necessary for national defense. But unless Congress wants to amend FIFRA, this job cannot be done by a different agency or branch of government. Though RFK Jr. has been more complimentary of glyphosate since his lawyer days, it does appear to be the first time someone who has literally sued Monsanto is in a federal cabinet, and this EPA has shown some desire to keep anti-pesticide “Make America Healthy Again” activists out of their crosshairs. Since the EPA does listen to public comments, it’s possible these calls from inside the house could have an influence, though of course actual peer-reviewed science must guide the process. Importantly, the EPA is also allowed to reclassify a chemical anytime they feel emerging science warrants it, this October deadline is simply a congressionally required one.
The EPA also does not need to give a firm “yes” or “no” on the question of whether glyphosate is a human carcinogen. As mentioned, there are various classes — definitely, probably, possible, unclear, or not likely — and glyphosate currently sits at the bottom. Even IARC did not call glyphosate a definite human carcinogen, so the EPA need not do that, but it would seem there is too much scientific evidence of carcinogenicity to leave glyphosate where it is.
As someone who regularly engages with scientists, I also find it odd that the EPA relies so heavily on science that is not peer-reviewed or available to the public in these evaluations. If there is actual contradictory data casting doubt on glyphosate’s carcinogenicity, it ought to be peer-reviewed, published in reputable journals, and available to the public. Scientists around the world could then replicate those experiments to see if they get the same results, or debate how they square with the rest of the scientific literature on the topic.
Setting aside Monsanto’s clear history of meddling with scientists and regulators, it is understandable that IARC has a different mandate than the EPA and similar agencies around the world. IARC’s mandate is to coordinate research on the possible causes of cancer and determine preventative strategies. In addition to glyphosate, IARC also classifies night shift work, hot beverages, and steak as “probable human carcinogens,” and processed meat, wood dust, and tanning beds as definite human carcinogens. These evaluations are absolutely based in science, and also likely elicit eye rolls from most Americans, myself included. The EPA, on the other hand, must weigh costs and benefits of their decisions, and one could argue the most used pesticide in the world supporting global food systems deserves the benefit of the doubt when IARC classifies it at a lower level than salami.
But a warning label is not a ban, it’s a free market tool to educate consumers about the product they’re purchasing. Farmers can still spray Roundup, but with a warning label, may do so more carefully or consider a transition to other products on a timeline that works for them. Though Roundup was a smart innovation 50 years ago, there are many targeted pesticides with far less safety risk today. The EPA may not have IARC’s expansive mandate to hunt down every known carcinogen, but they also don’t need to treat chemicals as “innocent until proven guilty.” If every formulation of every chemical is allowed to go to market for decades until the EPA finds overwhelming scientific evidence that prompts them to regulate it, they’ve failed to protect the public from environmental harm. It may require directives from Congress, but a far better system would adopt a more holistic approach than “innocent until proven guilty” that encourages safety innovation in the chemical industry, views new formulations of an unsafe chemical with significantly more skepticism, and uses pause buttons when science emerges casting doubt on a chemical’s safety, even if more research must be done.
The Supreme Court’s ruling on Monsanto v. Durnell shouldn’t be seen as controversial, partisan, or anti-environment. But it should be seen as important. Not because it challenged any science, but because it changed who bears the most responsibility for getting it right. So EPA, it’s on you now. Do your job.