“Design defect survives” is a reassuring headline and an empty one. It tells a person with lymphoma that a door is still open without telling them what is on the other side of it. What is on the other side is a specific, demanding set of proofs — and a Massachusetts order issued this month spells out exactly how demanding.
The Short Answer
A design-defect claim argues that the product was unreasonably dangerous as designed, independent of anything written on its label. To prevail, a Roundup plaintiff generally has to establish three things: that a reasonable, safer alternative design was available; that the product fails the governing state’s test for defectiveness; and that the exposure actually caused this person’s cancer, proven through expert testimony a court will admit. Nothing about the Supreme Court’s June ruling made any of that easier. What it did was remove the simpler theory that used to carry much of the load.
What Just Happened in Massachusetts
In Rubin v. Monsanto Co., No. 4:26-cv-40079, in the U.S. District Court for the District of Massachusetts, U.S. District Judge Margaret Guzman addressed the question every pending Roundup plaintiff has been waiting on: does Monsanto v. Durnell end the case? Reported on August 5, 2026, her answer was no — not automatically.
Judge Guzman read the Supreme Court’s decision as label-focused. As she put it, she believed Durnell “left open other state law claims, such as the other types of product defect and consumer claims that are not based on warning theories.” She also suggested there may be room for a theory built on warnings Monsanto could have given outside the EPA-approved label. Rather than dismissing, she gave the plaintiffs until August 28, 2026 to amend their complaint and identify additional expert testimony.
The other half of the order matters just as much. Judge Guzman was explicit that she was granting a chance, not a result: “This is an opportunity for the plaintiffs; it is not a foregone conclusion that it will be successful.” She also ruled that punitive damages are unavailable under Massachusetts law in this case. That is the honest shape of the post-Durnell landscape — narrower, harder, and still open.
What Durnell Preempted — and What It Did Not Touch
On June 25, 2026, the Supreme Court held 7–2, in an opinion by Justice Kavanaugh, that FIFRA’s preemption clause, 7 U.S.C. § 136v(b), bars state failure-to-warn claims that would require a pesticide manufacturer to add or change warnings on an EPA-approved label where the EPA has evaluated the risk. Our full breakdown of the holding is here: what the Monsanto v. Durnell decision actually decided.
Defense-side analysts read the boundary the same way plaintiffs’ lawyers do. Crowell & Moring’s client alert on the decision put it plainly: Durnell “does not foreclose all state law claims,” and claims not predicated on labeling requirements — design defect, manufacturing defect, advertising and promotional claims, and claims involving risks the EPA did not affirmatively consider — may remain viable. That is not a plaintiff’s spin on a loss; it is the consensus reading of a narrow holding.
Element One: A Reasonable Alternative Design
The core of a modern design-defect claim is comparative. Under the Restatement (Third) of Torts: Products Liability § 2(b) (1998), a product is defective in design when the foreseeable risks of harm “could have been reduced or avoided by the adoption of a reasonable alternative design.” The plaintiff, in most states, has to identify that alternative — not gesture at one.
In the Roundup litigation, the alternative-design argument has focused on the formulated product rather than glyphosate in isolation. Original Roundup formulations combined glyphosate with surfactants, including polyethoxylated tallow amine (POEA), and IARC’s 2015 review in Monograph Volume 112 examined glyphosate-based formulations, not the active ingredient alone. Our piece on Roundup versus other glyphosate products covers how formulation differences bear on an exposure history.
There is also a fact Bayer supplied itself. Beginning in 2023, the consumer version of Roundup sold in the United States no longer contains glyphosate; Bayer replaced it with different active ingredients and described the change as a step taken to manage litigation risk rather than an admission about safety. A marketed, functioning weedkiller without the ingredient at issue is directly relevant to whether a safer design was feasible and commercially available. We treat that carefully, and we explain why in the new “glyphosate-free” Roundup and the old product you actually used.
Element Two: The Defect Test — Which Depends Entirely on Your State
Product-liability law is state law, and states do not define “defective design” the same way. This is the part most national update pages skip, and it changes what evidence carries a case.
- California applies the two-pronged framework from Barker v. Lull Engineering Co., 20 Cal. 3d 413 (1978): a product may be found defective either because it failed to perform as safely as an ordinary consumer would expect, or because the risk of danger in the design outweighs its benefits.
- Massachusetts — the jurisdiction in Rubin — routes design claims through the implied warranty of merchantability, which the Supreme Judicial Court in Back v. Wickes Corp., 375 Mass. 633 (1978), described as the Commonwealth’s functional equivalent of strict liability. Fitness is judged by weighing factors including consumer expectations, the degree of danger, the feasibility and cost of alternative designs, and the adverse consequences of those alternatives.
- Many other states apply a risk-utility balancing test, a consumer-expectations test, or a blend, with their own gatekeeping rules on when a plaintiff must produce an alternative design at all.
Because the warning theory is off the table, the design claim now carries more of the case than it did a year ago — which means the governing state’s test matters more than it did a year ago. We walk through that shift in whether Durnell reaches state-court cases, and the theory itself in strict liability versus failure-to-warn.
Element Three: Causation, Proven by Experts a Court Will Admit
This is where the Massachusetts order should reset expectations. Judge Guzman did not simply say “proceed” — she required the plaintiffs to identify additional expert testimony. Changing the legal label on a claim does not lower the scientific bar, because causation is an element of every product-liability theory, warning-based or not.
A Roundup plaintiff must establish two layers. General causation asks whether glyphosate-based herbicides can cause non-Hodgkin lymphoma at all. That is where the published record does its work: IARC classified glyphosate as “probably carcinogenic to humans” (Group 2A) in Monograph Volume 112 in 2015, and the Zhang 2019 meta-analysis in Mutation Research/Reviews in Mutation Research reported a 41% increase in NHL risk in the most highly exposed groups (meta-relative risk 1.41, 95% CI 1.13–1.75). Specific causation asks whether this person’s exposure was a cause of this person’s cancer — a question about their products, years, frequency, and diagnosis.
Both layers come in through experts, and in federal court those opinions must satisfy Federal Rule of Evidence 702 and the reliability standard of Daubert v. Merrell Dow Pharmaceuticals, 509 U.S. 579 (1993). An unsupported exposure history is the most common way a scientifically sound claim fails, which is why we treat documentation as a first-week task, not a trial-prep task. See how to build the Roundup exposure record and what the human evidence on glyphosate and lymphoma shows.
Why This Ruling Is a Caution as Much as an Opening
Two things are true at once, and marketing pages tend to publish only the first. Yes: a federal judge, applying the Supreme Court’s newest pesticide decision, declined to end a Roundup cancer case and identified non-warning theories as the road forward. Also yes: the same judge called success “not a foregone conclusion,” put a short amendment deadline on it, required more expert support, and closed off punitive damages under the governing state’s law.
That mix is the realistic outlook for the roughly 3,900 cases pending in the federal Roundup MDL before Judge Vince Chhabria in the Northern District of California, where Bayer has argued that the remaining claims are repackaged warning claims that Durnell should sweep away. We covered that fight in Bayer’s push to dismantle the Roundup MDL. Rubin is one district judge’s view, not binding precedent elsewhere, and other courts may read Durnell more broadly. It is an early and useful data point, not a guarantee.
What This Means for a Case
Practically, the post-Durnell case is built differently from the case Monsanto faced in 2018. Less weight rests on what the label said; more rests on the formulation, the feasibility of a safer one, the governing state’s defect standard, and the strength of the expert record. That is a harder case to assemble and a much harder one to dismiss on a legal technicality, because it does not depend on a warning the EPA already reviewed.
It is also the theory this firm has built on from the start. Alex Alvarez, our Managing Partner and a Board Certified Civil Trial Lawyer, has litigated defective-product cases on strict-liability and design theories rather than warning theories alone — the approach that now matters most. Herb Borroto, M.D., J.D., our Medical-Legal Expert, reviews the pathology personally, because specific causation begins with confirming that the diagnosis is a lymphoma subtype the published science actually supports. Our overview of who qualifies for a Roundup lawsuit explains how that evaluation works.
If you have a pending case, the near-term question is whether your claims as pleaded rely on the label — and if they do, whether they can be amended. That is a conversation for your own lawyer, promptly, because deadlines in post-Durnell orders have been short.
Frequently Asked Questions
What is a design-defect claim in a Roundup case?
A design-defect claim says the product itself was unreasonably dangerous as designed — not that its warning label was inadequate. In a Roundup case that means the formulated glyphosate-based herbicide was defective in its composition, and a safer alternative formulation could have done the same job with less cancer risk. Because the theory does not ask a jury to rewrite an EPA-approved label, it sits outside the failure-to-warn preemption the Supreme Court applied in Monsanto v. Durnell on June 25, 2026.
Did the Massachusetts federal court say Roundup cases can still go forward after Durnell?
In Rubin v. Monsanto Co., No. 4:26-cv-40079, U.S. District Judge Margaret Guzman of the District of Massachusetts indicated the case is not necessarily over, stating that she believed Durnell “left open other state law claims, such as the other types of product defect and consumer claims that are not based on warning theories.” She also cautioned that this is “an opportunity for the plaintiffs” and “not a foregone conclusion that it will be successful,” and set August 28, 2026 as the deadline to amend the complaint and identify additional expert testimony.
What does “reasonable alternative design” mean in a Roundup lawsuit?
Under Restatement (Third) of Torts: Products Liability § 2(b), a design-defect plaintiff generally must show the foreseeable risks of harm could have been reduced or avoided by adopting a reasonable alternative design. In the Roundup litigation the argument focuses on the formulated product — glyphosate combined with surfactants — rather than glyphosate in isolation, and on whether a less dangerous formulation was technically and commercially feasible at the time the product was sold.
Do I still need expert testimony if my claim is design defect instead of failure to warn?
Yes. Changing the legal theory does not change the scientific burden. A Roundup plaintiff must still establish general causation — that glyphosate-based herbicides can cause non-Hodgkin lymphoma — and specific causation, that the exposure was a cause of this person’s cancer. That proof comes from qualified experts whose opinions must satisfy Federal Rule of Evidence 702 and the standard set in Daubert v. Merrell Dow Pharmaceuticals, 509 U.S. 579 (1993). The Massachusetts order requiring additional expert testimony illustrates the point.
Does the state where my case is filed change what I have to prove?
Yes, substantially. Product-liability law is state law. California applies the two-pronged framework from Barker v. Lull Engineering Co., 20 Cal. 3d 413 (1978), allowing a consumer-expectations theory or a risk-benefit theory. Massachusetts channels design claims through the implied warranty of merchantability under Back v. Wickes Corp., 375 Mass. 633 (1978), weighing factors including consumer expectations, the degree of danger, and the feasibility and cost of alternative designs. Which test governs affects what evidence matters most.
Bottom Line
“Design defect survives Durnell” is accurate, and by itself it is not useful. The claim that survives requires a plaintiff to identify a reasonable, safer alternative design, satisfy the governing state’s test for defectiveness — Barker in California, Back v. Wickes in Massachusetts, risk-utility or consumer expectations elsewhere — and carry general and specific causation through experts who can withstand a Rule 702 challenge. Judge Guzman’s order in Rubin is the first clear illustration of both halves of that reality: the door is open, and walking through it takes more, not less, than it did before June.
Nothing here is a prediction about any particular case or about how any court will rule, and past verdicts are never a promise of any future result. Rubin is one district court’s reading of a new Supreme Court decision and could be revisited or rejected elsewhere. This article is general legal and medical information, not advice on your own facts, and it does not create an attorney-client relationship. If you or a family member used Roundup and were later diagnosed with Non-Hodgkin Lymphoma, B-cell Lymphoma, Chronic Lymphocytic Leukemia, or Multiple Myeloma, a free, confidential case review is the right next step.