It’s Not About Who You Trust — It’s What They Were Given to Work With. How an unsupported claim about a viral infection ended up in New Zealand’s official case for glyphosate’s safety.
For years, this debate has been framed as a contest of authority.
On one side: the World Health Organization’s cancer agency. On the other: our own regulator.
In 2015, the International Agency for Research on Cancer (IARC) classified glyphosate as a “probable human carcinogen.” A year later, the New Zealand Environmental Protection Authority commissioned its own review. One toxicologist, working alone, reached the opposite conclusion: glyphosate was “unlikely to be genotoxic or carcinogenic to humans.”
So — who do you trust?
It’s a fair question. It’s also the wrong one to stop at.
Because when you actually trace where that 2016 report’s conclusions came from, this stops being a question of whom to trust. Nothing in the available evidence suggests Dr Wayne Temple knowingly relied on unreliable material. The larger question is whether the scientific literature available to him accurately reflected the underlying evidence. If influential review papers contained undisclosed industry authorship, unsupported factual claims, or inaccessible study summaries, those weaknesses could be inherited by any regulator who relied on them.
The EPA’s Ready-Made Defence
Before we go further, let’s give the EPA’s argument its due, because it’s not nothing.
IARC identifies hazard. New Zealand’s regulator assesses risk. These are genuinely different questions, and in theory, two honest reviewers could ask them of the same evidence and land in different places without either being wrong.
It’s a real distinction. When NZEPA’s report was challenged in 2017 by NZ public-health researchers, and again in a 2024–25 High Court judicial review, this is the ground the regulator has stood on.
But here’s what that defence quietly assumes: that the evidence being weighed was sound to begin with.
What if it wasn’t?
Three Ghostwritten Glyphosate Reviews, One Monsanto Playbook
In December 2025, a paper that shaped glyphosate policy for 25 years was retracted.
Williams, Kroes & Munro (2000) — “WKM2000” — had been cited by regulators worldwide, including New Zealand’s, as foundational evidence of glyphosate’s safety. Court documents released in 2017 showed Monsanto employees had ghostwritten large sections of it. Outside academics signed their names.
An internal Monsanto email from 2015 laid out the method without much shame: keep costs down, have company staff do the writing, let the credentialed scientists “edit & sign their names.”
Here’s what most coverage of the retraction has missed.
That same email identified two more review papers that would follow the same publication strategy.
- A genotoxicity review by Larry Kier and David Kirkland, published in 2013.
- A carcinogenicity review led by Helmut Greim, published in 2015.
Both landed in a different journal — Critical Reviews in Toxicology. Both are now under active investigation by their publisher, Taylor & Francis, over the same ghostwriting allegations.
This wasn’t one bad paper. It was a template.
The Co-Author They Wrote Out
Want to see how deliberate the “independence” was?
Disclosed emails show Monsanto’s David Saltmiras had substantially contributed to the 2013 Kier & Kirkland paper. He wanted co-author credit. One of the actual named authors, Kirkland, privately agreed he’d earned it.
He said no anyway. In writing, Kirkland explained that adding Saltmiras would mean the authors “would no longer be independent.”
Read that again. The independence wasn’t a fact about the paper. It was a decision about how the paper would look.
(For balance: Kirkland has since publicly disputed that the paper was ghostwritten, telling Retraction Watch that he and Kier had “sole responsibility for the writing and content.” The email record above is what readers can weigh that denial against.)
Saltmiras’s own 2015 performance review — also surfaced in litigation — lists among his achievements that he “ghost wrote cancer review paper Greim et al. (2015).”
He said it. On his own performance review.
The Claim That Wouldn’t Die
Researcher Jason MacLean has traced something worse than undisclosed authorship in the Greim paper: an apparent fabrication.
A mouse study (Kumar, 2001) had shown a statistically significant rise in cancer. Inconvenient. So Greim et al. (2015) explained it away — attributing the result to a viral infection that supposedly compromised the animals.
One problem: toxicologists including Peter Clausing and Christopher Portier say there’s no evidence that infection ever happened.
Didn’t matter. The claim spread anyway — adopted by the US EPA, the European Chemicals Agency, Germany’s BfR.
An unsupported explanation that was subsequently repeated by regulators and even cited in court.
Which raises the obvious question: did it stop at the border?
Where New Zealand’s EPA Comes In
It didn’t.
Dr Wayne Temple’s 2016 report — the one still underpinning New Zealand’s approval of glyphosate today — doesn’t just footnote Kier & Kirkland and Greim.
They are relied upon repeatedly throughout the report.
In the genotoxicity section, Temple devotes a full paragraph to summarising Kier & Kirkland’s conclusions before adopting them in his own assessment: glyphosate shows “a strong weight of evidence” of testing negative, with any positive results waved off as high-dose toxicity rather than genuine DNA damage.
A companion 2015 paper by Kier alone gets the same treatment — used directly to dismiss human biomonitoring data as uninformative.
And then there’s the viral infection.
Discussing the mouse data reviewed by “Griem et al.” (Temple’s report misspells the name — every single time it appears), Temple notes that a study showing a significant rise in lymphomas had its validity “questioned, due to the occurrence of viral infection which could have influenced survival rates.”
Not flagged as contested. Not qualified. Stated.
It resurfaces later, in the report’s own Discussion section, as one of the reasons offered for setting aside IARC’s animal findings altogether.
A disputed claim became part of the reasoning supporting New Zealand’s regulatory conclusion.
The Data Nobody Could Actually Check
There’s one more wrinkle, and it’s in Temple’s own words.
His report admits that IARC’s working group could only independently evaluate five of the nine rat studies referenced in the Greim review. The other four existed only as summaries inside Greim’s own paper — with, in Temple’s phrasing, “limited experimental data provided in the review article.”
So — what exactly was being weighed here?
Four studies that independent reviewers could not fully evaluate because only limited experimental data were available in the review article.
And WKM2000 — the paper retracted in December 2025 — isn’t a footnote either. Temple’s report reproduces its central number, a margin-of-exposure calculation, and quotes its now-retracted conclusion directly: that “under present and expected conditions of use, Roundup herbicide does not pose a health risk to humans.”
So Is It Really About Hazard Versus Risk?
Go back to the EPA’s defence for a second.
Hazard versus risk is a real distinction. It’s a legitimate way for two regulators to reach different conclusions from the same sound evidence.
But that’s the catch — the same sound evidence.
A viral infection that multiple independent toxicologists say never happened isn’t a difference in regulatory philosophy. It’s not a matter of how you weigh a risk. It’s an input error — inherited from a paper now under investigation for ghostwriting, and passed into New Zealand’s official record without anyone independently checking it.
You can debate hazard versus risk all day. It doesn’t resolve concerns about a conclusion that relied, at least in part, on a factual claim now challenged by multiple independent toxicologists.
Kier & Kirkland, Greim: ‘More Consequential Than the One That Got Retracted’
Here’s a detail worth sitting with.
Alexander Kaurov — one of the two researchers whose work led directly to the WKM2000 retraction — doesn’t consider Kier & Kirkland and Greim to be the lesser story.
Speaking to Retraction Watch in June 2026, he put it plainly: those two papers “are arguably more consequential for glyphosate regulation than the Williams paper.”
The US EPA’s own 2016 Glyphosate Issue Paper backs that up. For 17 genotoxicity studies and one carcinogenicity study the agency couldn’t access directly, it says outright: “data and study summaries provided in Greim et al. (2015) and Kier and Kirkland (2013) were relied upon for the current evaluation.”
Not background reading. The data.
What Happens When the Editor Says No
Regulatory Toxicology and Pharmacology retracted WKM2000 because its editor, Martin van den Berg, was willing to look at the evidence and act.
Critical Reviews in Toxicology hasn’t moved the same way. Its editor hasn’t meaningfully responded to the retraction request for Kier & Kirkland and Greim. The journal’s ethics office has offered little beyond confirming an investigation exists.
Worth noting: that editor is also a co-signatory on a piece defending the WKM2000 retraction’s critics.
One journal reviewed the evidence and concluded retraction was warranted. The other has so far chosen to investigate rather than retract.
The question that’s still open — for this journal, and for every regulator, including our own, that relied on these papers — is what happens when they aren’t.
Further reading: our review of Carey Gillam’s Whitewash covers the wider Monsanto playbook in more depth. For the pattern behind the pattern — how the same handful of scientists have manufactured doubt across tobacco, climate, and now glyphosate — Naomi Oreskes and Erik Conway’s Merchants of Doubt* (2010) is the essential read. Oreskes herself later became one of the two researchers whose work led to the WKM2000 retraction covered above.
*For your convenience, we provide links to Amazon.com. If you choose to purchase through these links, we may receive a small commission — at no additional cost to you. Your support helps us continue our work.
No More Glyphosate NZ is a citizen-led campaign demanding safer food, cleaner soil, and honest regulation.
Image Source & Attribution
The feature image on this page was created using AI-assisted image generation from an original concept developed by No More Glyphosate NZ and refined for publication in Canva.
AI is a useful creative tool for visualising complex investigative topics that cannot be meaningfully photographed, allowing us to illustrate ideas without implying that any specific scene or event actually occurred.
