Marketed on a large scale since the 1970s, glyphosate is one of the most used herbicides in the world. Monsanto's flagship product under the name Roundup, it has become a central agricultural tool, associated with genetically modified crops called “Roundup Ready”. For more than two decades, a scientific study published in 2000 in Regulatory Toxicology and Pharmacology served as a guarantee of the safety of this molecule.
Written by Gary M. Williams (New York Medical College), Robert M. Kroes (Utrecht University) and Ian C. Munro (Cantox Health Sciences International), it concluded that there were no risks to human health. But the journal announced its official withdrawal in November 2025, after the light of a serious conflict of interest involving Monsanto. The trust placed in this publication now raises serious questions about the independence of scientific expertise in public health decisions.
A reference scientific article at the heart of a biased validation system
In 2000, Regulatory Toxicology and Pharmacologya journal specializing in chemical risk assessment, publishes a study entitled Safety Evaluation and Risk Assessment of the Herbicide Roundup and Its Active Ingredient, Glyphosate, for Humans. Authored by Gary M. Williams (New York Medical College), Robert M. Kroes (Utrecht University) and Ian C. Munro (Cantox Health Sciences International), the article concludes that glyphosate poses no risk to human health. This work is immediately perceived as a rigorous and independent synthesis. It then becomes a world reference. It is cited more than 1,300 times in scientific literature, according to Google Scholar, and appears in numerous health agency reports.
This confidence is based on the academic status of the authors and the apparent solidity of the analysis. The document covers several aspects: acute toxicity, carcinogenic effects, impacts on the reproductive system and endocrine disruption. Based on internal Monsanto data – not published – the study ensures that glyphosate does not cause cancer or hormonal disorders. However, this exclusivity of sources poses a problem of methodological balance.
Regulatory agencies, including the US EPA, refer to this document to justify the authorization of glyphosate in many countries. Even if the agency claims today that it did not rely exclusively on this publication, it recognized its quality at the time. The problem arises from the fact that this type of journal, seemingly independent, plays a central role in structuring scientific consensus, thus influencing agricultural policies and public health standards on a global scale.
Direct involvement of Monsanto revealed by American justice
It was only from 2017 that the reality of the process of writing the 2000 study began to emerge. In lawsuits filed against Monsanto by non-Hodgkin's lymphoma patients, internal company documents are made public. These documents, including email exchanges between scientists and Monsanto executives, reveal that the firm directly participated in the writing of the article. But of course without being mentioned as co-author or official partner.
In an email dated 2015, William Heydens, toxicologist at Monsanto, clearly mentions the use of ghostwriting. He suggests writing a scientific article in-house, then having it signed by external researchers. He writes: “
Remember, this is how we handled the 2000 Williams, Kroes, and Munro paper “.
Other posts confirm that several employees worked for three years gathering data, structuring the article, and building relationships with the authors. Lisa Drake, head of government affairs at Monsanto at the time, celebrated the release as a major strategic success. She describes the study as “the reference” intended to “ defending Roundup and Roundup Ready crops around the world “. It justifies this approach within the framework of the internal plan called
Freedom to Operate.
The participation of Monsanto employees is not limited to simple logistical support. Their involvement is central in the selection of data and in the writing. However, in the article, the firm is only thanked for its “scientific support”. No mention of the active role of its employees in writing or designing the findings is evidently made. This concealment constitutes the major ethical basis justifying today the retraction of the study.
Partial data and a lack of methodological transparency
Beyond ghostwriting, the scientific analysis of the study shows a structural bias in the selection of data. The 2000 paper relies exclusively on unpublished internal studies carried out by Monsanto. It excludes academic publications available from 1999. This methodological choice greatly reduces the representativeness of the analysis.
The recent rereading of the article, motivated by Naomi Oreskes (Harvard University) and Alexander Kaurov (Victoria University of Wellington), highlights the absence of an explicit effort to integrate other data available at the time. In its retraction notice, the journal Regulatory Toxicology and Pharmacology states that “ the authors claim to know of other studies, unpublished and not accessible “. But they do not detail inclusion criteria or cross-validation attempts.
This opaque methodology greatly weakens the scientific value of the conclusions, particularly on the potential carcinogenicity of glyphosate. Previous scientific literature already contained warning signals about the chronic toxicity of the molecule. Furthermore, the exclusive dependence on a single source (Monsanto) prevents any critical comparison of the results.
For Alexandra Maertens, toxicologist at Johns Hopkins University, cited by Chemical & Engineering Newsthe lack of transparency represents a major methodological deviation. “ This type of publication should have clearly explained the role of the authors, the choice of sources, and their independence. This is not the case here “. By neglecting independent data and relying solely on results produced by the industry concerned, the study violates the fundamental principles of rigorous scientific evaluation.
A lasting impact on the regulation and public perception of glyphosate
Despite the revelations of 2017, the study remains active in scientific databases until its retraction in November 2025. In 25 years, it has influenced not only the public debate on glyphosate, but also major regulatory decisions. Its visibility in scientific articles, health agency reports and institutional documents has contributed to making the use of Roundup commonplace, particularly in major GMO crops.
For Naomi Oreskes, the impact of paper remains undeniable. It has been used as scientific justification in dozens of regulatory contexts and appears in Wikipedia articles. Sources sometimes used by artificial intelligence. This wide distribution has reinforced a widely shared perception of safety, despite the growing scientific controversies surrounding the molecule.
Since 2015, the WHO has classified glyphosate as “
probably carcinogenic to humans », based on animal experimental data. However, other agencies such as the EFSA (European Food Safety Authority) or the American EPA consider, for their part, that the product can be used without danger if it is applied according to the recommendations.
Meanwhile, Bayer, which has owned Monsanto since 2018, has already paid $11 billion to settle glyphosate-related litigation in the United States. More than 60,000 legal actions remain ongoing. Faced with this pressure, the retraction of the article adds a major argument to the complainants. It calls into question the reliability of the scientific bases which allowed the commercialization of this product on a large scale.
With an unwavering passion for local news, Christopher leads our editorial team with integrity and dedication. With over 20 years’ experience, he is the backbone of Wouldsayso, ensuring that we stay true to our mission to inform.
