A scientific review paper asserting that the weed killer glyphosate (sold under the brand name Roundup) “does not pose a health risk to humans” has been formally retracted-a full 25 years after it was published-amid substantial ethical concerns about industry manipulation.
The retraction arrives eight years after a 2017 court case concluded that staff at the chemical company Monsanto had been involved in ghostwriting the article’s safety assessment of the herbicide.
Retraction of the glyphosate/Roundup paper: what was claimed and why it matters
The now-withdrawn paper stated there was no evidence that Roundup causes cancer, leads to endocrine disruption, or is toxic to humans. It became one of the most frequently cited scientific publications connected to glyphosate.
Published in 2000 in Regulatory Toxicology and Pharmacology, the article listed Gary Williams, Robert Kroes, and Ian Munro as authors. With those conclusions now officially rolled back, the episode raises renewed questions about the evidential basis often used in arguments over Roundup’s safety.
Monsanto, Bayer, and the continuing safety claims about Roundup
Monsanto first introduced Roundup in 1974. The product and its maker were later acquired by Bayer in 2018, and Bayer continues to say the chemical is safe when used as directed.
At the same time, litigation has been extensive. By 2020, Bayer had paid US$10 billion to settle lawsuits tied to Roundup’s potential carcinogenicity, and more than 67,000 further cases were still pending.
What the journal said: academic integrity and ghostwriting concerns
In a retraction notice published in November 2025, Regulatory Toxicology and Pharmacology co-editor-in-chief Martin van den Berg wrote that the decision was driven by “several critical issues” viewed as undermining the paper’s academic integrity and its conclusions. Van den Berg also reported receiving no reply when attempting to contact Williams, described as the paper’s only surviving author.
Van den Berg noted that the article had long been treated as a landmark in debates about the carcinogenicity of glyphosate and Roundup. However, he wrote that uncertainty over which sections were written by Monsanto employees creates doubt about the reliability of the conclusions.
Key ethical and evidential problems identified
Among the “critical issues” highlighted in the notice were:
- Evaluations of glyphosate’s links to cancer and genetic toxicity relied only on unpublished Monsanto studies.
- The review left out numerous other long-term studies that had already been completed when the article was written.
- The lack of authorial independence was said to raise serious ethical concerns about the authors’ independence and accountability.
- The notice also pointed to non-disclosure regarding Monsanto employees’ involvement and the possibility of financial compensation paid to the named authors.
How glyphosate is used: Roundup Ready crops and intensive spraying
Glyphosate is among the world’s most widely used herbicides, bought by large-scale farmers and home gardeners to eliminate unwanted weeds.
In farming, glyphosate has frequently been marketed alongside Roundup Ready crops-plants genetically engineered to withstand glyphosate exposure. This group currently includes soya, maize, oilseed rape (canola), sugar beet, cotton, and alfalfa.
Because these crops are engineered to tolerate glyphosate, farmers can spray fields broadly, killing plants that lack built-in resistance while leaving the crop largely unaffected.
Wider concerns: human health, ecosystems, and the evidence base
Concerns continue to grow about the potential effects of glyphosate on human health, as well as its far-reaching consequences for wider natural and human ecosystems.
In 2015, the World Health Organization’s International Agency for Research on Cancer (IARC) concluded that glyphosate is a probable carcinogen, drawing mainly on animal studies. Other health agencies and organisations, however, have disagreed with that assessment.
Ultimately, determining whether glyphosate poses a genuine risk to humans will require rigorous, genuinely independent research.
The downstream impact of a highly cited retraction
Harvard scientist Naomi Oreskes found that the retracted paper is cited by more than 800 academic articles, as well as dozens of government documents and multiple Wikipedia entries. She also noted that many large language models now depend on such sources, which means disputed claims can be repeated widely even after concerns are raised.
Retractions do not automatically erase a paper’s influence. Citations and summaries can persist for years unless databases, guidance documents, and online references are updated-making careful correction and transparent disclosure essential whenever a prominent review is withdrawn.
Where the documents were published
The retraction notice and the original paper appear in Regulatory Toxicology and Pharmacology. An analysis reviewing the retracted paper’s impact was published in Science.
Comments
No comments yet. Be the first to comment!
Leave a Comment