Why the improper retraction of a 25-year paper matters for science – and for all of us
The importance of sound science can’t be underestimated. Science delivers innovative solutions for all aspects of society – and companies have the obligation to set a high standard in both the scientific rigor and integrity of their scientific studies.
Transparency is another important part of the equation. Whether in industry, government, or academia, science should be a transparent, collaborative and principled discipline. The scientific community and public at large should demand nothing less.
That’s why I chose to write this article and share my perspective:
When I learned about the circumstances under which the Williams et al. review paper had been redacted late last year, my first reaction was to speak up because I felt it was another example of lack of fair play undermining science. By that, I mean the double-standard which seems to apply to studies and science sponsored by companies. We have seen a lot of this in recent years – unscientific-based attacks on research solely because of the research sponsor. After letting my initial feelings of frustration subside, I still remain troubled about how things unfolded and wanted to share my thoughts.
What happened regarding the retracted Williams et al. 2000 paper?
Last year, scientific integrity, collaboration and agricultural innovation itself were severely threatened. This occurred when Williams et al, 2000 – a 25-year-old review paper that contained no original data – was retracted from the Journal Regulatory Toxicology and Pharmacology. The paper provided an analysis of primarily industry studies required by regulators, enabling the broader scientific community access to such data, which was not previously available.
The retraction was published by a single editor, without the required transparent evidence of a proper investigation, at the request of two academics who provided no new data and no substantive argument that the underlying science behind the paper was flawed. The fact is, the underlying studies of this review paper have not been called into question. Instead, the only questions presented were those previously raised by interested parties pursuing or supporting Roundup litigation.
Scientific integrity and transparency – core principles essential to scientific research, publication, and decisions around retraction – appear to be lacking in the case. This should be setting off alarm bells for the entire scientific community. I’m going to share why.
Why have these allegations resurfaced now and is this a coordinated litigation attack on science?
First, the timing is suspicious. The unfounded allegations regarding the authorship and provenance of the paper that led to this retraction had already been raised during Roundup litigation in 2017 – and were investigated and dismissed by both the author’s employer, New York Medical College, and the European Food Safety Agency (EFSA). The allegations then resurfaced at the same time that the Solicitor General of the United States submitted a brief supporting Monsanto’s position on the topic of federal preemption within the Roundup litigation before the Supreme Court, now scheduled to be heard on April 27, despite the fact that the retraction occurred quietly months earlier. Coincidentally, plaintiffs’ lawyers included the retraction in their opposition to the Solicitor General’s brief as part of their failed argument against Supreme Court review.
Second, the process does not appear to adhere to retraction guidelines. Because a retraction is so significant in scientific literature, full transparency is paramount. The retraction appears to be at the direction of only one of the co-editors of the journal, not both; the investigation does not appear to have made a sincere attempt to contact the paper’s authors or their employers; nor was there public notification of the investigation itself, falling far short of the rigorous editorial and scientific standards that would be customary for such a retraction. Moreover, the letter to the editor that prompted the retraction has never been made publicly available.
Third and perhaps most troubling, the renewed attack on the Williams et al. paper is not an isolated event. The two academics who initiated that retraction have also publicly requested the retraction of at least one other industry-supported paper, Kier & Kirkland 2013. This further demonstrates the retraction was not based on a concern for the scientific findings presented in the paper, but rather, motivated by interests that sought to cast the industry-sponsored paper in a negative media spotlight.
The timing of these events raises the question of whether multiple attacks against industry-sponsored public literature are mere coincidence or were orchestrated in anticipation of other expected milestones including Supreme Court review of one of the central issues in the Roundup litigation. In this environment, it will not be surprising to see further improper and unfounded attacks on industry-sponsored, scientifically sound literature. This is cause for concern, as retraction of public scientific literature should be science-driven, not agenda-driven.
How do regulatory agencies view this paper or its retraction?
Because Williams was a review paper and contained no original data, it had no impact on the conclusions of regulatory agencies. Regulators conduct their own independent evaluation of original industry data and studies; they look beyond initiatives and agendas. In fact, in 2023, the European Union renewed its approval of glyphosate without relying on the Williams review paper at all, and EPA commented recently that the retraction did not change its findings with respect to glyphosate’s safety. In other words, with or without review papers like Williams, the scientific consensus continues to be that glyphosate can be used safely and is not carcinogenic.
Why was the conclusion of the non-regulatory agency International Agency for Research on Cancer (IARC) different?
When IARC, a subset of the WHO and not a regulatory agency, reviewed glyphosate, only publicly available information was allowed to be used. As per this policy, studies used for regulatory review were deliberately ignored, and those comprised the majority of safety studies on glyphosate.
Under these flawed rules, critical data sets provided as part of the regulatory process were omitted, leading to an IARC conclusion in opposition to all other leading global regulatory bodies. In addition, the way IARC assesses carcinogenicity differs from every leading global regulatory body as well. IARC takes a hazard based approach, meaning that they do not look at real-world scenarios, taking into account actual risk potential. Conversely, regulatory bodies take a risk-based approach.
Interestingly, since glyphosate was assessed by IARC, it now recognizes the value of regulatory studies and data through a changed Monograph preamble guiding its scope and procedures, allowing unpublished regulatory studies to be taken into consideration. This change made it increasingly difficult for IARC to leverage the narrower body of published data as justification for hazard assessments. The procedural change may also be the reason behind IARC’s unwillingness to review their classification of glyphosate despite blatant divergence from regulatory agencies’ assessments.
Why is transparent industry-supported research important?
There is an unexpected twist to this story. One of the academics who prompted the retraction of the Williams paper is Professor Naomi Oreskes of Harvard – someone who, just a few years ago, publicly acknowledged Bayer’s transparency efforts. She specifically lauded Bayer’s approach of sharing study data, stating: "Making the adverse as well as the positive data available is an absolutely essential step and I think it's hugely important that Bayer is doing that.” Indeed, Professor Oreskes suggested that Bayer’s leadership on transparency could set standards for the entire industry, noting that “If Bayer can do it, then obviously other corporations can do it too.”
On this, Professor Oreskes is absolutely correct. Bayer sets the industry standard for scientific transparency, with our actions including:
- Public disclosure of scientific collaborations
- Adherence to international Good Publication Practice (GPP)
- Full transparency in regulatory data, including adverse findings
And if Bayer is being held to this standard – as anybody should be – then isn’t it fair to expect the same level of transparency from the journal regarding its investigation, or the litigation industry and the studies they fund and promote to continue to target and attack companies and approved products?
Why does Bayer invest so much in scientific research?
Criticism of industry-sponsored literature by litigation adversaries is disingenuous – they complain industry has not done enough research on product safety, while also criticizing reviews undertaken to put previously non-public data into the public domain, even when industry support is properly disclosed. This criticism misses a key point: these studies are vital to innovation and the continued availability of crop protection tools. In science, more data is always beneficial as it provides regulatory bodies and the scientific community with a larger body of evidence on which they can determine whether products are effective and can be used safely.
Industry-sponsored research must therefore co-exist alongside public literature within a collaborative and rigorous scientific community. Leading regulators worldwide have robust and detailed requirements for studies that must be met in order to renew a product registration or bring new innovations to the market. This research must meet stringent international standards to demonstrate products will not harm human health or the environment.
Companies fund these necessary studies to meet stringent regulatory requirements for crop protection products – investing as much as $300 million and as many as a dozen years from product concept to regulatory approval – not a burden the government or tax payers should shoulder. Academic institutions alone cannot fund agricultural innovation; public and private sector resources are essential to support these necessary studies. Given the daunting challenges to meet the needs of an ever-growing and urbanized population in times of nature-loss and extreme weather events, research in agriculture is declining. Against this trend, despite the unfounded pressures of the litigation industry, Bayer has kept its commitment to support the food system through science-driven innovation.
What can we do to ensure transparent science wins out?
It is vital, of course, that companies set a high standard in both the scientific rigor and integrity of their studies. Here, I am proud that Bayer’s commitment to transparency in research extends far beyond basic requirements – and far beyond what has been acknowledged by our critics.
Whether in industry, government, or academia, science should be a transparent, collaborative and principled discipline. That should be true for both scientific publication and scientific retraction. All such decisions should be based on sound scientific principles – not litigation interests, public policy, or other initiatives. The scientific community and public at large should demand nothing less.
The retraction of the Williams paper failed this standard, proceeding without transparent investigation. When retractions are potentially weaponized for litigation strategy rather than scientific integrity, the entire scientific enterprise suffers.
As the scientific community moves forward, it must recognize that transparency works both ways. Companies like Bayer that commit to unprecedented levels of data disclosure – making both favorable and adverse findings publicly available – deserve the same transparent, rigorous processes when their research contributions are challenged. Only through such mutual commitment to transparency and scientific rigor can we maintain the integrity of the research that underpins innovation in agriculture and public health.
Matthias Berninger, Senior Vice President for Public Affairs, Science & Sustainability
