What the Monsanto Papers tell us about corporate science

The Monsanto Papers make for fascinating reading, all the more since Monsanto constantly uses and abuses the moral prestige of science in its propaganda. See above and below for an example of a recent PR campaign by the company.

That Monsanto performs research is clear. The corporation spends about 10 per cent of its turnover in research & development to keep developing new agricultural technologies, and “believe[s] innovation has the potential to bring humanity’s needs in balance with the resources of our planet”.

But the Monsanto Papers show the company’s real, and rather troubling, approach to science and evidence. Revelations include confirmation that the company hardly tested the real-world toxicity of its products, actively avoided pursuing studies which might show unwelcome results, and ghostwrote the studies of supposedly independent scientists. The documents also show Monsanto systematically attacked scientists whose research threatened their profits, as aptly summarised in a 2001 email by a Monsanto executive:

Many executives in the pesticides industry probably believe their industry behaves as responsibly as possible in our capitalist world. Nobody likes to think of himself as a poisoner.

However, it is also true that the pesticides industry uses the ‘science-based’ argument to both hide its politics and lobby politicians. It routinely invokes ‘sound science’ in highly politicised ways to lobby for its interests, and dismisses scientific evidence that goes against its interests as ‘junk science’. For example Graeme Taylor, a director of the pesticides industry’s EU lobby ECPA (European Crop Protection Association), published several emphatic articles during the recent big battle around the EU’s re-authorisation of glyphosate in which he argued that public decisions should be based on “facts, not fear”, and that politicians’ job was to “look to the science”... yet using his own daughter throughout an emotive speech pretending he and his industry would, in fact, be saving the world from hunger.

So, beyond the interesting arguments of their lobbyists, a better question is: what sort of science are pesticide companies doing? The Monsanto Papers give us a rare glimpse.

‘Regulatory Science’ vs. Science

If science in the usual sense is free inquiry checked by peer review and reproducibility of results (for which data transparency and rigorous procedures are needed), the science practiced by pesticides companies is mostly of a different nature: the ‘regulatory science’ they refer to when defending their products’ safety is largely performed to comply with public regulations and international regulatory toxicology standards.

Pesticides companies spend considerable time and money trying to influence these public regulations and standards. A global science lobby group like the International Life Sciences Institute (ILSI), for instance, which is funded by dozens of large agribusiness and food multinational companies, and whose purpose is to influence regulatory agencies and their methods to make sure they do not look too closely into the hazards of industry products, received more than 20 per cent of its funding from Monsanto in 2012.1

For pesticides, this has major implications. Currently the public requirements for serious testing only concern the so-called active substance, not the formulations used in the real world (these are only superficially tested). This means that, for glyphosate-based herbicides, only glyphosate is really tested, not formulations such as Monsanto’s Roundup even though there is strong evidence that they are much more toxic than glyphosate alone (glyphosate is never used alone as a herbicide because it is inefficient). Academics, on the other hand, tend to look at the effects of formulations because this is what people and the environment are exposed to in the real world. But their work is usually considered second-grade by public regulators.

This impacts the research performed by the companies too. Internal Monsanto emails for instance show that the company has never tested Roundup for cancer: “you cannot say that Roundup is not a carcinogen... we have not done the necessary testing on the formulation to make that statement”. Neither did it test its glyphosate formulations for medium and long-term toxicity, or for teratogenicity (ie potential to cause developmental anomalies in fetuses).

However, disclosed documents from 2002 show that Monsanto was well aware of independent studies showing that formulations caused harm: “Glyphosate is OK but the formulated product (and thus the surfactant) does the damage”.

When the European Food Safety Authority (EFSA) and the European Chemicals Agency (ECHA) evaluated the toxicity of glyphosate, they also did not test glyphosate formulations, as EU regulations do not require them to. Only EFSA recommended that EU member states, who superficially test pesticide formulations, also test their genotoxicity (ie potential to damage genetic information within cells, causing potentially carcinogenic mutations): “the toxicity of formulations and in particular their genotoxic potential should be further considered and addressed”.

This is absurd: for 18 months, following the evaluation of the International Agency for Research Against Cancer (IARC) which had taken both pure glyphosate and formulations into account, the huge public controversy around the re-licensing of glyphosate in the EU revolved around the question of whether it was causing cancer or not to workers using glyphosate-based herbicides. But EU regulators could not answer that question.

Asking the ‘right’ questions, publishing the ‘right’ papers

That scientific studies funded by industry tend to deliver results benefiting industry – the so-called funding bias – is well-known and caused by two main factors. First, researchers working for companies tend to design their studies in ways that will benefit their sponsors. Second, companies funding research usually refuse to publish the outcome if it does not suit them, a particular type of publication bias. In the long run this can twist the available scientific literature on a given issue, which has big political consequences because literature reviews play a more important role than individual studies in informing public decisions.

But the Monsanto papers do not only provide evidence that the company did not test the real world toxicity of its products: they also show that the company actively avoided pursuing studies which risked producing results it did not want to see. For instance, internal documents show that the Dutch consultancy TNO was stopped by Monsanto from repeating a study. This study indicated that a glyphosate formulation was absorbed through the skin at a much higher rate than previously thought, which, in the words of a Monsanto executive, had the potential to “blow Roundup risk evaluations”.

Another example is the case of the work of another consultant, British toxicologist James Parry, who concluded in a 2001 study performed for Monsanto that, according to recently published studies at the time, “glyphosate is capable of producing genotoxicity” and suggested performing additional studies to double-check these findings. The response of Monsanto’s William Heydens at the time is worth quoting at length:

“We want to find/develop someone who is comfortable with the genetox profile of glyphosate/Roundup and who can be influential with regulators and Scientific Outreach operations when genetoxicity issues arise. My read is that Parry is not currently such a person, and it would take quite some time and $$$/studies to get him there. We simply aren’t going to do the studies Parry suggests.”

Other examples are Monsanto scientists acknowledging internally they have a problem with their product. For example in 2014 emails Monsanto scientists discuss the then-upcoming IARC review, saying:

“while we have vulnerability in the area of epidemiology, we also have potential vulnerabilities in the other areas that IARC will consider, namely, exposure, genotoxicity, and mode of action.”

Monsanto is a commercial company: the money it spends on research on a product it is already selling is used in a perspective of protecting profits against independent science, not improving knowledge.

When cutting costs leads to deception

This profit motive recurs throughout the Monsanto Papers, hardly surprising for the internal correspondence of a private company. But it leads to practices that are simply fraudulent in science, such as ghostwriting research studies.

This involves having a study written by company employees but authored by supposedly independent experts. Evidence of such practice has been disclosed for several important studies on glyphosate published in the scientific literature, such as the literature reviews Williams Kroes & Munro (2000) and Kier & Kirkland (2013), or a 2016 critique of the IARC evaluation. Even press articles such as op-eds published in Forbes by Henry Miller, a scientist, were ghostwritten by Monsanto employees. As Heydens explained in 2015 in an internal Monsanto email (see p. 203),

“A less expensive/more palatable approach might be to involve experts only for the areas of contention... and we ghost-write the Exposure Tox & Genetox sections.... [W]e would be keeping the cost down by us doing the writing and they would just edit & sign their names so to speak.”

Monsanto has of course disputed the ghostwriting accusations, but so have some of the scientists involved, such as David Kirkland, who argued both in press statements and during his answers to MEPs at a European Parliament hearing that he had no interest in accepting such a deal because it would ruin his reputation.

There is, however, evidence (see pt. 6) in the documents that David Saltmiras, a Monsanto scientist, did write drafts of the study he co-authored. A disclosed discussion between Kirkland, Kier and Saltmiras goes further: not only does it show that David Kirkland knew Saltmiras had co-written the study, but it even shows Kirkland refusing Saltmiras as a co-author as this would mean that “the authors would no longer be ‘independent’.”

In response, Saltmiras complains his constant ghostwriting deprives him of due credit as a scientist:

“I think my contributions on this now 2-year project may be considered deserving of recognition in co-authorship. I was prepared to let this slide, but in another candid discussion with Senior Monsanto management (several tiers higher than me), I was “strongly encouraged” to author some peer reviewed publications, since this is the fifth such Glyphosate related manuscript I have been involved with over the past few years without co-authorship.”

Kirkland replied saying the editor of Critical Reviews of Toxicology, the journal where the paper was published, would probably refuse:

“Given what he said about dissociating Larry and myself from industry during the responses to the referees comments, I would be surprised if he agreed to add an industry name to the authors.”

Saltmiras gave up:

“Then no need to bother the editor. The paper is Kier and Kirkland. Thanks for your patience.”

Attacking independent scientists

Monsanto does not hesitate to play dirty. Documents show that it went as far as it could to get the 2012 Séralini study2 retracted, orchestrated a campaign to destroy the International Agency for Research on Cancer after its critical evaluation of glyphosate as a probable human carcinogen, and that, more generally, the company maintains a permanent watch over the scientific literature to make sure that studies whose findings may threaten its business are undermined. This is made abundantly clear in the words of the previously quoted Monsanto executive who said: “Data generated by academics has always been a major concern for us in the defense of our products.” Intelligence gathering on NGO activities is also mentioned (see p. 302) in the documents.

What can we conclude from this brief tour through Monsanto’s internal discussions on the science around glyphosate?

First, that working for Monsanto seems to put its scientists in awkward positions. They have access to a lot of resources but need to keep cutting costs, at the risk of committing scientific fraud. They know their product in-depth but cannot publish what they know, and must actively fight others trying to find it. As scientists, it is difficult for them to get credit as co-authors because other scientists do not want to be associated with them. From the moment a product starts to be commercialised, their scientific competence goes into fighting the relevant science rather than advancing it.

Second, that these exchanges are yet another indication that one cannot be judge and jury, and that corporations have too much of a vested interest in the research they perform or sponsor on their products’ safety to be trusted as a reliable source for public decision-making. However, industry’s research remains the basis of public regulators’ evaluation today, in the EU and elsewhere. Nevertheless it would not be very difficult to charge industry for the cost of public evaluations and have the research for pre-market authorisations performed by independent researchers, following experimental protocols that take into account what we as citizens, and our environment, are actually exposed to, rather than the convenient fiction scenarios concocted by industry’s regulatory affairs departments and lobbyists. But this depends on political will, not science.

• 1. Interestingly, the food multinational Mars Inc. just decided to leave ILSI on the basis that it no longer wanted to “be involved in advocacy-led studies that so often, and mostly for the right reasons, have been criticised” – in this case referring to ILSI’s criticism of public guidelines recommending a limitation of added sugars in food products. See Breaking away from bad science? Mars to leave ILSI in transparency bid, Food Navigator, 8 February 2018
• 2. The publication, retraction and republication of a 2012 scientific paper by French researcher Gilles-Eric Séralini was hotly contested, with serious criticism pointing at both the study’s inconclusiveness and the fact that this wasn’t enough a reason to retract a paper.