Talk:Monsanto

Williams 2000 retracted

@KoA, Tryptofish, Dialectric, and David Tornheim: Williams 2000 has finally been retracted. Nosferattus (talk) 15:18, 29 November 2025 (UTC)

Kudos to you, Nosferattus, for finding this information. I fully agree with changing the page to reflect this new information, and I actually made an edit after yours, in which I made the language we use even stronger. This is, in my opinion, a definitive outcome, in terms of scientific integrity (or the lack thereof).

One thing about the edit I made, that I'd like you or other editors to check me on. I added 2025 as the date of the retraction, based on my understanding that the retraction has just come out recently. But I could not find a date on the retraction itself, so I just want to make sure that this really happened in 2025. Thanks again. --Tryptofish (talk) 01:36, 30 November 2025 (UTC)

Yes, it definitely happened this year, as there was no retraction notice there when we were debating this earlier. Also the Kaurov & Oreskes paper from September says that the paper hasn't been retracted. Nosferattus (talk) 02:02, 30 November 2025 (UTC)

The Kaurov & Oreskes paper has been online since July. There is normally another DOI published for a retraction e.g. doi:10.1016/j.yrtph.2025.105981 in the January 2026 edition of the same journal. I had a look through all the recent issues but couldn't see anything, so presumably is in the works. It hasn't been picked up by retraction watch yet either, so I presume it has only just been retracted. SmartSE (talk) 12:07, 30 November 2025 (UTC)

Le Monde reports and states it was retracted 28 November @Nosferattus Katzrockso (talk) 01:54, 4 December 2025 (UTC)

Retraction Watch published a story about it today . The retraction was specifically precipitated by Kaurov and Oreskes's paper and their subsequent letter to the editor; They also wrote to the editors of Regulatory Toxicology and Pharmacology to formally request the paper’s retraction. The editor's comments are very interesting, stating that the retraction "could have been done as early as 2017". Also contains a statement from Bayer on the matter. Katzrockso (talk) 01:58, 5 December 2025 (UTC)

@Smartse: I removed the Williams 2000 ghostwriting responses from Monsanto and NYMC since they are moot at this point and it seemed like excessive detail now that the case is finally settled. Do you think they are still important enough to include? These ghostwriting paragraphs are getting a bit lengthy. Nosferattus (talk) 16:34, 30 November 2025 (UTC)

When updating, I prefer to keep the existing content where possible, as it is still reliably sourced content. This was/is Monsanto's position for many years and while the paper being retracted is a clear line, it is still relevant to include their POV. SmartSE (talk) 20:45, 30 November 2025 (UTC)

I'm OK with restoring the response from Monsanto. However, I think that perhaps we should remove the sentence about the investigation by the university of the lead author. The sourcing for that describes the investigation as a "quick" one, and we now have strong sourcing to indicate that the university got it wrong. In my opinion, this ends up just being embarrassing for the university, and it only confuses the text for our readers. --Tryptofish (talk) 01:28, 1 December 2025 (UTC)

That seems like a decent compromise to me. Nosferattus (talk) 02:48, 1 December 2025 (UTC)

Yes fair point. I've removed it. SmartSE (talk) 21:31, 1 December 2025 (UTC)

That's what we really needed all along for sourcing rather than the months of discussion on this with lower quality sourcing. This would have been a good time to start content discussion and probably have something drafted up in a matter of days, but that's mostly moot now. Hopefully that acts as a good example of why we normally wait for sourcing like this, especially when it comes to WP:CRYSTALBALL.

The one caution I have in recent small change from the WP:BLP perspective is this recent change of the header from Allegations of ghostwriting to just saying Ghostwriting. We didn't actually gain any new knowledge on what transpired in 2000 unfortunately in that regard (aside from the paper itself being retracted), so we're not really in a position to move away from alleged at this point. There's a line between the issues with the study and us saying ghostwriting without the qualifier. The old section header threads the needle pretty simply on policy issues we have to navigate there. News sources being discussed here also are careful to use similar language on alleged, etc., so I'm not so sure the change can have consensus here. Other than that, the original language works. KoA (talk) 15:34, 5 December 2025 (UTC)

@KoA I read the previous discussion and saw many editors such as @David Tornheim arguing for the inclusion of the article. There has also been a new development, in that the Kaurov & Oreskes paper is mentioned in every article about the retraction. That the paper and its the authors were the direct antecedent, in addition to its high quality, is a strong reason to include the article.

Can you elaborate on how BLP applies here? Katzrockso (talk) 16:00, 5 December 2025 (UTC)

To be honest, there's been months-long discussions over that Kaurov paper in multiple talk pages, and consensus was that it was not needed. Yes, some editors wanted to include it and pushed for it, and others did not see the need for it. I'm not pinging any one of those per WP:CANVASS. The point is though that people are exhausted over that topic, and we already reached a resolution that did not include it. Resurrecting that only a month after that discussion concluded really isn't going to shed any additional light. The the retraction itself does not mention the paper at all, and the retraction really is the focus here. News sources like Retraction Watch mention other supplemental details, but not all those are going to be included.

As for BLP, that's just a matter of being careful on our end as we don't assert wrongdoing in Wiki-voice unless it meets a high threshold (see WP:BLPPUBLIC for examples). Here we can't outright say there was ghostwriting directly when it comes to Williams, but there were definitely unaddressed concerns by the journal that reached the level of retraction (and I want to be clear I'm not understating those concerns). The retraction didn't confirm there was ghostwriting though in part because of a lack of response by Williams, and that's a nuance we have to navigate. If the journal was able to confirm it, then it's easier for us to refer to the Williams paper as just ghostwriting. The seriousness of the retraction and us calling it ghostwriting are slightly different topics. This is for an entire section and not just the Williams paper though, so that language also let's us generalize things to include both cases of confirmed ghostwriting and instances where we have to stick to alleged qualifiers a bit tighter.

Just to be clear, this is all to say that the current version should have us at a rough baseline of what should be included for now, including the recent retraction. I got here a little late this week, but it looks like other editors had mostly settled on other relevant details, so hopefully we're at a point the topic can rest a bit. If someone really wants to add more details, I'd give it a WP:RECENTISM check in a couple months before adding (in addition to giving this talk page a break). Especially if new major details do come out later, that can be dealt with then, but right now it looks like we should have stable content finally. KoA (talk) 16:51, 5 December 2025 (UTC)

Several thoughts. I'm ambivalent about the section header. On the one hand, I'm receptive to being cautious about making accusations of ghostwriting in Wikipedia's voice. But on the other hand, when we are quoting the retraction statement verbatim as saying "lack of authorial independence" and "misrepresentation of contributions", I think we are out of the realm of this being merely an accusation.

About the Kaurov paper, please count me as one of the "exhausted" editors, and I don't think we need to include it. If the retraction statement from the journal had cited Kaurov explicitly, then I might want to include that as being the reasoning. I see that the Retraction Watch source says that the journal editor acted after receiving a letter from Kaurov and Oreskes, and that's not nothing, but it seems to me that "the story" here is the retraction itself, not Kaurov and Oreskes, who could be argued as simply being the latest in a long line of critics, rather than as people who fundamentally changed our understanding of what happened.

Now my biggest concern at the moment is the change of language, from what had been The paper, in Regulatory Toxicology and Pharmacology, was retracted on November 28, 2025 by the editor-in-chief, citing..., to The paper was retracted on November 28, 2025 following a lack of response by Gary M. Williams, the last surviving author of the paper, to concerns of the editor-in-chief of Regulatory Toxicology and Pharmacology. The unaddressed concerns cited.... What that accomplishes is adding the information of (1) Williams' name, (2) the fact that he is now the only surviving author, and (3) that he did not provide a response. That's more information, but I think it makes things worse, from a WP:BLP perspective. As I see it, this is the real BLP issue, not the section header. Not responding can be due to any number of things, and it's the absence of an action, rather than an action. (In context, it sounds like Williams may now be elderly.) And yet the wording makes it sound like this is an important part of what happened. I think in this case, less is more, and we should go back to the language we had before. --Tryptofish (talk) 19:38, 5 December 2025 (UTC)

So I think I'm on a fairly parallel track as you in your second paragraph at least in how I started looking at this. The key thing here is grounding ourselves on what the retraction itself says. The quotes you mention about authorial independence, etc. was a good approach for the text. The only piece I was seeing that seemed missing was how the retraction came about: 1. Journal raises serious concerns to the last remaining author, 2. No response from Williams, 3. Journal retracts the article because those concerns were not addressed. The question was how to put that into text hopefully with just a minor addition so the topic could rest.

Had Williams of responded, two main things could have happened. Williams could have provided clarification that showed ghostwriting, etc. did not occur, or the journal could have confirmed those issues did occur. Neither of those happened though. Instead because of the lack of response (and one is expected respond when there are serious accusations like this), the journal basically said that serious concerns were raised, and because they were not addressed, they had to retract the paper. The journal is very careful in most of their language to highlight both the lack of response and uncertainty on the individual concerns, so removing language to make it seem like this isn't an important detail. We don't need to get into why there was no response for examples like you highlighted as the journal just plainly stated that was an overarching important detail in the retraction. It unfortunately becomes another threading the needle situation (I wish it weren't) rather than a straightforward one with regards to academic publishing, but that's why I'm being super careful about BLP here. Had the journal been able to confirm some of those issues, we'd have an easier time here as editors in terms of moving beyond saying it's an accusation.

So for the content you mention in 1-3, the only thing I really started off with was that there was not a response from the author(s) and why the journal said that was important. That's the only key detail I was mostly focused on needing. KoA (talk) 20:24, 6 December 2025 (UTC)

Well, a lot of what we are now discussing here is about "how the retraction came about". One way of looking at that is that it was because Williams did not reply. Another is that it was because Kaurov and Oreskes sent the editor-in-chief a letter. And at this point, I think you can see where I'm about to go with this. As I see it, the real reason it came about was because of that letter, and there was going to be some sort of action taken by the editor-in-chief no matter what Williams did or did not do. Another editor has already reverted you, removing the stuff about Williams, and I agree with that edit. We can only speculate about why Williams did not reply and about how things might or might not have been different depending on whether he did. We don't have reliable sourcing about that, and we probably never will. The fact that the retraction statement is careful to specify that no response was received is because the statement is being appropriately careful and precise; it implies that there was, of course, a possibility for a response to have rebutted the criticism, but we have no reliable sourcing to indicate what would have happened in that alternative universe. What we have, definitively, is the retraction statement saying what it does, and it's straightforward about authorial non-independence and misrepresentation. They didn't retract it because there was a possibility of ghostwriting. It was because it was likely enough to take the extraordinary step of retraction. That's reliably sourced, and I don't accept that we should borderline-violate BLP in order to complicate that fact. --Tryptofish (talk) 23:00, 6 December 2025 (UTC)

I see where you're going with the line of thought you started with, but that shouldn't be relevant here. The retraction makes no mention of Kaurov and Orsekes, so when we're trying to focus on what the actual retraction says, we only need to focus on what the source says. We don't need to speculate about why Williams didn't reply either, only that it was a major contributor to why the paper was retracted. That's not my opinion, that's per the journal editor. I'm trying to be really careful we don't misrepresent the journal editor by glossing over that aspect of the issue our run ourselves into trouble with BLP on Williams. That's why I've been sticking with this issue a bit in the last two days because I don't want to see the existing new content scuttled because of this issue in the use of the source. I was hoping a high quality source like this would have made discussion easier on nailing down the key details.

I mentioned the scenarios above not to speculate, but highlight the three main bins retraction assessments fall under: 1. Major issues are confirmed or likely = retract. 2. Major concerns are not adequately addressed (or at all) = retract. 3. Concerns found not to be valid = no action. Any retraction is serious, but 2 happens because of the uncertainty it causes rather than confirmation of issues, which is worse than 1 in some ways. The journal is pretty clear this is a case of 2 in their statements, while I'm concerned that people inadvertently jumping on it being 1 can misrepresent what happened. It's not a matter of how likely misconduct was at that point in the process, but what degree of uncertainty is generated. That also means I'm coming at this from the perspective that the seriousness of this portion inadvertently ending up downplayed right now. I just want to make sure we get this right so issues don't propagate down the line. I guess to bring things to an end point, I'm just not seeing how reflecting the statements by the journal itself that Williams did not respond and inadequately/unaddressed issues led to the retraction are an issue, while I'm running into policy and guideline issues trying to navigate leaving it out. KoA (talk) 01:13, 7 December 2025 (UTC)

I wrote two separate replies to this, only for both to be deleted before I could send them earlier, but alas I will try to rewrite it again:

I disagree that BLP prevents us from writing that the section header "Ghostwritten" for several reasons:

1) The older version of the text, before the individual name of Williams was added (as noted by Tryptofish), did not accuse any living person of engaging in wrongdoing. Indeed, it only names the paper as ghostwritten and does not attribute this misconduct to any particular person, so it is very unclear how BLP is supposed to apply here at all.

2) There is no "crime" here that occurred ('researcher misconduct' at best might amount to something that is a crime, but there is no accusation that the behavior here rose to that level nor any logical connection that it does), so BLPCRIME cannot apply. You appeal to WP:PUBLICFIGURE, which once again speaks about BLP, and there is no BLP content here. In addition, I disagree that PUBLICFIGURE would even imply

3) There is even the bigger issue that "allegations" might be non-neutral (MOS:ACCUSE) and non-concordant with other discussions of similar academic misconduct on Wikipedia, which usually uses the term "controversy" (see Doris Kearns Goodwin#Plagiarism controversies, Stephen E. Ambrose#Plagiarism, where it becomes unclear if we can take on assumption that this class of header titles violate any BLP policy).

The former is my interpretation of BLP, but I am completely open to being wrong as this is a policy I am admittedly less familiar with than others (largely for interest, as I am rarely interested in writing about living people in most contexts), but perfectly willing to ask at e.g. WP:BLPN to clarify.

I will have to wait to reply to the rest until later Katzrockso (talk) 23:29, 5 December 2025 (UTC)

I agree. 1- the paper is ghostwritten, and doesn't have to name names. 2- Ghostwriting is not a crime. 3- The paper was retracted, that to me is case closed. It's now unreliable. Andre🚐 02:52, 6 December 2025 (UTC)

I agree with that, and I think that it's clear that we have proper sourcing to remove "allegations". I have made that edit. --Tryptofish (talk) 22:43, 6 December 2025 (UTC)

As for the Kaurov & Oreskes article, I strongly think that some mention of how exactly people came to the determination that the paper was ghostwritten and how it was retracted is important. As noted in the Retraction Watch and Le Monde articles: Their request “was actually the first time a complaint came to my desk directly,” and Why, given that these documents were made public by the press in 2017, did it take eight years for the article in question to be retracted?. I think the reader might question why there was this 8 year gap between the release of the emails in 2017 and the retraction in 2025. I certainly would, reading the section as is:

Monsanto internal emails made public in August 2017 during the Johnson v. Monsanto Co. litigation revealed Monsanto personnel referring to undisclosed ghostwriting of that and subsequent papers. A Monsanto spokesperson responded to the email release that Monsanto had provided only non-substantive cosmetic copyediting that did not alter the studies' conclusions. The paper was retracted on November 28, 2025. where the editor-in-chief of Regulatory Toxicology and Pharmacology cited "lack of authorial independence", "misrepresentation of contributions", "questions of financial compensation", and a variety of concerns over whether the conclusions presented were adequately supported by the scientific data.

The Chemical and Engineering News article also makes this clear But RTP acted only after Alexander Kaurov, an astrophysicist at the Victoria University of Wellington, and Naomi Oreskes, a historian of science at Harvard University, reported the issues to the journal.

Similar in Washington Post; Alexander Kaurov and Naomi Oreskes, who requested the retraction, published research in September that found the paper is in the top 0.1 percent of cited academic literature on glyphosate.

At a minimum I'd at least prefer we mention that the retraction came at the behest of Kaurov and Oreskes's intentional request for retraction, so it isn't inexplicable why there is such a gap between the emails and the retraction. As for the "months-long discussions over that Kaurov paper in multiple talk pages", I am well-aware of this, having read the discussions several times. I concur with David's comment that Despite walls of text, when it comes down to it, I still have not heard any valid reason to omit this key source prominently mentioned in The Signpost. I'm sure you are aware that WP:CCC. Katzrockso (talk) 09:20, 6 December 2025 (UTC)

I tend to agree with Katzrockso. It doesn't matter whether we think Oreskes and Kaurov are reliable for facts (but I think they are). The fact that their work led to the retraction is notable, encyclopedic, and DUE as a proportion of relevant sourcing relative to the topic. We can attribute everything to them since it is undoubtedly controversial and disputed, but that doesn't mean exclusion, it means attribution. There is no policy-abiding argument to entirely exclude them and none has been articulated in light of the changing circumstances (the retraction) which means that the prior consensus has reason to be revisited, and it seems it HAS changed along with the circumstances. If we must formalize that consensus we may, but I submit that KoA is simply out of step with the present consensus and should take a step back. Andre🚐 22:54, 6 December 2025 (UTC)

(edit conflict) I'm increasingly willing to entertain this, but I want to get it right. First, KoA and I have been working in this contentious topic area since the Jurassic period, and I want all editors to cut one another some slack. This stuff isn't as easy as it looks. As I see it, we have to parse out what is WP:DUE and what isn't. As I just said to KoA above, a lot of what we are now discussing is about how the retraction came about. Maybe that includes the lack of response from Williams, but I've said above that I think that would be too much to include. I'm ambivalent about the role of the letter, but I'm OK with including something, if we craft it thoughtfully.

Kaurov is an astrophysicist, and Oreskes is an historian of science. Neither is an authority on herbicides or on human toxicology. They criticized Wikipedia in their article, but that should not influence our decision here. We have very reliable sourcing that their letter was the primary reason the editor-in-chief decided to look into this. We also have very reliable sourcing that this was a new editor-in-chief, not the chief editor who had been there for about a quarter century since the paper was first published. (Did the previous editor get letters? No way for us to know.) How should we balance the importance of the letter against the importance of the change in editorship?

We can cite the significance of the letter to the sources we currently have. What exactly would we cite to Kaurov and Oreskes' paper? We don't need more sources for the allegations of ghostwriting that came in the wake of the email release. --Tryptofish (talk) 23:25, 6 December 2025 (UTC)

As I noted below, we should prefer academic sources rather than newspaper reports from journalists. A historian of science like Oreskes is 99 times out of 100 a better source for information about scientific publishing than a news journalist.

How should we balance the importance of the letter against the importance of the change in editorship? By following what the reliable secondary sources say about the causes of the retraction, not conducting original research into why the article was retracted (i.e. attributing it to the change in editor-in-chief or lack of response from Williams). Your original research may very well be right, but it does not belong in Wikipedia's text. Katzrockso (talk) 00:27, 7 December 2025 (UTC)

I don't understand why we are exclusively relying on newspaper reports about the ghostwriting when there is high-quality academic sources on the subject. Wikipedia prioritizes "academic and peer-reviewed publications" over newspapers and no compelling objections to the Kaurov and Oreskes source have been presented. Indeed, I would expect the exact best source would be published academic work from historians of science.

Some justifications to exclude said source in the past included It [the Williams et al. paper] is respectably published, not withdrawn or even subject to an expression of concern. and I'm against citing Oreskes as it is a piece of primary research and has had zero citations so far, making it UNDUE to include at the moment, both of which are no longer meaningful accurate. I think in light of the recent retraction, there is substantial reason to reconsider whatever previous consensus there may have been on the question of including this source (as you have noted). Katzrockso (talk) 23:17, 6 December 2025 (UTC)

Most of what you mention is already addressed in my earlier reply above to Tryptofish. The reality is that the caution about BLP and accusations of wrongdoing do not go away by just removing Williams name. Otherwise it becomes a sort of double-speak situation saying the authors aren't accused of anything because we didn't name them outright (even though we do either way). You also mention MOS:ACCUSE, but it outright states, alleged and accused are appropriate when wrongdoing is asserted but undetermined. Either way, the body of the text navigates/bypasses the accusation topic just fine, and it's only the section header's recent change that ran into it. I don't really see any strong reason to spend more time on the header subject either since it's not central to summarizing the retraction itself, and the section header is not defined on just this Williams paper discussion like I mentioned earlier. We're kind of getting into the weeds spending time on the header.

What you mention later on though are news sources, not the WP:MEDRS quality source (the retraction) being focused on here. As mentioned above, the retraction is highlighting the important details for us already. I'll echo Tryptofish's comment above on the Kaurov paper. It's just not mentioned in the retraction as being important here. The litigation however is mentioned in the retraction, and the content already links up to that. Consensus can change, but in WP:DEADHORSE situations, one really can't expect that horse to realistically move, especially on already taxed volunteer time. In those cases it's better to move on to what can be worked on.

That's why the focus here has just been on what the new retraction source itself adds rather than circling back to topics that already reached an end. The only thing currently missing from the summary based on the retraction's source is what you recently removed on Williams lack of response being a key aspect of the retraction. Other than that, we pretty much have the bases covered on what the retraction itself highlights as WP:DUE. My main concern here is by omitting the lack of response, we're putting our fingers on the scale as editors and disagreeing with the journal itself on a key intertwined reason as to why the paper was retracted. I don't know your background on academic publishing, but unaddressed concerns (besides substantiated ones) are why papers get retracted. I don't think we can get around not including the mention of no response from the authors, so how would you write it? KoA (talk) 23:17, 6 December 2025 (UTC)

It's simply not reasonable to characterize this as a DEADHORSE since in fact the circumstances have markedly and significantly shifted. That means revisiting. Andre🚐 23:38, 6 December 2025 (UTC)

Moving on from the header issue for now, since it is relatively minor in the grander scheme of things and it would be more productive to focus on more significant (textually) changes for now. We can revisit that at a later date.

WP:MEDRS are not required for any of the following:

• making statements about the nature of scientific publishing
• the existence of a publication from Kaurov and Oreskes
• the fact that they wrote a letter expressing concerns about the independence of the paper based on their research into the paper and its influence on scientific discourse
• the fact that the retraction occurred as a result of their letter

Indeed, we currently include zero MEDRS in this entire section, because we are not making claims about "biomedical information", but about the process of scientific publishing.

As for your argument about the lack of response from Williams, we currently have no reliable sources saying that the reason that the reason the paper was retracted because Williams did not respond to the concerns expressed by the editor. We have reliable sources stating that the proximate cause was the letter to the editor written by Kaurov and Oreskes and more specifically the ultimate cause was the fact that the paper was ghostwritten (has been retracted eight years after documents released in a court case revealed employees of Monsanto, the company that developed the herbicide, wrote the article but were not named as coauthors. from RW, Alexander Kaurov and Naomi Oreskes, who requested the retraction, published research in September that found the paper is in the top 0.1 percent of cited academic literature on glyphosate from WaPo, The disavowal comes 25 years after publication and eight years after thousands of internal Monsanto documents were made public during US court proceedings (the "Monsanto Papers"), revealing that the actual authors of the article were not the listed scientists – Gary M. Williams (New York Medical College), Robert Kroes (Ritox, Utrecht University, Netherlands), and Ian C. Munro (Intertek Cantox, Canada) – but rather Monsanto employees and When asked, van den Berg explained that he was not aware of the situation until the publication, in September, of an article by science researchers Alexander Kaurov (Victoria University of Wellington, New Zealand) and Naomi Oreskes (Harvard University) in the journal from Le Monde).

I'm sure the lack of response from Williams made it a much easier decision for the editor-in-chief, but to suggest that is the reason the paper was retracted is original research and should not be stated in the article in that manner. I suggest that the lack of response from Williams only introduces BLP material into the article when we are trying to avoid this. If we were to write a separate article (as Nosferattus suggested) on 'the Monsanto papers', it would be a warranted fact as we might have a much more detailed timeline. But given that it offers no explanatory role in any of the newspaper articles other than another piece of information, there's no good reason to include it in our short exposition here. Katzrockso (talk) 00:11, 7 December 2025 (UTC)

Thanks Katzrockso, that's helping highlight a few key misunderstandings. The big thing I'm seeing here is that you're making claims that contradict the journal itself, which violates WP:OR. To then claim that I'm violating OR for reflecting what the journal actually did say also runs into issues with that policy. That's not to point fingers, but to highlight I originally mentioned OR earlier because I was trying to get us to drill down into what the source alone actually did say to avoid inadvertently misrepresenting it. This kind of confusion is why I was trying to get that focus. The journal said what it said, and I made a recent comment above on the seriousness of this kind of retraction. I'm trying to make sure it isn't inadvertently downplayed because of misunderstanding.

Related to that, none of what you're quoting is from the MEDRS retraction article. As a side note, generally, we don't use news sources to comment on the quality of journal/MEDRS sources, but instead rely on other academic sources to do that. Usually if a MEDRS source doesn't mention something that lower-quality sources do, it usually isn't considered WP:DUE. That's why the retraction source made it a lot easier for us to discuss this topic since prior, we didn't have great MEDRS sourcing referring to the Williams paper issues. If you do quote from the actual journal, you're going to find items like This decision has been made after careful consideration of the COPE guidelines and thorough investigation into the circumstances surrounding the authorship and content of this article and in light of no response having been provided to address the findings. Claiming there are no reliable sources when that language is present throughout the summary-level content of the key source isn't helping with content assessment.

So when you look at the original edit and the source, can you point to what exactly was misleading and how you would have worded it differently based on the retraction source? That's not to get into other sources, but pin down what the issue is with respect to the main source. Otherwise, I pretty much paraphrased right from the journal, Concerns were raised regarding the authorship of this paper, validity of the research findings in the context of misrepresentation of the contributions by the authors and the study sponsor and potential conflicts of interest of the authors. I, the handling (co)Editor-in-Chief of Regulatory Toxicology and Pharmacology, reached out to the sole surviving author Gary M. Williams and sought explanation for the various concerns which have been listed in detail below. We did not receive any response from Prof. Williams. Hence, this article is formally retracted from the journal. . . KoA (talk) 02:28, 7 December 2025 (UTC)

OK, I'm curious as to where I made claims that "violate OR" (which in and of itself is not a problem: editors are allowed to conduct original research on talk pages so long as it does not violate other guidelines, original research is not permitted in articlespace) or "contradict the journal itself". Can you clarify?

As a side note, generally, we don't use news sources to comment on the quality of journal/MEDRS sources, but instead rely on other academic sources to do that Taking aside the misunderstandings of MEDRS professed earlier in the paragraph, is this not the best reason to include the Kaurov and Oreskes paper instead of the news article we are currently using? [a sidenote: I think you are misunderstanding the purpose of me providing that collection of quotes, which is in part my fault as I did not organize them in a particularly coherent manner, but we can get into that if necessary]

I don't understand what you are asking me to do, I agree that it is an accurate paraphrase of the content in the journal - and the current version of the article reflects that edit, bar removing the information that multiple editors agree has an issue with BLP. Something be true and verifiable is not a guarantee for inclusion (WP:VNOT) and indeed BLP here weighs against including the name of Williams in the article, as noted by Tryptofish already. My edit was minimal and simply removed the material that falls afoul to BLP. Katzrockso (talk) 04:19, 7 December 2025 (UTC)

I feel like this discussion is having difficulty getting to consensus, and I'm going to offer my advice on how to move ahead. It seems to me that the question we are dealing with is how to present what led to the retraction. What we really need is a specific suggestion about what, if anything, we should add to the page. I suggest that editors focus on making a specific proposal, formatted as I did in #Proposed revision, above. Please don't propose anything that has already been on the page and reverted. Focus instead on finding something that will get consensus. --Tryptofish (talk) 20:10, 7 December 2025 (UTC)

The aforementioned reasons ("this was already discussed before") to prevent the inclusion of particular material/sourcing should not be considered a robust basis for completely foreclosing a particular change. I hesitate to be accusatory, but I fear that Please don't propose anything that has already been on the page and reverted smells a bit much like WP:STONEWALLING. If I were objecting to a change on baseless grounds (which isn't exactly what's happening here, to be clear), I would hope that my reasoning-free opposition was not the grounds for rejecting a beneficial change to the encyclopedia. The only question that we have not formed a consensus on is the inclusion the Kaurov and Oreskes source, from what I can tell. Katzrockso (talk) 06:33, 9 December 2025 (UTC)

I was unclear; what I meant was not to propose what had just been reverted in the past few days. No need to accuse me of stonewalling. My main point is that editors should make a proposal of specific text (not just let's include or not include something). --Tryptofish (talk) 00:05, 10 December 2025 (UTC)

I don't see any issue with the proposed text here ;

Following a 2025 paper from researchers Alexander Kaurov and Naomi Oreskes that analyzed the influence of the article on scientific discourse, the Regulatory Toxicology and Pharmacology paper was retracted on November 28, 2025. The editor-in-chief cited "lack of authorial independence", "misrepresentation of contributions", "questions of financial compensation", and a variety of concerns over whether the conclusions presented were adequately supported by the scientific data.

Other than possible slight copyedits for clarity. That something has been reverted is not a irrebuttable reason to avoid implementing that text. Katzrockso (talk) 01:51, 10 December 2025 (UTC)

That makes it sound like the editor-in-chief read their paper and then decided to do the retraction, which clearly did not happen. He was reacting to the letter they sent him, not the paper that they published. --Tryptofish (talk) 00:43, 11 December 2025 (UTC)

Following researchers Alexander Kaurov and Naomi Oreskes' research that analyzed the influence of the article on scientific discourse, they sent a letter to the editor-in-chief of Regulatory Toxicology and Pharmacology outlining their concerns with the paper. The paper was later retracted on November 28, 2025, citing "lack of authorial independence", "misrepresentation of contributions", "questions of financial compensation", and a variety of concerns over whether the conclusions presented were adequately supported by the scientific data.

highlights the connection between the paper and their letter; After publishing the paper in September 2025, Kaurov and Oreskes decided the next logical step was to submit a retraction request for the Williams et al paper to RTP, assuming retraction requests had been issued in the past. They hadn’t. Katzrockso (talk) 01:24, 11 December 2025 (UTC)

What you just quoted there explains why Kaurov and Oreskes sent the letter, but it doesn't directly relate their paper to the decision to retract. The decision to retract was based on their letter, not on their paper. Please re-read what I already said here: , especially the last sentence. I'm friendly to indicating that the letter was the impetus for the retraction, but I want the language we use to be precise. It seems to me that Kaurov and Oreskes are significant here in terms of their letter, but I'm not seeing a valid reason to highlight their research instead. --Tryptofish (talk) 01:45, 11 December 2025 (UTC)

I apologize if the previous proposal insinuated that the Kaurov and Oreskes paper was "directly relate[d] ... to the decision to retract". That was my not my intention and the newest proposed addition does not contain such an implication. I agree that "[t]he decision to retract was based on their letter, not on their paper". But they wrote the letter as a result of their research.

As for why we should include their paper, there are many reasons:

1) Wikipedia prefers academic sources over WP:BREAKING news.

2) Wikipedia prefers peer-reviewed research over news reports

3) The Kaurov and Oreskes paper was mentioned in several of the aforementioned breaking news reports about the retraction of the article.

4) Their research into the Williams et al. 2000 is what caused them to write the letter, they didn't just wake up and decide to write a letter asking for its retraction. The letter was the result of their research into the influence of this paper on scientific and popular discourse.

You'll note that my most recent edit omits the fact that they published a paper, but simply refers to the fact of their research into broader influence of this paper, something prominently noted in many contemporary news reports about the retraction. I would suggest we cite their paper, but not mention the paper in the text of the article.

I have seen precisely zero substantive reasons to omit this paper and I will repeat another editors comment Despite walls of text, when it comes down to it, I still have not heard any valid reason to omit this key source prominently mentioned in The Signpost. Katzrockso (talk) 02:50, 11 December 2025 (UTC)

OK, at this point, I'm not hearing any other editors continuing to object to citing the paper, and I was arguing based largely on having seen such objections in the past. But if other editors aren't bothering to object, then I'm going to drop my objection in the interest of settling this and moving on. I think I agree with you that, based on other secondary sources, their paper has had influence and would be reasonable to include. I don't think there is a problem with news sources for what we are using them for here, but I've decided to agree with you about adding Kaurov and Oreskes to the group of cited sources. I'm going to implement a slight variation on your proposed edit now, with a bit of reordering and word-smithing. I hope this is OK. --Tryptofish (talk) 21:55, 11 December 2025 (UTC)

That's perfectly fine, please make it mesh better or whatever you see fit. I am open to other wordings that reduce the prominence of this source in relation to others or more succinct explanations in the aims of reaching a consensus, but I still strongly believe that its inclusion is warranted.

We can always reevaluate if their paper ends up being a flash in the pan or heavily critiqued or something, but I think the current state of knowledge we have means we should include it. Big part of my reasoning there is that it is the academic source that is the most 'independent' of the topic (i.e. hasn't been funded by organic groups) and by an eminently respected historian of science.

I don't think news sources are bad, I am just saying that we should prefer academic sources absent a very compelling reason otherwise, and when academic sources on a topic exist we should not exclude them in favor of breaking news sources (which lean more towards being WP:PRIMARYSOURCES given that they are temporally closer to the event than peer-reviewed books/academic articles that are more distantly removed). My only comment is I'm not sure we really need as many news articles as we currently have cited on some of the sentences. Perhaps WP:EVENTUALLY we will have an article on this topic itself (e.g. this whole ghostwriting scandal) and we can go into further depth and cite a million newspaper articles, but currently I don't see the need for so many (WP:CITATIONOVERKILL). But that's a much more minor detail that ultimately doesn't matter too much. Katzrockso (talk) 22:15, 11 December 2025 (UTC)

Actually, citation overkill was part of my previous concern. But, whatever. --Tryptofish (talk) 22:26, 11 December 2025 (UTC)

I would suggest cutting either the Science or Le Monde citation, as well as the NPR citation and the Bloomberg citation (short, paywalled and basically entirely about the later publications, not the 2000 paper). We can redistribute citations in the final two sentences another way - i.e. cite the retraction notice in the penultimate sentence.

I will boldly make these changes for now to show what it would look like. Katzrockso (talk) 22:44, 11 December 2025 (UTC)

My objection to including Kaurov and Orsekes still stood (even if I couldn't get to a computer in the last couple days). At the end of the day, the retraction source mentions nothing about them, so it is a WP:DUE issue as mentioned earlier. The whole point of this discussion was to focus in on what the retraction actually said. KoA (talk) 08:28, 12 December 2025 (UTC)

Once again, we cannot interpret the retraction ourselves, that is the definition of original research. We have to rely on what secondary sources say about/interpret the retraction, e.g. the news articles we have already included about the retraction (e.g. Le Monde, Retraction Watch and others I have cited here).

We can certainly focus on "what the retraction actually said" in the abstract sense of perhaps we think these secondary sources are so erroneous that we should omit them altogether, but we cannot substitute that for an original analysis of the retraction placed into wikivoice or article text.

My argument remains that by omitting the antecedent of the retraction (i.e. Kaurov and Oreskes' letter), we do the reader of the article a disservice in leaving the gap between the emails being published and the retraction unexplained, a question that features prominently in several news articles on the topic. That such a question and explanation should be omitted from our Wikipedia article is unjustified. Katzrockso (talk) 08:47, 12 December 2025 (UTC)

Once again, we cannot interpret the retraction ourselves. No one is doing that in-text except when you when you suggested earlier above we selectively omit what the journal said. Remember you were the one claiming that what the editor said didn't exist in sources despite it being directly pointed out to you, and now we're having to wrestle with that in content discussion. This is a secondary WP:MEDRS source. We are supposed to summarize it, and a pretty direct paraphrase is generally what you'd shoot for, especially since what it says (or omits) holds more weight than periphery WP:MEDPOP sources. The core source needs to guide us first rather than what editor preference or thought is.

With that, can you go back to my last Dec. 7 comment and answer the question of last paragraph? I'll include it here just to keep from having to jump back, but I was very direct in what I asked:

So when you look at the original edit and the source, can you point to what exactly was misleading and how you would have worded it differently based on the retraction source? That's not to get into other sources, but pin down what the issue is with respect to the main source. Otherwise, I pretty much paraphrased right from the journal, Concerns were raised regarding the authorship of this paper, validity of the research findings in the context of misrepresentation of the contributions by the authors and the study sponsor and potential conflicts of interest of the authors. I, the handling (co)Editor-in-Chief of Regulatory Toxicology and Pharmacology, reached out to the sole surviving author Gary M. Williams and sought explanation for the various concerns which have been listed in detail below. We did not receive any response from Prof. Williams. Hence, this article is formally retracted from the journal. This decision has been made after careful consideration of the COPE guidelines and thorough investigation into the circumstances surrounding the authorship and content of this article and in light of no response having been provided to address the findings.

That was asking you to do what Tryptofish already asked about content-building. You didn't really mention how you'd make that edit a more accurate paraphrase of the quote. Instead you mention BLP which wasn't really relevant because Williams was named in the source directly (and is mentioned in the citation anyways). Even if it was relevant, it was already mentioned that the name could be swapped with "author", so that wouldn't have been a good reason for entire removal either. Where do you see the paraphrase as inaccurate when you compare to the quoted source? I was hoping that question would have already led to some forward progress over the few days it took me to get back to a working computer. KoA (talk) 09:59, 12 December 2025 (UTC)

I'm going to move past the unevidenced insinuations that I made some claim that was 'directly pointed out' as false as to not dignify that characterization of the conversation.

That was asking you to do what Tryptofish already asked about content-building. You didn't really mention how you'd make that edit a more accurate paraphrase of the quote. As I responded then and I will repeat now:

I don't understand what you are asking me to do, I agree that it is an accurate paraphrase of the content in the journal - and the current version of the article reflects that edit, bar removing the information that multiple editors agree has an issue with BLP. Something be true and verifiable is not a guarantee for inclusion (WP:VNOT) and indeed BLP here weighs against including the name of Williams in the article, as noted by Tryptofish already. My edit was minimal and simply removed the material that falls afoul to BLP

We can paraphrase and quote all sorts of different aspects of the retraction notice, or indeed a multitude of different sources. What your question presupposes is that every piece of information in the specific quote you have selected needs to be summarized in the Wikipedia article, which is an unsupported presumption. Where do you see the paraphrase as inaccurate when you compare to the quoted source I suggest you refer to my previous comment once again, where I stated very specifically and in no uncertain words that I agree that it is an accurate paraphrase of the content in the journal. Katzrockso (talk) 10:36, 12 December 2025 (UTC)

Katzrockso, part of the problem is you're saying completely contrary things sometimes denying outright that the source even mentions what we're discussing and now saying it was an accurate paraphrase. The latter contradicts your edit summary when you removed the text too as you were implying that that the apparent accurate paraphrase of the journal is misleading. Less patient editors would have just dismissed that as not being able to contribute to WP:CONSENSUS and moved on. It's challenging to engage in consensus-building with that.

So now it boils down to you saying the proposed content was an accurate paraphrase, but you're claiming it's undue to mention the journal's summary reasoning for retraction. The problem is that there's nothing in the source saying it's undue, but instead the editor mentions it as key reason for retraction in their summary, repeatedly. Your argument is a very clear WP:DUE policy violation as well as WP:V when it comes an editor contradicting a source, so that's why we really can't give it traction in assessing consensus here. We're not talking about a tangential side sentence in a source that we'd normally dismiss under WP:ONUS like implied. Right now there's not a grounded WP:PAG reason for removal. I'm concerned that the whole addition on the retraction is in jeopardy at this rate, so that's why I've been working hard to get this problem of omission of what the journal highlights as key details addressed with what should have been a simple tweak rather than this long talk section. There are some ideas I have to break that log-jam, but I'd rather see what others have to say about the quoted material first. KoA (talk) 18:11, 12 December 2025 (UTC)

If you are going to make these accusations, I ask that you please back them up with diffs so I can attempt to understand what you are referring to. I read this comment and find a characterization of my contribution to this discussion that is entirely at odds with the posts I have made.

You have repeatedly stated (see , ) that I am "contradicting a source" but have failed to substantiate where and how I have done that. I am kindly asking you, again, as I did this in this comment to specify which comments of mine you are alleging contradict a source. If my comment implied something I do not believe or I said something incorrect, I am more than happy to clarify my statement or correct a factual error I made - I have made mistakes before and am perfectly willing to admit when I am wrong. But it is not fair to assume that my comments here mean I am "not able to contribute to WP:CONSENSUS".

Either way, I would ask that we either focus on content or support characterizations about other editors behavior with evidence to avoid casting aspersions. I am more than happy to continue discussing the merits of any potential change, but it is frustrating to see these unsupported insinuations about my comments here woven throughout more substantive disagreement. Katzrockso (talk) 21:28, 12 December 2025 (UTC)

You mention WP:FOC, which please remember is policy that applies to all of us. The main point is that we move on to content discussion, which I've been repeatedly trying to do here while trying to navigate the ideas you've proposed. At this point after referring back to your comment a few times, even with direct reply to that comment, I shouldn't need to quote you, we currently have no reliable sources saying that the reason that the reason the paper was retracted because Williams did not respond to the concerns expressed by the editor. There's no way to get around that such a statement directly contradicts the retraction source, so there was no way to discuss your content idea without bringing up that it misrepresented the source. I don't like being put in that position where content discussion is forced to comment on that.

Based on your recent comment though, the issue of being sourced should be moot now, so if you have questions on the above paragraph that may be better suited for usertalk from an FOC perspective. Instead, my last paragraph in the last comment was partly hoping for confirmation from you. If I'm understanding you correctly, you're claiming it's undue to mention the journal's summary reasoning for retraction (i.e., concerns and lack of response). Is that correct? This is at least helping move the needle on next steps. KoA (talk) 22:56, 14 December 2025 (UTC)

I will direct my comments on the FOC part to your talk page.

I do not believe that "directly contradicts the retraction source" or is otherwise misrepresentative. I recommend reading WP:PSTS and WP:PRIMARYUSE, where we can understand easily that the retraction notice is indeed a primary source of information about the retraction itself. More secondary sources like the newspaper articles perform the type of analysis/evaluation that allows us to state "the retraction occurred because of reason X" or "the retraction occurred because of reason Y", whereas we cannot state something like that based on the retraction notice itself unless said retraction notice explicitly stated that. On my reading of the retraction notice, I do not see any explicit statements of this form. Perhaps we can interpret the statements in the retraction notice, but this is interpretation that we need secondary sources to provide to avoid WP:OR.

I agree that the statement "Williams did not respond" is supported by the primary source as well as other secondary sources. I think this is indeed an accurate paraphrase of the retraction notice; following a lack of response by Gary M. Williams, the last surviving author of the paper. However, given its juxtaposition within the sentence, it improperly implies the the lack of response from Williams played some vital role within the reason for retraction or that had Williams responded to these concerns, the paper would not have been retracted. As I have explained, this may be entirely true and we can perform all sorts of analyses of the retraction notice to come to that evaluation. But this remains WP:SYNTH as it "impl[ies] a conclusion not explicitly stated by any of the sources". We have several sources that specifically identify the Kaurov and Oreskes letter to the editor as triggering the investigation by the editor into the paper. We do not have any that are analyzing the retraction and claiming it resulted from Williams lack of response.

Given that we do not have any secondary sources that make the implication of your text, in addition to the fact that there is no need to include the name of the surviving author (given that this would only introduce more sensitive BLP information), we need a much stronger rationale to include this information than the fact that the retraction notice itself mentioned it. Katzrockso (talk) 05:21, 16 December 2025 (UTC)

Ah I see. You're somehow thinking the secondary source (the retraction) is a primary source. That's not how it works. Otherwise, all review articles would be considered primary when they critique another academic source. What the rest boils down to though is that you're running into OR/SYNTH problems in the logic you're trying to apply when the source explicitly states the reasoning you're against including. That gives me enough for next steps at least. KoA (talk) 07:32, 16 December 2025 (UTC)

Yes, the retraction is a primary source for the information about the retraction. That is not really a negotiable idea. Katzrockso (talk) 08:08, 16 December 2025 (UTC)

Katzrockso is correct: a retraction notice is a primary source about the retraction itself. All sources are primary for some things, and the first-ever publication of journal's decision to retract a paper is primary for the fact that the journal decided to retract the paper.

@KoA, see the definition in WP:PRIMARY:

"A secondary source provides thought and reflection based on primary sources, generally at least one step removed from an event. It contains analysis, evaluation, interpretation, or synthesis of the facts, evidence, concepts, and ideas taken from primary sources. Secondary sources are not necessarily independent sources. They rely on primary sources for their material, making analytic or evaluative claims about them."

Then run down the list:

• Does this "provides thought and reflection based on primary sources"? No, this is the first-ever publication about the journal's decision to retract this article. Therefore it cannot provide "thought and reflection" based on any kind of prior sources (for the fact of the journal's decision to retract the article).
• Is this "at least one step removed from an event"? No, the retraction notice is is the event itself! (Again: for statements about the journal's decision to retract the article.)
• Does it contain "analysis, evaluation, interpretation, or synthesis" about the journal's decision to retract the article? No, this is the announcement of the decision itself, not a separate analysis, evaluation, interpretation, or synthesis of thoughts and reflections about the decision.
• Is this an independent source? No, and while Secondary sources are not necessarily independent sources, they usually are. This isn't: This is the journal's editors announcing the journal editors' own decision about the journal's contents.
• Does this "rely on primary sources for their material" about the journal's decision to retract the article? No, at the time of publication, this was the first-ever source about the the journal's decision to retract the article. It is the primary source (again: or statements about the journal's decision to retract the article. It could be a secondary source for other content, but not for statements about the journal's decision to retract this article).
• Does this source make "analytic or evaluative claims about" the decision to retract the article? No, this source is the decision to retract the article; it is not a separate analysis or evaluation of the decision to retract the article.

When this source is cited in support of claims about the journal's decision to retract the article, it is a primary source.

Wikipedia:Primary does not mean bad. This could be a reliable source to support multiple claims. It's just not a secondary source (or an independent one, for that matter).

BTW, if you are looking for a word to explain its value to other editors, you might consider choosing a word like authoritative instead. WhatamIdoing (talk) 02:58, 17 December 2025 (UTC)

What's frustrating is that is exactly what I did already in my reply from earlier this week. There was already a working edit that pretty much paraphrases the retraction. That could have been tweaked with suggested edits in talk, and I did ask Katzrockso to address the issues in the comment. So Tryptofish, when you look at the last paragraph in that Dec 7 reply of mine, how would you approach that question of attempting to pretty directly paraphrasing the journal with what I quoted? What's wrong with that original edit or how could it have been improved? I'm just not seeing much that falls out of alignment with the quote. KoA (talk) 08:57, 12 December 2025 (UTC)

KoA, you should recuse yourself from this discussion as you have a clear conflict of interest. The Kaurov & Oreskes paper discusses your repeated citation of the Williams 2000 paper on Wikipedia after it was revealed to be ghostwritten. Since you have a personal stake in whether or not the paper is cited (in order to avoid embarrassment), I think it would be best for you to recuse yourself from discussions that are directly related to that paper. Nosferattus (talk) 20:12, 12 December 2025 (UTC)

I agree. KoA should step back and not participate in this or related discussions. Andre🚐 22:37, 12 December 2025 (UTC)

I didn't take note of that detail when I read the paper. Nonetheless, I doubt that such inclusion would truly result in embarrassment since it is unlikely that the vast majority of readers would read the entire paper and take note of that small piece of information and subsequently connect it to a different username (it seems KoA uses a different username now than before). However, this might raise the appearance of a conflict of interest, though I am not familiar enough with the WP:COI policy to comment on that. I have seen some editors describe it very broadly. Katzrockso (talk) 22:45, 12 December 2025 (UTC)

I feared this was going to happen, and at this point I'm just going to tell the rest of you to work it out. I tend to feel that, after the role of K and O in the retraction, it's now reasonable to cite their paper, but I truly don't care strongly about citing it, one way or the other. I think there's a good case for including the fact that they wrote a letter to the editor-in-chief, so I'd like that to stay in, but I don't care that much about citing their paper. I'm satisfied with the content as it is now. KoA, please let me make a friendly suggestion that you just let it go. It's not worth going to battle over an inline citation.

But I also have something else to say to the rest of you. The claim that KoA has a WP:COI because of avoiding embarrassment is totally absurd. And there has been too much of a history of bogus COI accusations in the GMO topic area for me to let that pass. So if anyone wants to disagree with KoA on the merits, go right ahead. But the next editor who claims that KoA must not participate in this discussion: I am taking you to WP:AE with no further warning. I mean it. --Tryptofish (talk) 01:13, 13 December 2025 (UTC)

Yeah, the COI claims about "embarrassment" are definitely getting into WP:ASPERSIONS territory and just derails content discussion. I'll briefly clarify on that, but it's not content discussion that really belongs on article talk, so I'm collapsing that bit.

Extended content

For what it's worth, I think we did things (nearly) right here on the Williams paper, so the "embarrassment" comment wouldn't be appropriate even if this wasn't an article talk page. Prior to the retraction, it was a reliable source from a wiki-pespective (though older and not really used much anymore), and once the retraction happened, then we would have cleared any citations of it. That's all for my personal feeling on it and it's part of doing things even-handedly. We don't have a WP:CRYSTALBALL if a paper will be retracted or not, so I do think we jumped the gun a bit early, but it's moot at this time because even I would have been removing those citations at this point, which I made pretty clear in past talk discussions. Even the Signpost article mentioned aspects like MEDRS as to why we still had citations to the article prior to retraction. That's all meta to respond to a comment that never should have been made in a CT talk page though, so I'm just going to leave it at that.

Part of the problem discussion-wise I'm seeing here is that there was a persistent push to put mention of the K & O paper in. That's despite not being mentioned in the retraction at all. Issues were already brought up with it on the talk pages, and that secondary bit about K & O was inserted recently after there was a short period of time when no one spoke against it. Meanwhile, I'd been patiently not reinserting anything that was being discussed in edits I was proposing on the retraction itself above this sub-thread. I had been saying we need to get sorted on what the retraction itself has to say, only to have comments saying what the main source mentioned for primary reasons for retraction is somehow undue. There's a bit of a double-standard there in content building without even getting into the more inflammatory calls to leave above. I think part of the problem is the K & O topic kept getting reinserted into attempts to focus on what the retraction source alone had to say.

When it comes to mentioning the K & O topic itself, that is secondary to accurately describing the retraction source itself. That inclusion doesn't have consensus, but I'm not going to be the one to remove it right now. I really only have time for the retraction source discussion right now in this section. That's not how it should work, but that's all volunteer time really can allow for right now. KoA (talk) 21:39, 14 December 2025 (UTC)

I was not aware that KoA used a different username previously. I was only referring to Kaurov & Oreskes discussion of "KoA" specifically. What username did KoA edit under previously? Nosferattus (talk) 18:06, 13 December 2025 (UTC)

I thought the article said Kingofaces rather than KOA, so I must be misremembering. I read through some very old posts on this talk page before commenting, so I must have imagined that into the Kaurov paper Katzrockso (talk) 20:08, 13 December 2025 (UTC)

"Not working at Monsanto"

I'm concerned about this edit, that unilaterally changed what had been the recent consensus language: . My concern is over removing the words "not working at Monsanto". The edit summary says that the cited source, which is the paper itself (), does not substantiate that wording. Huh? There are three authors listed (click "show more" in the source to see their affiliations): Gary Williams of NY Medical College, Robert Kroes of RITOX in The Netherlands, and Ian Munro of Cantox Health Sciences. None of them were working at Monsanto, full stop. And it's relevant to include, because the whole point of the ghostwriting charge is that the paper represented that no one from Monsanto was writing it. Leaving that out makes no sense. --Tryptofish (talk) 00:54, 11 December 2025 (UTC)

I am sure they were not working at Monsanto and never worked at Monsanto, but the absence of an affiliation is not evidence of this. It would be improper WP:OR and WP:UNDUE information to make such a statement from authorial affiliations on a paper. Including it makes no sense as it makes the entire sentence structure awkward and sound completely unnatural, which is why I removed it, with the note that it can be added in another sentence or elsewhere in the paragraph with a source that substantiates it. We should not be introducing our own commentary when none of the articles that I read thought that this fact was relevant enough to include. This is precisely why Wikipedia relies on secondary sources and not our own interpretations of primary sources (i.e. the authorial affiliations of a paper). Katzrockso (talk) 01:12, 11 December 2025 (UTC)

I think you don't understand how scientific journals list authorship. --Tryptofish (talk) 01:22, 11 December 2025 (UTC)

I understand just fine, thank you. Katzrockso (talk) 01:25, 11 December 2025 (UTC)

Then sorry, but I absolutely do not understand your reasoning. The cited source states where those three people worked. All of the secondary sources treat this paper as having been made to appear to have been written by three scientists who were independent of Monsanto – and the ghostwriting evidence, the retraction, and everything, all treat the ghostwriting as having come from Monsanto people who were not listed as authors. To seriously entertain the notion that the three listed authors actually had appointments at Monsanto but were hiding it, which appears nowhere in the sourcing, seems inexplicable to me. --Tryptofish (talk) 01:32, 11 December 2025 (UTC)

How about "not known to be employees or affiliates of Monsanto"? Andre🚐 01:55, 11 December 2025 (UTC)

I am not entertaining that notion in any meaningful sense, but noting that it is a logical possibility (not stating that it is likely - I certainly believe it to be false). This is why we need to have secondary sources that analyze the authorial affiliations and note this fact for it to be included on Wikipedia. We don't analyze primary sources (authorial affiliations) ourselves, we rely on what secondary sources say about events and summarize that knowledge. None of the news articles I have read about this controversy have mentioned that the authors did not work for Monsanto, let alone placed this fact in such a prominent role as to be the very first sentence of their description of the controversy. That's why I have repeatedly stated that I am not opposed to this fact being mentioned provided a secondary source be provided that establishes it as WP:DUE and that perhaps it shouldn't be mentioned as a relative clause that impedes the flow of the sentence.

Please do not interpret this as me trying to promote or insinuate the idea that they were employed at Monsanto, I don't believe that at all. I was trying to increase the readability of the sentence (it read very unnaturally/awkwardly to me) and improve the compliance to Wikipedia policies.

As for Andre's suggestion, I would oppose that because it implies that there is some sort of basis for believing they were secretly hiding their employment at Monsanto that we haven't yet found. That implication would violate WP:SYNTH just as much as the original text does. We cannot make a conclusion that is "not explicitly stated by any of the sources", which includes the conclusion that they did not work for Monsanto. That is not explicitly stated by any of the sources that I can find. Katzrockso (talk) 02:09, 11 December 2025 (UTC)

I, too, think that Andre's suggestion, although well-intentioned, suffers from what it implies. I feel strongly that "not working at Monsanto" is needed because, as I said earlier, it's the necessary context for why this was ghostwriting (as opposed to, for example, a bad paper explicitly written by Monsanto employees).

But I decided to look for other sources, and I think this may break the logjam. We already cite two, that seem to me to confirm that the three authors were not working at Monsanto. From Le Monde: ...revealing that the actual authors of the article were not the listed scientists – Gary M. Williams (New York Medical College), Robert Kroes (Ritox, Utrecht University, Netherlands), and Ian C. Munro (Intertek Cantox, Canada) – but rather Monsanto employees (), and from Retraction Watch: Gary Williams, then a pathologist at New York Medical College in Valhalla, Robert Kroes, a toxicologist at the University of Utrecht in the Netherlands, and Ian C. Munro, a toxicologist at Cantox Health Sciences International in Ontario, Canada, were listed as the authors (). The first indicates where the three authors were working and clearly differentiates them from Monsanto employees, and the second joins the first in confirming the affiliations listed on the paper itself. With these two secondary sources, I think we have confirmation that there are two sources that consider the journal listing to be reliable, and that indicate that these three listed authors are not, themselves, the Monsanto employees who did the ghostwriting. --Tryptofish (talk) 22:51, 11 December 2025 (UTC)

The Retraction Watch source doesn't go as far as I would need it to, but the Le Monde source is better in that it explicitly contrasts the authors with Monsanto employees, rather than simply restating the author affiliations. I did a further search and found "The listed authors of the paper, Gary Williams, Robert Kroes and Ian Munro, were scientists who did not work for Monsanto" , which is the precise type of language I was looking for here and helps establish that the information has a much better case for being WP:DUE. I think we this description best fits in the third sentence of the paragraph that talks about Monsanto's response, but I wouldn't be opposed to it earlier. I just really do not like how the previous version flowed grammatically in the sentence; In 2000, three scientists not working at Monsanto published an evaluation of the safety in humans of glyphosate and Roundup reads as bad writing to me, e.g. the way that the relative clause is placed within the sentence. Perhaps because it disconnects the subject ('three scientists') from the verb ('published') so significantly. If you can propose an alternative version that doesn't have these grammatical concerns, I would be supportive and assuage my concerns about the fact that fundamentally I don't think the information is useful to the reader (i.e. most of the other articles do not report this information but I do not think they have omitted any vital information that would impact their readers understanding of the topic - indeed, I think ghostwriting already implies that the figurehead author(s) is a distinct actor from the 'true' author of the article - i.e. Monsanto can't ghostwrite for Monsanto). Katzrockso (talk) 06:51, 12 December 2025 (UTC)

@Tryptofish I added back the information citing The Guardian source and Le Monde here . I decided to put it in the second sentence, since I think it flows best there. I still don't think it is necessary/WP:DUE/particularly informative for readers but since you have advocated very strongly to include it, I'm OK with including it there. Feel free to propose some alternative wordings/structures, this just seemed to read fairly naturally to me. Katzrockso (talk) 21:34, 12 December 2025 (UTC)

Thanks so much! I changed the wording a bit, because I think it reads better this way, but if you'd rather word it differently, I'm happy to discuss it. --Tryptofish (talk) 01:02, 13 December 2025 (UTC)

@Tryptofish as I mentioned above, I really think the relative clause separating the subject from the verb (i.e. "three scientists [not working at Monsanto] published" - the [] separates three scientists from the verb published) doesn't work very well here and the information doesn't fit contextually very well in the first sentence. In the first sentence, we are explaining the paper and its conclusion, no mention of Monsanto. The second and third sentence we are explaining the background / context of the paper and its connection to Monsanto, which is why the authorial information would fit much better there as we can contrast the fact that they weren't employed at Monsanto with the observation that the emails indicated the paper was nonetheless ghostwritten. I'm open to other wordings that don't force the unnatural phrasing into the first few words of the paragraph. Katzrockso (talk) 04:06, 13 December 2025 (UTC)

I think we are in "you say to-may-to and I say to-mah-to" territory here. My feelings about which way flows better and preserves context better is the exact opposite of yours. (The first sentence doesn't have to otherwise mention Monsanto, because this page is about Monsanto. Having it in the first sentence gives context about who the "three scientists" who wrote the paper were. Doing it the other way results in something much more wordy, suggesting oddly that there is something unexpected about a spokesperson from Monsanto commenting about a paper written by three non-Monsanto people.) One thing that occurs to me, that might help a little, is to move the inline cites to the end of the sentence, so they don't interrupt it. I'm going to do that right now. Beyond that, if you really feel strongly about it, perhaps other editors here could offer a third opinion. --Tryptofish (talk) 01:05, 14 December 2025 (UTC)

it the other way results in something much more wordy, suggesting oddly that there is something unexpected about a spokesperson from Monsanto commenting about a paper written by three non-Monsanto people Concision is not always better when it leads to bad writing: interrupting subject-verb flow is rarely a good practice in scientific or encyclopedic writing. Even stating the fact in a separate sentence would be superior (i.e. The scientists were not employed by Monsanto.) This accords with both of the sources we are using, which introduce this fact in a separate sentence. Indeed, The Guardian provides the information about the conclusions made by the paper in a separate paragraph than the information about the authors. The Le Monde source is much more similar to the version that I proposed, which specifically contrasts the listed scientists on the paper with the 'actual authors' (i.e. Monsanto employees) [The disavowal comes 25 years after publication and eight years after thousands of internal Monsanto documents were made public during US court proceedings (the "Monsanto Papers"), revealing that the actual authors of the article were not the listed scientists – Gary M. Williams (New York Medical College), Robert Kroes (Ritox, Utrecht University, Netherlands), and Ian C. Munro (Intertek Cantox, Canada) – but rather Monsanto employees] and connects this to the emails. Once again, the sentence about the paper describing its conclusions and publication were in a separate sentence than information about the authorship. I don't know why we shouldn't following the separation/organization that our sources do.

This also goes back to my previous contention that this fact (that they didn't work at Monsanto) is rarely mentioned by articles, meaning it is hardly WP:DUE and does not provide additional useful information to readers [indeed, I think that the clause adds more confusion than clarity - why is it necessary to include this? Were there questions that they were employed by Monsanto? That is what the clause makes me wonder. Placement in another sentence resolves this anxiety, as it contrasts the authorship question with the ghostwriting information rather than retain the tension for longer]. To the extent that we have to include this information, as you are insisting, it should follow the organization/placement that the few sources that mention it do.

Also, I'm not sure what suggestion is being made by the alternative version.

As for moving the citations, that is not a good idea. Having a big mass of citations at the end does not clarify which source is supporting/verifying which portion of the sentence, which makes it more difficult for a reader to select which article to read if they wish to verify a particular statement in the sentence. Two little blue boxes do not interrupt the flow at all, as readers are not reading the little blue boxes but the words in the sentence. Katzrockso (talk) 08:42, 14 December 2025 (UTC)

Just wanted to leave a note that I saw this source from Carey Gillam had been added. Generally her articles are considered to have a COI (others here have run into that before) because of her heavy involvement in the organic industry, industry competitor, etc. when it comes to pesticide/GMO topics.

That shouldn't affect anything for the content from what I can gather reading this since Le Monde is used, but I'm paywalled at the moment on the Le Monde. A lesser issue there is that the author of that article is involved in the "Monsanto Papers" dispute with awards, etc. for their work on it. Kind of one I'd drift towards a different news article, but I left it for now so it hopefully doesn't upend the above. If there's something dependent on the Gillam article though, that content would need a different source. KoA (talk) 04:16, 15 December 2025 (UTC)

The only reason The Guardian source has been added is because it supports the claim about the authorial information, refer to the thread above. If you can find an alternative source that explicitly makes this claim, feel free to change it out. Otherwise, there is no policy issue with using even a biased or non-independent source for uncontroversial claims; reliability exists in a specific context. Katzrockso (talk) 05:01, 15 December 2025 (UTC)

Katzrockso, please undo your reinsertion of the source, I already talked to you about ArbCom expectations on your talk page when it comes to continuing to reinsert new content, especially for something as simple as this one. When there was a clear problem with the COI source on the talk page, the course of action isn't to just reinsert it. We're not citing Monsanto for anything, nor are we citing those from the organic industry either, and that was already covered on this talk page. This isn't something to just push ahead on. The Le Monde source already covers the existing text based on your quote above, so that's why I pretty clearly said I was leaving that source even if it had some lesser issues that could have otherwise warranted looking for another source. That was supposed to be a small tweak and move on moment rather than extending this section even more.

I would say though that the source already gives you a different framework to paraphrase. Just list the names and/or their affiliation like the sources do. If for some reason the the COI source is vital to something, that is a red flag if since independent reliable sources should be guiding us, not what we want to include as editors. That's just my suggestion after reading over this, but the existing text should be fine as a default if the Gillam source is removed, which is one of the simplest resolutions currently on this long talk page. KoA (talk) 19:46, 15 December 2025 (UTC)

I'm going to agree with KoA about removing that source. I didn't recognize the COI issue when I went along with adding it to satisfy the concern about verifying that the scientists didn't work there, but I remember now that there has been a lot of previous discussion about that author. But, as I have said repeatedly, this is such a patently WP:BLUESKY kind of thing, when one takes it in context, that we really should not need any sources beyond the retracted paper itself, to explain who the authors of that paper were affiliated with. We've still got the Le Monde source, and there comes a point where common sense should not be confused with original research. --Tryptofish (talk) 00:47, 16 December 2025 (UTC)

I have removed the source, pending a later discussion of this topic. I do not agree that the Le Monde source is sufficient, as I explained at length, The Guardian source is necessary to establish a bare minimum of WP:DUEness for this information. That you personally believe this information is important enough to include in Wikipedia is not established by the body of secondary sources on this subject and removing the only source that explicitly describes it because of a supposed COI issue (once again, COI sources are perfectly fine to establish non-controversial facts and there is no policy basis to remove a source for this reason) only highlights that this information is not WP:DUE.

As for KOA's comment, I would say though that the source already gives you a different framework to paraphrase. Just list the names and/or their affiliation like the sources do We have already discussed this issue multiple times and to the extent that we are implying anyone engaged in scientific misconduct/ghostwriting, this would be unnecessary / introduce BLP sensitive content. I would support including the university affiliations if it wouldn't lengthen the text so much for minor information [like e.g. if it were one university] - we are not discussing the authors themselves at length, so I have no clue why we would need to include the lengthy construction "Ritox, New York Medical College, Interek" or even longer to clarify. Katzrockso (talk) 05:02, 16 December 2025 (UTC)

First, I want to say thank you, for the self-revert about that COI paper.

Now about the issue of sourcing for where the three authors worked, I'm going to try to explain my reasoning with the following, which one can think of as A Modest Proposal. I'm afraid that we are going to have to remove all mention of Kaurov and Oreskes from this page. That's because we need, and seem not to have, a reliable secondary source that establishes that the research they did, the research paper they wrote, and the letter they sent to the journal editor, were all done by just those two people, without anyone else contributing. Please understand, I believe that they did the work themselves, and I doubt very much that there were any other, uncredited, contributors. But we need to have sourcing if we are going to credit the two of them with having had such an important influence. I realize that their research paper lists just the two of them as authors. But we need a secondary source to confirm that. It's not enough to have reliable secondary sources saying that they were the authors. We also need a reliable secondary source that says, explicitly, that there were not any other authors who contributed to their research. We cannot give them credit without such a source. --Tryptofish (talk) 22:44, 16 December 2025 (UTC)

Your analogy doesn't work for many reasons, but I will list four:

1) I never suggested removing the Williams et al paper because there were no secondary sources suggesting that they were not employed at Monsanto.

2) The basis for excluding this information is that it is not WP:DUE for inclusion based on the fact that it is not mentioned in secondary sources. This objection has not been rebutted, but ignored numerous times, in addition to the reference to WP:VNOT.

3) I am against the inclusion of unnecessary information that would describe Kaurov and Orekses as "not employed at Monsanto" (or whatever other unencyclopedic descriptors) just as much as I am against the inclusion of this unnecessary information. I see no benefit to the reader by the inclusion of this clause and believe it only confuses the reader.

4) You are conflating the absence of verifiable information about the fact that the authors of the Williams et al paper were not employed by Monsanto with the positive statement that they may have been employed by Monsanto. Those are not analogous. Removing verifiable information does not give any credence to the negation of the verifiable information. For example, if I added the clause "Naomi Oreskes, a historian of science", it would not give credence to the idea that Oreskes is not a historian of science to remove that clause. It would simply mean that it was otherwise unsupported or undue. Katzrockso (talk) 23:13, 16 December 2025 (UTC)

Part of my use of that explanation was to speak to other editors, not necessarily to change your mind, which increasingly looks like it will not happen. And I think it's pretty obvious what I was seeking to get across.

1) So what? You actually did make an edit removing the language that said that Williams et al. did not work at Monsanto.

2) Nobody rebutted the argument about DUE based on secondary sources? I sure did!

3) I also explained how omitting the information would confuse the reader.

4) I understand perfectly well that your preference for removing the language about them not working at Monsanto was not based on your belief in the positive statement that they actually might have worked there. It was based on your misunderstanding of WP:LIKELY.

I think it's time for you to drop this, and move on. --Tryptofish (talk) 23:10, 17 December 2025 (UTC)

My thoughts were already made known above and elsewhere on removing K & O, and I won't rehash any of that (and also having only so much time for all the open threads). Instead Tryptofish, I just wanted to say after reading your Dec. 16 comment (and later), you gave a very through response on a different angle for this conversation, and I do think it at least matches the level of scrutiny Katzrockso has been asking for elsewhere. KoA (talk) 00:44, 18 December 2025 (UTC)

I agree that the 'three scientists not working at Monsanto' wording is awkward, and that a bracketed [not working at Monsanto] is an odd construction. Better flow would be "three scientists, who were not Monsanto employees, published a review...". While the inclusion may not be strictly necessary, it does help clarify the relationship between listed authors and Monsanto which could be more complex, as it is in the Industrial Bio-Test Laboratories section below.Dialectric (talk) 23:52, 14 December 2025 (UTC)

I asked for a third opinion, and I got it. Thanks, and done: . --Tryptofish (talk) 23:58, 14 December 2025 (UTC)

Just a heads up that Kaurov and Oreskes also submitted a retraction request for Kier & Kirkland 2013 due to alleged ghostwriting by Monsanto. That paper, which examines genotoxicity, doesn't appear to be currently cited on Wikipedia. It's mentioned in some of the other sources about Monsanto ghostwriting, but doesn't seem to be as important as Williams et al. 2000. Nosferattus (talk) 08:29, 11 December 2025 (UTC)