Talk:Glyphosate/Archive 18

Archive 15

Archive 16

Archive 17

Archive 18

Archive 19

Archive 20

Archive 21

Source 164

Please remove the text section that is related to source 164 - it has been shown to be an article with severe conflicts of interests:

Emails from within Monsanto showed that Murphey sent a slide deck of talking points and a suggested narrative to Reuters reporter Kate Kelland asking her to write a story that accused Aaron Blair, who was the chairman of the IARC working group on glyphosate, of concealing data that would have changed IARC’s conclusion on glyphosate. Murphey told Kelland in an April 2017 email that it was “vitally important information that needs to be reported.” He also told her to treat the information he sent her as “background,” meaning she should not mention she got the story idea and materials from Monsanto. Kelland then wrote the story Monsanto wanted. A deposition of Aaron Blair indicated the accusations laid out in the story were false, but Kelland did not include a copy of the deposition with her story. The story was promoted by Monsanto and chemical industry organizations and Google advertisements and was picked up and repeated by media outlets around the world. Murphey said in his deposition that he put no undue pressure on Kelland, and Monsanto believed the story to be valid and important. “Once I provided the initial information to — to Ms. Kelland, she was free to do with that information what she saw fit,” he said. “And the decision to investigate a story and ultimately — ultimately publish it was her decision, and the decision of her editors at Reuters.”

This is part of the text on this page that Wikimedia Foundation has been made aware of with much more supportive evidence from the history of this page.BillyHatch2020 (talk) 00:52, 31 March 2020 (UTC)

We definitely can't use USRTK here as a source due to their own conflicts of interest and general unreliability in this topic. Reuters is a reputable source, so if there were issues with that article, it would give a statement, retraction, etc. Reporters get sent press releases, etc. all the time, and they're expected to take that information and figure out what they want to do with it in terms of fact-checking or use. If something was amiss with that portion of the process (this source doesn't really address that), then that's where the focus would need to be where we would begin content discussion here. Kingofaces43 (talk) 02:02, 31 March 2020 (UTC)

I agree that USRTK is not the source to use here. I suggest that the original court documents should be used as the source and also the detailed response to the Reuters article should be included from IARC, as a reason to remove the Reuters source and text. This is not about the use of a Press Release, it is about an orchestrated attack on an independent scientific body by a company trying to protect its product, following the previous tactics used by the tobacco industry. I suppose it depends what side of history such information wants to be on. Sources below:

References

BillyHatch2020 (talk) 23:48, 31 March 2020 (UTC)

Advertising Controversies Addition

Editors please add the following information to the Advertising Controversies section:

On 27 March 2020 Bayer (Monsanto) settled claims in a proposed class action alleging that it falsely advertised that glyphosate, the active ingredient in Roundup Weed & Grass Killer, only affects plants with a $39.5 million deal that included changing the labels on its products. Monsanto falsely claimed through its labeling that glyphosate, the active ingredient in Roundup, targets an enzyme that is only found in plants and would therefore not affect people or pets. According to the settled suit, that enzyme is in fact found in people and pets and is critical to maintaining the immune system, digestion and brain function. More sources available if required. — Preceding unsigned comment added by BillyHatch2020 (talk • contribs) 00:31, 31 March 2020 (UTC) BillyHatch2020 (talk) 00:34, 31 March 2020 (UTC)

As I mentioned in the edit summaries in other articles, that doesn't really belong here since it's a product/company specific claim. We have the Roundup article for that. Kingofaces43 (talk) 02:03, 31 March 2020 (UTC)

Thank you for your kind directions to a new editor:). However, it is true that the claim that was settled is specifically about the action of glyphosate in the Roundup product and how glyphosate affects enzymes in plants, animals and humans. In my opinion and in the opinion of a broad group of my scientific and legal advisers such a settlement belongs here as information regarding a false claim made by the producer about the active ingredient glyphosate as per the Law360 and Bloomberg source. Please also review your deletion in Glyphosate-Based Herbicides based on these facts. BillyHatch2020 (talk) 23:57, 31 March 2020 (UTC)

Pronunciation?

Please give the pronunciation for "glyphosate". Various online dictionaries are inconsistent. Bhami (talk) 03:19, 29 June 2020 (UTC)

@Bhami: I see User:Nardog just added two variants, cited to Merriam–Webster online. If there are others you have seen, please post links so we can see if there are more alternatives to include. DMacks (talk) 03:43, 29 June 2020 (UTC)

...and now a third variant, cited to two other dictionaries. Would be nice to know what Monsanto thinks. DMacks (talk) 04:15, 29 June 2020 (UTC)

Algal blooms

Another effect of glyphosate that could be explored in the article is its potential for encouraging toxic algal blooms due to the phosphorus component of the compound. https://www.mcgill.ca/newsroom/channels/news/new-cause-concern-over-weedkiller-glyphosate-295385 — Preceding unsigned comment added by 134.41.64.173 (talk) 21:46, 31 August 2020 (UTC)

This is a piece of primary research (papre here) and we need to wait until it has been mentioned in review articles to demonstrate WP:WEIGHT. Also, per the paper Although still a minor source of P relative to fertilizers, P inputs from glyphosate use have now reached levels comparable to those from sources for which P regulations were initiated in the past SmartSE (talk) 08:13, 1 September 2020 (UTC)

Semi-protected edit request on 2 October 2020

Toxicity

According to this (https://www.nature.com/articles/s41598-019-42860-0) article, published in Nature, the currently allowed amount of glyphosate needs to be reconsidered.

The dose in the experiment — half the amount expected to show no adverse effect — produced no apparent ill effects on either the parents or the first generation of offspring.

But writing in the journal Scientific Reports, the researchers say they saw "dramatic increases" in several pathologies affecting the second and third generations. The second generation had "significant increases" in testis, ovary and mammary gland diseases, as well as obesity. In third-generation males, the researchers saw a 30 percent increase in prostate disease — three times that of a control population. The third generation of females had a 40 percent increase in kidney disease, or four times that of the controls.

More than one-third of the second-generation mothers had unsuccessful pregnancies, with most of those affected dying. Two out of five males and females in the third generation were obese. Tomvanwijck (talk) 14:17, 2 October 2020 (UTC)

​Are there any review articles or textbooks that mention this? – Thjarkur (talk) 14:29, 2 October 2020 (UTC)

​Hi Þjarkur, it is published in Nature, for the rest I know of Science Reports and of ScienceX. The latter, ScienceX is where most of the text comes from: https://medicalxpress.com/news/2019-04-health-effects-popular-weed-killer.html. This is my first edit request, so I hope I am doing things right. Let me know if you need more information. Tomvanwijck (talk) 15:03, 3 October 2020 (UTC)

​It's still too much of a primary study even though published in Nature, see WP:MEDRS, we try not to mention the results of primary studies, instead we try to wait for reviews / meta-analyses / textbooks. – Thjarkur (talk) 15:07, 3 October 2020 (UTC)

​Okay, then I think you are looking for things like these 25 citations where the article is mentioned in other articles: http://citations.springer.com/item?doi=10.1038/s41598-019-42860-0 correct? Tomvanwijck (talk) 16:33, 3 October 2020 (UTC)

I took at look at the first one of those, doi:10.1017/err.2020.16, as an example. "Once through the human epidermis, POEA-based GBHs also promote the movement of glyphosate through cellmembranes, where it can then damage DNA via oxidative stress or some form of chromosomal or genetic aberration, including possibly epigenetic changes." is the only statement cited to the Kubsad lead ref. And the only purpose of this statement is to focus on the transdermal transport properties as an effect of substances other than glyphosate itself. Therefore, this does not seem to be a MEDRS-level secondary review on the nature of glyphosate anywhrere close to the level of RS we have for other content. This suggests ref-cite-counting is not going to be a suitable approach. Instead, as Þjarkur said, we need an actual specific MEDRS-level review article covering the novel aspects of glyphosate itself. DMacks (talk) 17:12, 3 October 2020 (UTC)

"but death has been reported after deliberate overdose of concentrated formulations".  This is in the section headed Glyphosate alone but the source clearly refers to a formulation with other chemicals.  It is in the wrong section!  — Preceding unsigned comment added by 2.102.212.142 (talk) 23:59, 4 January 2021 (UTC) 

Misleading paragraph in introduction on carcinogenicity

I think the fourth paragraph in the introduction of the article is misleading and incohesive with regard to carcinogenicity to humans. I think the main takeaway should be, as stated later in the article, "The consensus among national pesticide regulatory agencies and scientific organizations is that labeled uses of glyphosate have demonstrated no evidence of human carcinogenicity."[99]

2601:600:8180:FE30:585D:5361:DF20:E9C2 (talk) 19:20, 7 January 2021 (UTC)

No, it's perfectly appropriate to highlight the WHO findings in the lead section, per WP:LEAD. If anything, the carcinogenic characteristics of the usual impurities should be given greater weight throughout the article. Binksternet (talk) 22:03, 7 January 2021 (UTC)

I'm not sure why you're bringing up the WHO since it's not even part of the content discussion the IP brought up. For the rest of your statement, please remember that Wikipedia is not a place to advocate for massive content violations of WP:DUE or WP:FRINGE like that. Nothing has changed in the real-world since the topic quieted down to justify even more undue focus on the subject. Glyphosate is still a boogeyman pushed by quacks in the real world[1][2][3], so we still need to be wary of that WP:POV out there when it comes to content discussions. Kingofaces43 (talk) 03:01, 8 January 2021 (UTC)

This is actually a valid concern as the third sentence from the German group having selective quoting can confound what they were saying for some readers. Focusing on just the quote alone makes the authors' statement seem much more uncertain than the surrounding context in that source and what others say about it. Just flipping the sentence around makes it a bit clearer for the time being, so I went ahead and did that. Schinasi is also somewhat cherry-picked to be lead-worthy and does come across as WP:FALSEBALANCE, but that's maybe better left for another day when the overall article is updated.

It wasn't brought up yet, but the fifth paragraph could also use some work in terms of WP:DUE. Normally we wouldn't give prominence to the IARC (other arms of the WHO disagree with them) at the start of that paragraph as a blanket statement of what they did right at the beginning given the minority or even fringe view status of that. Usually minority views come second within context. Better to let other sources put an introductory sentence in on that point, and there have been sources that discuss how the IARC decision has been either out of alignment with nearly all other agencies or taken out of scope.[4] The IARC does need to be described though, but it'll be better to have multiple good secondary sources organized to tackle that one, so that's something for later. Kingofaces43 (talk) 03:05, 8 January 2021 (UTC)

Needed Addition to Human Toxicity Section

Glyphosate residues in food may adversely affect the human gut microbiome.

"Glyphosate is the most common broad-spectrum herbicide. It targets the key enzyme of the shikimate pathway, 5-enolpyruvylshikimate-3-phosphate synthase (EPSPS), which synthesizes three essential aromatic amino acids (phenylalanine, tyrosine and tryptophan) in plants. Because the shikimate pathway is also found in many prokaryotes and fungi, the widespread use of glyphosate may have unsuspected impacts on the diversity and composition of microbial communities, including the human gut microbiome. Here, we introduce the first bioinformatics method to assess the potential sensitivity of organisms to glyphosate based on the type of EPSPS enzyme. We have precomputed a dataset of EPSPS sequences from thousands of species that will be an invaluable resource to advancing the research field. This novel methodology can classify sequences from nearly 90% of eukaryotes and >80% of prokaryotes. A conservative estimate from our results shows that 54% of species in the core human gut microbiome are sensitive to glyphosate."

Lyydia Leino, Tuomas Tall, Marjo Helander, Irma Saloniemi, Kari Saikkonen, Suvi Ruuskanen, Pere Puigbò, Classification of the glyphosate target enzyme (5-enolpyruvylshikimate-3-phosphate synthase) for assessing sensitivity of organisms to the herbicide, Journal of Hazardous Materials, 2020, 124556, ISSN 0304-3894, https://doi.org/10.1016/j.jhazmat.2020.124556. Chrys Ostrander (talk) 07:04, 11 February 2021 (UTC)

"Primary sources should generally not be used for medical content ..." Please see WP:MEDRS for additional info. Thanks JimRenge (talk) 08:14, 11 February 2021 (UTC)

Incomplete information in 'Residues in food products' section

Glyphosate residues in food is a huge issue. It deserves a much more comprehensive treatment than the current entry. I offer a smattering of supporting information. I'm not an expert, but the current entry is woefully inadequate and needs much improvement. For example:

"...A total of 7955 samples of domestic (n = 2122) and imported foods (n = 3622) as well as those of unknown origin (n = 2211 samples; insufficient information on the label to determine country of origin) were collected and analyzed for a range of residues and analytes, including glyphosate and AMPA [aminomethylphosphonic acid, a glyphosate metabolite], between 2015 and 2017. Food samples included a wide variety of fresh and processed fruit and vegetable products, grains (e.g., wheat, corn, oats, barley, buckwheat, and quinoa), beverages, pulses (bean, pea, lentil, chickpea), soy products, infant foods, and ready-to-eat/frozen meals..."

"...Of the samples tested, 3366 samples (42.3%) contained detectable glyphosate residues, and only 46 samples (0.6%) were determined to be noncompliant with applicable Canadian regulations. The noncompliant samples included 15 samples of buckwheat flour/groats (0.11 to 2.1 ppm), 7 samples of whole millet (0.22 to 1.5 ppm), 7 samples of rye flour (0.20 and 5.9 ppm), 4 samples of millet flour (0.11 to 0.46 ppm), 3 samples of chickpea flours (4.2 to 13 ppm), 2 samples of grapefruit juice (0.11 and 0.15 ppm), 2 samples of mixed grains (0.19 and 0.52 ppm), 2 samples of beans (8.4 and 8.6 ppm), 2 samples of fresh mushrooms (0.11 and 0.21 ppm), 1 sample of fresh apples (0.20 ppm), and 1 sample of fresh limes (0.12 ppm). Of these food types, only beans and chickpeas are subject to a 4 ppm MRL; the remaining products are subject to the 0.1 ppm GMRL. The 46 noncompliant samples include 15 organic products and 31 conventionally grown products..."

"Data obtained and analyzed by Radio-Canada show that more than a third of samples tested by the Canadian Food Inspection Agency between 2015 and 2018 contained residues of glyphosate...

"...Between 2015 and 2018, the Canadian Food Inspection Agency (CFIA) tested no less than 12,767 food products of all kinds. We obtained and analyzed these results: 37% of the samples contained glyphosate residues, but only 0.5% of them contain it beyond the allowable limits...

Percentage of samples that contained glyphosate residues: Wheat bran 96%, Pizzas 90%, Wheat flour 88%, Oatmeal 85%, Crackers 84%, Fresh pasta 84%, Canned pasta 83%, Insect products 82%, Cookies 81%, Dry pasta 80%, Rye flour 80%, Oat bran 76%, Oats 75%, 74% pastry mixes, Couscous 73%, 70% chickpea flour, Wheat products 68%, Fruit nectars 67%, Lentils 67%, Bean flour 67%.

The Radio Canada report also cited the following: "...A joint study by Environmental Defense and Équiterre corroborates Health Canada's findings.

"Very common products like hummus, children's cereal, macaroni and cheese, crackers and cookies contained glyphosate residues, but did not exceed allowable limits:

"47% of products made from beans, peas and lentils; "37% of cereal-based products; "32% of children's cereals; "11% of soy products.

In 2018, the Environmental Working Group (EWG) released results of scientific studies it had commissioned on glyphosate residues found in oat-based products.

"Products tested by Anresco Laboratories in San Francisco included 10 samples of different types of General Mills’ Cheerios and 18 samples of different Quaker brand products from PepsiCo, including instant oatmeal, breakfast cereal and snack bars. The highest level of glyphosate found by the lab was 2,837 ppb in Quaker Oatmeal Squares breakfast cereal, nearly 18 times higher than EWG’s children’s health benchmark...

"...The tests detected glyphosate in all 28 samples of products made with conventionally grown oats. All but two of the 28 samples had levels of glyphosate above EWG’s health benchmark of 160 parts per billion, or ppb."

"EWG’s benchmark for total glyphosate consumption is 10 micrograms per day. The benchmark was developed based on a cancer risk assessment for glyphosate and includes an additional tenfold children’s health safety factor. EWG’s benchmark corresponds to a one-in-one-million cancer risk and is significantly lower than both the EPA's dietary adult exposure limit of 70,000 micrograms per day, for an adult weighing approximately 150 pounds, and California’s No Significant Risk Level of 1,100 micrograms per day."

Tests commissioned by EWG by Anresco Laboratories of products purchased between January and March 2020 revealed widespread residues in hummus products."...Of the 43 conventional, or non-organic, chickpea and chickpea-based samples [of hummus] tested, more than 90 percent had detectable levels of glyphosate. Over one-third of the 33 conventional hummus samples exceeded EWG’s health-based benchmark for daily consumption, based on a 60-gram serving of hummus (about four tablespoons). One sample of hummus had nearly 15 times as much glyphosate as EWG’s benchmark, and one of two tests from a sample of conventional dry chickpeas exceeded even the Environmental Protection Agency’s too-permissive legal standard. EWG also tested 12 samples of organic hummus and six samples of organic chickpeas. Most contained glyphosate, but at much lower levels than their conventional counterparts: All but two were below our scientists’ health-based benchmark, although one dry chickpea sample had the highest average level of all our samples. Glyphosate use is not permitted on organic crops, so these samples may have been contaminated by the chemical drifting from nearby conventional crop fields, where it was likely sprayed as a pre-harvest drying agent..."

It should be pointed out that "Federal government standards for pesticides in food are often outdated, not based on the best and most current science. The EPA’s standards for pesticides and other chemicals are also heavily influenced by lobbying from industry."

--Chrys Ostrander (talk) 20:59, 10 February 2021 (UTC)

References

What you have missed out though from reference 1 is Health Canada determined that there was no long-term health risk to Canadian consumers from exposure to the levels of glyphosate found in the samples of a variety of foods surveyed. The high level of compliance (99.4% of samples with the Canadian regulatory limits) and the lack of a health risk for noncompliant samples indicate that, with respect to glyphosates, the food available for sale in Canada is safe. Just because something can be detected does not make it dangerous. The Environmental Working Group is an activist organisation and not a reliable source of health related information. SmartSE (talk) 21:21, 10 February 2021 (UTC)

It looks like the 'Residues in food' section has been updated since my comments but it contains an error that seems to be a gratuitous and erroneous swipe at organic agriculture. It is incorrect, a gross exaggeration of what is contained in the referenced source "Analysis of Glyphosate Residues in Foods from the Canadian Retail Markets between 2015 and 2017," reference #94: "Of the products that exceeded MRLs, one third were organic products." No. Out of 7955 products tested (organic and non-organic combined) "There were 46 noncompliant samples ... There were a total of 1504 products labeled as organic (on the product label or accompanying documentation) that were sampled... The analysis indicated that 75% of the samples of products labeled as organic did not contain detectable glyphosate residues (as compared with 51% for conventionally grown products) ... Fifteen samples [of organic products] contained glyphosate residue levels that exceeded the GMRL of 0.1 ppm (13 buckwheat and 2 millet)..." That calculates to 0.0019% of the total sample set that exceeded MRLs, not one-third!!! The study reports "134 [organic] samples ranged from 5.2 to 99.3% of the applicable MRLs for the given food commodity..." In other words, even those organic products did not exceed the MRLs, although some came close. Really begs the question, WTF? We know that the USDA has fallen down in terms of ensuring that imported organic foods are really organic, especially when it comes to grain crops. Canadian certifiers likely face the same challenges. All of the organic products in the study that exceeded the GMRL of 0.1 ppm were imported grains. The study postulates that some of the blame for the residues on organic products lies on unintentional contamination. "Organic products may have detectable levels, as samples may be contaminated by drift from adjacent farms where pesticides are used, environmental contamination (e.g., pesticides remain in water or soil), or transfer from contaminated trucks, bins, or other devices used in product storage or transport." Chrys Ostrander (talk) 06:31, 11 February 2021 (UTC)

The source says The 46 noncompliant samples include 15 organic products and 31 conventionally grown products 15/46 = one third. You seem to be confusing detectable traces with the MRL and also that the text added to the article refers only to those 46 samples which had residues over the MRL. SmartSE (talk) 16:40, 11 February 2021 (UTC)

Mexico legality status- Semi-protected edit request on 2 March 2021

In the other countries section, we should mention the new Mexico ban that went into effect this year.

https://www.reuters.com/article/us-mexico-herbicide-idUSKCN25902N 174.250.34.12 (talk) 21:24, 2 March 2021 (UTC)

 Done. Volteer1 (talk) 12:18, 3 March 2021 (UTC)

Dewayne Johnson award update Semi-protected edit request on 3 March 2021

The award was reduced to 21.5 million last year [5], this should probably be noted after the first reduction. 2604:3D08:1782:ED00:4A5:9C93:C5F8:DD63 (talk) 06:48, 3 March 2021 (UTC)

 Done. Volteer1 (talk) 12:37, 3 March 2021 (UTC)

GLYCOPHOSPHATE

The correct word is and has always been GLYCOPHOSPHATE and not this garbage made up crap word nonsense you are utilizing here. FIX YOUR STUPID TITLE!!!!!

ALSO DO NOT DONATE TO THIS GARBAGE MARXIST WEBSITE WHICH SPREADS DISINFORMATION!!!! DONATE TO ARCHIVE.ORG which has all the accurate files and where you can proof that this is called wait for it... GLYCO-PHOS-PHATE (for the intellectually challenged of you, who can't pronounce a simple word when they see it!) See dummy below me — Preceding unsigned comment added by 134.41.64.173 (talk) 21:46, 31 August 2020 (UTC) — Preceding unsigned comment added by 108.178.137.210 (talk)

We Marxists at Wikipedia love reliable sources. Can you please provide one to support your claim about the word GLYCOPHOSPHATE? HiLo48 (talk) 09:11, 14 January 2021 (UTC)

Glyphosphate already redirects here. I think it is sometimes listed under that name in the UK. The compound isn't an organophosphate though. We could note it under the 'Other names' section of the chembox?--Project Osprey (talk) 10:18, 14 January 2021 (UTC)

Not quite the "LA Times"...

In Glyphosate § California Office of Environmental Health Hazard Assessment there is this citation:

"Federal judge rules against California's attorney general over..." LA Times. Retrieved 14 August 2018.

I don't think the |newspaper= parameter has been set correctly there; the publication appears to be named "Agweek". (Also, it looks more of a magazine than a newspaper?) I was going to WP:Be bold and change it myself, but the big warning message over the editing box was scary.... —2d37 (talk) 10:14, 6 April 2021 (UTC)

 Done, with this edit. Zefr (talk) 17:12, 6 April 2021 (UTC)

Recent sources

Based on the discussion above, I did a PubMed search to see what the most recent WP:MEDRS-compliant reviews have been saying about glyphosate and non-Hodgkins lymphoma (NHL). I did a search for reviews, systemic reviews, and meta-analyses from 2019–2021, using only "glyphosate" as the search term: [6]. I then looked to find the returns that dealt with lymphoma in humans, excluding purely animal studies and studies of anything else (although there may well be other topics in the search results that could be of interest for this page). I did not find any authored by Séralini, but I do not know if any authors are associates of his, and I do not want to cite any sources that are non-independent in this way.

• There is one that concludes that there is an association of glyphosate with NHL: [7]
• There is one that finds some such association at high exposure levels, but presents it with a lot of nuance: [8].
• There is one that does not conclude that, says that the data is contradictory, and concludes that the safety has yet to be determined: [9].
• There is one that concludes that there is not convincing evidence for carcinogenicity: [10].
• There is also one that discusses the relationship of Monsanto's corporate conduct and regulation: [11].

My initial reaction is that the issue is currently not a settled one, and that we might want to revise the page along those lines. (This is not the same as saying that it does cause NHL.) --Tryptofish (talk) 17:32, 19 May 2021 (UTC)

There has already been extensive discussion about source #1 and it is cited in Glyphosate#Cancer. https://pubmed.ncbi.nlm.nih.gov/29136183/ Is from 2018 but uncited and also examined the same data that #1 used but came to opposite conclusions. I'm not sure that there is much to add to the article though to be honest - we already summarise this in the lead as "the available data is contradictory and far from being convincing" and this position does not seem to have changed much in the last 12 years (since the source in the lead was written). SmartSE (talk) 18:01, 19 May 2021 (UTC)

Thanks: you clearly have a better memory than I do, because I was part of that extensive discussion (woops!). Well, if we look over the most recent literature and come to the same conclusions, then at least it was a worthwhile exercise. --Tryptofish (talk) 18:08, 19 May 2021 (UTC)

I'll carve out a little about those two sources. Basically Zhang et al. is a meta-analysis that included the long-term Agricultural Health Study data or Andreotti et al. for the most recent iteration.

Andreotii et al. is a primary source, so that's why it's uncited, though it was clear on no statistically significant effects. Editor hat off for a second to say the AHS is probably one of the more rigorous pesticide/health related datasets out there if analyzed correctly (though easy to misuse). Zhang et al. included some other studies and came to opposite conclusions in the analysis. I had been uncomfortable about current language in the article because it seems to misrepresent Andreotii by omission even though it mentions that data was included.

I had been sorting through some of these same articles lately (though slowly still). One Tryptofish mentions is Kabat et al., and they address the Andreotii/Zhang train of thought pretty well giving an overview of issues with Zhang and other meta-analyses. I have to reread that one since there is a lot of current info in it. They also cite this article reviewing issues with Zhang, which also led me to this mini-review. I had been checking out who was citing Andreotti/Zhang recently, so I'll mostly keep that focus in this subthread. I was going to bring it up somewhere down the road, but I hadn't really had all my thoughts organized on it yet. KoA (talk) 02:03, 20 May 2021 (UTC)

I just read over Glyphosate#Cancer, and it seems to me that the language there, especially at the beginning of the section, goes farther than that in terms of saying in Wikipedia's voice that there's essentially no evidence. To some extent, it's couched in terms of effects at the dosages that consumers are exposed to, which is OK insofar as it goes, but it may make the literature sound more settled than it really is. --Tryptofish (talk) 18:13, 19 May 2021 (UTC)

I don't have much time tonight, so I'll just mention two main things.

In terms of WP:MEDORG we have a lot a higher-level sources than individual meta-analyses, so overall, there shouldn't be anything significantly changing until they change course. I'm not seeing much change in the meta-analysis suite of citations at the broad overview level, but potentially more to add for where current discussion is.

I had been looking at some of these articles you initially listed, so here's a few quick notes I had written down:

1. Zhang et al. (mentioned in my above comment).
2. Kabat et al. (mentioned above) has a lot of detailed info within I initially found useful for commentary across multiple studies. Not sure if you got a hold of the full text (took me awhile even with university access).
3. Agostini et al. The 1.2. GLY classification section could be useful, though it fairly mirrors what we currently have. 3.1.2. Tumor cells is mostly cell line stuff that's difficult to use. 3.2.1. Cancer lists a a bunch of studies (mostly older), but I wasn't sure what content could be crafted from that at the time. I did copy this quote Currently, most agencies that assessthe risk of exposure to GLY do not consider it carcinogenic to humans,but the consensus among the scientific community regarding GLY's safety has yet to be established. from it as a main summary. The first part makes sense, though I'm not sure how to weight the second statement on overall safety (which goes outside the scope of this conversation a bit). We have a lot of other sources saying other routes of exposure are pretty minimal, and this review was focusing primarily on cell lines without bringing the next steps of higher level research in.
4. Donato et al. was on my to-read list because Kabat et al. discussed it a bit too. They do mention a publication bias issue similar to the previously mentioned Krump article. This seems to be a recurring criticism of the meta-analyses suggesting a positive correlation.
5. Caiati et al. seems to get into some "Monsanto Paper" stuff, which is in the fringe realm. I don't have full access right now, but they cite both Seralini and Seneff extensively as well as lots of similar blogs, so that's a pretty big red flag right now until I can check out what they say in the whole paper.

That all might be unneeded at this point, but that's the gist of what I had written down at the time. I'll try to get up to speed on the overall conversation here in the coming days. KoA (talk) 01:17, 20 May 2021 (UTC)

Possible changes

Just as a tentative suggestion, very much subject to debate, I'd like to workshop, here, possible changes to Glyphosate#Cancer:

Currently The consensus among national pesticide regulatory agencies and scientific organizations is that labeled uses of glyphosate have demonstrated no evidence of human carcinogenicity. The Joint FAO/WHO Meeting on Pesticide Residues (JMPR), the European Commission, the Canadian Pest Management Regulatory Agency, the Australian Pesticides and Veterinary Medicines Authority and the German Federal Institute for Risk Assessment have concluded that there is no evidence that glyphosate poses a carcinogenic or genotoxic risk to humans. The EPA has classified glyphosate as "not likely to be carcinogenic to humans." One international scientific organization, the International Agency for Research on Cancer, classified glyphosate in Group 2A, "probably carcinogenic to humans" in 2015. There is weak evidence human cancer risk might increase as a result of occupational exposure to large amounts of glyphosate, such as agricultural work, but no good evidence of such a risk from home use, such as in domestic gardening. According to a systematic review and meta-analysis published in 2016, when weak statistical associations with cancer have been found, such observations have been attributed to bias and confounding in correlational studies due to workers often being exposed to other known carcinogens. The review reported that studies that show an effect between glyphosate use and non-Hodgkin lymphoma have been criticized for not assessing these factors, underlying quality of studies being reviewed, or whether the relationship is causal rather than only correlational. Writing for the Natural Resources Defense Council environmental advocacy group, Jennifer Sass criticized the influence exerted by Monsanto on research about glyphosate safety, and noted that the review was funded by Monsanto. A meta-analysis published in 2019 looked at whether there was an association between an increased risk of non-Hodgkin lymphoma in humans and high cumulative exposures to glyphosate-based herbicides. The analysis used the most recent update of the Agricultural Health Study cohort published in 2018 and five case-control studies published in 2019. The research found a "compelling link" between exposures to glyphosate-based herbicides and increased risk for non-Hodgkin lymphoma.

Revised As of 2020, the evidence for whether or not long-term exposure to glyphosate can increase the risk of human cancer remains inconclusive. There is weak evidence human cancer risk might increase as a result of occupational exposure to large amounts of glyphosate, such as agricultural work, but no good evidence of such a risk from home use, such as in domestic gardening. According to a systematic review and meta-analysis published in 2016, when weak statistical associations with cancer have been found, such observations have been attributed to bias and confounding in correlational studies due to workers often being exposed to other known carcinogens. The review reported that studies that show an effect between glyphosate use and non-Hodgkin lymphoma have been criticized for not assessing these factors, underlying quality of studies being reviewed, or whether the relationship is causal rather than only correlational. Writing for the Natural Resources Defense Council environmental advocacy group, Jennifer Sass criticized the influence exerted by Monsanto on research about glyphosate safety, and noted that the review was funded by Monsanto. The consensus among national pesticide regulatory agencies and scientific organizations is that labeled uses of glyphosate have demonstrated no evidence of human carcinogenicity. The Joint FAO/WHO Meeting on Pesticide Residues (JMPR), the European Commission, the Canadian Pest Management Regulatory Agency, the Australian Pesticides and Veterinary Medicines Authority and the German Federal Institute for Risk Assessment have concluded that there is no evidence that glyphosate poses a carcinogenic or genotoxic risk to humans. The EPA has classified glyphosate as "not likely to be carcinogenic to humans." One international scientific organization, the International Agency for Research on Cancer, classified glyphosate in Group 2A, "probably carcinogenic to humans" in 2015. A meta-analysis published in 2019 looked at whether there was an association between an increased risk of non-Hodgkin lymphoma in humans and high cumulative exposures to glyphosate-based herbicides. The analysis used the most recent update of the Agricultural Health Study cohort published in 2018 and five case-control studies published in 2019. The research found a "compelling link" between exposures to glyphosate-based herbicides and increased risk for non-Hodgkin lymphoma.

References
References

All that I did was to switch the positions of the first and second paragraphs, add a new first sentence to the new first paragraph, and add a second source to the sentence after it. In this way, I placed the science before the regulatory agencies instead of after, and began with sources that are more current than what we had before. There are no other changes here. The effect is to make the overall tone of the section more tentative than what it was before, which I think is more consistent with the sources, and the lead. --Tryptofish (talk) 19:10, 19 May 2021 (UTC)

Looking back at this, I feel that an extended part of the (revised) first paragraph (currently the second paragraph) spends as awful lot of text on a review from five years ago, only to say that it has been criticized: starting with "According to a systematic review and meta-analysis published in 2016..." and ending with "...and noted that the review was funded by Monsanto." That's the kind of thing that comes across to me as outdated and superfluous. It could be replaced by a single sentence along the lines of A 2021 meta-analysis of glyphosate exposure and non-Hodgkin lymphoma cautioned that results can be biased by "assumptions made about both exposure level and latency period.", and be sourced to this: [12]. --Tryptofish (talk) 21:44, 19 May 2021 (UTC)

Revised (ALT) As of 2020, the evidence for whether or not long-term exposure to glyphosate can increase the risk of human cancer remains inconclusive.As of 2020, the evidence for long-term exposure to glyphosate increasing the risk of human cancer remains inconclusive. There is weak evidence human cancer risk might increase as a result of occupational exposure to large amounts of glyphosate, such as agricultural work, but no good evidence of such a risk from home use, such as in domestic gardening. A meta-analysis published in 2019 looked at whether there was an association between an increased risk of non-Hodgkin lymphoma in humans and high cumulative exposures to glyphosate-based herbicides. The research found a "compelling link" between exposures to glyphosate-based herbicides and increased risk for non-Hodgkin lymphoma. A 2021 meta-analysis of glyphosate exposure and non-Hodgkin lymphoma cautioned that such results can be biased by "assumptions made about both exposure level and latency period." The consensus among national pesticide regulatory agencies and scientific organizations is that labeled uses of glyphosate have demonstrated no evidence of human carcinogenicity. The Joint FAO/WHO Meeting on Pesticide Residues (JMPR), the European Commission, the Canadian Pest Management Regulatory Agency, the Australian Pesticides and Veterinary Medicines Authority and the German Federal Institute for Risk Assessment have concluded that there is no evidence that glyphosate poses a carcinogenic or genotoxic risk to humans. The EPA has classified glyphosate as "not likely to be carcinogenic to humans." One international scientific organization, the International Agency for Research on Cancer, classified glyphosate in Group 2A, "probably carcinogenic to humans" in 2015.

References
References

Taking that a little further, I thought of this alternative. It makes that replacement, and combines the former third paragraph into it (with one sentence of that third paragraph removed). This also puts that former third paragraph into context. We could perhaps either do it, or the version above. --Tryptofish (talk) 22:09, 19 May 2021 (UTC)

Overall, I like the approach you're taking towards refining some of the information in your most recent iteration above. One issue I have is related to your comment of In this way, I placed the science before the regulatory agencies. Regulatory agency or respected scientific organization assessments are generally a higher tier of weight with often more comprehensive scope than a single study rather than organizations and "science" (I'm pretty sure you're implying published journal review literature there) being two distinct things. They aren't really something to separate out though.

So with that said, we can't really have two conflicting paragraphs when the regulatory agencies, etc. haven't really changed course. That's in part why they are listed first. Order isn't the hill I'd die on though, but where I do have concerns is the addition of As of 2020, the evidence for whether or not long-term exposure to glyphosate can increase the risk of human cancer remains inconclusive. sourced to Agostini et al. That creates more uncertainty than what the totality of secondary sources you listed above really mention as I mentioned in an earlier comment, and I do have concerns about the context of Agostini being pulled from.

Looking at Agostini, this is the main relevant quote Currently, most agencies that assess the risk of exposure to GLY do not consider it carcinogenic to humans, but the consensus among the scientific community regarding GLY's safety has yet to be established.. The first part is on cancer, and more or less agrees with other sources that most organizational reviews do not consider glyphosate an elevated risk of carcinogenicity. The latter part is on overall safety though, not cancer, so I'd table that part for now just to keep this conversation focused (the abstract gives a little more context to this though about human cell line experiments).

So as I mentioned earlier, nothing really seems to have changed significantly in the literature other than saying we have more recent studies that can give commentary. The way null vs. alternative hypothesis testing works in safety testing means that either evidence of an effect (e.g., cancer risk) is conclusive or inconclusive, but you can never say a lack of effect is conclusive (null hypothesis). The burden lies on demonstrating a significant conclusive effect rather than the other way around just by nature of how science works (and a pain to try to teach to freshman bio students). I have some ideas I'm still trying to formulate from all the new secondary sources you mentioned as well as some of those I found, but if I was going to tweak your version to be something I'd be content with in terms of the overall sources, I'd just cut the first sentence to As of 2020, the evidence for whether or not long-term exposure to glyphosate can increasing the risk. . . from your current first sentence. A lot of text just to reach that conclusion, right? That at least keeps the language related to inconclusive tight. KoA (talk) 01:48, 22 May 2021 (UTC)

Just briefly (), from Agostini's abstract: While GLY toxicity is clear in human cells, epidemiological studies investigating individuals exposed to different levels of GLY have reported contradictory data. Therefore, based on currently available in vitro and epidemiological data, it is not possible to confirm the complete safety of GLY use, which will require additional comprehensive studies in animal models and humans. Looking at that again, I do note that they are talking about toxicity as opposed to carcinogenicity. However, I'm not aware of contradictory published results in epidemiology except in connection with cancer.

But, that said, your revision of the first sentence is something I can agree with entirely, and I think it's an improvement. --Tryptofish (talk) 19:32, 22 May 2021 (UTC)

I think this has sat here for quite a long time, and I'd like to go ahead and implement this in the most recent form, subject, of course, to further revisions. Does anyone have any objections? --Tryptofish (talk) 18:59, 12 June 2021 (UTC)

 Done: [13]. --Tryptofish (talk) 20:09, 14 June 2021 (UTC)

The Guardian source

I have returned the information from The Guardian because it well within our RS sources guidelines. Gandydancer (talk) 17:52, 12 May 2021 (UTC)

@Gandydancer: The policy KoA referred to was WP:DUE and I am inclined to agree that the content added goes into excessive detail for this article. The 2019 article only has two sentences that refer to glyphosate and is a summary of the 2016 article. It would seem better suited to International Life Sciences Institute, with one or two sentences here briefly summarising it. The detail like mentioning the organisations involved in the study and the precise amount of funding is definitely excessive for this article. SmartSE (talk) 18:39, 12 May 2021 (UTC)

(edit conflict) Gandydancer, please remember WP:ONUS policy, and that is one of the expectations when WP:1RR was imposed here. Content shouldn't be blanket reinserted like that, especially when the issues I brought up were entirely unaddressed.

To reiterate the issue I brought up in the edit summary, The Guardian or really any newspaper is not appropriate for reporting on results of scientific papers. The paper in question was authored by the director of USRTK, an organic industry front group (among all its other fringe components), and I've been consistently clear that author affiliation is a major red flag for usability of scientific papers here. If Bayer/Monsanto had a similarly authored paper that was covered by a few newspapers, the response would still be the same.

Even if you took everything as described at face-value instead, then that's just one industry group making claims about another industry group, and that's really scraping the bottom of the barrel in terms of WP:DUE. If the study's contents are something worth mentioning, leave it to WP:INDEPENDENT uninvolved reliable sources for scientifically published content. Then we're at the point of tailoring something specific for at least some article like International Life Sciences Institute as SmartSE mentioned rather than the mishmash I originally removed for that reason. If there were sources documenting serious issues in the EFSA decision that does affect this article rather than what amounts to standard hand-waving language from a known fringe group, that should be discussed. There's way too much going on here to just curtly say The Guardian is reliable in response. KoA (talk) 19:06, 12 May 2021 (UTC)

• Mostly agree with KoA, Guardian is not an RS for scientific validity, and a lot of this belongs at the ILSI article if anywhere. A tiny part is appropriate to Glyphosate however, and Guardian would be an RS for that: The conflict of interest/funding/other employment question. Invasive Spices (talk) 20:19, 12 May 2021 (UTC)

(edit conflict) If it were any general topic the Guardian can be ok, but the WP:FRINGE aspect of USRTK and uncritical use by the Guardian of them runs into other policy issues outside of just being an RS or not.

One could get into how The Guardian is more of a "use with caution" source in terms of what's DUE in agricultural topics where hot button pseudoscience comes up (e.g., articles like this related to the Seralini affair), but we wouldn't need to delve into that for this conversation really. Instead assuming we did use it, the sub headline from the Guardian we'd paraphrase here would be something to the effect of International Life Sciences Institute used by corporate backers to counter public health policies, says opposing industry front group.

That kind of reads like fluff on its face, so that's why I'd rather see a source pointing out issues on their own rather than mostly taking the stance that here's what this one industry publication has to say. It's basically a protection against two fronts of industry groups with that approach. It gets to be an even more tangled web because we do use the EFSA source in terms of WP:MEDRS nearby. If there was actually something wrong with the findings, we would need to be clear and pointed about that rather than the casting a cloud of doubt regardless of finding validity that fringe groups like USRTK typically do. If there isn't something wrong with the findings, then we cannot distract from the validity of the findings, even tangentially. That's the difference between:

1. "Guy working for industry group X was found omitting studies that should have been included for analysis, and EFSA had no checks in place for industry representatives." vs.
2. "Guy working for industry group X was involved in EFSA's decision process." (with no mention of if industry reps are expected to participate with certain checks required)

The first is generally what we'd be looking at for content at this page if such sources existed. The second can be technically true but easily misleading to readers and sometimes purposely used to vaguely insinuate 1 to readers, which fringe advocacy groups often use as a tactic. I wasn't really intending to unload that much and get too far ahead, but hopefully that gives an idea of all the bullets I'd be trying to dodge in crafting content on this here specifically. KoA (talk) 22:30, 12 May 2021 (UTC)

How about you suggest a replacement text containing the content that you think can be supported by the available evidence? While I agree that fringe sources are not reliable, The Guardian is relatively reliable for the statements it actually makes, which I adhered to closely in what I had added. Do not forget that Monsanto has been known to bribe government officials, and this claim is backed by the US Department of Justice (https://www.justice.gov/archive/opa/pr/2005/January/05_crm_008.htm, also see https://www.nytimes.com/2007/03/07/business/worldbusiness/07iht-monsanto.4824776.html). Note that it is also basic research ethics to disclose conflicts of interest, so it is dishonest for the chair of the 2016 panel (who is vice-president of ILSI) not to disclose the large amount of donations from Monsanto and Croplife International to ILSI. This failure of Monsanto's researchers to disclose secret funding from Monsanto is also well-known (https://www.theguardian.com/environment/2020/mar/12/revealed-monsantos-secret-funding-for-weedkiller-studies-roundup). Lim Wei Quan (talk) 06:40, 15 May 2021 (UTC)

WP:ONUS/WP:BURDEN say otherwise.

I already addressed what is needed for appropriate sourcing. If it's a case of someone or source trying to assert a variation of the Monsanto shill gambit, we generally can't take that on face value because of how often fringe groups hand-wave that as a tactic regardless of validity. Instead, we'd really need sources saying that 1. the person actually had a significant influence (rather than just a designated person to represent industry input), and that something was out of line with the decision. What we have right now in sourcing is just too sparse to give any sense of WP:DUE. KoA (talk) 17:57, 15 May 2021 (UTC)

I have no idea what your motive is. Are you trying to say you do not even believe that Monsanto gave ILSI 1 million to influence the EFSA panel conclusion via Alan Boobis? If you do, I'm asking you to help with including this information. Is it not enough that the parliament withheld funding from EFSA at one point due to precisely that undisclosed conflict of interest, showing that the parliament thought it was out of line? If you do not, why not? Lim Wei Quan (talk) 03:06, 16 May 2021 (UTC)

My motive is to follow Wikipedia policy and guideline. Beyond that, please be mindful of WP:ASPERSIONS. It's been pretty thoroughly explained already the current issues and what is lacking with current sourcing. Because this gets into a WP:FRINGE topic, editing is a bit more strict in this area on all fronts. KoA (talk) 17:51, 16 May 2021 (UTC)

Then why use the term "shill gambit" (which implies that it is false)? I have no problems with strict editing guidelines. What I do not understand is why you keep phrasing your responses in such a way as to cast doubt on the facts of the matter. Lim Wei Quan (talk) 18:43, 16 May 2021 (UTC)

Lim Wei, The Guardian would be just fine if one were to criticize WallMart, BP, Conoco, Nestle, etc., in other words every single giant corporation, except for Monsanto. They get special treatment here. Gandydancer (talk) 20:59, 16 May 2021 (UTC)

Oh, really? By whom? --Tryptofish (talk) 21:01, 16 May 2021 (UTC)

Although I know all the evil things this kind of companies do, and I would not be surprised if there were Monsanto employees watching the wikipedia pages, do you have statistical evidence that Monsanto is getting special treatment here? Lim Wei Quan (talk) 05:48, 17 May 2021 (UTC)

Because the shill gambit is such a common fringe talking point in sources for the GMO/pesticide subject that we're basically stuck as editors trying to sort through the mess that causes in dealing with due weight when dealing with fringe advocacy groups claims vs legitimate COI. Please don't project that muddy water situation as me "casting doubt on facts". Instead, that means the bar is raised in terms of how we can use certain sources, especially popular press, and that we can't overextend sources even if they are trying to insinuate something.

As mentioned previously (which addresses most of your questions), the current sourcing really doesn't have enough meat to craft any content specific for this article. If someone has appropriate sources that legitimately go the next step of questioning the validity of the EFSA finding (related to COI or not) rather than stopping short, that would be worth discussing here. Until that, most of this discussion is moot for this article at least. KoA (talk) 18:35, 17 May 2021 (UTC)

(edit conflict) I should explain that I've been around this topic on Wikipedia (GMOs, pesticides, and the companies associated with them) for a depressingly long time. We've had an ArbCom case, and I was the filing party. So I've spent an awful lot of wiki-time just trying to make WP:NPOV edits but being told or implied to that maybe I'm working under some deep dark agenda. I'm not a Monsanto employee, nor am I in any way influenced by them. And ArbCom determined that it's disruptive and sanctionable to imply otherwise about editors without evidence. (That's where the "shill" language comes from.) So it gets very, very "old" to keep hearing this kind of suggestion, in place of simply saying "I disagree about this edit and here's why." --Tryptofish (talk) 18:53, 17 May 2021 (UTC)

@KoA: Thank you for clarifying. I understand what you are saying about fringe science, and I agree with a high standard for citation. However, your use of the term "the Monsanto shill gambit" in response to my edit is still wrong, because (as per your own cited definition of "shill gambit") this gambit becomes fallacious when conflict of interest has not been demonstrated. In contrast, there is sufficient evidence here for the conflict of interest. Also, if you want a high standard for citation, then you should not support inclusion of that EFSA's panel report in this glyphosate article because it is an unreliable source. Lim Wei Quan (talk) 13:48, 18 May 2021 (UTC)

Because this can be such a sensitive point in discussions on these topics, I feel the need to clarify that I think that the kind of COI we are referring to here is the alleged COI of EFSA in making their decision, not to be confused with WP:COI. --Tryptofish (talk) 18:05, 18 May 2021 (UTC)

Gandydancer, sorry, my friend, but Grauniad plus USRTK is fringe. The Grauniad is 100% beard-and-sandals organic macrame woo friendly.

Full disclosure: I live close to George Monbiot, would happily buy him a pint, and subscribe to the Grauniad. Whether or not I own a beard, sandals, or any combination of the above, is left as an exercise for the reader. Guy (help! - typo?) 21:09, 12 May 2021 (UTC)

• I've thought some more about this, and I think that there are two separate issues that should be treated separately: the degree to which The Guardian is a WP:RS, and the degree to which the disputed edit satisfies WP:DUE. I agree with the point that The Guardian, like most news sources, should be used very cautiously for things of a scientific nature that would fall within the scope of WP:MEDRS. So I wouldn't want to use it as a source for saying something about whether or not glyphosate has certain health effects. But here, the content pertains to alleged corporate misconduct, so the science/MEDRS concerns do not apply. It really boils down to a due weight consideration. This is a subjective matter, but I think it's reasonable to examine the degree to which The Guardian is reporting on the view of an industry critic organization as opposed to something that will be of more-than-passing significance in terms of actual effects on policy. I also think it's reasonable that such material may perhaps be more appropriate to the page about the company, than to the page about the chemical: putting it on this page is like saying in Wikipedia's voice that, because of an accusation against the company, there must be something worrisome about the chemical. And evaluating content that way does not entail some sort of supposed editorial bias in favor of Monsanto – the same would be applied to Walmart, for example, if someone were to propose content about the corporate culture for a page about some product that one can buy there. --Tryptofish (talk) 21:57, 16 May 2021 (UTC)

As Tryptofish suggests, this article is not the most appropriate place for extended content about the International Life Sciences Institute / reported conflicts of interest involving ILSI. The first place for such content would be the International Life Sciences Institute article, or European_Food_Safety_Authority#Criticism. One sentence here on the conflict of interest is appropriate if we have multiple refs which clearly mention a Glyphosate approval-related COI.Dialectric (talk) 01:44, 17 May 2021 (UTC)

One sentence here seems reasonable though I'd say both of the two you mention rather than one or the other.Gandydancer (talk) 02:11, 17 May 2021 (UTC)

@Dialectric: I would much appreciate if you can help with that, because I do not want to cause any trouble with my edits that have been considered unsuitable. Lim Wei Quan (talk) 10:15, 17 May 2021 (UTC)

A much briefer version is definitely worth discussing. As noted by Dialectric, it would need ample sourcing and would have to actually be about something wrong with the approval of glyphosate, as opposed to an unsuccessful attempt to influence that approval. --Tryptofish (talk) 18:53, 17 May 2021 (UTC)

@Tryptofish: I agree that The Guardian cannot be a source for scientific claims, but as you say the claim here is not about the health-effects of glyphosate. However, I am unsure about what you are trying to say regarding "editorial bias in favor of Monsanto". In my opinion, the original phrasing of the article itself (especially in the introductory paragraphs right at the top) strongly implies that it is reasonable to believe that glyphosate in itself is not carcinogenic, by mentioning that the EFSA concluded in November 2015 that it is unlikely to be genotoxic. A lay reader with no scientific training would obviously give significant weight to EFSA's conclusion. But the evidence clearly shows that EFSA was influenced by Monsanto via ILSI. This implies that the current version of the article is actually giving an extremely misleading impression to its readers. To emphasize, those introductory paragraphs are effectively giving Wikipedia's approval to EFSA as being a legitimate source of information regarding glyphosate, and thus implying that their panel conclusion should also be given weight more or less on par with the WHO's classification of glyphosate as probably carcinogenic. So while I of course agree with you that we should not imply something via "ad hominem", it is not at all ad hominem if it is directly related to the veracity of the 'scientific' claims. Does what I am saying make sense to you? I think we have similar views, but I wanted to clarify exactly why I think this COI issue is actually relevant to whether the wikipedia article portrays an accurate picture. Anyway thanks for sharing your thoughts! Lim Wei Quan (talk) 10:15, 17 May 2021 (UTC)

In case it wasn't clear, I'm trying to say that, overall, there is not such an editorial bias. I'd want to be satisfied that EFSA actually was materially influenced in their conclusion, as opposed to there having been an attempt to influence them. Once we get into statements about carcinogenicity, what I'd rather do is take an approach that is not one of qualifying the EFSA conclusion, if there is more recent science that makes it outdated. In other words, if there are recent WP:MEDRS sources that change the current scientific understanding, then the page should reflect that, and that's far more useful than a back-and-forth over earlier science.

For personal reasons, I've pretty much ignored this page for about a year, and am just now paying attention again. Before I left, I was becoming aware of some then-recently published MEDRS-compliant scientific reviews that were re-evaluating glyphosate and non-Hodgkin's lymphoma. I'd have to do some searching to convince myself one way or the other about whether, as of now, the scientific consensus has (or has not) changed. But if it has, then the page should be revised based on that. And that would make any controversy over earlier science more of an historical matter. --Tryptofish (talk) 18:53, 17 May 2021 (UTC)

Well, EFSA definitely was influenced in their conclusion by Monsanto! According to a report for the European parliament, EFSA based its decision on an assessment by the Germany’s Federal Institute for Risk Assessment (BfR), which had actually copy-and-pasted tracts from Monsanto studies; dozens of pages are identical to passages in an application submitted by Monsanto on behalf of the Glyphosate Task Force (GTF)! (https://www.theguardian.com/environment/2019/jan/15/eu-glyphosate-approval-was-based-on-plagiarised-monsanto-text-report-finds) Such plagiarism from Monsanto is undeniable evidence of influence. But of course I agree with you that it would be even better if we can get more recent scientific sources to settle the issue more definitively. I just want to reiterate that in my opinion there is an unfair bias in the article, not bias towards Monsanto per se but rather an inaccurate portrayal of the relative weight of EFSA's conclusions against the classification by the IARC. That is, if we have sources A,B of information that make completely opposite claims, but A is not known to have COI whereas B is known to have many COI infractions in history, it would be improper to just say "A says X. In contrast, B says not X." and just leave it at that. Lim Wei Quan (talk) 13:48, 18 May 2021 (UTC)

This: [14] may perhaps be a better source (also mentioned in the next talk section). --Tryptofish (talk) 19:21, 19 May 2021 (UTC)

If I understand correctly, this content was rejected because of a risk that covering suggestions of problems in the process of EFSA's decision (e.g., people with COI were involved) could lead people (unfamiliar with the inner workings of bodies such as EFSA) to assume problems in the conclusion of that process (e.g., the people with COI unduly influenced the conclusion), even if actually there were procedural safeguards against letting people with COI unduly influence the conclusion. However, it seems to me that such allegations of problems in processes, with and without explicit suggestions that they led to problems in conclusions, will continue floating around elsewhere on the Internet, and being read by people unfamiliar with such processes, even if Wikipedia ignores the allegations.

Would it be better that Wikipedia address such allegations directly? I mean something like "It has been noted that [...allegations...].^[1][2] However, involving industry representatives in decisions that will affect their industries is standard practice, and procedural safeguards exist to prevent such representatives from unduly influencing such decisions.^[3][4]" (or whatever is true verifiable — I certainly am not familiar with such processes).

—2d37 (talk) 08:34, 8 July 2021 (UTC)

I note that Glyphosate already contains some coverage that, at least to my untrained eyes, looks like it might present the same problem of highlighting issues (which might sound like problems) with the process behind a decision without showing that such issues led to problems in the decision itself.

1. sections [...] used by EFSA were copy-pasted from a study done by Monsanto — but did EFSA not verify the claims before publishing them?
2. a scientist advising the IARC had a COI — but did IARC not have measures in place to prevent people with COI from unduly influencing its decisions?

—2d37 (talk) 08:46, 8 July 2021 (UTC)

Section removal

Hi @JimRenge: I think this info certainly belongs in the article. Also I don’t know what “out of context” means in this case. Invasive Spices (talk) 22:56, 8 July 2021 (UTC)

I'm not sure that it is particularly relevant in the grand scheme of things. It's also worth noting that the interaction of Al and glyphosate is a WP:FRINGE theory of Stephanie Seneff about how glyphosate is supposed to cause all manner of diseases: [15] scirp . org/ html/5-3000951_53106.htm (blacklisted) [16]. SmartSE (talk) 09:25, 9 July 2021 (UTC)

Yeah, at least as the section stood, it wasn't particularly informative. If there was relevant non-fringe background on why chelation is ecologically relevant for this chemical rather than it simply being capable of it, then that could be worth fleshing out. KoA (talk) 15:12, 9 July 2021 (UTC)

When I saw the removal, I wondered whether the problem might be that the existing text did not make clear how this aspect of the molecule's chemistry (and of course, the chemistry is relevant to the page) can relate to ecological effects on soil chemistry (perhaps nutrient deficiencies that might affect crops). If that had been the only issue, I would suggest returning the material but making the relevance clearer to the non-scientist reader. However, I'll defer to Smartse and KoA about the fringe aspects of it, as I know less about that than they do, and I certainly agree that we should not include fringe material. --Tryptofish (talk) 18:50, 9 July 2021 (UTC)

• I'm not a chemist but my impression is that chemical characteristics are a normal thing to include in articles, and Stephanie Seneff did not appear anywhere in the removed text. I'm not familiar with the journal (Journal of Inorganic Biochemistry) but it looks good. It's an Elsevier and (if this were other subjects) the IF would be good enough. Invasive Spices (talk) 17:45, 10 July 2021 (UTC)

A vast amount of verifiable information is available for Glyphosate. Wikipedia is written in summary style and the encyclopedia is not an indiscriminate collection of information. An encyclopedic article should present information in context with explanations referenced to academic mainstream sources. Unfortunately we can not ask the editor who introduced the sentence "Glyphosate chelates aluminum." why he thinks that this information is an improvement ([17], [18]). SmartSE ´s research shows that the sentence was probably copied from an article by Anthony Samsel and Stephanie Seneff [19] omitting the context: "Glyphosate also chelates aluminum,[230] and it has been reasoned that this enables aluminum to get past the gut barrier more readily through direct analogy with the situation with arsenic, which is also a 3+ cation.[193]" Invasive Spices, please see WP:NOTEVERYTHING, WP:INDISCRIMINATE, WP:UNDUE and WP:ONUS. JimRenge (talk) 01:02, 11 July 2021 (UTC)

• We're not talking about "everything", we're talking about chemistry information in a chemical article. The continued attempts to connect that person to research (and text here on WP from that research) unconnected to her - neither of which mention her - is... odd. At the most this doesn't seem to belong as its own section unto itself, but should have been in ==Chemistry== instead. Invasive Spices (talk) 18:59, 11 July 2021 (UTC)

When it comes to "everything' or WP:INDISCRIMINATE, there's nothing absurd about this really. That a chemical simply can act as a chelator really doesn't approach WP:DUE in the sense of establishing the importance of that. For an encyclopedia, we're instead asking if this chemical is an important chelator in some system. Is it something that chemists, etc. specifically make a point of, or is it a minute detail that gets thrown in back of a compendium? That there is fringe stuff related to that shouldn't be a distraction, but it does further necessitate the need for the content to really establish why it's important and worth inclusion. That's why I mentioned above the route that could be taken to flesh out such content.

But to address fringe stuff like Seneff more directly, we do take care in similar fringe science subjects, such as climate change denial. That significant fringe proponents exist in the subject often means we take a bit more care in tailoring content, so the carefulness going on here is pretty standard. KoA (talk) 19:20, 11 July 2021 (UTC)