Talk:Glyphosate

IARC

Why is the IARC classification even mentioned? Only the monograph has been published, the data hasn't been released in full. It's also come under heavy criticism for being misleading as it refers to doses which applicators are exposed to (not consumers) and the 2A classification doesn't mean much (plenty of compounds we ingest are 2A - formaldehyde in fruit, caffeine, ibuprofen, etc). Furthermore, 3 other UN agencies are convinced it does not pose a human health risk. You can find× a series of statements from experts worldwide about the IARC classification here and comments from a pesticide expert here. There is also a GLP post here with sources enclosed. 205.193.114.245 (talk) 17:00, 11 August 2015 (UTC)

It happened and it is significant. We discussed it extensively here - please see the archives. We generally don't cite the GLP as it is an advocacy site. Every one here is also aware that the IARC judgement isn't binding on any county's regulations; so far impact has minimal. As for me I am interested to read the full report when it publishes. For now everything is fine and reflects the consensus of the editors here. Jytdog (talk) 17:31, 11 August 2015 (UTC)

Perhaps the wording can be changed to make the situation more clear? Just a brief note on the controversial nature of the ruling, or the insignificance of the classification. Don't use a GLP link, sure, but the editors are aware that the classification is highly controversial and has been inappropriately bandied about by antis? 205.193.114.245 (talk) 19:18, 11 August 2015 (UTC)

We do our best to keep content encyclopedic and not get into the nitty-gritty tit-for-tatting that goes on. I think everybody sane is waiting for the actual report to come out and for the scientific community to react to it. That will probably be the next shoe to drop. That, or the EU's next move (whether they accept the German report as the EU statement on glyphosate or not; they are probably waiting for the rest of the IARC story to unfold as well tho.) Jytdog (talk) 19:51, 11 August 2015 (UTC)

Per WP:MEDORG, statements and syntheses by reputed medical organization like the WHO are considered reliable secondary sources for human health related content. Therefore, it's a reliable source for the statements that it makes. SageRad (talk) 18:31, 30 August 2015 (UTC)

the key thing is that this is one of several, and the only one that takes this stance. Per NPOV we look at all the reliable, recent, sources. The german report is of equal validity and possibly more, b/c they published all their work and all we have from the IARC is the summary published in the Lancet. I am not proposing any changes, just describing how we work with sources. Jytdog (talk) 18:54, 30 August 2015 (UTC)

Actions to remove G

Where belongs content like this Supermarkets and garden centres ban Roundup weedkiller suspected of causing cancer. This appears to be a European afford, but the legal section of the article doesn't exactly covers that kind of removal. prokaryotes (talk) 15:42, 26 August 2015 (UTC)

Prokaryotes, note that there is a whole section on legal status that contains information like this. It's been under development for a while. SageRad (talk) 16:33, 26 August 2015 (UTC)

Recent deletes

I am opening this space for dialogue on two recent edits that removed claims.

Jytdog here removed a section that was sourced to a recent primary article.

I here removed claims that were sourced to very old review articles.

SageRad (talk) 15:23, 30 August 2015 (UTC)

My edit was subsequently reverted here by an editor without an account. I reverted it back. I am quite happy to discuss this with editors here, but an IP address allows no way to notify or contact the user to enter discussion if we're going to use the BRD sort of editing. This talk page entry existed when that revert was made, and the editor in question could have commented here, but did not. We're going to have dialogue with integrity if we're going to do this. SageRad (talk) 16:48, 30 August 2015 (UTC)

I'm fine with these changes. Jytdog (talk) 17:03, 30 August 2015 (UTC)

Thanks, Jytdog. It would be good to do a good search for the most recent review articles on health effects or lack thereof of glyphosate-based formulations. I'll get on that when i find time. SageRad (talk) 17:18, 30 August 2015 (UTC)

Skyring, would you explain your reinsertion of a year 2000 review article as the source for a large claim about human health, as per WP:MEDRS? Thanks. If it's about finding newer sources then i agree, and let's get on that task, but not worry about whether this is in the article until then. If a source is not good for a claim then it's not good for the claim. All your edit comment said was "Restored sourced material" but you ought to have known by my previous edit comment why i deleted it, and could have checked the talk page too. SageRad (talk) 17:28, 30 August 2015 (UTC)

… an IP address allows no way to notify or contact the user to enter discussion if we're going to use the BRD sort of editing … That's ridiculous! Did you even try? --Pete (talk) 17:31, 30 August 2015 (UTC)

Instead of getting contentious, can we talk about the content itself? SageRad (talk) 17:34, 30 August 2015 (UTC)

I'd just like you to acknowledge that IP editors are valuable and welcome contributors to our project. If you don't understand this, you're missing a lot of the point of Wikipedia, "the encyclopaedia that anyone may edit". --Pete (talk) 17:48, 30 August 2015 (UTC)

Skyring/Pete: I second the suggestion to return to discussion of content, not of editors and how Wikipedia works. This is how boondoggles begin. --Tsavage (talk) 18:17, 30 August 2015 (UTC)

Looks like we're back on track with this point. I don't care about this particular content so much, but I certainly care that we're all using the same rules, so as to avoid misunderstanding. There are some editors with years of service who choose not to open an account, and there are those who are fresh to the project, just dipping their toes in the water. I think we treat IP edits more warily than those from people we know, but still, their contributions are welcomed, and if anyone disagrees, they may talk to Jimbo, who started this thing off. --Pete (talk) 03:44, 31 August 2015 (UTC)

I do certainly acknowledge that unregistered users are valuable. It's just that when there is an edit that requires dialogue, they often are not found on the talk page and i didn't think there is a way to notify them. I just left a notice on their talk page in case they want to take part in this dialogue. SageRad (talk) 18:05, 30 August 2015 (UTC)

I thought the edit summary was clear enough. You claimed the ref was out of date but did not update with a newer ref. If the content needs updating that is fine but should not be removed until a new source is used. After all until you have found this new research how do you know it contradicts the current content? 2601:645:C201:3840:6429:FD90:FD31:AF5D (talk) 02:03, 31 August 2015 (UTC)

Thank you for showing up. I will replace the text with a new link now. SageRad (talk) 02:42, 31 August 2015 (UTC)

The old and new should stay, so that we can see how things developed.prokaryotes (talk) 03:13, 31 August 2015 (UTC)

Yobol deleted my addition. I added back my addition, with the two old ones, which are favorable to glyphosate. All the review articles are valuable, and the work that editors did to summarize them is valuable. We need more recent review articles. SageRad (talk) 03:30, 31 August 2015 (UTC)

Yobol, i don't understand why you deleted this review article while you let the other ones stand, which were favorable to glyphosate. SageRad (talk) 03:42, 31 August 2015 (UTC)

You mean the 2000 and 2002 reviews I deleted before you made this message? Yobol (talk) 03:47, 31 August 2015 (UTC)

Toxicity Section Cleanup

Last night, some editing regarding toxicity resulted from crossed messages and misunderstandings, much of which was on my part. For example, when Yobol deleted a section saying that glyphosate wasn't mentioned in the source, i re-added the Shinasi review article here and two other review articles that had been in the passage previously, though Yobol had been referring to another statement and source. Sorry about my misunderstanding here.

Yobol is also correct in that some statements and sources are used in two sections, in the same way.

I think that speaks to an underlying problem with the toxicity section as a whole, and i'd like to ask for ideas on how we can clean it up. The problem as i see it, is that toxicity is divided into three sections: glyphosate itself, glyphosate-based formulations, and additives toxicity. This makes it rather complicated when a source speaks to both glyphosate and glyphosate-based formulations. Do we include it in both section? Should we do away with the sections altogether and have a mass of information on toxicity of glyphosate and formulations? Or perhaps just get rid of the section on additives toxicity and combine that information into the formulation toxicity section? On e more problem: at the same heading level, we have a section called "Government and Organization Positions on Glyphosate Toxicity" which is right below the glyphosate toxicity section. Is that right or should that be a subheading under glyphosate toxicity? And to top it all off, in each section, we have subsections on human, soil, and other animals and ecological toxicity.

So, it's a mess, and perhaps that's ok because the reality is in fact a mess, too. But maybe we can clean it up somewhat, simplify the structure. Suggestions?

SageRad (talk) 12:54, 31 August 2015 (UTC)

I think the structure is a problem; we probably should combine the human toxicities sections in some way, having multiple disjointed sections seems confusing to the reader. (And we probably need to be explicit if we're going to use terms like "formulation toxicity" what the actually means and how that differs from "human toxicity" in the section above it). Yobol (talk) 13:57, 31 August 2015 (UTC)

I have just been editing in the toxicity section and I found the same problems. I eventually decided to be very careful and report only on technical grade glyphosate initially, but then on the formulated forms. I think they should be kept separate as there are different toxicity effects due to different ingredients in the formulated forms.DrChrissy ^(talk) 13:16, 1 September 2015 (UTC)

"Glyphosate can be lethal to non-human mammals."

I recently introduced "Glyphosate can be lethal to non-human mammals." and a similar sentence about fish and aquatic animals. These were almost immediately reverted. I have a feeling this could very quickly develop into an edit war so I am trying to avoid any more ongoing drama here. Should such statements be included? I believe they should. The problem with simply giving an LD50 dose and then summarising by saying that it has low toxicity as that it conveniently covers up that 50% of the animals died. Shock - the substance can be lethal! In my opinion, a clear statement of the fact that the substance can be lethal should be included.DrChrissy ^(talk) 13:28, 1 September 2015 (UTC)

Well, having an LD50 or LC50 value clearly implies that the substance is lethal at some level of acute toxicity. SageRad (talk) 13:42, 1 September 2015 (UTC)

But does the average reader know that? A sentence explaining that was also reverted at the same time, but I have reinserted it.DrChrissy ^(talk) 13:54, 1 September 2015 (UTC)

The dose makes the poison. Water can be lethal to non-human mammals. Air can be lethal to non-human mammals. Fresh fruit can be lethal to non-human mammals. Sitting still can be lethal to non-human mammals. Running can be lethal to non-human mammals. The content explaining LD50 is OK to keep; I won't argue about that. Jytdog (talk) 14:18, 1 September 2015 (UTC)

I hear you on that, Dr Chrissy, but i think it looks good how it stands now. I made one minor edit to specify "acute toxicity" as that is what LD50 and LC50 indicate, as opposed to other sorts of toxicity which can be cumulative and long-term outcomes based. SageRad (talk) 14:33, 1 September 2015 (UTC)

Persistence in water? Access to Journal of Applied Toxicology

My university credential will not allow me to access this journal, oddly. I'd like to read this review paper to update the section on persistence of glyphosate in water. This is a 2014 review and the current statement is based on a very old source.

I also do see this primary research paper that reports half life of 47 to 315 days in seawater in various conditions. This is much longer than the persistence states in the article for pond water. Does anyone have access to the review article and wish to update this section with current state of the science? SageRad (talk) 14:45, 1 September 2015 (UTC)

Toxicity expansion

So DrChrissy on the one hand understands that WP readers are ignorant about toxicity to the point that we need to explain what LD50 is (see above) but at the same time, seems to think that these same readers will be able to make any kind of sense out of content like 'In a study of rats and mice fed diets of containing 0%, 0.3125%, 0.625%, 1.25%, 2.5%, or 5.0% glyphosate for 13 weeks, endocrine effects..." or "Pregnant rats given 3,500 mg/kg/day glyphosate by gavage (stomach tube) on gestation days 6-19 suffered effects including both increased maternal mortality and the number of fetal skeletal abnormalities" (3,500 mg/kg/day is a huge amount of glyphosate). Are we providing readers with baby-step context, or not? It needs to be one or the other.

Also, making major changes to a controversial article is something that is best discussed before implementing. Not necessary, but best, so we don't end up in this place. Jytdog (talk) 15:38, 1 September 2015 (UTC)

In your massive deletion, you also deleted my very carefully worded statement referencing a recent review article by Kimmel et al. of rabbit feeding studies. I would definitely like to reinstate that part. Do you have any objection to it? I think it's clear and useful in that section as it rounds out toxicity to not only be about acute single exposure:

A review of unpublished short-term rabbit feeding studies reported severe toxicity effects at 150 mg/kg/day and NOAEL doses ranging from 50 to 200 mg/kg/day.

References

I do agree about 3500 mg/kg/day being a massive dose, near to the LD50 dose for rats. I would like to ensure that the toxicity section addresses different aspects of toxicity: acute, short and medium term (like 30 to 90 days), and long term exposures. These tend to produce different results and are all required for a complete toxicity picture of a chemical in relation to a population. SageRad (talk) 15:44, 1 September 2015 (UTC)

We should pay attention to weight, and the toxicity section should contain the best and most recent science about toxicity in relation to each class of organisms, and should survey acute as well as short and long term exposure. We should *not* "throw everything at the wall and see what sticks", but should instead do this somewhat deliberately, so that we give a complete but not overly long picture of toxicity in each subsection. SageRad (talk) 15:49, 1 September 2015 (UTC)

I do think the 3,500 mg/kg/day feeding study is extraneous here, and ask Dr Chrissy if it would be ok to remove it. The review i added does speak to toxic effects at much lower levels in short-term feeding studies. SageRad (talk) 16:06, 1 September 2015 (UTC)

I'm not convinced why the content I included should be removed on the basis of your edit. First, one study discusses rats (rodents) and the other uses rabbits (lagamorphs); they have extremely different digestive systems. Second, your review is of unpublished studies, mine is of a published study. Third, the content I inserted gave specific details of the affects, yours is more general. I'm not suggesting your material be removed, I just do not see how it can replace the material I inserted.DrChrissy ^(talk) 16:16, 1 September 2015 (UTC)

Do you think you can find research -- hopefully a review article -- that would report on effects on rats at lower doses? The representivity of the dose level is my main concern here. 3,500 mg/kg/day is so close to the LD50 level for rats that it seems superflous to me to show that there are toxicity effects at short-term exposure at that level. By the way, the review article i listed also speaks to rat feeding studies. Let me take a look and see what that reports. SageRad (talk) 16:27, 1 September 2015 (UTC)

Indeed, table 3 in the Kimmel review shows NOAEL from 300 mg/kg/dayup to >= 1000 mg/kg/day, and reports "At 1000 mg/kg/day and higher, animals in three of the six studies showed signs of lethargy, as well as respiratory and gastrointestinal distress." Deaths also reported in the 3500 mg/kg/day dose. SageRad (talk) 16:32, 1 September 2015 (UTC)

• Comment I also find this dose very strange, limits for humans vary around 10-20mg. However, i did not read the study, but the parts from the article here didn't explained the extremely high dosage.prokaryotes (talk) 16:36, 1 September 2015 (UTC)

Good research SageRad. Regarding high amounts, it is OR to dismiss data unless we can verify why this should be done. The fact is that these doses have been tested (on several occasions) and the reports resulted. We should report these without inferring our own opinion on them.DrChrissy ^(talk) 16:44, 1 September 2015 (UTC)

Maybe the authors elaborate on these amounts, some more context would be nice.prokaryotes (talk) 16:55, 1 September 2015 (UTC)

Ecological relevance of doses is one of the requirements for showing relevance in toxicological literature. I'd be looking comments on how likely specific doses would be reached or in what kind of situations. Kingofaces43 (talk) 17:02, 1 September 2015 (UTC)

The fact sheet does not give further details. Going to the original source will mean we are then using a primary source which several editors on here are opposed to. These doses must be biologically relevant, otherwise, we have scientists killing animals for no reason.DrChrissy ^(talk) 17:05, 1 September 2015 (UTC)

Afraid the assumption that the dose must be relevant can get into OR territory. Lab studies can often pick a range of doses in order to see an effect, and that can often go beyond normal exposure (or sometimes under if poorly designed). If that weren't the case, ecological relevance wouldn't be a criteria in assessing the applicability of toxicological findings. Kingofaces43 (talk) 17:44, 1 September 2015 (UTC)

It is OR to delete data on the basis - "I have a hunch that the doses were too big". Do not forget that the source I am using is a secondary source and is the National Pesticide Information Center. Are you suggesting this is an unreliable source.DrChrissy ^(talk) 17:57, 1 September 2015 (UTC)

Please don't misrepresent my comments DrChrissy. I never commented on the source. I was responding with basic background on toxicological research in response to Prokaryotes and your discussion on the doses mentioned. Sometimes ecological relevance is mentioned in sources, and other times you can be at the other extreme where someone is just doing a lab assay trying to push the dose as high as they can to reach an LD50. It depends on the scope of the study and was intended to help in figuring out what was going on in this specific instance. Kingofaces43 (talk) 23:44, 1 September 2015 (UTC)

DrChrissy I will ask you again, are you writing for experts or the general public? You appeared to start out with WP:TECHNICAL very much in mind, and then turned a corner and started dropping loads of pretty detailed tox study content into the article that takes an expert to make sense of. So who are you writing for? Am not questioning the source which is fine and was already used in the article. Primarily the issues here are WP:TECHNICAL and WP:WEIGHT. Thanks. Jytdog (talk) 19:13, 1 September 2015 (UTC)

Jytdog, i reverted your massive edit and i don't feel at all bad in doing so. How do you think it's ok, at this point where we are actively discussing the toxicity section, and specific references and claims, to completely scramble the whole article and completely delete a subsection, and many references and statements that we've been talking about? You often admonish people to go slower, and now i am asking you to do the same. Please, talk about major changes, as it's clear that we have many eyes on deck here. SageRad (talk) 19:25, 1 September 2015 (UTC)

SageRad please self-revert. This big expansion of the tox section without prior discussion, and adding all this WP:TECHNICAL detail, in the same breath that we are describing things in babytalk, makes no sense. Thanks. Jytdog (talk) 19:43, 1 September 2015 (UTC)

I hear your concerns about oscillating in voice, but i don't think that the wholesale deletion and shuffling of the article is a good way to address it. You even deleted my very simple reference to the 2013 Kimmel review article, and the whole section on effects on non-human animals. Let us please talk together about the big picture of this article and what we want it to look like. I would ask DrChrissy and Prokaryotes and everyone else to take a deep breath and discuss the larger picture, what the article should cover, and with what sort of tone. Let's pretend we're adults sitting around an editorial table, and discuss this like people with a common goal. SageRad (talk) 20:29, 1 September 2015 (UTC)

Well said SageRad. For my part, the changes in technical/non-technical tone might come about because when I move content that has been in the article for some time, I am often reluctant to change it out of respect for the previous editor. That is why the complex sentence occurred. I did not enter this content, I moved it with minor tweaks. If someone wants a simplification of this, a simple message on here or an appropriate tag is fine. I just do not get on with the "delete first and ask questions later" approach. As for the structure of this tox section we appear to have several variables which need to be included. Glyphosate/glyphosate formulations, acute/chronic toxicity, the range of animals studied (how do we divide this vertebrate/invertebrate, major taxonomic groups) and effects (LD/reproduction problems/carcinogenicity/tetrogenicity). Might a table be the best way to deal with this?DrChrissy ^(talk) 20:50, 1 September 2015 (UTC)

I went ahead and fixed the voicing concerns and updated some of the content from sources just to streamline things for now. That was a perfect opportunity to wikilink to describe a term rather than define it in article. I'm not sure at the moment if/how we should rattle off various LD50's, but that's about all I can do tonight. Kingofaces43 (talk) 00:39, 2 September 2015 (UTC)

• sorry gang, was hammered at work today and had no time for this. i see there is a new EWN case. Will check in there. Jytdog (talk) 00:48, 2 September 2015 (UTC)

Understandable as we're all volunteers and can't always respond to each crisis in a sardine can immediately. I'm out for the evening, otherwise I would have removed a lot of the content added per WP:STATUSQUO so it could be discussed from a weight perspective. I'll see what I can do to chime in once morning rolls around. Kingofaces43 (talk) 01:21, 2 September 2015 (UTC)

• DrChrissy I will ask you again. You took into account that people won't know what an LD50 is. So such a person happens to be pregnant, and reads "Pregnant rats given 3,500 mg/kg/day glyphosate by gavage (stomach tube) on gestation days 6-19 suffered effects including both increased maternal mortality and the number of fetal skeletal abnormalities" What will she make of that? (that is a real question). Let me also ask you, what do you make of that data with regard to human toxicity? Thanks. Jytdog (talk) 10:53, 2 September 2015 (UTC)

Don't be daft - you know very well that if I make any comment on humans I will almost certainly be in breach of my topic ban.DrChrissy ^(talk) 11:06, 2 September 2015 (UTC)

Let me follow up on that. Why did you direct the question about human toxicity at me, rather than at other editors or the Talk page in general (that is a real question)?DrChrissy ^(talk) 11:43, 2 September 2015 (UTC)

DrChrissy I directed the question at you, because you, specifically, decided to dramatically expand the tox section. You know as well as I do that we do tox studies on animals in order to inform things like minimum exposure levels for humans. So please speak to that. Jytdog (talk) 17:23, 2 September 2015 (UTC)

We conduct many tox studies to look at the toxicity of a substance in the target species, non-target species, and environmental toxicity as a whole. Why else would we examine the LD50 of glysophate in bees?DrChrissy ^(talk) 17:58, 2 September 2015 (UTC)

Last chance to back away from this. Jytdog (talk) 18:53, 2 September 2015 (UTC)

What is going on here? What is this ultimatum about? I recently added the review article that gave results about toxicity assessments in rabbits. The study was done in order to assess health risks in humans, and it did in fact satisfy WP:MEDRS requirements, being a review article in the relevant field, but i used it in the section on toxicity to non-human animals, as it definitely provided that knowledge and was reliable. Other studies do test toxicity to non-human animals solely for the sake of knowing about the effects on those non-human animals, or to study other biological and biochemical dynamics that are not essentially related to human health but may also provide tangential knowledge that will help our understanding of the chemical in relation to human health down the line. There is not a clear distinction on this question, and why is Jytdog's tone that of an ultimatum here and pressing DrChrissy specifically on this? Is Jytdog insinuating that DrChrissy is presenting data on non-human animal toxicity in a way that incorrectly makes a synthesis to human health effects? If so then please directly state that, Jytdog. I sense innuendo in the language and i would much prefer if people have an issue to plainly state it outright. SageRad (talk) 19:01, 2 September 2015 (UTC)

I too am at a loss as to what I am supposed to back away from. Is this more intimidation?DrChrissy ^(talk) 19:09, 2 September 2015 (UTC)

Great, folks, now the article is locked out for editing except by admins. Let's really try to be adults here and not edit war like this any more. Let's talk and make incremental changes, and if we want to make sweeping changes, definitely talk about it first, knowing that the article is actively edited by several people who care about it. Anyway, i am glad that the current frozen version is at least acceptable, from the most recent changes by Kingofaces43 -- i am glad the most recent edits aren't something outlandish or extreme, and didn't delete much material or seem to be partisan. SageRad (talk) 12:11, 2 September 2015 (UTC)

Sorry, but I can't agree that all the changes are acceptable.DrChrissy ^(talk) 12:19, 2 September 2015 (UTC)

I don't mean that i think it's all perfect, but at least it's not frozen on the massive deletion by Jytdog yesterday. I think it's at least frozen at a reasonable state. SageRad (talk) 12:46, 2 September 2015 (UTC)

OK. Fair enough. I can agree with that. Perhaps we could discuss the structure of the section - I mentioned above that multiple variables will need to be incorporated. Let's take this slowly. I suggest the two major headings at == level should be "Acute Toxicity" and "Chronic Toxicity".DrChrissy ^(talk) 12:52, 2 September 2015 (UTC)

Yes, we can use this time for discussion, which is a benefit of the freeze, i suppose.

I suppose that if we are going to structure it by three levels of variables (glyphosate versus formulation, type of toxicity, and class of organism) then one question is: In what order do we present these variables? To date, the top-level has been glyphosate versus formulation, then organism, and then class of toxicity. You're proposing that class of toxicity should be the top level variable? SageRad (talk) 15:18, 2 September 2015 (UTC)

Don't forget that the top heading at the moment is "Toxicity". I was suggesting this could be replaced with two headings "Acute Toxicity" and "Chronic Toxicity". I am easy about this - I can easily see why we might want to have "Glyphosate" and "Formulation", then "Acute" and "Chronic" within both of these, and then animal groups. In situations such as this (a closed page) I have seen a sandbox set up so that we can edit and discuss without affecting the main article. Should we do this?DrChrissy ^(talk) 15:41, 2 September 2015 (UTC)

Sandbox sounds like a great idea. We have 8 days to play in it. SageRad (talk) 15:52, 2 September 2015 (UTC)

I have created the sandbox here.[1] What I did not realise is that because the page is closed, I could not lift the in-line references and markup, only the text as it appears in the article. So, this might be a bit more difficult than I anticipated.DrChrissy ^(talk) 16:36, 2 September 2015 (UTC)

Thank you, DrChrissy. I found a way to copy the source -- there is a new tab called "View Source" now that "Edit" is gone. SageRad (talk) 16:41, 2 September 2015 (UTC)

The point of page protection is not to continue making edits elsewhere, but to reach consensus on edits by discussion. Just putting that out there for now to make sure the former doesn't preclude the latter. A sandbox is fine for providing example edits for discussion, but it won't necessarily be the consensus version at the end. Just making sure that is clear as page protection begins. I have some ideas for trimming down this new content into a much more concise form that doesn't lose important toxicological information, but I won't be able to revisit this discussion on it until Saturday. Kingofaces43 (talk) 22:24, 2 September 2015 (UTC)

I agree fully with reaching consensus through discussion, but i think the sandbox would give us a way to discuss complex issues more easily by making edits and then asking questions about them. I should be less conflict-causing because the edits are not actually presented to the public. We ought to learn from and talk with each other to reach consensus. I do support talking before making edits even in this sandbox, and that is what we've been doing in the section below. We're starting by discussing overall organization of the section -- the three layers of variables and how to best present them in a concise and even-handed way. SageRad (talk) 01:20, 3 September 2015 (UTC)

Page protected discussion

Seeing that non one has really attempted discussion since the lock-down, I'll get the ball rolling now that I'm more available for a bit. The contested content is the content added in the section changed from Fish and Amphibians to Non-human mammals.[2]. Various content was added on LD50s, and that was opposed on a WP:TECHNICAL and WP:WEIGHT basis. There was some explanatory language on what an LD50 was, but that has been fixed in the most current version,[3] so those concerns shouldn't need to be addressed here. Kingofaces43 (talk) 17:19, 5 September 2015 (UTC)

I agree with Jytdog above that rattling off various LD50s is too technical and unneeded for encyclopedic content. In pesticide articles when we're talking about intended use, we don't start listing what various concentrations are used on account of WP:NOTHOWTO in additional to TECHNICAL. We have a similar parallel here where readers aren't particularly gaining knowledge on the topic with inclusion of LD50 values. WP:NOTJOURNAL also comes to mind as we aren't summarizing the literature as a journal does, but rather describing what's relevant for encyclopedia readers. Common practice in pesticide articles is to list the overall findings of particular sources for readers. If a reader is very toxicological minded, they are a bit outside the scope of the intended audience here and would be expected to dig into the sources if they want specific LD50 values. There are also WP:MEDRS concerns with reporting animal studies used as proxies for human toxicology.

1. With that all in mind, looking at this diff,[4] there are a few easy ways to fix the concerns with the newly added content. The first two paragraphs can be reduced to the conclusion statements from the sources by a summarization is in line with both sources originally used where the more technical information is in the osurce for whoever wants to read it more in-depth:

Mammalian toxicity is considered to be low to very low toxicity due to low LD_50s; signs of glyphosate toxicity in animals typically appear within 30 minutes to 2 hours following ingestion of a large enough dose, and include initial excitability and tachycardia, ataxia, depression and bradycardia but severe cases can develop into collapse and convulsions.

2. The carcinogenicity content in the third paragraph is probably best left out for now due to WP:WEIGHT we have multiple high quality sources saying there is not a significant general concern for carcinogenicity, so it's probably best to leave the current events to unfold in the position statements section first. These are rat studies used for human health aspects per WP:MEDRS. If it were more of a ecological impact study I'd be more open to it here, but it's very squarely on medical lab rat testing.

3. The first paragraph of the aquatic life section can be made concise like the mammals section. Just say:

For freshwater fish, glyphosate is generally considered "slightly toxic to practically non-toxic."

4. The second paragraph of wood frogs can probably go. It's not a toxicological effect (belonging in the formulation section if it was), and it looks like a misread of the source as it's listed as an indirect behavior effect. A really interesting lab study (equivalent to in vitro cautions with medical research) when you dig into it, but not really a good fit for this section.

5. The third algae-related paragraph (another lab study from the same lab in the previous paragraph) also doesn't seem to have a good fit with the toxicology section, but rather in a broader ecological effects section.

6. Likewise, the fourth paragraph on persistence belongs in the environmental fate section.

This should take care of any WP:JARGON or weight issues for this section. Kingofaces43 (talk) 17:20, 5 September 2015 (UTC)

Kingofaces43, note that it is not true that noone has attempted to discuss this section since the lockdown. I have done so, at length, and invited you and Jytdog to do so as well. DrChrissy and i have indeed been editing in the sandbox extensively and asked for the input of others. I hope that you can recognize that for you to say this after all these days, during which you were present in Wikipedia, is kind of not cool, or else you just didn't notice, even though i pinged you about it, so you ought to have noticed. I hope you will work with the sandbox version that we've put so much effort into.

As for your comments, on a brief skim, i have to say that i do think that a few LD50 or LC50 values are appropriate in each section. That gives an idea of the relative potency of the chemical or the formulation in relation to the type of organism. Having the hyperlink to the Lethal Dose article gives the reader an easy way to understand what this means, if they are unaware. SageRad (talk) 18:30, 5 September 2015 (UTC)

I also agree the LD50 information is extremely important - this helps us understand how glyphosate may have different adverse effects at different levels of the wildlife food chain. This is why I have very recently included LD50 data on aquatic invertebrates (potentially the lowest food-chain level and therefore arguably the most important level).DrChrissy ^(talk) 18:36, 5 September 2015 (UTC)

No one was really focusing on a discussion outlining what exactly needed to be worked on, and that was part of my concern with the sandbox approach where people would just go ahead and start adding whatever they want. At this point, can the above edit proposals be considered acceptable for a minimum amount of text? That was the intent to give something instead of the wholesale removal that was originally going on. If someone wants more, we can just keep ratcheting the content more as discussion goes on, but let's at least agree on a minimum first. Kingofaces43 (talk) 02:24, 7 September 2015 (UTC)

Let's play in the sandbox

DrChrissy created a sandbox copy of the article that we can use to work out ideas while the article is frozen. I created this new talk page section to talk about it. SageRad (talk) 16:44, 2 September 2015 (UTC)

Great! Welldone! We learn something everyday!DrChrissy ^(talk) 16:59, 2 September 2015 (UTC)

I've been thinking about this section, and i propose that we organize the section still in a similar way to how it already was. I think the top-level division should be glyphosate technical versus glyphosate-based formulations. Under that, different classes of organisms. Under that, the same pattern of reporting acute toxicities and then long-term toxicities. The organization seems simple and logical to me. We can eliminate the small section on the adjuvants and include that information in the GBF section, as that section mainly treats the adjuvant toxicity anyway. We can include the adjuvant-only information as an aid in de-aggregating the effects of glyphosate versus the adjuvants.

Overall, i think we should strive for clarity, concision, and an even-handed voice that speaks in a good summary level of detail, not going into any one point too much, and using links more than explaining terms like NOAEL and LD50.

It can be done. I would like to invite the proactive participation of Jytdog and Kingofaces43 here, especially, to help in the principles-based discussion on re-framing the section. Please explain how you would like to see the section turn out, in terms of purpose and organization. Thank you. SageRad (talk) 01:30, 3 September 2015 (UTC)

By the way, check out my edit in The Sandbox, where i organized the toxicity section according to the principles i outlined above. Note how i have reduced the number of section headings and the complexity, by merging the adjuvants section with the Glyphosate-Based Formulations section. Note i've renamed that section, to the most common term used in the literature. SageRad (talk) 01:43, 3 September 2015 (UTC)

I did a lot more editing to make the section more streamlined and logical, and added a figure on LC50 for a fish. I expect that when the article is re-opened, this sandbox will be useful for editing the section. SageRad (talk) 13:06, 3 September 2015 (UTC)

I want to state it plainly that i expect that when the article is open for editing again, the sandbox version of the toxicity section would be a good candidate for wholesale copy and paste into the main article. We have this time, and we have this process available to us to work things out in the sandbox, and through discussion here. I hope that those who have raised the most objections will take part in the process of reworking the section, so that there is indeed consensus when the article is reopened for editing.

I have done a lot of work already on the section. I think i have done it in a very NPOV fashion. I'm not pushing any agenda, and seek only to streamline and better organize the section. It is a lot of work, and i think the section already shows quite a marked improvement. I would like to also focus on the content as well as the organization -- for example, making sure we have the most up-to-date sources on various aspects of toxicity, as well as filling out some details in some sections that are anemic. SageRad (talk) 13:26, 3 September 2015 (UTC)

I see that DrChrissy has been working on the toxicity section in the sandbox, and i am thankful for the work. SageRad (talk) 11:20, 4 September 2015 (UTC)

And thank you back!DrChrissy ^(talk) 13:05, 4 September 2015 (UTC)

User:SageRad, we are approaching the time when the block will be lifted on this page. Between us, we have made significant beneficial changes in sandbox version. However, we have received little input from others. Please could you ping others requesting they look at the changes and make comments/changes where appropriate. I would do this myself, but I am banned from Jytdogs' Talk page. Thanks in advance.DrChrissy ^(talk) 14:41, 5 September 2015 (UTC)

Jytdog and Kingofaces43 and Prokaryotes and Jusdafax and Yobol, DrChrissy and i have made extensive changes in the sandbox to the toxicity section, and by our estimation have improved the section greatly. We hope to do a copy and paste when the editing is open again. We have invited Jytdog and Kingofaces43 to help with this or at least to provide input in the text, and to date have seen no input. The invitation was made, though, in the above talk page text. We've put in a lot of work. There may be small quibbles here or there, but overall i think the organization is greatly improved, and the content is much more well presented now. Please take a look and provide any input. SageRad (talk) 18:25, 5 September 2015 (UTC)

I cleaned it up some. But it remains a mess. For instance the reproductive toxicity section is still presented as though anyone gives a rat's ass about the effects on rat reproduction. That data is gathered to help understand possible human toxicity. This requires a lot more work, to make it a useful and meaningful section on toxicity. Jytdog (talk) 18:56, 5 September 2015 (UTC)

Jytdog, what is the section you refer to as "the reproductive toxicity section"? There is no section named that. And secondly, who gives a "rat's ass" about effects of a chemical on reproductive health? A lot of people do. What section are you speaking of, what particular content, and why would is it not useful to a reader? SageRad (talk) 11:59, 6 September 2015 (UTC)

@SageRad: I agree with Jytdog that there are significant issues with undue use of primary sources there. There also appears to be significant duplication of toxicity sources in both human toxicity sections, which needs to be rectified. Yobol (talk) 00:15, 6 September 2015 (UTC)

What can we do to keep the two human toxicity sections from repeating themselves? Yobol (talk) 00:49, 6 September 2015 (UTC)

User talk:Yobol please could you explain what you mean by "undue use of primary sources" and which PAGs you are using for this.DrChrissy ^(talk) 11:29, 6 September 2015 (UTC)

Indeed, there will be an issue if editors are insisting that no primary sources can be used in this section. I personally think it's better to use secondary sources, and to use them well, but that primary sources are sometimes useful and relevant to the content. That is also what Wikipedia guidelines state. I can see an argument about level of detail and relevance, as well as the reliability and generalizability of conclusions of primary sources being lesser than secondary sources. But i also see an issue in the section of bias, in that among secondary sources, it seems that some people like to select the most benign (in the sense of showing little harm) secondary sources, and then to present them in the way that makes them seem most benign at that. In other words, i see a sort of systemic bias in the use of secondary sources, as if this is a courtroom and there is an advocacy for the premise that glyphosate and GBFs are safe and harmless. Yes, let us prefer secondary sources, but let us use them fairly and without bias, so we use the latest couple of reviews on a topic and then present them fairly (not cherrypicking conclusions that seem to show safety). SageRad (talk) 11:35, 6 September 2015 (UTC)

I also like to know why Yobol removed 9 studies, i looked at the first which has 215 cites, thus appears credible - not undue. prokaryotes (talk) 11:36, 6 September 2015 (UTC)

Primary sources can be used, with care. In this case I am following WP:SCIRS. Put another way, a google scholar search for "glyphosate toxicity" brings up >30,000 sources. If we include one primary study, there would be a WP:WEIGHT problem if we do not include ALL primary studies (at that point, editors are cherry-picking specific primary studies to highlight). Clearly our toxicity section cannot include >30000 sources, so we need to find the most relevant information to include, which would be from secondary sources like books and reviews. I agree that we need to include all high quality secondary sources, summarize them appropriately and weight them in accordance to their prominence in the literature. Haphazard (or deliberate) placement of specific primary studies is not helpful. Yobol (talk) 13:37, 6 September 2015 (UTC)

We go by how many cites a study received, and 215 is a lot. prokaryotes (talk) 14:35, 6 September 2015 (UTC)

Indeed, secondary sources help us to organize what is included in the section, to prioritize what to report. Still, even with secondary sources, it is possible for them to be cherrypicked, and presented in a way that favors an undue conclusion as to the safety, or to the harm, of glyphosate. I see it go both ways, and i hate it. Like a secondary source can have conclusions that are complex, and show some danger, and makes several good observations about toxicity, and then an editor with an industry bias will summarize it as "... report that there is very little toxicity in normal use patterns" or some such 7-word summary that completely omits more complex statements about potential harms, etc. I think it takes some stepping back to see one's own agenda, and then to make a conscious effort to really read a study and gain the sense of what it truly says, and then summarize it accurately. Let's avoid cherrypicking, both on the level of which study to include, as well as what conclusions to report from the study. SageRad (talk) 15:46, 6 September 2015 (UTC)

No, we absolutely do not go by how many cites a study received...Andrew Wakefield's pile of manure "study" has been cited 2163 per Google Scholar (an extraordinarily high number), but that does not make it any less a pile of dung. I have no problem not cherry-picking of high quality secondary studies, and of course agree that we need to neutrally represent the entirety of a source rather than specific portions that might. Trying to involve primary studies, however, will necessarily involve cherry picking specific studies (as we cannot cite all of them), and should be a non-starter. Yobol (talk) 17:44, 6 September 2015 (UTC)

Recent edits of mine that have been deleted include content on aquatic invertebrates. We decided at Talk that we should include content for the major animal taxa. I have researched these and using RS, inserted content with in-line references.User:Yobol why have you deleted this?DrChrissy ^(talk) 18:00, 6 September 2015 (UTC)

We should include material related to anything found in high quality secondary sources. Are you saying there are no secondary sources, at all, describing the effects of glyphosate on aquatic invertebrates? Yobol (talk) 18:05, 6 September 2015 (UTC)

I have not found any. Have you? If you have, please insert these. If not, perhaps a suitable tag is less disruptive than deleting the content.DrChrissy ^(talk) 18:15, 6 September 2015 (UTC)

If material isn't important enough to have been found in a secondary source, it probably does not deserve any WP:WEIGHT here as this means the overall scientific community has not noticed its contents. Yobol (talk) 18:17, 6 September 2015 (UTC)

WP:Weight says nothing of the kind.DrChrissy ^(talk) 20:07, 6 September 2015 (UTC)

Well or course, WP:WEIGHT is not supposed to spell out in excruciating detail every situation imaginable. In a universe where there are literally thousands of primary sources, how do we know which pieces of information is important (and which are wrong, unimportant, etc)? Well, the ones the scientific community finds important and has filtered for us - the ones in secondary sources. How do we build a good neutral article? We choose high quality sources (such as recent review articles, textbooks, etc) and we summarize them, being careful to give due weight to their prominence of each argument in that literature. How do we build stilted and non-neutral articles? We try to jam in irrelevant primary sources that may or may not have any lasting significance because we as editors decide that it is important. Yobol (talk) 22:02, 6 September 2015 (UTC)

Again, cites are a good indicator for importance. prokaryotes (talk) 22:07, 6 September 2015 (UTC)

If it's so important and well cited, then it shouldn't be a problem finding a quality secondary source citing it and using that source, because the secondary source will place the primary source in context to other primary sources. Yobol (talk) 22:11, 6 September 2015 (UTC)

It depends, but most often you want the primary study, when citing from it. Secondary sources are useful when it comes to reviews of the literature, and again to have some sort of impact measure.prokaryotes (talk) 22:28, 6 September 2015 (UTC)

That, of course, is the point. We want to give our readers high quality information, that is synthesized by the scientific community to place those primary sources in context and to let us know that they are important. Especially important primary studies can and are routinely discussed, but we know they are important because secondary sources say they are, not because individual editors feel they should be. Yobol (talk) 22:33, 6 September 2015 (UTC)

• Repeated discussion This discussion is basically the continuation from hereprokaryotes (talk) 22:28, 6 September 2015 (UTC)
• I agree. This is disruptive and will be reported if it is not stopped.DrChrissy ^(talk)

I'm sorry, what is disruptive? My good faith efforts in explaining my views on how to improve this page? Do you think labeling other editors' explanations as "disruptive" is any way to move forward in a collaborative environment here on Wikipedia? Yobol (talk) 23:12, 6 September 2015 (UTC)

As indicated above by Prokaryotes, this is a continuation. It is disruptive to bring up issues that have been discussed before.DrChrissy ^(talk) 00:00, 7 September 2015 (UTC)

Wait a minute, I'm supposed to know what has been discussed before when I haven't participated in those discussions, and I'm not allowed to say anything similar to what anyone else has said before, or it's "disruptive"? Yeah, go ahead and "report" that anywhere you want and see how fast the WP:BOOMERANG flies. Yobol (talk) 00:05, 7 September 2015 (UTC)

stucture

I set up the structure of separate tox sections for glyphosate alone, additives alone, and formulations, as these can be, and have been, studied separately. The new content mushes them back together somewhat - there is now content in the glyphosate-alone section that is about formulations. I think this structure is useful, but before I fix this, i wanted to ask -- should we maintain this structure? Jytdog (talk) 14:32, 7 September 2015 (UTC)

To be honest, there's been so much editing in the sandbox that's it's hard to make heads or tails of what's being added/removed. Could you link a diff to be sure of what you're referring? Kingofaces43 (talk) 15:08, 7 September 2015 (UTC)

It is all over the place in the glyphosate alone section. What I am asking here is, if people are OK with the divisions that exist in the article and in the sandbox. Jytdog (talk) 16:00, 7 September 2015 (UTC)

I am happy with the structure in the sandbox version, although I am in favour of having/re-instating a separate "additives" section in Toxicity. Yes, there is now content in the glyphosate-alone section that is about formulations, the content just seems to flow better that way. However, if they can be separated, please go ahead.DrChrissy ^(talk) 16:28, 7 September 2015 (UTC)

Oh crap the additive toxicity section went away. That is not good. POEA is a problematic actor in formulations and that needs to be discussed. DrChrissy you again added a bunch of content based on primary sources. argh. Can you please use reviews? Jytdog (talk) 16:45, 7 September 2015 (UTC)

Contentious material/changes need to be discussed with consensus before it is added. I still would like to see a reason why we have such duplication regarding the human toxicities; it just looks weird having the same material in two different places twice, and I wonder if we could adjust one section or another to not have basically multiple repeated sentences. Yobol (talk) 16:58, 7 September 2015 (UTC)

Yobol, I set up the separate sections because the toxicity of glyphosate alone, is very different from the formulations. Formulations contain POEA and other agents that make them really toxic. The content in our article has muddied that some and the sandbox is a bit worse on that front. Jytdog (talk) 17:03, 7 September 2015 (UTC)

Note that i have edited the toxicity section in the sandbox version to change the structure, and i went into detail about what i did and why here on the talk page. I agree with Jytdog on the need for separate sections for glyphosate and formulations, of course. In the sandbox version, i removed the "additives only" section and combined it with the section on full formulations, as that seemed to make a lot of sense and reduce the complexity. To do that, the lede of the formulation toxicity section contains text that explains the distinction and mentions POEA among other things. I also moved the statements of organizations on toxicity to follow the reporting sections on toxicity. If there is material about formulation toxicity in the glyphosate-only section then that is a mistake in categorization that can be corrected. I do think that additive toxicity can be discussed well in the formulation toxicity section, however, and having 2 instead of 3 toxicity subsections is an improvement in simplifying. SageRad (talk) 16:20, 14 September 2015 (UTC)

Primary sources and reviews

Jytdog and Yobol are both stating that we should not be using primary sources in this article but instead using reviews. Which PAGs are you using for these statements please?DrChrissy ^(talk) 17:08, 7 September 2015 (UTC)

I said we should not be giving undue WP:WEIGHT to primary studies that have not been evaluated in the scientific literature through a secondary source (often a review, but university textbooks and the like would also be good sources), per my discussion above in this thread which took place yesterday, which you abandoned abruptly when you called my explanations "disruptive". As I noted previously, there are literally thousands of primary sources which could theoretically be used in this article, and deciding what information is important to include in this article will require the use of secondary sources. If multiple high quality secondary sources single out particularly important primary studies, we can of course include a discussion of those primary studies then. Since our goal is not to use specific sources, but to find high quality sources to determine what to put in the article, your insistence on using likely inferior sources is puzzling. Yobol (talk) 17:16, 7 September 2015 (UTC)

Every policy and guideline says that we should use secondary sources and urges us away from using primary sources:

• While WP:OR allows primary sources to be used, it is "only with care, because it is easy to misuse them";
• WP:NPOV says "Neutrality assigns weight to viewpoints in proportion to their prominence. However, when reputable sources contradict one another and are relatively equal in prominence, describe both approaches and work for balance. This involves describing the opposing views clearly, drawing on secondary or tertiary sources that describe the disagreement from a disinterested viewpoint."
• WP:VERIFY, in a section called "Original Research", says "Base articles largely on reliable secondary sources. While primary sources are appropriate in some cases, relying on them can be problematic. For more information, see the Primary, secondary, and tertiary sources section of the NOR policy, and the Misuse of primary sources section of the BLP policy." Jytdog (talk) 18:01, 7 September 2015 (UTC)

Arbcom, requests for cases

A request for an Arbcom [5] case and a AE request to apply pseudoscience discretionary sanctions [6] have been filed that may affect this article. All editors wishing to make a comment should visit the pages linked to. AlbinoFerret 17:07, 8 September 2015 (UTC)

Seralini review

Page protection went away, and right away we are back at it. Everymorning just added the following, in this dif which I reverted and have pasted below for discussion.

A 2015 review found that glyphosate may be toxic below the lowest-observed-adverse-effect level that has been assigned to it by regulators, and that its effects may include "teratogenic, tumorigenic and hepatorenal effects."

References

Introducing a review by Seralini et al, in this way, was not a good thing to do, on several levels. We need to discuss whether to mention this at all, and if so, how, given Seralini's very clear record of producing work that most of the scientific community rejects, as described in Seralini affair. As this article is among several currently being considered as part of an Arbcom case referenced above, I suggest we all hold back from making bold/significant edits until that is resolved. Jytdog (talk) 16:35, 9 September 2015 (UTC)

Per WP:V, reliable sources need to have a reputation for fact checking. Seralini certainly does not have that. We should not use it. Yobol (talk) 03:37, 11 September 2015 (UTC)

You are not in the position to question the author or the study, based on your personal assessment. Because a previous study involving one of the study authors was questioned doesn't mean we censor. prokaryotes (talk) 03:45, 11 September 2015 (UTC)

Actually, you will note that most of the review's authors are the same as the authors of the disputed Seralini study. They certainly do not have a reputation for fact checking, having been widely criticized by the scientific community. We generally avoid authors who have a poor history of scholarship. Yobol (talk) 03:58, 11 September 2015 (UTC)

Their paper has been republished and there are no indications that these authors have a poor history of scholarship (as you put it). prokaryotes (talk) 09:15, 14 September 2015 (UTC)

There is indeed a lot of critique of Seralini out in the ether. I have not personally evaluated it, though i know that there is much criticism of his work, though that seems to be expected toward anyone whose research challenges the GMO industry in general. Another research group has also been criticized lately in the "skeptic blogosphere" if you will, for the following paper:

Bøhn, Thomas, et al. "Compositional differences in soybeans on the market: Glyphosate accumulates in Roundup Ready GM soybeans." Food Chemistry 153 (2014): 207-215.

In this instance, i have indeed read the paper, and read some of the critique in blogs, and i have to say that the critiques i have seen are quite weak and agenda-oriented. The principle i would like to invoke is that presence of critique of a paper or research group does not necessarily invalidate the paper or research group as being a source of information worthy of inclusion in an article. It may alternatively indicate that there exists a motivation amongst a group of people to discredit the paper or research group, for agenda-based reasons. Likewise, there is a paper cited in this article that i believe would be quite the equivalent of a Seralini review but in the pro-industry direction. In other words, strong hints of agenda-based slanting of presentation and analysis of evidence. That paper is this review article that has been used to support several claims in this article, i believe:

Duke, Stephen O., and Stephen B. Powles. "Glyphosate: a once‐in‐a‐century herbicide." Pest management science 64.4 (2008): 319-325.

While some critique Seralini as having it out for glyphosate, others critique that paper as being a hagiographic brochure for glyphosate. It goes both ways, and agenda is agenda. It's not symmetrical, however, and we must be aware of the sociological dimensions of power/knowledge dynamics in a context in which there are deep economic interests vested in a certain reading of reality by science. SageRad (talk) 13:57, 14 September 2015 (UTC)

thanks for your thoughts sagerad. this is a bit of an apples and oranges thing. the discussion about PMID 24491722 (the "Compositional differences in soybeans on the market") is a primary source which we generally steer clear of regardless of what they say. PMID 26282372 (the seralini paper that was added and is the subject of this thread) is a review - a secondary source - so clears that hurdle. The discussion about it, is a) whether it is reliable and b) if we use it, how, given Seralini's publication history. PMID 18273882 (The Duke paper you cite) is also a review. (btw the reason we use PMIDs all the time, is pubmed classifies papers as to whether they are reviews or not - looking at the pubmed abstract is the quickest way to classify a source as primary or secondary. Jytdog (talk) 14:08, 14 September 2015 (UTC)

Yes, indeed, i am aware of that distinction and how to use PubMed to find out what is a review versus a primary research article. However, that's not so much material to my analogy. It's more about the principle that just the presence of critique of a paper does not negate the paper's potential use in a Wikipedia article as a source. We must also evaluate the soundness of such critique. There is much chatter in the blogosphere. Of course, Seralini has been a lightning rod and there has been much written about the history of his work and papers, but this in itself is not a sole reason to exclude his work. It is a reason to be cautious. Where there is smoke there is often fire, but where there is smoke there is sometimes a smoke bomb and not fire. SageRad (talk) 14:37, 14 September 2015 (UTC)

Yes indeed. I have been hammered over the head that RS should be considered on a case-by-case basis. Testing whether other WP editors adhere to this landed me a topic ban. We must remember here that the article is published in an Elsevier journal. Are we really trying to say that Elsevier are (again) publishing sub-standard work?DrChrissy ^(talk) 16:35, 14 September 2015 (UTC)

Per WP:SOURCE, there are several components that go into the reliability of the source - the type of source, the author, and the publisher. I don't think anyone feels Elsevier is a bad publisher, I think my argument is that the authors (Seralini and coauthors) has been widely critiqued (for instance, by highly reliable sources such as the EFSA) for their previous work in this area in particular, so they do not have the reputation for fact checking needed to be considered reliable. Again, we generally avoid authors who have a poor reputation for scholarship (which should be obvious without being stated), so this source should not be used. Yobol (talk) 17:02, 14 September 2015 (UTC)

Do you not think that a publication house, journal, editor-in-chief and peer reviewers would have all been extra careful before publishing this? It is not for WP editors to judge this article unless concern has been expressed elsewhere and can be verified.DrChrissy ^(talk) 17:38, 14 September 2015 (UTC)

(e/c)Haven't the slightest clue if anyone paid any extra attention, nor does it have any relevance to the discussion of whether the author has a reputation for fact checking. We can, and do, judge sources all the time on criteria to see if they are reliable. You seem focused on only one of the pillars of what makes a source reliable (the publisher) while I am focusing on another (the authors). It is clear that the authors do not have a reputation for fact checking, having been widely criticised in the scientific community on the very topic of the review in question (glyphosate toxicity). Again, we generally avoid authors with reputation for poor scholarship/fact checking. This is per WP:V our core policy on sourcing material. I am baffled by the insistence on using a source authored by a person with a poor reputation for fact checking. Yobol (talk) 18:02, 14 September 2015 (UTC)

The controversy about the Seralini paper and Seralini group makes it clear that one must take special care to evaluate the reliability of the source. However, the very existence of controversy does not, in itself, provide enough reason to exclude a source. It is clear to me in a brief review of the "Seralini affair" as it's called, that there was an industry-centered push against his publication. The fact that the EFSA criticizes it does not in itself mean that it is unreliable. The EFSA is also questionable, to me, as a source on this topic, given the recent EFSA/BfR report on glyphosate, written in close cooperation with the Glyphosate Task Force, an industry group that exists to promote interests of merchants of glyphosate and related HT crops. In other words, there is a sociology of vested interests at work here. An analogy would be a case wherein Wikipedia editors may wonder about including a paper by Michael Mann on the topic of climate change. Mann has been voluminously criticized for his work in climate science that supports the notion that human-caused climate change is real. He has been criticized by people allied with the fossil fuel industry, which has a vested interest in a certain scientific conclusion regarding climate change. In fact, there was something analogous to the "Seralini affair" called the Hockey Stick Controversy that is a little further back in the past than the Seralini affair, that may be instructive in this instance. In summary, my main point is that there are indeed vested interests in the real world that do generate the appearance of controversy perhaps moreso than pure unbiased science might. I haven't personally delved into the whole "Seralini affair" enough to have an opinion on it, but the very fact of a controversy does not automatically exclude a paper of research group completely. SageRad (talk) 17:49, 14 September 2015 (UTC)

This is not a "controvery" (which smacks of the dubious point scoring of the Teach the Controversy movement which basically tries to make both sides of a "controversy" equally valid), it is a straight forward assessment that high quality sources, and that the scientific community by and large have found Seralini's work dubious. We should not be using sources created by authors with poor reputation for fact checking (and I find it odd to come here appearing to argue Seralini is reliable without looking in-depth at the Seralini affair to see the opposition his research has found in the scientific community). (As an aside, Both sides of the GMO lobbying wars have tried to recruit scientists, so attempts at stating only one side is using dubious practices would appear to be a somewhat naive position.) Yobol (talk) 18:02, 14 September 2015 (UTC)

You misrepresent my position. I did not say that Seralini is reliable. I simply said that existence of a controversy does not automatically mean that he is not reliable. I also did not engage in a "Teach the Controversy" fallacy. SageRad (talk) 18:08, 14 September 2015 (UTC)

Yobol, you are also misrepresenting my position. There is no insistence on my part that we use this source. I am arguing that we should be looking at sources on a case-by-case basis and not automatically excluding them on previous publication history. My point above is that the paper has gone through a peer-review process which will have involved fact checking.DrChrissy ^(talk) 18:34, 14 September 2015 (UTC)

Except that no one is saying that it should be excluded because of a "controversy", it should be excluded because the author has previously been widely criticized by the scientific community on the very subject they are now writing on and therefore this source (due to its authorship) does not have a "reputation for fact checking" required by our core policy on how to determine reliable sources. It has been pointed out that the publisher is only one of several independent factors that goes into the reliability of a source, per WP:SOURCE, and continually bringing up the publisher when no one is objecting on the grounds of the publisher (when the authorship is the point of contention) does not appear to be a fruitful avenue to pursue. Yobol (talk) 18:45, 14 September 2015 (UTC)

What I am saying is that if science authors do not fact check appropriately, this is usually picked up by the editorial team and peer reviewers, and the paper will not be published until this is attended to. This means that there are in fact, at least two levels of fact checking in published science article. In this particular case, given the history of the authors, I am sure that both the authors and the publishers would have increased their efforts in fact checking to avoid any further controversy. This means WP should be careful about using the source, but it does not mean it should be automatically excluded.DrChrissy ^(talk) 19:03, 14 September 2015 (UTC)

You will note that these authors' initial paper that generated the controvery was published, and subsequently was retracted after the scientific community had a chance to read it. The fact that it was published the first time does not make it reliable by the mere fact of publication, clearly, since it was subsequently retracted. Peer reveiew is not magic pixie dust that makes a source reliable. An editorial process is necessary for a source to be considered reliable (thus requiring it to be published), but it does not address the issues with the poor reputation for fact checking of the authors. Again, you are making an argument about the publisher, when the issue is with the reputation of the authors. The authors do not have a good repuation for scholarship or for fact checking, and therefore should not be used here. Yobol (talk) 19:12, 14 September 2015 (UTC)

And we are not here to sprinkle WP POV dust. I can agree with almost everything you say above, but I would replace the last 4 words with "....be used carefully here".DrChrissy ^(talk) 19:29, 14 September 2015 (UTC)

This is not "careful" use of the source, it is using the source as if it is as reliable as every other source in this article, which it is not. It should not be used in such a fashion, as it is not reliable for use as such. I am still baffled by the insistence that we use this source at all, given its authorship deficiencies and the presence of numerous higher quality sources. Yobol (talk) 19:33, 14 September 2015 (UTC)

Note that according to PubMed, it seems to be the only general toxicology-related review on the topic in English since Mink (2011). SageRad (talk) 19:43, 14 September 2015 (UTC)

• Some quick thoughts. :First, it seems like nobody is arguing that we should treat the Seralini review like would would any other. That is a good thing. We are on pretty uncharted ground, with regard to how exactly to treat it. which comes down to two questions; the "Weight" question - how much (if any) to give content based on this source, in this article; and the attribution question (how exactly do we contextualize any content we do generate from this source). Both of those are hard. I do suggest folks hold off on working them out until the arbcom case is done. What comes out of that case will likely be helpful to resolving this issue and others. Jytdog (talk) 19:47, 14 September 2015 (UTC)

(e/c)Yobol, if you are aware of higher quality sources, why not share these? That would be the collegiate approach.DrChrissy ^(talk) 19:48, 14 September 2015 (UTC)

I count 9 sources already used in this article that seem to be high quality secondary sources regarding human health and glyphosate. None of them appear to have the issue that the Seralini review has regarding the authorship. Why we would want include the Seralini review as if it is on par with these other sources continues to baffle. Yobol (talk) 19:54, 14 September 2015 (UTC)

The paper discusses toxicity in non-human animals - do you have any better quality sources for content on this subject-matter.DrChrissy ^(talk) 20:10, 14 September 2015 (UTC)

page protection

i just requested full protection for this article again. This kind of aggressive editing, while this is about to go before Arbcom. Crazy. Jytdog (talk) 21:18, 14 September 2015 (UTC)

Please, folks, especially DrChrissy and Jytdog -- let's slow down, let's not get another page protection slapped on here, and let's discuss -- in GOOD FAITH -- intended edits.

Indeed, we did a LOT of work in the sandbox, and there was ample opportunity to discuss all the way along. I did a lot of work in there. I made some very good edits, and so did others. Let us not throw those out.

However, on the other hand, let us seriously take a few hours to pose the question about how to use that work in the sandbox, and let's get some agreement if possible about importing it to the live page.

Let's not edit war, and let's talk. Please.

To make any progress, we will need to talk here in good faith. We need to show up and do the work. We need to speak truly and with honesty, and not wikilawyer. We need to work for the good of the article, and not for the benefit of any point of view. SageRad (talk) 21:30, 14 September 2015 (UTC)

I am acting in good faith. I was assuming that the lack of comments for several days meant we were all happy with this.DrChrissy ^(talk) 21:31, 14 September 2015 (UTC)

(edit conflict)DrChrissy, deciding that the discussion is done and dumping your sandbox into the article as you did here is disruptive. Please don't do that. There are several open questions about that material, per the discussion(s) above. Jytdog (talk) 21:32, 14 September 2015 (UTC)

(edit conflict)Jytdog, I reverted your deletion of a large amount of text, including what appear to be sourced material. I take no position on whether it should remain or not, but many people have warned you about mass deletions and the need to discuss matters first. Please do so. GregJackP Boomer! 21:38, 14 September 2015 (UTC)

GregJackP you evidence no understanding at all of the context here. We will just need to work this out at Arbcom. Jytdog (talk) 21:42, 14 September 2015 (UTC)

Arbcom doesn't decide, deliberate, or rule on content disputes, only conduct. Viriditas (talk) 22:56, 14 September 2015 (UTC)

Yes and the conduct here is making it impossible to work out content. Jytdog (talk) 22:58, 14 September 2015 (UTC)

Well, we've done it again, folks. Page protected until 28th of September. And the toxicity section is a complete mess. The outline, in itself, is a complete mess. Back to the sandbox. Anyway, the structure i would like to see is to begin with glyphosate toxicity, then glyphosate-based formulations toxicity, then the section on organizational statements on toxicity of glyphosate. And the section on formulation toxicity can cover the material formerly covered in adjuvants toxicity. So... any issues with this? SageRad (talk) 23:08, 14 September 2015 (UTC)