Talk:Glyphosate/Archive 11

Archive 5

←

Archive 9

Archive 10

Archive 11

Archive 12

Archive 13

→

Archive 15

RFC: Appropriate use of NON-WP:MEDRS primary study.

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

---

per WP:SCIRS and WP:VNT, is an appropriately framed mention of this NON-WP:MEDRS primary study (Mailin Gaupp-Berghausen, Martin Hofer, Boris Rewald, Johann G. Zaller. Glyphosate-based herbicides reduce the activity and reproduction of earthworms and lead to increased soil nutrient concentrations. Scientific Reports 5, Article number: 12886 (August 2015)doi:10.1038/srep12886), recently published by Nature.com acceptable? Semitransgenic _talk. 23:02, 5 December 2015 (UTC)

• Comment – the only way to appropriately frame this source is to state that its design is seriously flawed and hence no conclusions can be drawn from it. Boghog (talk) 20:45, 5 December 2015 (UTC)

on that blog, the author states, "I’m speculating here, but my guess is...I’m not an earthworm expert, but I would guess.....," there also appears to be a COI (Arysta LifeScience, BASF, Bayer CropScience, Dow AgroSciences, DuPont, FMC, Hatch Act Funds – USDA, Loveland Industries, Monsanto, NovaSource, Repar Corporation, StateLine Bean Cooperative, Syngenta, amongst others, are listed as funders). I see a disclaimer, but why, as Wikipedia editors, should we perceive this POV to be a more reliable source than Nature.com? Is it an acceptable reason to exclude the study Semitransgenic _talk. 22:44, 5 December 2015 (UTC)

Yes, they should. The blog points out an obvious problem with the study, the lack of an essential control (where the plants are killed without use of herbicide). Studies lacking appropriate controls are worthless. Boghog (talk) 23:12, 5 December 2015 (UTC)

Is it "common sense" to understand about controls, in/appropriate controls, etc, or is this a scientific editor behaving as a reviewer and showing OR?DrChrissy ^(talk) 23:21, 5 December 2015 (UTC)

The OR policy is not an excuse to turn your brain off. As I have already stated above, prohibitions against OR only apply to Wikipedia articles, not to talk pages. It is appropriate to use OR on talk pages to judge the appropriatness of a source. And it doesn't take a expert to realize that this particular study did not rule out an obvious alternative explanation (worms are not happy when you take away their food source). Boghog (talk) 23:43, 5 December 2015 (UTC)

I'm not sure I accept your reasoning, this is not a WP:MEDRS matter. There is nothing to say primary sources cannot be used if the source is reliable, which it appears to be. WP:SCIRS suggests that "primary sources should be used when discussing a particular result...When citing a primary source, be especially mindful of the policy on undue weight...An individual primary source should never be cited or juxtaposed so as to "debunk" or contradict the conclusions of a reliable secondary source...," accordingly, I don't see such an issue mentioning the findings in an appropriate fashion. Semitransgenic _talk. 00:12, 6 December 2015 (UTC)

I never claimed this is a MEDRS issue and I am not rejecting the study because it is primary. I am rejecting the studying because it is so fundamentally flawed. It is appropriate for editors to discuss the quality of sources on the talk pages and to apply common sense in deciding whether a particular source is reliable. This one clearly is not. Boghog (talk) 07:35, 6 December 2015 (UTC)

What some blog author thinks or doesn't think about some scientific publication in nature.com is completely irrewlevant. If one wants to point out potential weaknesses or problems of that study it normally requires a properly published scientific review or at very least a statement by a reputable scientist. Generally I see no problem with using a study published in Nature in the article on glyphosate and I don't see any WP:MEDRS as we're not dealing with a human health issue or human health claims.--Kmhkmh (talk) 00:44, 6 December 2015 (UTC)

Blogs authored by established experts may be considered reliable. The author of this blog post is a reputable scientist that works in the field. Finally no one has claimed this is a MEDRS issue. Boghog (talk) 07:48, 6 December 2015 (UTC)

The author of that blog is a pesticide promoter with a strong POV. SageRad (talk) 08:40, 6 December 2015 (UTC)

He is still qualified to comment on the quality of the study and his criticism of the study are spot on. It is not OK to run a study without proper experimental controls. Boghog (talk) 09:17, 6 December 2015 (UTC)

granted he may be qualified, but doesn't the stated COI perhaps undermine our confidence here? Wouldn't it be more appropriate for you to present a WP:RS that refutes the methodology of this study? Lacking said secondary refutation, it would seem that given the source of the study is WP:RS, saying, "according to a recent study, by x et al (date), published in x, such and such" would be admissible under WP:VNT (if undue weight is not given, and it is not framed in such a manner that it reads like a refutation of another source). Semitransgenic _talk. 13:46, 6 December 2015 (UTC)

We should not be adding material based sources whose experimental methods are seriously flawed. Andrew Kniss' blog supports the doubt, but we don't need his blog post to come to that conclusion. We can independently come to the same conclusion. How do you know that the negative effects glyphosate on earthworms was not simply to killing plants rather than a direct action of the glyphosate on earthworms? Why didn't the study include an appropriate control for plant killing? How can the authors possibly justify their conclusions based on a study without proper controls? Once a source is determined to be reliable, OR policy dictates that we should simply state the conclusions of reliable sources. The OR policy does not prohibit applying common sense and editorial judgment to determine whether a source is reliable and appropriate to begin with. Boghog (talk) 14:27, 6 December 2015 (UTC)

Wikipedia lacks the necessary expertise to ascertain the degree to which this study might be flawed. For an editor to state it is flawed, based on his/her speculative analysis, is WP:OR. Kniss's blog does not constitute a WP:RS, and, his speculative conclusion, might actually be flawed (especially in light of his COI). Nature.com is WP:RS, and per WP:VNT, appropriate mention of the study is justifiable. On balance, I see little harm in including it. Semitransgenic _talk. 15:51, 6 December 2015 (UTC)

I have already stated this several times, but the message is not getting through. OR policy does not apply to talk pages nor to evaluating the reliability of sources. Wikipedia does rely on common sense and editorial judgement to evaluate the reliability of sources. There is no question that the study is flawed and we do have the required expertise right here to make that determination. A study which lacks appropriate controls is worthless. Basing articles on worthless sources damages Wikipedia. Here is a simple, straightforward question that anyone should be able to comprehend and to answer: How do you know that the negative effects glyphosate on earthworms observed in the study was not simply to killing plants rather than a direct action of the glyphosate on earthworms? Please answer the question. Boghog (talk) 16:02, 6 December 2015 (UTC)

unfortunately, owing to my lack of expertise in the areas of weed biology, ecology, and Lumbricus terrestris, it would be inappropriate for me to offer a speculative opinion on this matter. Semitransgenic _talk. 16:25, 6 December 2015 (UTC)

But surely you understand for the need for scientific controls. Boghog (talk) 16:27, 6 December 2015 (UTC)

Affirmative, however, what I understand is irrelevant, it's not my role to speculate. Under existing guidelines, the inclusion of the source is legitimate if it is worded correctly. Semitransgenic _talk. 16:37, 6 December 2015 (UTC)

E/C @Boghog Please could you state who you believe has the required expertise and how you have come to this conclusion. I will assume anyone you do not mention means you do believe they do not have the required expertise.DrChrissy ^(talk) 16:32, 6 December 2015 (UTC)

Anyone that understands the importance of experimental controls (I and believe that includes everyone here) has the required expertise. However some here don't want to apply that expertise because it leads to conclusions contrary to to the position that they have already staked out. Boghog (talk) 16:50, 6 December 2015 (UTC)

So how do we distinguish between experts choosing not to use their knowledge, and experts who are not really experts?DrChrissy ^(talk) 18:57, 6 December 2015 (UTC)

As it has always been done, by consensus. This thread has gone on long enough. We can avoid all these problems by insisting on secondary sources. Boghog (talk) 19:32, 6 December 2015 (UTC)

Basically, we have two chocies. Either critically evaluate primary sources ourselves, or wait for secondary sources to do the evaluation for us. The later is simpler and much more reliable. Boghog (talk) 06:42, 7 December 2015 (UTC)

• Comment. It's slightly odd for this RfC to say the source was published by "nature.com" - while this is technicaly true it sort of implies this is a Nature source, when in fact the journal is Scientific Reports, a rather less prestigious open access journal and just one of the many publications within the Nature stable (and one which has published some rather dodgy stuff in my view). Alexbrn (talk) 08:48, 6 December 2015 (UTC)
• Do not use the source. Wikipedia is meant to reflect accepted knowledge, and generally that is found in high quality secondary sources. Of course sometimes primary sources are useful for painting a fuller picture, but they should generally be in line with the base accepted knowledge established by stronger sources. In this case, it is right that editors exercise their judgement in assessing the proposed source, and in my view there is sufficient doubt over its quality that it would be better for the Project if it was ignored. We can always revisit this information if and when better sources on this aspect of the topic appear. Alexbrn (talk) 06:41, 6 December 2015 (UTC)
• Yes, it is acceptable. Should we use it? I don't know. That's an editorial decision, but it's completely acceptable by the guidelines of Wikipedia. If it is used, then it must be represented accurately, to state that it was glyphosate-based formulations that were tested, and not glyphosate alone, and that the formulations applied included Roundup Speed, which contains another active ingredient (pelargonic acid). That should go without saying that sources must be represented accurately. SageRad (talk) 08:41, 6 December 2015 (UTC)
• Use secondary sources. We can either spend infinite time arguing (or warring), source by source, point by point, over each change, or we can adopt a convention that allows everybody to get on with things. The waste of energy on this (talk) page is incredible. The best chance we have to make a better WP is to go with secondary sources on controversial articles. Trying to figure out whether a given primary source is acceptable is OR on its face. We already know that one group of editors will celebrate any anti-GMO/anti-pesticide piece and attack any pro- piece, while the other side will do the reverse. The details of any of these disputes are stupefyingly boring in their repetitiveness. Use your heads, folks. Life is short! Lfstevens (talk) 15:48, 6 December 2015 (UTC)

perhaps you have a point, but I fail to see why such staunch resistance to a tiny mention of this study exists when it is perfectly clear that it can be included under existing guidelines. It's the principal of the matter, there is actually a middle ground between including it, and suppressing it outright, but the culture here is so hostile, that no one is interested in reaching a compromise; which is, as far as I'm concerned, worth highlighting, even if it is "stupefyingly boring" for onlookers. Semitransgenic _talk. 16:00, 6 December 2015 (UTC)

You say I have a point, but then you immediately go back to your position. The problem with this subject is that it is so inflamed that we have entire WP articles discussing the validity of individual studies. We know that there are study authors/groups that are pushing political agendas (on both sides). We are not competent to adjudicate their reliability. I have successfully added material on a primary anti-pesticide study (actually I got Jytdog to do so). I now regret that. I apologize for the "hostile" climate. Everyone involved with this topic is feeling abused, including me. Thank you for sticking with it. I still hope we can find a modus vivendi, even though nobody is pleased. Lfstevens (talk) 16:55, 6 December 2015 (UTC)

I thought the point of an RfC was to get fresh views (which is why I commented). But I agree, in a vexed topic surely if a source looks even a bit iffy the safest thing to do is to avoid. Surely there are enough strong sources on this topic to write a decent article!? Alexbrn (talk) 17:04, 6 December 2015 (UTC)

Comment: IMO, to favor practical discussion, some form of proposed wording should be included in this RfC.

Meanwhile, the overarching issue here is that, from what I've seen in our debates over science and medical content, the fallback position presented as the voice of middle ground reason, that we should generally choose to exclude all but the most conservative review sources, would seem to introduce a serious, systemic bias in our coverage, in that the most "reliable" positions on modern technologies are continually qualified and often reversed or otherwise brought into question - at times within a few years, methods with the approval of "reliably mainstream" review are discounted, products are withdrawn, the probably safe is found significantly unsafe. Therefore, to suggest that we regularly avoid ongoing coverage that continues to develop a particular topic from what we determine is its last most stable current position, would seem to be contrary to the role of a modern, digital, uncensored, real-time-edited general encyclopedia, which should (and is expected to) provide the most up-to-date, balanced and verifiable information available.

As difficult as reasoned, balanced coverage of newer findings in the primary research area may be to accomplish in our situation - an editorial method without appointed expert arbiters, using an unruly consensus-seeking mechanism applied case by case - inclusive coverage is exactly what Wikipedia is built on, and this is clearly spelled out in our core policies. While those policies remain, we must be open to all reliably sourced information, broadly speaking, and strive to reconcile it with fairly balanced coverage, no matter how inconvenient and frustrating that way at times may be. --Tsavage (talk) 17:37, 6 December 2015 (UTC)

• Yes, this particular source qualify as WP:RS, whereas http://weedcontrolfreaks.com (used to disprove the source) is questionable. This is very simple. I agree with SageRad. Writing essays is not the way to undo policies, such as WP:RS and WP:NPOV. My very best wishes (talk) 18:46, 6 December 2015 (UTC)

@My very best wishes: – you have got to be kidding. Did you read the blog or are you applying a rote formula (blogs automatically bad, peer review automatically good)? Please reread the blog very carefully. The problem with the paper is that it has not used appropriate controls and therefore no conclusion can be drawn from it. A really, really bad paper. Boghog (talk) 19:28, 6 December 2015 (UTC)

The problem with the blog is that it is a blog. My reading of WP:RS is that blogs are the lowest of the low in terms of reliability. Please drop this. You are just wrong.DrChrissy ^(talk) 19:51, 6 December 2015 (UTC)

Self-published expert sources may be considered reliable when produced by an established expert on the subject matter, whose work in the relevant field has previously been published by reliable third-party publications WP:BLOG. What we have in this case is an established expert in the relevant field. At the very least, it raises serious questions about the primary source which we need to take into the account. Boghog (talk) 20:05, 6 December 2015 (UTC)

Are you talking about this source? It was linked in the RfC question above. It looks like an article by Nature.com to me. My very best wishes (talk) 20:08, 6 December 2015 (UTC)

The journal in question is not Nature (journal), but Scientific Reports which is part of the Nature group of journals. This particular journal has been rather controversial as it has accept additional fees for expedited review and several of the journal editors have resigned over this controversy. But that is besides the point. Even in the highest quality journals, questionable papers are sometimes published. That is what happened in this case. Please carefully read the blog. It is written by an established expert in the field and he makes a number of excellent points. Boghog (talk) 20:18, 6 December 2015 (UTC)

So, you are trying to disprove something published by RS based on a blog post by expert. OK, except that I do not really know who that expert is. My very best wishes (talk) 20:25, 6 December 2015 (UTC)

The author is Andrew R. Kniss. Again, please read the blog. The study left out a critical control. Not all peer reivewed articles are reliable. Boghog (talk) 20:36, 6 December 2015 (UTC)

I wish to acknowledge some valid points made by Boghog. (1) We can discuss things on talk pages in a frank manner and that in itself doesn't cross the WP:OR guideline which applies to articles. (2) The study is indeed not on glyphosate alone, but on glyphosate-based formulations, and includes Roundup Speed, which has a second active ingredient besides glyphosate, and GBFs also include some rather potent and often biocidal ingredients like POEA which are not even considered "active ingredients" by the common usage, so if the content is included, it must make this clear. (3) Nature Scientific Reports is not Nature itself, yet it does have a high impact factor and editorial process with integrity. (4) Andrew Kniss has some valid points. However, that said, we must also recognize that this is not Andrew Kniss' field of expertise, and that his critique is not peer-reviewed. Also, on the point of "no control" that depends on the question the researchers are asking. If the question is, "What is the effect of these glyphosate-based formulations on earthworms in a small ecosystem that is similar to that of agricultural fields?" then they did indeed have a control. It is clear that the death of the plants in the experimental group was part of the causal network. This was an ecological question. Kniss and Boghog are assuming that the question is only "What's the effect of the glyphosate-based formulations, all other things being held equal?" whereas it seems the researchers' question was "What's the effect of the glyphosate-based formulations on worms in a small ecosystem that simulates that of a typical agricultural field?" So, i hope this may help to untangle the knot here. I think the question of how glyphosate-based formulations affect worms is a relevant question. I'm not sure if this study is the ideal one to answer the question, but it seems like it could be useful to readers, as long as it is described in full with the caveats mentioned. SageRad (talk) 23:35, 6 December 2015 (UTC)

Use secondary sources The seriously flawed Gaupp-Berghausen 2015 study is an excellent example of pitfalls of using primary sources. While it is debatable whether the blog is reliable or not, the blog has become an important part of the process of checking the reliability of publications. These doubts will be taken into account when review articles are written. I agree with Lfstevens. We can spend endless unproductive time debating the merits of various primary sources. It is much safer and more efficient to rely on high quality secondary sources. A suprisingly high percentage of scientific work cannot be repeated. It takes time for other scientists to digest, reflect, and to write review articles that undergo a second round of peer review. The result of this process is much higher certainty about the results. Boghog (talk) 19:57, 6 December 2015 (UTC)

I am not familiar with this subject, but can see what you are talking about [1]. This paper itself looks to me as a scholarly source, and I think it should be treated as such, unless the paper has been officially retracted... My very best wishes (talk) 20:25, 6 December 2015 (UTC)

It does appear scholarly, but serious question have been raised about its reliability. In any case, it would be much better to wait for a secondary source that reviews this work. Boghog (talk) 20:40, 6 December 2015 (UTC)

Sure, the secondary sources are better, but I do not understand why this is so hotly debated, given that negative impacts of pesticides are generally well known. Telling that it has no any negative environmental impact would be an "extraordinary claim".My very best wishes (talk) 20:57, 6 December 2015 (UTC)

• This is an herbicide, not a pesticide. No one claimed this particular herbicide is 100% safe, but the available reviews indicate it is relatively safe. Because this safety claim is back up by reliable secondary sources, it is not at all extraordinary. Boghog (talk) 21:12, 6 December 2015 (UTC)
• It is also important to point out that the primary source directly contradicts a older secondary source:

Giesy JP, Dobson S, Solomon KR (2000). "Ecotoxicological Risk Assessment for Roundup® Herbicide". Reviews of Environmental Contamination and Toxicology. Reviews of Environmental Contamination and Toxicology. 167: 35–120. doi:10.1007/978-1-4612-1156-3_2. ISBN 978-0-387-95102-7.{{cite journal}}: CS1 maint: multiple names: authors list (link)

and repeating what Smartse said above: This secondary source from 2010 summarises the reviews and is clear about the low toxicity of glyphosate to earthworms. Boghog (talk) 21:31, 6 December 2015 (UTC)

Sure, these reviews qualify as secondary RS. But it does mean anyone should eliminate other well sourced materials about harmful effects of the herbicide - as in "bee section" just below - I agree here with DrChrissy and Montanabw. My very best wishes (talk) 02:15, 7 December 2015 (UTC)

You will at least acknowledge that because the Gaupp-Berghausen 2015 primary source contradicts the conclusions of the best avaiable secondary sources, that it is not appropriate to cite this primary source? Also it is important to note that this is a highly controversial topic both outside (see for example the Séralini affair) and inside Wikipedia (see GMO arbitration request and it generally is agreed that Glyphosate is a Core GMO article), it is best to stick to high quality secondary sources. Because of the tempation to cherry pick, primary sources can easily be abused. Everything becomes toxic, if dosed high enough. The critical question is glyphosate toxic at concentrations that are found in environment. This is a difficult question to answer and it is best to wait for secondary sources to analyze and integrate the avaiable data. Boghog (talk) 06:34, 7 December 2015 (UTC)

What secondary sources are you talking about? If you are talking about the blog, then no. If you are talking about reviews, then it matters that the original paper has been published later than reviews, so that could be a new scientific finding in the paper. If reviews refer to an original publication X, one should use reviews. If they do not, one should use publication X. I do not have time for this, but quick look at Séralini affair shows one of the problems. It tells "genetically modified maize, as well as glyphosate". GMO and herbicide(s) are completely different subjects, scientifically speaking. The reason for them to be combined are probably interests of certain industrial corporations... My very best wishes (talk) 13:58, 7 December 2015 (UTC)

New sources are not automatically better than old. Per WP:SCIRS: An individual primary source should never be cited or juxtaposed so as to "debunk" or contradict the conclusions of a reliable secondary source, unless the primary source itself directly makes such a claim. The primary source in question, PMID 26243044, does not cite the secondary source and hence does not directly make such a claim. On that basis, we should exlucde the primary source, even if it is newer. Glyphosate and glyphosate resistant plants are not completley different, they are complimentary. If one is doing saftey testing, is it reasonable to test the combination because the combination is commonly used in agriculture. The reason why they are combined is not only interest to Monsanto, but also farmers who are looking for higher yields. Boghog (talk) 17:47, 7 December 2015 (UTC)

• Do not include An outlier result, in a primary paper which is published in a low impact journal which is criticised in "pop=sci press" (which includes blogs) probably shouldn't be included until picked up in a secondary source. SPACKlick (talk) 08:02, 7 December 2015 (UTC)

Arbitrary break 1

Boghog is repeatedly referring to the Weed Control Freaks blog post by Andrew Kniss to call the study in question "fatally flawed". To correct a couple things, the study in Nature Scientific Reports is not necessarily "fatally flawed" just because a pesticide promoter (Kniss) said so on his blog "Weed Control Freaks". I'm sorry to be posting similar information several times, but it's needed because Boghog has been placarding his claims everywhere, so this message needs similar visibility as to not get lost in the shuffle. Please see my longer comment in the previous section of this talk page on whether the study had "no control" as Kniss and Boghog are saying, for it is not the case. If the question is ecological and not solely about effect of GBF on earthworms directly, then there is a control. Secondly, note that the 2000 secondary source that's been cited in this discussion repeatedly to claim that there's no effect on earthworms actually does mention Springett et al. and does report that "glyphosate applied to the soil at the prescribed rate was reported to have caused significant reductions in the rate of growth and maturation of earthworms at lesser concentration than that selected for the TRV." Therefore, these two arguments (secondary sources don't support, and Kniss says the study has "no control") are both flawed. Please consider these points of fact. SageRad (talk) 14:48, 7 December 2015 (UTC)

OK, farmers should not remove weeds by what ever means because it will hurt worms. Got it. Boghog (talk) 18:57, 7 December 2015 (UTC)

it is important to known whether the effect glyphosate is direct toxicity on worms or indirect (through glylphostate killing weeds). If the effect is direct, then the study is relevant to this article. If the effect is indirect, this study would be more appropraite for weed control. The "environmental" control does not answer this critical question. The Springett and Grey study was exluded for methodological reasons which is the same as saying it is unreliable. Boghog (talk) 16:51, 8 December 2015 (UTC)

Boghog Re removing weeds by what ever means will hurt worms: not true. instead of sarcasm respond to Sage Rad's arguments. --Wuerzele (talk) 08:25, 8 December 2015 (UTC)

@Wuerzele: How do you know removing weeds will not hurt worms? With out an appropriate control, there is no way to distinguish between glyphosate directly hurting the worms or indirectly hurting the worms by removing their food supply. Boghog (talk) 16:51, 8 December 2015 (UTC)

@Boghog: The study hypothesized that killing the plants "would stimulate earthworm activity and reproduction due to the increased availability of dead plant material that earthworms can use as food source." In addition, the study "provided extra food for earthworms in all treatments (i.e., dried chopped hay spread over the soil surface)." Why are we joining Kniss in arguing against this study with personal speculation, that seems to be more armchair guessing at anything that may raise some sort of conversational, circumstantial doubt, than an expert challenge to specific details and conclusions? --Tsavage (talk) 10:16, 9 December 2015 (UTC)

@Tsavage: The blog is an expert challenge to specific details and conclusions. Have you read it from beginning to end? Regardless, we should wait for secondary sources that discuss this study before drawing any conclusions. Boghog (talk) 10:23, 9 December 2015 (UTC)

Yes, I've read it from beginning to end.

• Kniss concedes his lack of expertise in half of the primary subject, earthworms. He then claims an absence of control based on his own interpretation of the study objective and earthworm behavior.
• The other claim is that the result is invalid because one RoundUp formulation contained a second herbicide, pelargonic acid. PA is a naturally occurring fatty acid found in most plants, and an FDA GRAS substance used among other things as a food additive; according to one source, "Toxicity tests on non-target organisms, such as birds, fish, and honeybees, revealed little or no toxicity." Another source says, "glyphosate-based, ready-to-use weed control products often contain pelargonic acid (PA) at a concentration equivalent to that of the glyphosate." All of which of course doesn't eliminate PA, however, this study as titled is simulating real-world conditions for "glyphosate-based herbicides," specifically "the most widely used glyphosate-based herbicide Roundup", and not simply glyphosate, so the presence of PA is not unaccounted for.

From what I can make out, we would not be discussing any of this if not for Kniss' non-expert argument being introduced as a way to discredit the study, which does appear to explore new territory not covered in existing sources. And there's a fundamental difference between describing and drawing conclusions, as made clear in core policy (WP:PRIMARY). --Tsavage (talk) 10:57, 9 December 2015 (UTC)

I think it is obvious to both experts and non-experts that scientific studies need controls to eliminate other explanations. Without proper controls it is impossible to determine the cause of the effect. The control Kniss suggested is an obvious control. However it gets messy for non-experts to debate the merits of this argument. That's why we need to wait for a secondary source that reviews this study. Boghog (talk) 11:10, 9 December 2015 (UTC)

As a general concept, controls are being argued here at a high school general science level, it's not difficult to grasp. Regarding the specifics in this case, you say, "it gets messy for non-experts to debate the merits of this argument," meanwhile, Kniss declares that he is non-expert and only guessing ("I'm not an earthworm expert, but I would guess"), and says he can't find any literature to back him up on his guess ("I wasn’t able to find any studies that specifically evaluated the impact of plant removal on earthworms"). I haven't taken anything out of context. So why are we to consider that expert commentary? --Tsavage (talk) 11:52, 9 December 2015 (UTC)

Lack of an obvious control is common sense commentary. In addition, Kniss is an expert on the use of glyphosate to kill weeds. One of the other points he raises is the amount of glyphosate they applied is about an order of magnitude too high to be relevant to a field situation. Hence we do have expert commentary questioning the amount of glyphosate used in the study. Boghog (talk) 12:32, 9 December 2015 (UTC)

The study explains the application amount: "In total for all applications, 176.12 ml m−2 of herbicide was applied which is 53% lower than the recommended plant-based application rate of 1000 plants l−1 for ‘Roundup® Speed’ and 62% lower than the recommended dose of 800 plants l−1 for ‘Roundup® Alphée’ (Monsanto Co., St. Louis/Missouri, USA)." Apparently, they're following instructions, and if they'd misfollowed, presumably Kniss would have caught that directly, rather than introduce a novel calculation and a new application rate of his own (a rate based on what? for which products?). So, another questionable criticism.

This seems a bit unnecessarily painfully drawn out, as I'm simply following up on a point that others have made, that Kniss' post is not a reliable critical source for our purposes in this case, and that should be discussable without explicitly examining the blog post, line by line.

As it stands, you're proposing that the opinion of Kniss, a self-declared non-expert in the main focus of this study, earthworms, in a non-rigorous blog post discussion where he literally "guesses" that his central critical argument is valid, should outweigh the peer review of a Nature Publishing Group journal, to the point where we should consider that study fatally flawed. That does not seem reasonable, or RS - we should just stick to reliable sources, including waiting for an independent, rigorous negative assessment of this study by subject experts, until then, it seems fine for careful use per WP:PRIMARY. --Tsavage (talk) 14:07, 9 December 2015 (UTC)

SageRad - The 2000 source discusses why Springett and Grey isn't reliable: "for these and other methodological reasons, the study was not included in this analysis". Yet another example of why we need to wait for reviews rather than including primary research findings. SmartSE (talk) 17:36, 7 December 2015 (UTC)

The 2000 source doesn't call Springett at al. (1992) unreliable. They excluded the study from their results for other reasons, isn't that the case? SageRad (talk) 18:19, 7 December 2015 (UTC)

That's a polite way for the authors to say that the results weren't reliable. SmartSE (talk) 18:45, 8 December 2015 (UTC)

Science isn't polite. If they wished to say the study was flawed, they would have. From what i glean, they excluded all studies that didn't use natural soil because that's the criteria they set up for their methodology and their particular question, but that is not the same as saying the study was unreliable. Please quote whatever description you may be referring to from the review article to make you think that they say it's unreliable. SageRad (talk) 12:35, 9 December 2015 (UTC)

Here is a quote from the 2000 review which makes it clear that the Springett and Gray 1992 study is unreliable. They didn't exclude the study because the soil wasn't natural. They exlcluded the study because (1) the population size was too small, (2) variability in the study was not adequately reported, and (3) the relative toxicity of Roundup to insecticides was not consistent with other studies.

• Giesy JP, Dobson S, Solomon KR (2000). "Ecotoxicological risk assessment for Roundup® herbicide". Reviews of Environmental Contamination and Toxicology. 167: 35–120. doi:10.1007/978-1-4612-1156-3_2. ISBN 978-0-387-95102-7. In the first study, glyphosate applied to soil at the prescribed rate was reported to have caused significant reductions in the rate of growth and maturation of earthworms at lesser concentrations than that selected for the TRV [toxicity reference value] (Springett and Gray 1992). Only six worms were included per treatment, and variability among worms and treatments was not thoroughly characterized. Results of the assay indicated that RU [Roundup Herbicide] was as toxic or more toxic than several insecticides, which is not consistent with other literature. For these and other methodological reasons, the study was not included in this analysis.{{cite journal}}: CS1 maint: multiple names: authors list (link)

Boghog (talk) 04:37, 10 December 2015 (UTC)

• Do not include (obviously as I removed it). If we are to have any hope of editing this article in a collaborative manner without arguing over every single sentence ad infinitum, we need to stick to secondary sources. While MEDRS may not apply, the justifications for requiring high-quality sources apply equally to agrochemicals. While the criticism of this specific source on Kniss' blog is of interest (and strengthens the case for excluding the source per SCIRS: "Blogs by relevant subject matter experts may be useful in talk page evaluation of the relevance of very new results") this wasn't the reason I removed it and nor is it required to justify exclusion. The content made some extraordinary claims about both earthworm reproduction and nutrient leaching and primary sources should not be used to contradict the established knowledge in secondary sources. SmartSE (talk) 17:54, 7 December 2015 (UTC)

Where is your justification that agrochemicals require these overly-high quality sources? If I want to write about the harm that glyphosate/Roundup causes to fish, amphibians, non-human mammals, where are the PAQs indicating secondary sources must be used?DrChrissy ^(talk) 18:39, 7 December 2015 (UTC)

There aren't, but given the controversy, don't you think that is something we should strive for? SmartSE (talk) 18:45, 8 December 2015 (UTC)

Arbitrary break 2

• Comment It's absurd to use a 2000 review as the principal reference for any topic, unless there is nothing more recent available. I see one from 2014, "Pesticides and earthworms. A review" by Céline Pelosi , Sébastien Barot, Yvan Capowiez, Mickaël Hedde, Franck Vandenbulcke from 2014 in Agronomy for Sustainable Development Volume 34, Issue 1 , pp 199-228. OpenAccess at [2] I found it by the subtle method of looking at the references in the Nature Scientific Reports paper. Didn't any of the people disputing about that article actually read it all the way to the end, where the references are listed?

On the general matter, of course a primary study in a good journal can be used to supplement a review. The question is rather if the blog criticizing it can be used. If the author is not an authority on this subject, our rules are quite clear it cannot. Unfortunately, this may give a false impression, but if the criticism are valid they will appear in something actually citable in a year or two and can be added then. DGG ( talk ) 18:33, 7 December 2015 (UTC)

You read my mind - before reading your post, I made my own post below regarding the source you mention above!DrChrissy ^(talk) 18:46, 7 December 2015 (UTC)

Thank you so much, DGG. Always preferable to simply look for better sources. I missed that one because apparently the journal is not indexed by PubMed, which i used for searching for review articles. SageRad (talk) 18:55, 7 December 2015 (UTC)

I actually found it independently of the nature.com paper by using Google Scholar.DrChrissy ^(talk) 19:04, 7 December 2015 (UTC)

It is not absurd to use a 2000 source if the source is still the best one avaiable. I did see this review, but didn't think anything of it, because the only conclusion the authors came to was: We believe that responses observed at infra-individual or individual levels have an impact on higher organisation levels (populations, communities) but there is currently no strong proof. I agree that the blog is not the final word, but it does give us sufficient doubt about the primary source that it is prudent to wait until review that discusses the primary source is published. Boghog (talk) 19:33, 7 December 2015 (UTC)

Oh, yes, this review from 2014 is an excellent source. And it tells a number of times that glyphosate adversely affects earthworms. How significant that adverse effect was, according to this review, is another matter that needs to be examined more closely by people who are interested in this subject. My very best wishes (talk) 19:48, 7 December 2015 (UTC)

@My very best wishes: – How significant that adverse effect was – exactly. Everything is toxic if dosed high enough. Also where exactly did it say that glyphosate was toxic to worms? All I see are statements like Glyphosate may affect cocoon hatchability and Despite these data, information is lacking on the pesticide effects on earthworm reproduction and growth. They did list conclusions of primary studies, but they did no critical evaulation of individual studies. At a minimum, a review should evaluate primary studies to determine whether appropriate methods were used and exclude those that did not. This "review" did no such thing. The only definitive conclusion this review offered is that we need more data. Boghog (talk) 20:17, 7 December 2015 (UTC)

c Boghog Re Everything is toxic if dosed high enough. that is not a uniformly correct premise.it is old school toxicology. Many substances are toxic at certain low and high concentrations. Plus developmental windows. --Wuerzele (talk) 08:25, 8 December 2015 (UTC)

@Wuerzele: The fact that laboratory animals when exposed to low doses of toxins often do better than vehicle control animals (hormesis or the j curve) in no way invalidates what I said above. I was referring to the effect at high doses, not low doses. The dose makes the poison is not old school toxicolgy, it is still very relevant. Boghog (talk) 06:31, 9 December 2015 (UTC)

Yes, at least at the first glance, none of these reviews tells something like "glyphosate is bad for the environment". But whatever it tells (like "glyphosate may affect cocoon hatchability") can be quoted here. In addition, the exclusion of recent research publications that can be reasonably viewed as RS seems rather questionable to me. My very best wishes (talk) 20:51, 7 December 2015 (UTC)

Your arguments are becoming absurd. We are finding secondary sources and now you are asking for significance levels! This is a technical interpretation, way, way beyond "common sense". Furthermore, how do you know that the authors did not conduct a critical evaluation of individual studies. That is precisely why you are insisting on using secondary sources because we assume (correctly or incorrectly) that secondary sources have made such a critical evaluation. You can not have this both ways.DrChrissy ^(talk) 20:43, 7 December 2015 (UTC)

The 2000 review explcitly includes and excludes various sources and explains why. The 2014 review does not. The former is a high quality sytematic review which comes to definitive conclusions. The later is superficial review whose only definitive conclusion is that we need more data. Boghog (talk) 21:02, 7 December 2015 (UTC)

The title of the 2014 review is "Pesticides and earthworms. A review". Read the conclusions carefully - they relate to pesticides! Glyphosate is a herbicide, and nobody here is suggesting we use content from the Conclusions. DrChrissy ^(talk) 21:12, 7 December 2015 (UTC)

To ask the obvious question, if the review is about pesticides, why are we even discussing it all? In reality, the review is about both pesticides and herbcides (glyphosate is mentioned 22 times) and its conclusions are equally vague about both. Boghog (talk) 21:22, 7 December 2015 (UTC)

We are discussing it because you incorrectly insisted on secondary RSs. These have been provided. You now appear to be rejecting these for reasons you are making up as we go along.DrChrissy ^(talk) 21:37, 7 December 2015 (UTC)

• The term "pesticide" includes herbicides. SageRad (talk) 21:49, 7 December 2015 (UTC)

@Boghog:: A new question from the "I'm not an expert on earthworms, but" camp: Are the secondary sources you're citing referring to the same type of earthworm as the study in question, and does that matter? The study here states that most glyphosate earthworm studies to date were conducted in lab conditions with epigeic species that "commonly do not inhabit agroecosystems," while this study was conducted in greenhouse conditions with anecic and endogeic species that "are indeed frequently found in agroecosystems." Also, the study that weedcontrolfreaks cites to support-in-passing the idea that plant removal may have been the source of the effects, says in the abstract: "A significant decrease in earthworm population density of 63% and in total earthworm biomass by 84% was the single most prominent response to the reduction of plant species richness, largely due to a 50% reduction in biomass of the dominant `anecic' earthworms." So ecological categorization would seem to matter, and again, this study used anecic and endogeic species, so even on the surface, that study doesn't seem to squarely apply.

I'm not trying to argue science here, I'm pointing out how murky it gets when we try to argue for exclusion by attempting to trump with other, "better" sources, scientific or pop-sci, things can quickly become original interpretation, which of course is totally off-policy. Better to get context and weight right than try to cut things out entirely without a sufficiently supported reason.

Perhaps a section, Recent developments or similar, might be useful. --Tsavage (talk) 22:01, 7 December 2015 (UTC)

No reply? My impression is that expert opinion on Wikipedia is best suited to actually clarifying questions like these, directly in Talk page discussion. --Tsavage (talk)

@Tsavage: Sorry for not responding sooner, but during the middle of the week, I am busy in real life. You raises some excellent points. Yes, the type of earthworms used in studies clearly matters. I also agree that things do get murky when we try to interpret sources. Hence given that questions have been raised about the methodology used in the primary source and given that this is a highly controversial topic, wouldn't it be better to wait for secondary sources to review the methods and conslusions of the primary source? Per WP:PSTS, it is generally agreed that secondary sources are more reliable therefore preferred over primary. If we do mention the primary source, then I think we should also mention that the methodology of this source has been questioned as is done in the German Glyphosat Wikipedia article. Boghog (talk) 06:15, 9 December 2015 (UTC)

"I think we should also mention that the methodology of this source has been questioned" - That certainly seems like the thing to do if there's an appropriate critical source. As editors have pointed out, the Kniss critique itself appears to be flawed. He self-declares his non-expert status and then goes on to "guess," and he questions assigning the effects to glyphosate in isolation, when the study is clearly about glyphosate-based herbicides, and specifically, about the RoundUp brand, following manufacturer instructions for ready-to-use products. To note that the methodology may be flawed, we'd need a reliable source for that. And thank you for the reply. --Tsavage (talk) 11:25, 9 December 2015 (UTC)

Absolutely! But no reply even though, incredibly, this Kniss blog post is still being heralded below. FS! --Elvey^(t•c) 23:58, 12 January 2016 (UTC)

Still, crickets.--Elvey^(t•c) 19:23, 13 January 2016 (UTC)

• Use secondary sources, per long-standing practice. Dumpster-diving the primary sources is not only dangerous, it's completely unnecessary when a subject has as many secondary sources as this one does. Guy (Help!) 08:26, 8 December 2015 (UTC)
• Yes the above mentioned Berghausen,Hofer study from 8/2015 even though primary can be used. there is no evidence in the peer reviewed literature that it is flawed, as has been amply argued above. On top of that one can add the 2014 secondary source "Pesticides and earthworms. A review" by Céline Pelosi , Sébastien Barot, Yvan Capowiez, Mickaël Hedde, Franck Vandenbulcke from 2014 in Agronomy for Sustainable Development Volume 34, Issue 1 , pp 199-228.

Sarcasm ("Dumpster-diving") again used by an opposing editor, rogue administartor User:JzG is WP:UNCIVIL and has no place here. it is WP:Disruptive. Therefore, I think the RfC should be closed soon or modified to include the 2014 review. --Wuerzele (talk) 08:39, 8 December 2015 (UTC)

You really want to go down that route? It's a very bad idea. Wikipedia always prefers secondary sources, the use of primary sources is discouraged to reduce the tendency to synthesise from them. This is documented in WP:RS, one of our longest-standing guidelines. Also, it's spelt "rouge". Guy (Help!) 08:54, 8 December 2015 (UTC)

(edit conflict) @Wuerzele: See WP:ROUGE, and actually read WP:DE. Being critical of arguments, sources and sourcing approaches is not "disruptive" or incivil. WP:CIVIL does not require obsequiousness. And admins are entitled to the same editorial rights as you are; disagreeing with you doesn't make them "rogues". Accusing people of being rogue admins without diffs to prove it is, however, incivil; see WP:ASPERSIONS.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  14:06, 25 January 2016 (UTC)

@SMcCandlish: I modified your post to remove the ping, because in the intervening time, that editor has been topic banned by ArbCom from GMO pages. --Tryptofish (talk) 21:09, 25 January 2016 (UTC)

• Comment. I just checked the policy here. It tells:

1. "Articles should rely on secondary sources whenever possible...", meaning that primary research papers can be used, but should be quoted directly rather than interpreted by wikipedians.
2. "Material such as an article, book, monograph, or research paper that has been vetted by the scholarly community is regarded as reliable, where the material has been published in reputable peer-reviewed sources or by well-regarded academic presses", meaning that primary research papers are great.
3. "Completed dissertations or theses ..." meaning that even PhD theses which were not published in any journals (!) are good. My very best wishes (talk) 17:07, 8 December 2015 (UTC)

• Comment and further to the above, this is a direct quote from WP:RS "Wikipedia articles should be based mainly on reliable secondary sources, i.e., a document or recording that relates or discusses information originally presented elsewhere." I have highlighted the word "recording" as this means if an author in the Introduction or Discussion section of a research paper discusses previous work, this makes the source secondary for that content.DrChrissy ^(talk) 17:18, 8 December 2015 (UTC)

• Per WP:PRIMARY, this source seems fine for use, subject to the information selected and the wording. It is published in a reputable journal, and no independent, expert argument against it has been brought up. Also, the general claim is not exceptional, as other studies indicating generally comparable effects have been cited in this discussion. (I also find this RfC to be poorly formed, it should have at least a draft of the proposed content, rather than present the source for blanket approval.) --Tsavage (talk) 03:01, 9 December 2015 (UTC)

• Do not include per WP:PRIMARY, WP:RS, and our science sourcing guidance in WP:MEDRS and WP:SCIRS. All relevant policies and guidelines strongly recommend against the use of a primary source like this in scientific topics. That a primary source can be used in some instances according to policy doesn't allow carte blanche use of them. Being a controversial topic, editors are expected to raise source quality. It's been a longstanding expectation on this talk page that we rely on secondary sources and not use primary sources because of how easily they can be cherry-picked or misinterpreted. Kingofaces43 (talk) 06:25, 10 December 2015 (UTC)

• Include this primary study per our present guidelines. Our guidelines were set up so that we could avoid going through this argument over and over. Kingofaces43, you keep saying that there is a longstanding agreement on this talk page that we rely only on secondary sources. I've been watching this article for a long time and I can't recall that decision. I'd appreciate it if you'd point it out to me. This is an argument that you have brought up at other articles as well, honey bees for instance. Gandydancer (talk) 14:57, 10 December 2015 (UTC)

• Not an acceptable source. It's a primary source. Moreover it describes an experiment done without a control: if you kill all the vegetation, the earthworms are affected, that is unremarkable. A controlled experiment could compare the effect on the earthworms of killing the vegetation using glyphosate, with killing it mechanically, say by using a sickle or by uprooting it. And contrary to what has been said above, it was not published in Nature. Maproom (talk) 08:41, 11 December 2015 (UTC)

• Do not include. This is undue WP:WEIGHT to a primary study. There are literally thousands of primary studies published every week, we do not include every one of them here unless there is a secondary source that determines they are significant. Yobol (talk) 02:26, 14 December 2015 (UTC)

• Do not include. Per WP:RS, WP:PRIMARY, and WP:WEIGHT. Primary studies are manifold and require non-trivial coverage in secondary sources before their merit may be assessed. — ArtifexMayhem (talk) 11:12, 14 December 2015 (UTC)

• Include - This is sorta a silly RfC because the person who started the RfC didn't seem to point to exactly what content we were thinking about adding. Since the question seems to be, can we use this primary source, the answer seems to be an obvious yes. Primary sources can be used. Look at WP:PRIMARY. This RfC should be re-done and show exactly what content the proposer is suggesting we add with the reference in question. NickCT (talk) 20:21, 14 December 2015 (UTC)

• Yes It can be used as a source. Looks like another attempt to block non-MEDRS-sources. The Banner talk 20:17, 10 January 2016 (UTC)

• Include this primary study per our present PAG, WP:RS, WP:PRIMARY, and WP:WEIGHT, which were set up so that we could avoid going through this BS over and over. Policy is clear.

We MUST NOT WP:IAR no matter how shrilly or repetitively Boghog demands that we do so. We must not IAR when it serves to protect the reputation of big companies with deep pockets. But I do not and have not claimed to know why Boghog demands that we IAR. --Elvey^(t•c) 18:57, 11 January 2016 (UTC) (revised for clarity for those who have poor English comprehension skills)

@Elvey: What shred of evidence do you have that I am trying to protect the reputation of big companies with deep pockets? Another plausible explanation is that I do not like studies that appear not to contain appropriate scientific controls. Per WP:ANALYSIS, secondary sources are preferred because WP:PRIMARY sources are easily abused and as a consequence lead to WP:WEIGHT problems. Boghog (talk) 20:51, 11 January 2016 (UTC)

No, I'm saying that you (in demanding that we accept a blog as a MEDRS-compatible) are demanding that we IAR. My evidence of that is your comments on this page. I say that we must not IAR when it serves to protect the reputation of big companies with deep pockets. But I do not and have not claimed to know why Boghog demands that we IAR.

@Elvey: Really? protect the reputation of big companies with deep pockets? Evidence on this page? You need to be more explicit. The only "rule" that I am suggesting that we follow is WP:ANALYSIS which prefers secondary sources. Boghog (talk) 21:25, 11 January 2016 (UTC)

Elvey, you can disagree with Boghog with saying that Boghog is motivated by working for companies. When you make an edit to this talk page, you see a message right at the top of the edit screen, about the fact that the Arbitration Committee has applied discretionary sanctions to this and other pages. One of the findings in that arbitration case is that editors must not make unfounded accusations that other editors are "shills" of companies. If you continue to make this accusation without proof, you will be in violation of those sanctions, so please stop. --Tryptofish (talk) 21:58, 11 January 2016 (UTC)

i have made no such accusation. Your reading comprehension skills appear to be frustratingly poor. I say that we must not IAR when it serves to protect the reputation of big companies with deep pockets. But I do not and have not claimed to know whether or suspect that Boghog seeks to protect the reputation of big companies with deep pockets. Now, let's get back to the topic - the problem is the insistence that we give significant weight to a blog. Policy doesn't support that. Policy does support giving weight to a primary study when that's the best evidence we have, and it' in freaking Nature.com for crying out loud. --Elvey^(t•c) 01:35, 12 January 2016 (UTC)

The problem is that the primary source in question (PMID 26243044) contradicts earlier secondary sources (doi:10.1007/978-1-4612-1156-3_2 and this secondary source from 2010). In addition, the blog questions the reliability of this primary source. When sources contradict each other, per WP:ANALYSIS, it is better to wait for a more up-to-date secondary sources that reviews the available evidence. Also repeating what I have stated above, the journal in question is not Nature (journal), but Scientific Reports which is part of the Nature group of journals. This particular journal has had a rather controversial history. Boghog (talk) 07:33, 12 January 2016 (UTC)

Again, my judgement is that http://www.nature.com/articles/srep12886 is to be included. We disagree. Andrew Kniss' blog post is extremely poor quality. It is NOT a reliable source. Giving up defending it (e.g. the "As I have already explained elsewhere, this particular blog is a reliable source" claim): good; saying it was never suggested : well, I leave it for others reading this thread (e.g. the closer) to judge the accuracy of that claim. You are now saying the problem is something else. Given the quality of your initial evidence your ancient backup evidence doesn't impress, even though, of course, [review > secondary > primary], generally. I don't consider the nature.com mud sling significant in this context; nature.com remains more than torso, head and shoulders above Kniss. --Elvey^(t•c) 17:07, 12 January 2016 (UTC)

Repeating what I have already written above, blogs authored by established experts may be considered reliable. Below you give an example of where a high quality blog was a reliable, useful source. What evidence do you have that Andrew Kniss' blog post is extremely poor quality? And where exactly have I given up defending the blog? I have been very consistent in stating that (1) the blog raises doubts about the study, (2) the primary source in question contradicts prior secondary sources, and (3) because of these doubts and conflicting sources, it is better to wait for a more up-to-date secondary source to review the available evidence. Boghog (talk) 18:39, 12 January 2016 (UTC)

If I gave you credit you didn't deserve, let me know; it seemed you'd given up defending it. I'm not going to reargue further as I see the argument has been had and archived.--Elvey^(t•c) 19:16, 12 January 2016 (UTC)

It may have been mentioned before, but it's worth mentioning again that WP:PARITY covers this situation. Often times a study that has been rejected by the scientific community just gets ignored rather than having write-ups about it in journals. That's why blogs by experts are considered reliable in these instances. Kingofaces43 (talk) 19:32, 12 January 2016 (UTC)

One entertaining bit of evidence that Andrew Kniss' blog post is extremely poor quality: the high quality source says "Eight weeks after planting, the vegetation in half of the mesocosms was treated with a lower-than-recommended dose of glyphosate-based herbicide." Kniss says "the amount of glyphosate they applied is about an order of magnitude too high to be relevant to a field situation". The peer-reviewed source is trustworthy, so I believe either Kniss' writing is flat out untrue, or misleading. It doesn't provide enough info on sources and calculations behind his numbers to be credible. A good answer convinces readers that it's trustworthy and credible by providing reliable sources, and being reproducible. Roundup sold at retail often doesn't have the same formulation or strength as the Roundup products sold to farmers.--Elvey^(t•c) 19:23, 13 January 2016 (UTC)

@Kingofaces43:: This seems new to the discussion, but I don't see how the guideline, WP:PARITY, applies here in the way you describe. You appear to be referring to:

"In an article on a fringe topic, if a notable fringe theory is primarily described by amateurs and self-published texts, verifiable and reliable criticism of the fringe theory need not be published in a peer reviewed journal."

In this case, the study appears in a peer-reviewed journal, Scientific Reports, by the publishers of Nature. There has been some criticism here of the quality or standing of Scientific Reports, so I tried to check on that: I understand impact factor is not the end all of measures, still, SCImago Journal Rank categorizes SR as Multidisciplinary and ranks it at #4 out of 109 journals in that category, behind Nature, Science and Proceedings of the National Academy of Sciences of the United States of America. I would think that helps qualify the study in question as being described in a reputable scientific journal, and not by amateurs, in self-published texts. PARITY goes on to say:

"Fringe views are properly excluded from articles on mainstream subjects to the extent that they are rarely if ever included by reliable sources on those subjects."

Is Scientific Reports not a (highly ranked) reliable source for the subject, as Nature or Science would be? I'm trying to follow what seems to me to be esoteric reasoning. --Tsavage (talk) 23:01, 12 January 2016 (UTC)

"#4 out of 109 journals in that category, behind Nature, Science and Proceedings of the National Academy of Sciences", you say? Another excellent point - which will be studiously avoided - if past experience is any guide!--Elvey^(t•c) 23:58, 12 January 2016 (UTC)

Yup, crickets.--Elvey^(t•c) 19:23, 13 January 2016 (UTC)

(edit conflict)Journal title is largely what's esoteric here for the most part. Regardless of where something is published, a flawed or otherwise not accepted study within the scientific mainstream often won't be pointed out by other journal articles (often ignored instead), but instead pointed out by experts on blogs, and other sources per WP:PARITY. That's largely repeating what I said before though explaining where blogs come in and how it's inappropriate to dismiss such a source solely because it is a blog. Kingofaces43 (talk) 00:03, 13 January 2016 (UTC)

It's enlightening to compare this study to the similarly fatally flawed study discussed here: https://en.wikipedia.org/wiki/Talk:Valsartan/sacubitril/Archive_1#Don.27t_be_too_promotional. Despite hundreds of lines of discussion, any criticism of that fatally flawed study was kept out of the article for many many months. In this/that case, a very high/low quality blog had accurately/inaccurately portrayed the study as of very poor design. In addition, the NY Times had reported on the flaws. In that case, one of the the main editors who battled successfully against User:Nbauman to keep all criticism out of the article for a year was eventually exposed as a sockpuppet: Formerly 98. The other one is still at it.

--Elvey^(t•c) 18:57, 11 January 2016 (UTC)

I am really confused by your comments. So you agree that the study is of very poor design? Boghog (talk) 20:51, 11 January 2016 (UTC)

Elvey, you were already warned by MastCell, so remember you need to focus on content and not continue to pursue personal disputes such as alleging sockpuppetry on unrelated pages after being told to drop the stick at the SPA talk page in addition to your warning. Kingofaces43 (talk) 15:20, 12 January 2016 (UTC)

Someone once said, "The way Kingofaces43 speaks above is evidence of a battleground mentality." Seems quite apropos. --Elvey^(t•c) 17:38, 12 January 2016 (UTC)

Whoops, no. Flawed, yes, but not seriously. Clarification attempted. My point is to point out that when a seriously flawed study design was the best evidence supporting efficacy of Entresto, a drug that would predictably be a zillion dollar blockbuster, it was approved despite the serious flaws, and consensus at wikipedia was for an article on the drug to remain scrubbed of any mention of the evidence of the serious flaws, for over a year. --Elvey^(t•c) 19:16, 12 January 2016 (UTC)

• Do not include. I wasn't going to comment here in this RfC-that-will-apparently-go-on-forever, but I decided to comment now, given that comments have apparently resumed. I think that this page has become overly-long in its discussions of seemingly every primary study that has ever found anything negative about glyphosate. Parts of the page read more like some student's book report than a Wikipedia encyclopedia page. For me, the WP:WEIGHT issue is not as simple as this one source, and the RfC is very badly formed anyway, but rather there are simply too many such sources being presented at much too great length. The page should make more use of WP:Summary style. --Tryptofish (talk) 21:51, 11 January 2016 (UTC)

Tryptofish is wearing an advocate hat and an (askew) admin hat in these discussions, and pushing for the same side while wearing both hats. --Elvey^(t•c) 01:35, 12 January 2016 (UTC)

@Elvey: Tryptofish is not wearing an admin hat (askew or otherwise), because he is not an admin. I am, though, and please consider this a formal warning under the discretionary sanctions: your tone needs to change, and you need to start being more thoughtful and a lot less shrill and combative if you want to continue to participate in this topic area. MastCell ^Talk 01:55, 12 January 2016 (UTC)

Struck. I see that you are concerned enough about my tone to issue a formal warning both here and on my talk page. I don't appreciate the false accusations of having made unfounded accusations. Especially the second accusation, which was made after I made it extra clear that I had not intended to and felt I had not made the accusation I was accused of having made. If my tone fails to hide the fact that the false accusations piss me off, please suggest edits to my comments that would address your concern, MastCell. And in the interests of appearing to be balance, I would appreciate it if you addressed the users who made the false accusations. Willing to do that? --Elvey^(t•c) 03:39, 12 January 2016 (UTC) Hello, MastCell? 19:16, 12 January 2016 (UTC)

I'm not entirely clear to which "false accusastions" you refer. You wrote: "We MUST NOT WP:IAR in order to protect the reputation of big companies with deep pockets, no matter how shrilly or repetitively Boghog demands that we do so." A reasonable person would interpret that comment as an accusation that Boghog is acting as a corporate shill. I don't see evidence of false accusations here—except perhaps on your part. MastCell ^Talk 02:38, 13 January 2016 (UTC)

My reading comprehension skills are such that trying "to protect the reputation of big companies with deep pockets" would be considered to be having a COI, broadly construed (since it would be strange to do that altruistically on behalf of those deep pockets). And my memory was just jogged to notice that there is an existing community topic ban about that. --Tryptofish (talk) 03:23, 14 January 2016 (UTC)

• Do not include. It is a primary source of a study run without controls to generate the unsurprising result that creatures do not thrive if you remove their food source. We have good sources to support a claim that a herbicide kills vegetation. To make a confected argument that a herbicide is thus a vermicide is not something we should support. Not unless we have a more reliable source. I think some editors are not here to help the project, but rather to conduct their own political fight in a scientific arena. Science isn't about voting and waving banners vigorously and having your opponents ejected from your rallies. --Pete (talk) 16:02, 12 January 2016 (UTC)

• It's complicated. I agree entirely with the general notion that such a source can be included "per our present PAG, WP:RS, WP:PRIMARY, and WP:WEIGHT" as someone said above; WP:POLICY is policy. While I'm a big fan of WP:MEDRS, it has its flaws and is often over-interpreted and applied as if it's a policy handed to us by WP:OFFICE. It's a guideline and it does not auto-trump other guidelines, much less policies. Part of WP's job on all controversial topics is to teach the controversy, and the principal way to do that is to correctly identify the (usually primary-sourced) politicized arguments, attribute them, never state what they say as fact, contrast it with the opposing view, and keep the "this is what the debate is" material separate from "this is what the scientific consensus is" material. All that said, our most central policy is WP:CONSENSUS. Just because a source qualifies for potential inclusion does not mean that it must be included, and it's entirely legitimate for consensus to conclude against using this source, even within the proper bounds of WP:PRIMARY, if problems with the source's underlying credibility have been identified, either by external reliable sources, or by our own examination of the sources' reputability.

  I think it's utterly absurd that a Nature article is being treated with this level of suspicion, absent them retracting it, or a journal of equal calibre publishes work that directly refutes it (and even then we should probably just give both views). Some blog by someone who says "I'm no expert" is not worth our time in consideration of forming that consensus. Some funding from Monsanto might be, but not necessarily. Just because some money was contributed by a party doesn't mean they had any input into the output, and a journal like that is cautious about such things.

  In the end, the important things here to remember are that WikiProject Medicine is a wikiproject (i.e. some editors agreeing to collaborate and nothing more, as a matter of policy). Per multiple WP:ARBCOM decisions, no wikiproject can control an article, as most topics are within the scope of more than one, and editors not associated with a scope-claiming project have 100% the same editorial rights. Consensus is based on WP:COMMONSENSE application of policy. If the legitimate consensus here, based on rational reasons, is that this source should not be included, then it should not be included, and that isn't some MEDRS conspiracy, it's editorial consensus, so get over it. Many science articles here rely far, far too much on primary studies, because people who become WP editors from scientific academia come from a world where the "sexy" material is primary studies, and literature reviews are boring homework. They have to learn to adjust and follow WP's own sourcing standards.

  Finally, the supposition that primary research papers are secondary sources because journals have editors is wrong here. WP has its own strict definitions of source levels. Journal editors and review boards do enough peer review to determine if a paper seems sane and interesting enough for them to publish, after they tweak it a bit; no more. It does absolutely nothing to bolster the reliability of the research, data gathered by it, and claims and conclusions that the authors derive from it. In many cases, papers are accepted because they're destined to be controversial, not because they're sure things. Secondary-level sourcing is what other researchers' attempts to reproduce and build on that primary reearch, and what literature/systematic reviews, provide (often in the same types of journals). Primary claims are always primary, period. The same publication can contain primary, secondary, and tertiary material. The same paper can contain all three as well (much of the background research in any good primary research paper is secondary, and data pulled from elsewhere, e.g. what the currently accepted species epithet for something is, is tertiary). PSTS is always contextual on Wikipedia. Learn this, know it, feel it. It applies to all sources. If a piece of journalism is entirely secondary except that it presents the journalist's subjective conclusion at the end, that particular claim is primary. If a reliable-source book a year later draws on that article, among other sources it does analysis/synthesis with, and says that the journalist's suspicion was prescient and dead-on, then it's secondary.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  14:00, 25 January 2016 (UTC)

RFC closure

I see no new editors opinions and no new arguments. continued repeating of same comments is unproductive. Since this issue is contentious, I will request formal closure by an uninvolved administrator.

• Boghog, Alexbrn, King, Jzg, SPACKlick, Lfstevens and Smartse oppose the source mentioned in the RFC question.
• DrChrissy, SageRad, myself, myverybestwishes, tsavage and Semitransgenic support the source
• Kmhkmh commented without expressing yes/No --maybe they can add that to their section ?
• Aside from the RFC question DrChrissy, SageRad, myself and DGG support the 2014 review first mentioned by DGG. --Wuerzele (talk) 02:28, 9 December 2015 (UTC)

I requested closure as repeats and strawman arguments continue, and a discussion of alternative sources (2014 review) outside of the RFC make this section increasingly difficult to review.

Please do not add RFC comments in this section.--Wuerzele (talk) 05:48, 10 December 2015 (UTC)

The request for closure is inappropriate at this time, especially going on only five days. RFCs typically last 30 days, and users requested by Legobot typically don't come in until some time has passed already. Even with the topic bans of multiple editors coming soon, there's no rush here or agreement to end this early. Kingofaces43 (talk) 06:08, 10 December 2015 (UTC)

The request for closure is appropriate at this time. The stalling is inappropriate, especially since no new editors no new opinions and no new arguments have surfaced.

Re "no rush": By King's advice the 3 deletions of sourced content over the last wee, that probably need >than 3 Rfc's times 30 days place us into spring 2016.

There is an obvious rush to delete instead of flagging. King ignores this, , Hogbin was the last Smartse ignored the advice, JzG ignored it, Boghog ignored it here and [3] and here creating a doublestandard. This is disearnest.--Wuerzele (talk) 09:43, 10 December 2015 (UTC)

• Comment I had hoped to offer my thoughts but time is very short these days what with the holidays coming up. I thought that I still had time... Gandydancer (talk) 14:45, 10 December 2015 (UTC)

You do. Weeks, in fact. — ArtifexMayhem (talk) 14:51, 10 December 2015 (UTC)

• Comment I cannot do the controversy as it stands justice, which would be good reason for recusing myself, but I think that part of my problem is that the article is an unnecessary and undesirable conflation of concerns. I say as much below and propose that the article should be separated into at least two articles, one dedicated to matters of controversy and the other(s) to matters of material fact. If there is any general feeling that there is merit to the suggestion I shall be willing to contribute, but otherwise I cannot see myself being of much value in the issue. JonRichfield (talk) 11:59, 11 December 2015 (UTC)
• Reminder: This is not a vote. In any RfC, the closer will weigh not the number of supports or opposes, but their grounding in WP:PAG. 500 "I like it" !votes can be refuted a singly solidly argued "it's WP:OR". In this case the applicable PAGs are WP:PRIMARY, WP:UNDUE and potentially WP:SYN. Insistence on the use of the primary source to "balance" reliable secondary sources speaks volumes. Guy (Help!) 12:32, 13 January 2016 (UTC)

Editors flagged as topic banned: Why?

Editors of some comments have been marked as topic banned. What is the purpose of this? IOW, do the reasons for the bans call into question their opinions? Is what is presented in each of the so-marked comments likely to be interpreted differently in light of the editor's ban, and if so, is that justified? If these comments were post before the editors were banned, then what is the relevance now - should all of their comments, or at least, their comments that could be considered current, going back, say the span of the ArbCom case, or perhaps in all non-archived related Talk pages, or for some other arbitrary period, be also marked? Maybe I'm missing something here...? --Tsavage (talk) 00:15, 18 December 2015 (UTC)

I did that, and I'm fine with discussing it. As I understand it, it's pretty simple: if someone is topic banned from a topic, there is a consensus that their opinions are not considered helpful, based on conduct before the ban was issued and reflected in edits made before the ban. The topic bans were issued in the recent ArbCom GMO case. I didn't delete any comments, or even strike through them, but it seems to me that at a minimum, this information should be available to whoever determines the consensus when the RfC is closed. I only did it in this section, because there is a consensus to be formally determined, as an RfC. I didn't do it in other talk sections, although a case can be made for doing so. Where you ask about the time frame of before and after the ArbCom case, talk page edits made after the topic bans were issued would be struck or reverted and would result in blocks. Here, these are edits made shortly before the bans were issued, and which would be subject to interpretation while the bans are in effect. --Tryptofish (talk) 00:31, 18 December 2015 (UTC)

There's also the more pragmatic issue of people following up comments or questions and getting no response. Also a lot of edits and proposals were made in the closing days when the likely outcome was already clear. Guy (Help!) 01:11, 18 December 2015 (UTC)

I disagree with the position that a ban in the GMO case can be read as 'there is a consensus that their opinions are not considered helpful'. Bad behavior and bad opinions are not equivalent, and arbcom's findings target disruption/behavior issues in the topic area, not incompetence. Dialectric (talk) 01:21, 18 December 2015 (UTC)

Some topics bans had already passed before the large push from those editors on the talk page in various areas. We also shouldn't be engaging in WP:PROXYING for these editors or their opinions they tried to bring up at the close of the case. A topic ban simply means the editor has no weight in discussion here. Kingofaces43 (talk) 01:23, 18 December 2015 (UTC)

I'm starting to see that this discussion about process is becoming, itself, a sort of proxy argument, with editors lining up to vote in favor of not labeling the edits. I am not arguing that the editors were "incompetent", for goodness' sakes. I'm saying that they are now banned from contributing to a consensus. And all of their comments are still here. --Tryptofish (talk) 01:27, 18 December 2015 (UTC)

I don't care one way or another whether the names are tagged as banned, and I have no interest in voting on the issue. I do care if past comments from recently banned editors are being dismissed solely on the basis of their ban, when the comments were made before the ban was finalized. Dialectric (talk) 02:05, 18 December 2015 (UTC)

Putting the tags there is not the same thing as dismissing the comments outright. Whoever assesses the consensus of the RfC will be able to see everything, including what we are discussing here, and place however much or little weight on any given comment that they deem appropriate. You would be absolutely right to object if anyone were blocked or otherwise sanctioned for having made the edits before the ban was finalized. But assessing the weight of a comment is not a punishment. It is an assessment of community consensus at the time the RfC is closed, and the close will necessarily be after the bans have come into effect. --Tryptofish (talk) 02:13, 18 December 2015 (UTC)

I think this is outright bizarre, on top of being mean spirited. No, there's no consensus that a now topic banned user's opinions were unhelpful. Would you please stop this nonsense and self revert whatever scarlet letters you have imposed. Minor4th 02:18, 18 December 2015 (UTC)

It is not bizarre. It is not intended to be mean spirited. It is not nonsense. They are not scarlet letters. --Tryptofish (talk) 02:21, 18 December 2015 (UTC)

Then why flag them if not to draw attention to their topic ban and encourage discounting their opinions? I believe it is al the things I mentioned. Please revert and allow the closing admin to evaluate consensus without such influences.Minor4th 02:28, 18 December 2015 (UTC)

For those worried about the tagging, remember that the RfC will be closed based on a consensus of views. If the tagged views are held by a significant number of editors, they will be picked up and supported and expanded upon by other contributors. If no one agrees with and is willing to support them, then they do not reflect consensus anyway. For the record, I have not looked at the topic or formed a view, beyond that an ongoing debate about the tagging is unnecessary (at best) and counter-productive (at worst). EdChem (talk) 02:29, 18 December 2015 (UTC)

The essential concern is not over whether the tagging affects the RfC, it is over the tagging. From the little I understand of ArbCom, it is a forum for behavioral issues, it does not speak to content, and its sanctions are not punitive, they are intended to alleviate problems. Please correct me if I'm wrong. If I'm not, then how is labeling banned editors relevant to anything ongoing? And how is it not disruptive behavior to continue to highlight those editors? I'm not posting for the sake of arguing, and certainly not to defend anyone, I am sincerely, profoundly disturbed by this behavior, and more so by an environment that condones it. --Tsavage (talk) 02:48, 18 December 2015 (UTC)

You appear to be saying that I am being disruptive, and that's a serious thing to say. I have, throughout this discussion, been polite and been open to discussing the issue, even when other editors have called me mean-spirited, nonsensical, and so forth. Here, I will go now to the AN link given in #FYI, below, and ask that an uninvolved administrator take a look, and either revert me or not. --Tryptofish (talk) 02:54, 18 December 2015 (UTC)

[4]. --Tryptofish (talk) 03:02, 18 December 2015 (UTC)This editor has an axe to grind

The tag just above was added by User:Semitransgenic: [5]. Uninvolved editors may want to consider whether the accusations of being bizarre, mean-spirited, and nonsensical, that were directed at me, should in fact be directed elsewhere. --Tryptofish (talk) 17:14, 18 December 2015 (UTC)

you should perhaps apply the same logic to yourself and indicate you were a named party in the arbitration if you are so concerned about balance. How about you get this endorsed by the arbs first, at least then there is some weight behind it, right now it's just you stirring shit. Semitransgenic _talk. 18:06, 18 December 2015 (UTC)

The Arbs have already issued the topic bans, so their decision already exists. They also applied discretionary sanctions, so those apply to how you are addressing me, and WP:NPA is very relevant there. I asked yesterday for uninvolved administrators to review my edit, and to revert it if it were contrary to process. As of this time, no administrator has reverted it. --Tryptofish (talk) 18:15, 18 December 2015 (UTC)

Personally, I don't think marking editors posts with topic banned was a wise decision and will only escalate issues in a pretty raw topic. AIRcorn (talk) 06:11, 19 December 2015 (UTC)

I respect the fact that you are not one of the POV pushers here, and I have come to the conclusion that your point is a good one. I therefore self-reverted the tags. At this point, I feel that it is more important to decrease the drama, and whoever closes the RfC can still see this discussion here and use it to assess the fact that there were comments by editors who were subsequently topic banned. At the same time, I want to point out some things. My request at AN for any uninvolved administrator to revert me has been open a couple of days, and no admin has reverted me. What I did was correct in terms of policy and in terms of arriving at a proper RfC outcome. I'm reverting it, instead, because it is more important to have peaceful editing, and reverting does not really harm anything. I will also point out that I have acted in good faith and been entirely polite to other editors throughout this entire discussion. My civility, to a considerable extent, has not been reciprocated. The comment just below personalizes and assumes bad faith on my part, in a way that I have not done to anyone else (does anyone in their right mind really think that I did this for some sort of "satisfaction"?). There will almost certainly be further DR about GMOs, and other editors in this discussion will be held accountable for their conduct. --Tryptofish (talk) 18:01, 19 December 2015 (UTC)

Arbitration takes a lot of time and energy, for what purpose? to resolve a dispute. Wouldn't it be better to satisfy yourself with the blocks and simply bury the hatchet? This tagging stunt appears hysterical, you really think their views exert the required influence to sway consensus? For someone who once claimed "I cannot overstate how much I want to put the entire experience of the GMO case behind me and return to editing content that I enjoy editing" don't you think maybe it's time you let this whole thing go? Wasn't one "error of judgment" enough? Or maybe this is just about you starting that DS "office pool"? Semitransgenic _talk. 13:46, 19 December 2015 (UTC)

• Summary. I don't think it's been stated outright, so for the closer, DrChrissy, Wuerzele, and SageRad had their ArbCom topic bans pass before this RfC started, but they were not implemented until shortly after. How the closer deals with those editors comments, especially when there was such a push by some of them when they knew their topic ban was looming, is up to them. Kingofaces43 (talk) 15:11, 12 January 2016 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.