Talk:Circumcision/Archive 58

Definitions
Re this edit re-inserting a comment about the distinction between therapeutic and non-therapeutic circumcision: I oppose including this information. For one thing, as Jake already pointed out, it's unsourced. (Per WP:BRD and the controversial template, please don't re-include reverted material without discussing and getting consensus first; and also, please don't insert unsourced information in this already extensively edited article.) This could perhaps be remedied; a source supporting the statement could perhaps be found; but I would nevertheless continue to oppose its inclusion. The reason is that different sources might use different definitions, and this material seems to me to purport to tell the reader how to interpret phrases encountered while reading other sources: outside Wikipedia's bailiwick, in my opinion, and non-NPOV if some sources use different definitions. If we need to make a distinction about how the words are used in this article itself, maybe we might need something like that; but I don't think we do; if you think so, please point out what parts of this article would seem ambiguous without such clarification, and we can consider how best to clarify. Thanks. ☺ Coppertwig (talk) 13:43, 29 August 2009 (UTC)
 * Completely agree. I'm not so sure, though, that we could find a solid source for the statement that "Neonatal circumcision is not considered medically necessary and is therefore categorised as non-therapeutic" &mdash; I'm sure that it is very rare for neonatal circumcision to be necessary, but I find an assertion that this is never the case to be very hard to believe.  The sentence would be more believable if it is rewritten as "Non-therapeutic neonatal circumcision is not considered medically necessary and is therefore categorised as non-therapeutic", though of course that is circular.  Actually, the whole paragraph is somewhat dubious, too.  Look at the previous sentence, in which therapeutic circumcision is defined as "where there is a medical need to circumcise".  That seems a little dubious to me.  What about situations in which circumcision is medically advisable but not actually needed?  If we had a source then we could at least attribute the claim, but as you point out, Coppertwig, the following sources may have their own definitions. Jakew (talk) 13:57, 29 August 2009 (UTC)

(ec) I do not understand what is controversial about these medical definitions. Don't give me WP:BRD Coppertwig. This was not newly inserted material and just reverted. These definitions have enjoyed consensus in the article for months. Jake removed them and I reverted him. Why do we not want to discuss the distinction between therapeutic and non-therapeutic? Certainly its on topic. Easy enough to provide sources for sources for a medical definition but unnecessary in my opinion. Garycompugeek (talk) 14:03, 29 August 2009 (UTC)
 * I'm sorry about the way I mentioned BRD here. I didn't know when the material had been added, and jumped to the conclusion that it was probably recent; perhaps I should have checked if I was going to say something like that.
 * If you provide a source, that may help. However, this article is not about the definition of a phrase, but about circumcision.  You might try Wiktionary: I think it accepts definitions of some phrases.  Possibly a Wikipedia article with the phrase being defined as its title might be a more appropriate place to include such a definition, if an argument can be made for having such a page. You haven't refuted the concerns I raised in my comment above. ☺ Coppertwig (talk) 14:28, 29 August 2009 (UTC)
 * I haven't refuted them because I was unaware that there was conflicting medical defintions of therapeutic and non-therapeutic circumcision. Garycompugeek (talk) 14:34, 29 August 2009 (UTC)
 * I guess what I don't really understand is we are not debating whether circumcision is necessary here but the meaning of the word therapeutic. Garycompugeek (talk) 14:38, 29 August 2009 (UTC)
 * After you find a source with one definition, maybe other people will find sources with conflicting definitions. In any case, I don't like to include this information without a source; I'd like to study the wording of the source to make sure the definition is accurately conveyed in all nuances, if the material is included (which I still oppose anyway, per above). ☺ Coppertwig (talk) 14:53, 29 August 2009 (UTC)
 * Non-therapeutic and not-medically necessary - isn't that cosmetic surgery? Mish (talk) 16:59, 29 August 2009 (UTC)
 * It depends on what definitions you're using, and also on the actual reasons for a particular circumcision (including, eg., religion, prevention of disease, etc). Some people use "non-therapeutic" to mean "not as treatment for a medical condition", while others use a broader meaning (something like "neither a treatment nor prophylactic against future conditions").  And if that wasn't enough, "cosmetic surgery" isn't particularly well-defined either: some use the term to mean procedures performed for any reason other than medical, while others use the term to mean procedures performed for cosmetic reasons only. Jakew (talk) 17:55, 29 August 2009 (UTC)

The Word Mutilation (Uncircumcised & Intact)
I just went to the female circumcision page (http://en.wikipedia.org/wiki/Female_circumcision) and noticed that the word "mutilation" is mentioned 87 times (that's right, go count it yourself). But we cannot use that word here? --201dan (talk) 15:01, 26 August 2009 (UTC)
 * Thoughts on this 87:1 bias would be appreciated, instead of side-tracking into the topic below (replying to this topic below). Thank You.--201dan (talk) 18:03, 26 August 2009 (UTC)

Please comment on this (above... the most important point).--201dan (talk) 16:06, 26 August 2009 (UTC)

Additionally, the offensive word "uncircumcised" is used 22 times on this page, while the neutral word "intact" is only used 3 times. --201dan (talk) 15:09, 26 August 2009 (UTC)
 * Actually, intact is considered offensive by most, where uncircumcised is neutral. See many discussions about this over the years in the archives. -- Avi (talk) 15:31, 26 August 2009 (UTC)
 * Indeed. Jakew (talk) 15:35, 26 August 2009 (UTC)
 * Completely disagree with both of you. How can you say "intact is considered offensive by most"?  By whom are you referring to Avi?  Any sources to back up your claim? Garycompugeek (talk) 17:50, 26 August 2009 (UTC)
 * No. Not Indeed. The DEFAULT position is intact (which includes most of the world). Uncircumcised implies "missing an important procedure" or "having something undone". I'm sorry, but that's a fail on both accounts. But that you're focusing on the less important point (the more important point is the 87 instances of mutilation mentioned on the female circumcision page above, and only 1 on the male circumcision page)... well, I'm not shocked, let's say that much. —Preceding unsigned comment added by 201dan (talk • contribs) 15:53, 26 August 2009 (UTC)
 * There are two separate issues here: whether circumcision (and, by extension, non-circumcision) is right or wrong, and whether language used to describe various states is neutral or not. What we're discussing is the latter question &mdash; how neutral are the various terms that can describe a specific type of penis.  The specific type of penis that we want to describe is one that hasn't been circumcised.  For the purpose of this article, we're not interested in whether it has been pieced, tattooed, subincised, or whatever.  Consequently, even if we were to ignore the non-neutral connotations of the term, the word "intact" is too vague for our purposes.  The word "uncircumcised", however, describes what we mean exactly, it is a well-established term that can be found in most dictionaries, and is composed of the prefix "un-" (meaning "not") and the word "circumcised".
 * Regarding "mutilation", that term often appears in reliable sources discussing female genital cutting, as many authors consider it to be a form of mutilation. It should therefore be no surprise that the term frequently appears in quotations, etc.  In contrast, very few reliable sources describe circumcision as mutilation.  Jakew (talk) 16:09, 26 August 2009 (UTC)
 * Ok... so let me get this right. It's ok to use the word "Mutilate" 87 times for female circumcision page, and only once on the male circumcision page, because of what you consider normal use of language? First off... the difference in language isn't 87:1 ...and I shouldn't have to say any more, but I'm sure you'll drag the obvious out forever to avoid fixing the issue. The word mutilate is used for both procedures very often, and ONLY SLIGHTLY more for female circumcision... it's not 87:1--201dan (talk) 16:26, 26 August 2009 (UTC)
 * And, while I'm not the best at organizing these wiki discussions, I think it would be best if this was under the topic above, to keep this on topic, and not confused with the intact/uncircumcised discussion (they are separate). I see that you are trying to mask the issue by replying to it under the wrong section, and I do not appreciate it. Thanks. —Preceding unsigned comment added by 201dan (talk • contribs) 16:33, 26 August 2009 (UTC)
 * Thank you for illustrating these glaring violations of NPOV. You make excellent points.  It is difficult to remedy with editors who refuse logic and babysit the page 24/7. Garycompugeek (talk) 17:50, 26 August 2009 (UTC)
 * I have no issue with presenting facts, but this bias is way out of control.--201dan (talk) 17:56, 26 August 2009 (UTC)
 * Again... This is a secondary topic that I added, which isn't nearly as important. Please focus on the main topic above.--201dan (talk) 18:06, 26 August 2009 (UTC)
 * "The specific type of penis that we want to describe is one that hasn't been circumcised." Interesting. Surely the specific type of penis that we want to describe is one that is typical, in contrast to one that is atypical (i.e., circumcised).  The description 'uncircumcised' renders a certain normativity to 'circumcised' which doesn't exist.  References to penises in their natural state (including in images) need to be as 'typical', because those that are circumcised depart from the naturally occurring, typical, penis. Mish (talk) 18:50, 26 August 2009 (UTC)
 * The subject of this article is circumcision, not typical vs atypical penes (which may not necessarily be meaningful even in another context, since what is "typical" can depend on context). Consequently the distinction most often made in the article is between penes that are circumcised vs those that are not.  Jakew (talk) 18:56, 26 August 2009 (UTC)
 * Interesting flipped logic. Nice try... but that kind of a rationalization won't fly in the real world. The default is intact, and the procedure creates a circumcised penis. This isn't difficult... or hardly worthy of debate. If I used your logic here on another topic it would fall apart just as easily.--201dan (talk) 19:51, 26 August 2009 (UTC)
 * Personally, I find 'mutilation' a difficult term because of its emotive overtones, and while I agree that it is important to describe significant views from reliable sources that deal with male circumcision as mutilation, I don't think citing the usage on a different article helpful. Intersex surgeries are considered by many intersex people and commentators as 'mutilation', but it is not a helpful term, and it is increasingly avoided.  There is only one mention of mutilation within Intersexuality, although the procedures involved are often far more invasive and traumatic than male or female circumcision (but not as common). Mish (talk) 18:50, 26 August 2009 (UTC)
 * I also don't feel that the term "mutilation" is best. However, instead of seeking neutrality, this page has focused on the emotive word "uncircumcised"... which implies that something needs to be done, or something has been undone, which is both incorrect, offensive, and unbalanced considering that most of the worlds men are intact (not UNcircumcised).--201dan (talk) 19:53, 26 August 2009 (UTC)
 * But I would mostly love it if someone could try to rationalize the 87:1 ratio... that would be fun... ;)
 * How about we stop playing war and get real... is that possible?--201dan (talk) 19:54, 26 August 2009 (UTC)

If I have my arm amputated, before the surgery was done, would you classify that arm as unamputated? Would you walk up to someone who is about to have amputation surgery and say "Hey your arm is unamputated?" or "Hey your arm is intact?". Garycompugeek (talk) 19:32, 26 August 2009 (UTC)
 * Personally, I'm not in the habit of commenting on the existence of people's arms. :-)  If it was important to highlight the fact that it hadn't been amputated, the most sensible choice of word would be "unamputated". Jakew (talk) 19:44, 26 August 2009 (UTC)
 * That's silly. Nobody uses language that way, at least in that context.--201dan (talk) 19:59, 26 August 2009 (UTC)
 * Are you Undead? Or are you Alive? Both are valid, but nobody would use the prior unless they were talking about something that was once alive, dead, and then alive again.--201dan (talk) 20:02, 26 August 2009 (UTC)
 * Mutilation is a charged word, I'd say it's not neutral in this context unless it's directly quoting a notable source. Intact is, however, neutral.Simonm223 (talk) 19:52, 26 August 2009 (UTC)

I know little about the subject, but surely the wrong way to go about this would be to debate what terminology is neutral and what isn't. The right way would be to look at what terminolgy neutral RSs use and, assuming there is reasonable consistency, adopt that. --FormerIP (talk) 20:08, 26 August 2009 (UTC)
 * I only used the word "neutral" because Jake focused on that word several months back (the last time I visited). There is no way the term "mutilation" is used 87 times more for female circumcision than male circumcision in academia... and if someone can provide a percentage chart based on good values I'd love to look at it. I'd say it's possibly 2-3 times more... 4 times at most. 87? Not a chance.--201dan (talk) 20:13, 26 August 2009 (UTC)

This (intact vs. uncircumcised) has been discussed multiple times over the years (as far back as archives 7 & 8) and each time consensus is that "uncircumcised" is neutral and "intact" implies that circumcised people are flawed. Secondly, changing direct quotations, especially to further some POV, is pretty much vandalism. We do not change what people said or printed to fit particular political viewpoints on wikipedia. Thank you. -- Avi (talk) 20:18, 26 August 2009 (UTC)
 * Quite honestly Avi, I don't care how many times basic language (meaning) has failed here... it's hardly an excuse to continue to fail. Intact means whole, while "UN" implies undone or Needs To Be Done (EXTREMELY offensive at the highest possible level!). There is nothing offensive about Intact. You know this, and everyone here knows this just as well as I do. --201dan (talk) 20:25, 26 August 2009 (UTC)
 * I see neither uncircumcised nor intact as being offensive. Suggest using both words for variety and as dictated by flow.Simonm223 (talk) 20:32, 26 August 2009 (UTC)
 * Antonyms of "intact" include "broken, damaged, defective, harmed, hurt, injured, violated", which could well be perceived as offensive. Regardless, as noted previously, "intact" is too imprecise.  For example, consider a penis that is uncircumcised but tattooed.  Is it intact?  What if it is pierced?  What if it is subincised?  What if a glansectomy has been performed?  In each of these cases, one could argue that it is incorrect to describe it as "intact".  Consequently, if a source describes men as "uncircumcised" then we know that their penes are uncircumcised, but we do not know that their penes are intact, and hence we cannot describe them as such. Jakew (talk) 20:51, 26 August 2009 (UTC)


 * I guess in the case of "intact" it's a mere question of functionality, e.g. does a circumcised penis still "work." Can it pee? Check. Can it get an erection? Check. Can it produce babies? Check. So the comparison to an arm ,as given above, is out: an amputated arm can do... pretty much nothing. In addition, the wound caused by circumcision isn't that big of a deal... Let's say someone takes a knife, cuts you across your cheek, and in the process you are left with a half-inch scar; would you then say that your face is not "intact" anymore? Hardly... It might've hurt like sh*t, it might not look pretty to some people, but in order for someone to consider your face "not intact," you'd need call on the flamethrower-and-chainsaw squad to give it a really good work-out. Seb az86556 (talk) 21:19, 26 August 2009 (UTC)
 * If your basing your thoughts on this in regards to functionality then I have to disagree. There are many negative consequences of circumcision so many would disagree with you that a circumcised penis is fully functional. Garycompugeek (talk) 21:28, 26 August 2009 (UTC)
 * I do realize many would disagree, and I have to say that I myself have to ask why there should be a foreskin in the first place if it didn't serve any purpose. I simply find the phrase "not intact" too dramatic in a way. When I hear "not intact" (for example about a car, TV-set, cellphone) what I understand is "this thing never works." That's obviously not true. Seb az86556 (talk) 21:39, 26 August 2009 (UTC)

I have no problem using uncircumcised if the source uses it. However arbitrarily saying that is the only way to refer to a penis that has not been circumcised is not NPOV. Whole, intact, natural or just penis describe a penis that has not been circumcised also. Garycompugeek (talk) 21:40, 26 August 2009 (UTC)


 * Would "cut" and "uncut" be more neutral? The discussion is about "intact" and "circumcised", not "intact" and "unintact", is it?  Why does the fact that some people are intact imply that people who are circumcised are flawed?  I don't see why we would refer to people who have not had a procedure by citing the procedure when referring to them.  If we talk about people who have had cosemtic surgery, such as rhinoplasty, we do not refer to those who have not had such surgery as having 'unrhinoplastied' noses, or even 'intact' noses.  We refer to them as people who have not had such surgery, people who have a nose (while others have a nose that have had rhinoplasty).  So, rather than applying an inappropriate label, perhaps it would be better to simply describe them as people who have not had these procedures, rather than ascribe them as a non-existent category of people who have not a particular procedure.  To do so is effectively pathologise the normal (intact penises) - when in actuality, it is the act of circumcision that creates an ideopathic artifice, a circumcised penis.  Whether a penis is intact is irrelevant - there are two types of penis, one is a penis, the other is a circumcised penis, and one type happens not to have been circumcised. We need to refer to penises, and penises that have been circumcised. Mish (talk) 22:34, 26 August 2009 (UTC)


 * Trouble is, Gary, "whole", "natural", and of course "penis" can also describe a penis that has been circumcised, too. Depending on what meaning is understood by the term, the same can also be said of "intact".  Jakew (talk) 08:44, 27 August 2009 (UTC)


 * Responding to WP:NPOVN post – I'd say use frequencies of occurrence in books and scholarly sources as a guide:
 * Books mentioning "circumcision" and "intact penis" scholarly publications 72 books, 119 scholarly sources
 * Books mentioning "circumcision" and "uncircumcised penis" scholarly publications 520 books, 434 scholarly sources
 * Both terms are used; on the basis of this google-based frequency analysis, the ratio seems to be roughly 5:1 in favour of "uncircumcised penis"  JN 466  23:31, 26 August 2009 (UTC)
 * "Uncut penis" occurs, but more rarely: books, scholar  JN 466  23:33, 26 August 2009 (UTC)
 * Good investigation. Uncircumcised is more neutral, more accurate, and much more widely used in reliable sources. Jayjg (talk) 23:41, 26 August 2009 (UTC)
 * As for "mutilated": the description of male circumcision as a genital mutilation does occur in reliable sources. As a POV, it is perhaps more common outside the US. I don't think it would be undue weight to mention it in this article. Some source samples:  etc. etc.  JN  466  00:23, 27 August 2009 (UTC)


 * By that reckoning, as there are 15 'uncircumcised' references, with the only 'intact' reference being in a quotation, at two or three 'uncircumcised' references should be changed to 'intact'. Mish (talk) 00:30, 27 August 2009 (UTC)
 * As noted, the term "intact" has certain problematic connotations, and if that weren't enough, is also imprecise. Consequently in most cases it would be inappropriate for us to use it, the exception being inside of direct quotes. Jakew (talk) 08:44, 27 August 2009 (UTC)
 * That is essentially an OR argument, and not borne out by the sources. It's not inappropriate to use the term "intact" occasionally; per NPOV, we even have to use it occasionally, following the idea of "representing viewpoints in proportion to their prevalence in the most reliable sources".  JN 466  10:36, 27 August 2009 (UTC)
 * I disagree. Firstly, if one were to sum up the role of WP editors, a good start would be "finding neutral ways to describe ideas in reliable sources", and those ways would necessarily include selection of neutral terminology.  As a (hypothetical) example, even if half of all reliable sources described a person as a "complete bastard", it would be inappropriate for us to use the same description except in the form of a direct quote, since doing so would violate NPOV and BLP.  Put another way, although OR is forbidden in the context of ideas about the subject, OR is often unavoidable when deciding how to apply WP policy.  Secondly, and please correct me if I'm mistaken, there are currently fifteen uses of the word "uncircumcised" in the article.  Of these, four are found in direct quotes (and hence cannot be changed), and the remaining eleven use the same language as the cited source.  That is important, since to argue that some of these usages should be changed is to argue that we should use a different term from that used by the source.  Finally, as I've noted previously, "intact" has a broader meaning than "uncircumcised", so we have no way of knowing that sources referring to "intact penes" actually mean the same thing as sources referring to "uncircumcised penes".  Jakew (talk) 11:16, 27 August 2009 (UTC)
 * This indicates a deeper problem in the article, reflecting a deeper deficit; while the one source cited using 'intact' within a quote counterbalances those quoted sources using 'uncircumcised', the fact that there are eleven references that use 'uncircumcised' and none that use 'intact' confirms the bias in the selection of material - there need to be two sources cited that specifically use 'intact', reflected by using 'intact' in this article discussing them, otherwise the article is not reflecting the 1:5 relationship of articles that use 'intact' rather than 'uncircumcised'; hopefully, addressing this by using two such articles will help to balance the article further beyond simply terminology - by presenting a specifically 'intactivist' view of circumcision as would be required under WP:NPOV. Mish (talk) 11:35, 27 August 2009 (UTC)
 * MishMich makes a strong point. However I would like to caution against article bloat and / or using references to push either PoV.  As I said previously I find neither uncircumcised nor intact to be particularly offensive terms. I still think using both, as dictated by flow, and keeping in mind that 5:1 ratio would be the best way to handle this.Simonm223 (talk) 14:19, 27 August 2009 (UTC)
 * Completely agree. I'm not saying we should switch the balance of terms, but other terms must be represented per source ratio. Garycompugeek (talk) 15:11, 27 August 2009 (UTC)
 * You seem to be making the assumption that for every source using the word "uncircumcised" in the context of a specific fact, there is an equivalent source using the word "intact", and thus they are interchangeable. Unfortunately, that doesn't seem to be the case.  14 of the 15 instances of "uncircumcised" in the article are in the medical section (understandably, since this term is often used in a comparative context).  To assess how easily the corresponding sources might be replaced, I conducted a PubMed search for "circumcision intact NOT female" (the purpose being to find all uses of "intact" in the context of circumcision, but excluding references to "female circumcision").  I found a grand total of twenty references (for comparison, I found approximately nine times as many - 171 - when searching for "circumcision uncircumcised NOT female").  This is not encouraging, and it would be advisable to think carefully before citing any of these just because of their choice of words, as it is important to ensure neutrality with respect to ideas.  I did note that a common usage of the word "intact" was in the form "with an intact prepuce", which is at least more precise (and probably less offensive) than "intact penis".  Jakew (talk) 16:56, 27 August 2009 (UTC)
 * No I don't think anyone is making that assumption at all Jake, at least I'm not. We are simply saying that different terms need to be mentioned because the sources mention different terms.  It is not for us to judge which terms to use in the article although the frequency should be weighed according to available sources.  Garycompugeek (talk) 19:12, 27 August 2009 (UTC)
 * Jakew. Please don't assume what you think I assume. I am quite happy for the 'uncircumcised' sources to remain.  My concern is the exclusion of the balancing 'intact' sources (i.e. at least two more) from the article - in whichever section it is most relevant to have them.  Seeing you know so much about this, feel free to select what you feel are the most significant, relevant, representative and reliable 'intact' sources and bring them to the table, then we can work out which section(s) they would be best suited for. Mish (talk) 22:31, 27 August 2009 (UTC)
 * As usual, when a valid point is made - in this case that there is a deficit of text relating to two sources that discuss 'intact' penises' - there is no further response, and the discussion focuses on something else. This 'either my point is carried, or else I will ignore contradiction' methodology seems to be prevalent in this article's discussion page.  Please can we have on record that we agree that there should be two more sources that discuss 'intact penises', given some editors are keen to ensure that information is finely tuned to represent fair weight. Mish (talk) 17:05, 29 August 2009 (UTC)
 * I agree about the stonewalling and need for more terms to describe a penis that has not been circumcised. Garycompugeek (talk) 17:16, 29 August 2009 (UTC)
 * I think sources need to be selected for due weight of ideas. To select them based on the language they use to describe a certain kind of penis seems rather absurd.  But unless and until someone proposes adding specific sources to the article, there doesn't seem much more to say.  Jakew (talk) 17:48, 29 August 2009 (UTC)
 * I think the onus is on those who wish to include a type of source to go off and find such sources, which we can then discuss. We're volunteers here; no one is required to do research they find uninteresting or unnecessary. It seems possible to me that sources saying "intact" might be more likely to be the type of sources we wouldn't use, e.g. self-published, primary sources etc.  In that case, there would be no need to fix the proportion of language.  I don't think there's any requirement to use vocabulary in proportion to how often it's used in sources. (Can someone find a guideline about that?)  Instead, I think the standard on Wikipedia is to find a reasonably NPOV wording, use it as a title for an article, and then usually use that word for the thing when it appears in other articles.  (That's why the link to FGC in the lead is called "female genital cutting".) There wouldn't be an article on "uncircumcised" or "intact", though, so that doesn't work here. ("un-male-circumcised"?)  In any case, if we do use the wording in proportion to sources, I think it would be in proportion to sources fitting Wikipedia's standards for reliable sources, or perhaps sources notable enough to include as references in this article, not just all sources. I agree with Jake's points about precision and about using the language in the particular sources verifying particular facts, since use of a different term might convey a different meaning.  Maybe if people find sources for expanding the social, history, advocacy etc. sections then the word "intact" might naturally flourish; or it might not, depending on the sources and also depending on the need here for precision, as Jake points out. Please at least propose specific changes before accusing anyone of "stonewalling". ☺ Coppertwig (talk) 18:22, 29 August 2009 (UTC)
 * The idea that we should vary terms within an article so that they appear in proportion with their published frequency seems very strange. Doesn't it invite confusion?  This article could end up with many terms (more than the current "uncircumcised" and "intact" recommendations) all referring to the exact same thing.  It seems like we should pick one term and use it consistently throughout the article.   AlphaEta  03:21, 30 August 2009 (UTC)


 * There are a couple of reasons I don't want to play that game. First, reliable sources tend to require a subscription or one-off payment - as you say, most of us are volunteers, not paid professionals in associated  fields, so that is not viable.  Unless I have a personal sub already, or my institution has a sub, or the publisher allows free downloads, I cannot access all articles.  Not all are in that situation, so the onus would be on those who have such access to do that (in good faith) to ensure balance and NPOV.  The second reason is that given what I have seen already in this article, should such sources be presented, some other excuse will be found to exclude them.  So, I will test this out with one journal I do have access to.  A reliable source, not self-published, but by the BMJ:
 * "A review of the scientific literature, however, reveals that the actual effect of circumcision is the destruction of the clinically demonstrated hygienic and immunological properties of the prepuce and intact penis."
 * PM Fleiss, FM Hodges, and RS Van Howe, Immunological functions of the human prepuce BMJ:Sexually Transmitted Infections 1998;74:364-367
 * That is one example - somebody else can come up with another. Mish (talk) 22:53, 29 August 2009 (UTC)

Live long and Prosper
Spock was one of the most notable pediatricians of the 20 century. His book The Common Sense Book of Baby and Child Care has sold over 50 million copies and been translated into 39 languages. We even have a section about circumcision on his page (I didn't put it there :). Jake reverted a quote I placed in the article. Considering his prestige, I find his opinion on circumcision very notable and feel our readers will too. Thoughts, comments? Garycompugeek (talk) 21:32, 2 September 2009 (UTC)
 * I reverted your addition for several reasons, Gary:
 * The quote that was added was "My own preference, if I had the good fortune to have another son, would be to leave his little penis alone". In other words, this isn't even his recommendation to other parents, but a statement of what his own hypothetical parenting choice would be.
 * The source cited by Milos for the quote (their ref 28) dates from 1989. In the world of circumcision recommendations, that's quite dated.
 * Even if it had been a recommendation, we simply don't have space to include recommendations re circumcision from everyone who has written a baby book. We don't even have space to include more than a representative sample of medical organisations' policy statements, let alone those of individuals.
 * It was added to the "Ethical issues" section, but did not appear to be an ethical issue. So it appeared to be misplaced, and also disturbed the balance of material in that section.
 * Jakew (talk) 08:19, 3 September 2009 (UTC)

Rate of STDs

 * (reply to MishMich's comment of 22:53, 29 August 2009 (UTC)) Interesting article. Read and enjoyed. Thanks for finding it. Perhaps it or some of the articles it cites could be added to the Medical analysis of circumcision article (whether or not also used here). It's not really a review article; it's labelled "debate", and doesn't discuss criteria for selecting studies nor enumerate specific results of the studies. It claims that "rigorously controlled studies have consistently shown that circumcised males are at greater risk for all major STDs than males whose penises are intact." This seems to contradict what I vaguely remember reading in other articles cited here and at Medical analysis of circumcision. It gives five references to support this statement.  The first two find no significant differences in the rate of contracting sexual diseases.     The third study found a higher rate of nongonococcal urethritis among circumcised males.   The fourth found a higher rate of genital warts among circumcised males.   The fifth says "Neither increasing age nor lack of circumcision was associated with HSV-2 infection." . These articles don't seem to me to support their statement, though they might support some weaker statement. ☺ Coppertwig (talk) 23:55, 29 August 2009 (UTC)
 * I think your summary is fair, Coppertwig. Fleiss et al. should be regarded as an opinion piece.  We should generally favour stronger sources such as meta-analyses and systematic reviews (per WP:MEDRS). Jakew (talk) 08:10, 30 August 2009 (UTC)
 * Thanks for proving me right. This is why I won't play the game. There is no policy against citing opinions expressed in WP:RS, in this case a research review, as long as the view is cited as such.  It also demonstrates the bias in the article by filtering out peer-reviewed papers from non-US sources that use terminology such as 'intact' as 'opinion pieces'. Mish (talk) 09:44, 30 August 2009 (UTC)
 * Yes, opinion pieces can in principle be cited, but generally we prefer to cite higher quality sources (such as those I outlined). In cases such as this article it is questionable whether citing an opinion piece is the best use of limited available space, especially when one considers the impact of policies such as WP:UNDUE. Jakew (talk) 11:23, 30 August 2009 (UTC)
 * I agree that the paper can be used as a source of opinion. I suggest the following changes: In the section Other sexually transmitted infections where the first sentence is "Studies evaluating the ability of circumcision to reduce the incidence of other sexually transmitted infections have reached conflicting conclusions" I suggest changing this to "Studies evaluating the effect of circumcision on the incidence of other sexually transmitted infections have reached conflicting conclusions."  I also suggest adding a sentence at the end of the paragraph, saying "Fleiss et al. cite studies finding higher rates of STDs among circumcised men as compared to intact men." ☺ Coppertwig (talk) 13:37, 30 August 2009 (UTC)
 * I wouldn't object to changing the first sentence, but I would object to citing Fleiss et al as you describe, for the following reasons. Firstly, it gives rather a lot of weight to what is only an opinion piece, and quite an old one at that.  I've just searched PubMed for "circumcision (std OR sexually transmitted)", using the "limits" tab to limit results to meta-analyses or reviews.  My search identified 77 items.  I did not examine every result, but checked most of those on the first page, of which none argued that circumcision increases the risk of STDs.  Consequently, I think citing Fleiss et al. as one of six papers would give that paper undue weight, though it might possibly be more appropriate in a more detailed place.  Secondly, it's inaccurate: several of the studies cited by Fleiss et al actually found no significant differences.  What is true is that Fleiss et al. claim that they show higher rates of STDs in circumcised men.  Finally, if we use the term "intact" in this context it must be in a quotation, otherwise it would violate NPOV  Jakew (talk) 13:58, 30 August 2009 (UTC)
 * OK, I accept your reasoning. I've changed the first sentence. ☺ Coppertwig (talk) 14:08, 30 August 2009 (UTC)
 * The problem with this is that by cherry--picking sources in this way, you are still excluding the 20% of sources that use the word 'intact'. What the status of such papers is not the issue, it is reflecting the sources, and this discrimination is WP:OR.  Do you have sources that suggest this specific report is wrong in its attribution of immunological qualities?  The issue is not whether we like what the authors say, but whether it is in a reliable source.  The BMJ is one of the two most prestigious medical journals and publishers in the UK, and therefore the equal (or more) to any other source in this article.  You need to justify why you wish to refuse a peer-reviewed paper from an internationally renowned publication beyond your consideration of it as an opinion piece.  If the age is the problem - then it should appear as noting that at one point the foreskin was seen as having immunological properties, potentially compromised by its removal, and if there is more recent evidence that has 'disproved' this view, that then needs to follow on from that. Mish (talk) 16:38, 30 August 2009 (UTC)
 * The question is not whether authors are wrong, but whether their viewpoint is significant enough to be included in the section. To frame the same question another way, would including their viewpoint in a small section give undue weight to that viewpoint, making it appear more significant than it actually is?  Please bear in mind that this is a relatively small section, currently consisting of a single paragraph citing five sources.  That means that, unlike a larger block of text, it is quite susceptible to undue weight: even adding a relatively small amount of material can shift the balance of the paragraph quite dramatically.  So an important question is, how representative is this viewpoint of the literature in general?  As I've already explained above, I performed a basic PubMed search that indicated that this viewpoint is that of a small minority at best.
 * To correct you on another matter, Fleiss et al. was published in Sex Transm Infect, which is not the BMJ, though it is published by the BMJ group. Jakew (talk) 17:02, 30 August 2009 (UTC)
 * To correct you, it is clear throughout what I said that this is a BMJ publication. Both the Lancet and the BMJ have specialist publications.  That doesn't mean they are not a BMJ or Lancet publications, does it?  This is not the first time you have sought to exclude British input into this article, where we have different perspectives to the USA. Mish (talk) 18:04, 30 August 2009 (UTC)
 * Paul Fleiss resides in California, as does Frederick Hodges. Robert Van Howe currently works (and, I presume, lives) in Michigan, though I believe he was in Wisconsin at the time of publication of that paper.  What "British input" are you referring to?  Jakew (talk) 18:08, 30 August 2009 (UTC)
 * Interesting isn't it? Obviously North American LGBT academics are not the only people who have to look to British journals to publish their papers.  Suggests the bias against certain perspectives in the USA is endemic, and not confined to some editors here. Mish (talk) 20:33, 30 August 2009 (UTC)
 * Actually, Mish, it's quite common for authors to publish in journals that are based in another country (for example, I'm British, and have published in American, Dutch, Irish, as well as British journals). So it's probably unwise to read too much into this; certainly there's no basis for assuming that this is indicative of bias.  Anyway, this is somewhat off-topic, so I suggest we continue this discussion on one of our talk pages, if anywhere. Jakew (talk) 20:53, 30 August 2009 (UTC)
 * I don't know what to make of this then, because I had been WP:AGP and assumed this was down to some cultural thing. Mish (talk) 09:26, 3 September 2009 (UTC)
 * No, it's about preferring review articles over other types of articles, per WP:MEDRS. ☺ Coppertwig (talk) 00:22, 5 September 2009 (UTC)

Redundancy
As a result of recent edits, the following has been added to the start of the "Policies of various national medical associations" section:

"Most guidelines make a distinction between therapeutic and non-therapeutic circumcision. Therapeutic circumcision, where there is a medical, religious, or ritualistic need to circumcise, is less controversial. Other elective neonatal circumcision is categorized as non-therapeutic.[here the AMA are cited]"

The first sentence appears to be original research, but the second and third sentences are of particular interest. I dislike the passive voice in "...circumcision is categorized...", and I was tempted to change it to "The AMA categorises other elective neonatal circumcision as...". However, later on, in the "United States" subsection of the same section, we find:


 * "The American Medical Association supports the AAP's 1999 circumcision policy statement with regard to non-therapeutic circumcision, which they define as the non-religious, non-ritualistic, not medically necessary, elective circumcision of male newborns."

So here we have an example of redundancy: citing the same source in two different sections, to say essentially the same thing. We need to decide which should remain in the article. I would suggest the "United States" one. Jakew (talk) 20:47, 2 September 2009 (UTC)
 * I see what you mean Jake. I just changed it a bit to reflect source better (I hadnt seen your post here yet). Garycompugeek (talk) 21:20, 2 September 2009 (UTC)
 * Gary, you've changed it to read "Non-therapeutic is elected routine neonatal circumcision". I'm afraid I don't even understand what that sentence means.  I presume that "elected" is a spelling mistake for "elective", but even so this still doesn't make any sense: "elective" has almost the opposite meaning from "routine", so the sentence contradicts itself.  Which is intended?  And I don't understand why it is a better summary of the source, which states: "This report is confined to circumcisions that are not performed for ritualistic or religious purposes. In this case, the term "non-therapeutic" is synonymous with elective circumcisions that are still commonly performed on newborn males in the United States."  This seems to say that within the context of non-religious, non-ritual circumcision, non-therapeutic effectively describes elective circumcision.  However, since the AMA were careful to specify that context it cannot be removed. Jakew (talk) 21:38, 2 September 2009 (UTC)
 * Yes a conundrum... how can we have elective routines? Hmmmm.. circumcision is elective but done on a routine basis.  I repaired spelling and tried to tweak flow but paradox remains... maybe this isn't the best source to use.  I just wanted to supply medical definitions to foward paras... well they were already there until you removed them. Garycompugeek (talk) 21:56, 2 September 2009 (UTC)
 * Since you seem to agree that this is confusing at best, I'm going to remove the paragraph for the time being. Jakew (talk) 08:00, 3 September 2009 (UTC)
 * Incidentally, the source doesn't say "routine". Instead, it says "commonly performed". Jakew (talk) 09:22, 3 September 2009 (UTC)
 * Please don't insert material which states or implies that religious circumcisions are therapeutic, unless there is discussion and agreement on the talk page and support by reliable sources. For example, I oppose this wording mentioned above: "Therapeutic circumcision, where there is a medical, religious, or ritualistic need to circumcise, is less controversial.". ☺ Coppertwig (talk) 17:56, 5 September 2009 (UTC)

Re earlier reports
Please use accurate edit summaries; e.g. say "etc." at the end if your summary doesn't describe everything you did. Here some material was deleted, I believe, although the edit summary doesn't say so and it's hard to see in the edit since material was also moved. I think the material about "earlier" studies (re HIV) was deleted. At the moment I have no comment as to whether it should or shouldn't be deleted; I'm just noting this here in case I (at a later time) or others want to consider it. ☺ Coppertwig (talk) 20:35, 24 July 2009 (UTC)

(ec) A recent edit to the HIV section has made a number of changes that require discussion.

A non-exhaustive list of the changes:
 * According to the edit summary, "moved observational studies". It is unclear from the diff what was moved and why.
 * Addition of "A recent mathematical model found that scaling up male circumcision would have a considerably lower impact than condom use or antiretorviral therapy (ART) coverage on new HIV infection rates and on death rates in men in South Africa", citing . There seems to be some confusion here - the title in the citation is that of the IAS abstract, but the link is to an AIDSMap article.  The abstract is preferable.  More importantly, if this is mentioned then other modelling studies should also be mentioned.  I'll search for the references.
 * Addition of "A randomised controlled trial in Uganda found that male circumcision did not reduce male to female transmission of HIV. The authors say they could not rule out the possibility of higher risk of transmission with men who do not wait for the wound to fully heal.", citing Wawer et al. "Say" should be past tense, preferably "stated".
 * Addition of four sources to the sentence "Some have challenged the validity of the African randomized controlled trials, prompting a number of researchers to question the effectiveness of circumcision as an HIV prevention strategy." Come on, we only site three sources when giving the mainstream view ("As a result of these findings, the WHO and the Joint United Nations Programme on HIV/AIDS (UNAIDS) stated that male circumcision is an efficacious intervention for HIV prevention but should be carried out by well trained medical professionals and under conditions of informed consent (parents consent for their infant boys)").  To cite twice that number when giving the view of a small minority is undue weight.  Two is acceptable, three is borderline, six absurd.  Jakew (talk) 20:38, 24 July 2009 (UTC)


 * I'll go through your points in turn Jake.
 * You know what was moved because I moved them to circumcision and HIV. It is fairly obvious to anyone who has circumcision and HIV on their watch list.
 * Fine. If you can find a model which contrasts the different approaches to HIV prevention I'd be happy for them to be included.
 * I have no objection to using the abstract as opposed to the article.
 * I don't see anything wrong with say . But I have no qualms about the changing of a minor word.
 * Regarding weight. I don't see a problem with the number of references.  We do not mention their names like we do the WHO, UNAIDS or CDC so it is not as if we are giving them equal prominence.  The section tends to focus on developing countries such as Africa and doesn't say anything about developed countries so there are NPOV issues there. I personally think that that is more of an issue. So in all I think you are making a fuss over nothing as usual.  Tremello22 (talk) 20:52, 24 July 2009 (UTC)


 * Ah, it wasn't obvious from the edit summary that you meant moving to another article, rather than moving within the section.
 * Regarding "say", there are two problems. First, the previous sentence uses "found", and it doesn't make sense to use the past tense in one sentence and the present tense in another.  For consistency, "said" would be preferred.  But technically, "said" is not strictly appropriate for written communication.  Hence why I suggest "stated".  Actually, thinking about it, a few other changes are needed: "The authors say they could not rule out the possibility of higher risk of transmission with from men who do did not wait for the wound to fully heal before engaging in intercourse."
 * Finally, as noted, the number of references is disproportionate. If there were nothing to make a fuss about then I assure you I would not make a fuss.  Do you think any of the newly-added references are better quality than the original two?  If so, we can cite those instead.  Otherwise, I'm going to revert the additions.  Jakew (talk) 20:59, 24 July 2009 (UTC)
 * Regarding say I have no problem with your improvement.
 * Regarding the additional references. Sorry but you haven't exactly explained why it is disproportionate.  A few reasons why I think you are wrong.  1)The views of the WHO, CDC and UNAIDS are based on the 3 randomised control trials.  Discussion of the trials already takes up most of the section.  2)The additional refs are not taking up any additional space.  Each are interesting and have different things to say.  3)The views of the WHO, CDC and UNAIDS take up much more space so it is not undue weight at all.  4) We don't name the sources like we do the WHO, UNAIDS, and CDC. 5) We don't quote them like we do the WHO, UNAIDS and CDC.
 * So i am perplexed as to why you are so eager to revert a perfectly good edit to the article. You have also conveniently ignored what I wrote about the section being focused on developing countries like Africa and lacking info on circumcision's effect in developed countries. Tremello22 (talk) 21:17, 24 July 2009 (UTC)
 * Regarding "say", that's okay, then.
 * Regarding the additional references, I'm sure I've already explained this. To use three refs to discuss a mainstream point of view and then to use twice that number to discuss a minority point of view is out of proportion.  To address your comments in turn:  1) The views of the six sets of authors are not exactly formed in a vacuum either.  They reference the RCTs and observational studies, too.  2) Yes, they do take up additional space.  The fact that it's in the references section rather than the body of the article does not mean that it is nonexistent.  Also, all sources are potentially interesting and have different things to say; that's not by itself a compelling reason to tolerate undue weight.  3) Your argument here makes no sense: the views of the CDC, etc, ought to take up more space than those of a minority.  4) and 5) seem not to make sense.
 * So I'm not going to revert the entire edit, but I will partially revert it, to correct the undue weight problem. I'll also link to the abstract, and will adjust the "say" sentence. Jakew (talk) 21:29, 24 July 2009 (UTC)
 * Re disproportion: I think the idea is that we could probably find large numbers of refs giving a similar POV to the WHO etc., and we don't list all those, so per WP:UNDUE we shouldn't just list all refs we can find to support an opposing POV.  I think usually in this article we don't list all possible refs but just state main points and give a small number of refs to support those points.  Again, more detail can be given in the subarticles.
 * I don't have time to look at all the articles on my watchlist, and I appreciate informative edit summaries. Also, rather than simply putting an edit like this straight into the article, please first post a description of it on this talk page (similar to the description of your edit which Jake gave above) and wait for discussion, as requested by the template at the top of this talk page which says "Please discuss substantial changes here before making them". Also, when moving material from one page to another, to ensure GFDL compliance please follow the instructions at WP:SPLITTING; I corrected this with this edit.  I would appreciate it if someone would quote on this talk page exactly what is the material that was deleted from this page (i.e. moved to the other page); it's hard to tell from the diff because material was apparently also moved from one part of this page to another in the same edit.  I'm sorry if this seems like too many little rules, but a little effort by one person can make things easier for several other editors. ☺ Coppertwig (talk) 00:58, 25 July 2009 (UTC)
 * I agree with Coppertwig's points above. I'd also add, re disproportion, that citing multiple refs can create the impression that there is more support for an idea than is actually present.  To use an extreme example (deliberately extreme for purpose of illustration), consider the following sentence: "Some people think the earth is a sphere,[1] whereas others think it is flat.[2][3][4][5][6][7][8][9][10][11][12][13]"  Assuming you knew nothing about planetary geometry, and had to guess which was the mainstream view, and which was that of a minority, how would you guess? Jakew (talk) 08:42, 25 July 2009 (UTC)


 * This sentence is not supported by the reference: "The authors could not rule out the possibility of higher risk of transmission from men who did not wait for the wound to fully heal before engaging in intercourse." Further more, something like this should be apparent from the results (First Follow up vs. other follow ups) and adjusted as a con-founding factor, and should have been mentioned in the Interpretation section of the report. I think the sentence should be removed, anyone else agree? --Intactguy (talk) 19:30, 24 August 2009 (UTC)
 * Agreed and removed. Garycompugeek (talk) 19:51, 24 August 2009 (UTC)
 * Please read the complete paper. The second sentence of the discussion section reads "Furthermore, we cannot exclude the possibility of higher HIV transmission in couples who resumed intercourse before complete healing of the surgical wound." The paper is available with free registration at the Lancet, so you may check for yourself. -- Avi (talk) 20:00, 24 August 2009 (UTC)
 * (ec) In the "discussion" section of the cited source, we find: "Furthermore, we cannot exclude the possibility of higher HIV transmission in couples who resumed intercourse before complete healing of the surgical wound." Could either of you please explain how this fails to support the sentence? Jakew (talk) 20:02, 24 August 2009 (UTC)
 * Page 235, column 1 to be precise. -- Avi (talk) 20:03, 24 August 2009 (UTC)
 * Ah yes. I see it now. Didn't realize I had to register to see the entire article.  Splendid.  Thank you for reverting me Avi. Garycompugeek (talk) 20:43, 24 August 2009 (UTC)
 * No problem :) -- Avi (talk) 20:48, 24 August 2009 (UTC)
 * Sorry, I didn't check backs sooner, I too didn't notice I had to log in to view the full paper. But now that I have, I see another sentence that I would think should follow up the sentence I questioned. From the paragraph that starts: "We were disappointed that.... *However, it should be noted that we did not find a difference in HIV incidence rates between women from the intervention group and women from the control group at 12 months and 24 months, substantial periods of time after wound healing appeared complete.*" This is relevant to the discussion of whether or not the healing was a factor. I would like to see a condensed version of this sentence added.   --Intactguy (talk) 18:57, 15 September 2009 (UTC)
 * There comes a point at which a study is given too much coverage, at the expense of other research. I think we're nearing that point now.  I'm concerned that we should be careful to avoid writing an entire essay (forgive my exaggeration to make my point) about this study. Jakew (talk) 08:46, 16 September 2009 (UTC)

As mentioned above, other modelling studies should be included if Lima et al is included. As promised, I've performed a search for refs. The initial results include the following:



Jakew (talk) 08:35, 25 July 2009 (UTC)


 * Re:disproportionality, Jake asks "how would you guess which was the mainstream view?". I think you should give the reader a bit more credit.  The reader can see from what is written that major organisations are advocating it.


 * three refs to discuss a mainstream point of view and then to use twice that number to discuss a minority point of view is out of proportion. We do not actually discuss what was said in these critical review studies, I think that is the point I was making.


 * the views of the CDC, etc, ought to take up more space than those of a minority And they still do with the added references.


 * Yes, they do take up additional space. The fact that it's in the references section rather than the body of the article does not mean that it is nonexistent I never said it was nonexistant.  I said because they are simply references they only take up the space of 2 characters each as they are represented by numbers.


 * all sources are potentially interesting and have different things to say; that's not by itself a compelling reason to tolerate undue weight. I also said they had different things to say. As far as I know the WHO, CDC and UNAIDS are all saying the same thing.  That is the main point which you have conveniently ignored.


 * 4) and 5) seem not to make sense. What I was saying here was that the names of the authors of the critical studies are not mentioned. We also do not discuss what the criticisms actually were; with the WHO, CDC and UNAIDS, we actually discuss what they said.


 * Very few of the mathematical models you mention fit the criteria of contrasting MC with other methods of HIV prevention. Also, some aren't using the data from the RCT's. The mathematical model I included provides additional information to the reader. A mathematical model that says less people will be infected with HIV is obvious to anyone who has just read that MC reduces HIV infection by up to 60%, so is just taking up space. Tremello22 (talk) 21:18, 25 July 2009 (UTC)


 * Regarding my question re guessing the mainstream view, I'd be grateful if you'd actually answer my question. The example I chose did not name any organisations at all.  I think it one would have to conclude that, in that example, the viewpoint that the earth is flat is given more weight.  I agree that naming organisations can help to reduce this undue weight, but I don't think that excuses poorly chosen references (including the number of references used to support each point).  Regarding your next point, I also agree that we don't discuss the content of these critical articles; again this may lessen the effect of undue weight, but it does not eliminate it.  Moving on, when you say "they are simply references they only take up the space of 2 characters each" you seem to ignore my point that additional space is taken up in the references section - you're only counting space in the body text.  Next, I don't think it's correct to say that the WHO, CDC, etc all say the same thing.  They certainly have similarities (and we summarise an important similarity in the sentence describing their collective views), but they also have differences.  Similarly, the critical articles have similarities (they're all critical of the RCTs and/or conclusions or decisions based upon them), but have differences too.
 * Regarding mathematical modelling studies, I wasn't aware that we had actually established the criteria that you describe. Nevertheless, I think that the findings of Hallett et al. and Podder et al. seem largely comparable to Lima et al., and others may be useful as well. Jakew (talk) 21:49, 25 July 2009 (UTC)

Okay, here's an initial draft of a paragraph describing modelling studies. This only includes Lima, Podder, and Hallett:


 * Modelling of the population-level impact of circumcision has shown mixed results. Podder et al. reported that circumcision would reduce, but not eliminate HIV, stating that reduction was more effective when circumcision was combined with anti-retroviral drugs than with condoms.  Disease elimination was considered feasible when all interventions were combined.  Hallet et al. predicted "dramatic" reductions in HIV if circumcision were scaled up alongside behaviour change programmes.  Lima et al., however, predicted that circumcision would have a "limited" impact in reducing HIV (3-13% reduction in HIV cases and 2-4% reduction in deaths); its impact would be "overshadowed when combined with the other interventions".

Jakew (talk) 09:35, 26 July 2009 (UTC)


 * We are discussing 2 separate issues here. I'd appreciate if we could concentrate on one thing at a time.  So let us 1st discuss the amount of critical reviews that we can agree on and what impression we are trying to create.
 * Regarding my question re guessing the mainstream view, I'd be grateful if you'd actually answer my question. The example I chose did not name any organisations at all. I think it one would have to conclude that, in that example, the viewpoint that the earth is flat is given more weight Your use of this extreme analogy is a red herring.  That is why I did not answer your question.  Please stick on topic.  Your example did not name organisations but the HIV section does.  And people can see that 3 major organisations are advocating it.  People can also see what the results of the trials were - that HIV was reduced by up to 60% through circumcision.
 * Why are the references poorly chosen?
 * Regarding your next point, I also agree that we don't discuss the content of these critical articles; again this may lessen the effect of undue weight, but it does not eliminate it. You haven't actually given a good reason why there is undue weight.  Here you are saying that it does lessen the effect of undue weight but how are you calculating it is undue?  If it is undue how are you calculating the extent to which it is undue?
 * Regarding space. I thought your argument was that more space means that the reader gets the impression that the anti-circ POV is more valid than it is in reality. The reader can only see the body text.  So I don't see how a few extra small numbers creates a POV imbalance that justifies the fuss you are making.
 * I don't think it's correct to say that the WHO, CDC, etc all say the same thing. They are advocating it.  So what we have is a whole paragraph focusing on the positive results of these trials. Then we have these major organisations saying "on the basis of these results we advocate it... (or words to that effect).
 * One thing that may be causing the problem here is that the section heading, 'Medical aspects' restricts and determines the criteria for inclusion. I wonder if that is deliberate on your part?  We seem to focus too much on the studies themselves and therefore restrict critical commentaries or the social implications of these studies.  I mean, say if we wanted to discuss Africa in general and its attitude to circumcision, there doesn't seem to be a place for that.  We tell the reader of these studies that circumcision reduces HIV but we don't mention anything about where circumcision is going to be used to prevent HIV.  Nor do we mention where circumcision fits in with other HIV prevention strategies. Tremello22 (talk) 19:49, 26 July 2009 (UTC)


 * As I acknowledged, it is an extreme example, but I'd still be grateful if you'd answer my question. It is an important question, I believe, because it gets to the fundamental issue of whether the number of references cited in support of a point can influence the reader.
 * Regarding "poorly chosen references", I thought the parenthetical comment immediately after that phrase explained it: "(including the number of references used to support each point)"
 * As I've said above, devoting twice the number of references to a minority point of view than to a mainstream point of view is disproportionate, ie., undue weight. I don't understand why further explanation is required.
 * Regarding the amount of space, I don't understand your argument that the "reader can only see the body text". I've just viewed the page twice, once as a logged-in reader, and once as an IP user, and both times the references were visible.  Do you find different results?  I should also say that this has nothing to do with the validity of a POV, which we do not even try to judge, but instead with it's prominence.
 * I don't quite understand the relevance of your assertion that the WHO, CDC, etc "are advocating it". Jakew (talk) 20:15, 26 July 2009 (UTC)


 * It seems I am not getting through to you, so maybe we should get an outside view on things. What do you say? Tremello22 (talk) 20:18, 26 July 2009 (UTC)


 * Jakew? Tremello22 (talk) 20:56, 26 July 2009 (UTC)


 * Tremello: re discussing the social implications: perhaps you could put on this talk page some suggested text, and if we agree that adding it would be due weight, then we can discuss where it would go, changing section headings if necessary. ☺ Coppertwig (talk) 20:59, 27 July 2009 (UTC)

The topic is circumcision for HIV prevention. One wikipedian thinks the addition of more critical references will create a NPOV problem ; another member thinks it will enhance the article. See above discussion for more info. Tremello22 (talk) 21:08, 26 July 2009 (UTC)
 * RFC
 * Given that the request concerns the representation of viewpoints in the scientific literature, I think we might get more feedback if we changed the RFC from (Society, sports, law, and sex) to  (Maths, science, and technology).  (Note: I've added "."s to try to avoid confusing the rfc bot.  14:01, 27 July 2009 (UTC))  I'm not certain, but I presume that it can be simply changed, and the bot will update as needed.  If nobody objects, I'll make the change. Jakew (talk) 07:52, 27 July 2009 (UTC)
 * Why don't we manually list it at both? -- Avi (talk) 07:55, 27 July 2009 (UTC)


 * Note, this is listing incorrectly at the moment and so I have deleted the template above to avoid confusion (I came from the RFC monitoring page). I apologise if you had this under control already. If you follow the guidance of the RFC template, the tag should look like ((rfctag | soc | sci)) followed immediately by the question. I suggest you make this a new sub-section and re-write the question.—Ash (talk) 14:18, 27 July 2009 (UTC)

A sample of the literature
The following represents the first two pages of results found when performing a search for 'circumcision hiv' using the PubMed search engine. The search was conducted today, within about an hour of this post. Although this is non-random (PubMed usually lists results in rough reverse chronological order), it should be reasonably non-biased, and probably less biased than any selection by a less systematic method would be. Two pages is a fairly small number of articles; if anyone wishes to add to the table that would be appreciated.

Note: the purpose of this sample was to identify and classify views specifically regarding circumcision and HIV. Consequently, the "classification" column indicates whether the viewpoint expressed was positive that circumcision reduced the risk of HIV or not. Thus in this context, "positive" does not indicate that circumcision is a good thing generally; it has a more specific meaning than that. I've included the rationale for each classification. In some cases definite classification was judged to be impossible from the data available; in these cases I have left it as unclassified.


 * See table at /Sample PubMed
 * Note, table moved for readability in accordance with wp:RTP.—Ash (talk) 17:07, 27 July 2009 (UTC)

(I'm inserting this here because it is expected that it will be useful to those responding to the Request for Comment, which suggests that readers 'see above'.) Jakew (talk) 16:42, 27 July 2009 (UTC)

Another sample of the literature

 * (ec) The issue here is whether listing 6 references to support a minority viewpoint gives too much weight to that viewpoint when only four references are listed for the mainstream viewpoint. A related issue is whether 6 consecutive superscripts unduly clutter the article and slow page loading time. (I believe page loading time is heavily influenced by the number of ref tags.) The edit in question is here, listing 6 refs after the words "Some have challenged the validity of the African randomized controlled trials, prompting a number of researchers to question the effectiveness of circumcision as an HIV prevention strategy".
 * As an attempt to determine what amount of weight to put on the two viewpoints (supporting or challenging the results of the randomized controlled trials), I'm doing a Google Scholar search for "circumcision HIV prevention randomized" from 2007 to 2009. I'm marking results as "supportive" if they seem to support the idea that there is evidence that circumcision reduces HIV, and "challenging" if I feel that they support the wording Tremello had added to the article with 6 refs. In this analysis, therefore, I'm marking results as "supportive" if they agree that there is evidence of protective effect but raise other concerns about taking action in response to such evidence.  An analysis of the first 20 of 1400 hits:
 * First two hits are two randomized controlled trials (RCTs) in 2007.
 * 3rd hit is irrelevant (trial of diaphragm and lubricant gel)
 * 4th hit: (supportive); a modelling study concluding "Large-scale uptake of male circumcision services in African countries with high HIV prevalence, and where male circumcision is not now routinely practised, could lead to substantial reductions in HIV transmission and prevalence over time among both men and women." (Nagelkerke et al 2007 )
 * 5th hit: (supportive) Weiss et al 2008, "Male circumcision for HIV prevention: from evidence to action?" says "Responding to the conclusive evidence that male circumcision offers significant protection for men from HIV infection, several countries..."
 * 6th hit: (not accessed)  Male circumcision to cut HIV risk in the general population. The Lancet, Volume 369, Issue 9562, Pages 617-619 M. Newell, T. Bärnighausen 2009.  I can't access the full text or abstract at the moment so I can't assess it.
 * 7th hit: (supportive) Sullivan et al 2007 "It is now clear that male circumcision can be efficacious for men in reducing their risk of HIV acquisition through sex with women"
 * 8th hit: (irrelevant perhaps) Westercamp et al 2007 . Perhaps not directly relevant; seems to be about how people feel about being circumcised, but seems to me to assume that the evidence is supportive.
 * 9th hit: Kigosi et al 2007 Again not directly relevant.
 * 10th hit: (not accessed) Male circumcision and HIV/AIDS: challenges and opportunities. The Lancet, Volume 369, Issue 9562, Pages 708-713 S. Sawires, S. Dworkin, A. Fiamma, D. Peacock, G. Szekeres, T. Coates. I can't access the abstract or full text at the moment so I can't assess it.
 * 11th hit: (challenging) Potts et al "The largest investments in AIDS prevention targeted to the general population are being made in interventions where the evidence for large-scale impact is uncertain." Science 9 May 2008: Vol. 320. no. 5877, pp. 749 - 750 DOI: 10.1126/science.1153843
 * 12th hit: (challenging?) "However, the protection of MC may be partially offset by increased HIV risk behavior, or “risk compensation,” especially reduction in condom use or increases in numbers of sex partners." Kalichman S, Eaton L, Pinkerton S (2007) Circumcision for HIV Prevention: Failure to Fully Account for Behavioral Risk Compensation. PLoS Med 4(3): e138. doi:10.1371/journal.pmed.0040138
 * 13th hit: (supportive) "Male circumcision could have substantial impact on the HIV epidemic and provide a cost-effective prevention strategy if benefits are not countered by behavioral disinhibition." AIDS: 23 April 2007 - Volume 21 - Issue 7 - p 845-850 doi: 10.1097/QAD.0b013e3280187544
 * 14th hit: (supportive) assumes that "Male circumcision reduces the risk of HIV acquisition by approximately 60%" but raises concerns about safety.
 * 15th hit: (supportive) Rennie et al Journal of Medical Ethics 2007;33:357-361; doi:10.1136/jme.2006.019901 Copyright © 2007 by the BMJ Publishing Group Ltd & Institute of Medical Ethics.
 * 16th hit: (not accessed) Gray and Wawer Randomised trials of HIV prevention The Lancet, Volume 370, Issue 9583, Pages 200-201 R. Gray, M. Wawer I can't access the full text or abstract at the moment so I can't asses it.
 * 17th hit: (supportive) Journal of Sexual Medicine Volume 4 Issue 4i, Pages 838 - 843 Published Online: 10 Jul 2007 Vardi et al.   Acknowledges evidence for HIV prevention but says more evidence is needed about health risks and benefits.
 * 18th hit: (challenging?) Millet et al. JAIDS Journal of Acquired Immune Deficiency Syndromes: 15 December 2007 - Volume 46 - Issue 5 - pp 643-650 doi: 10.1097/QAI.0b013e31815b834d Study fails to find evidence that African results apply in the United States, but would have expected to see such results if the African results applied directly.  However, it also states that the RCTs "found that circumcision conferred a strong protective effect", so it's apparently not directly challenging the results, only whether the results apply in the US.
 * 19th hit: (irrelevant) "Effect of aciclovir on HIV-1 acquisition in herpes simplex virus 2 seropositive women and men who have sex with men: a randomised, double-blind, placebo-controlled trial"
 * 20th hit: (supportive) Weiss, 2007 Current Opinion in Infectious Diseases: February 2007 - Volume 20 - Issue 1 - p 66-72 doi: 10.1097/QCO.0b013e328011ab73 (incidentally: mentions keratin in the abstract: this increases the notability of the keratin explanation, of which I had already argued for mention in the article).
 * In this analysis I count 8 articles (marked "supportive") which appear to me to agree that the RCTs are evidence that circumcision can have a protective effect against HIV, and two (marked "challenging?") which could perhaps be considered as challenging those results or their implications in some way but which I don't think go so far as to support the words Tremello inserted into the article, and one marked "challenging" which I have classified as challenging the validity of the results although I have only accessed a one-sentence summary.
 * This analysis suggests to me that the supportive interpretation of the results of the RCTs should be given considerably more weight in the article than opposing views, and that possibly the words Tremello added might not represent a sufficiently notable viewpoint for inclusion in this article but that a more nuanced opposing POV should probably be included (e.g. maybe not claiming that the results are not valid, but raising concerns about safety, other effects, whether there is sufficient reason at this time to take action, whether it applies in the US etc.) I also looked at a few of the references Tremello gave to support that point, and I think "challenge the validity" may not be an accurate representation for at least some of them, or may need to be made clearer as to what is being challenged (the results? the trials themselves in some sense? the conclusions of the authors of the trials?): For example, the ref by Garenne says "The trial was certainly well conducted, and it nicely confirmed observational studies, which came to the same conclusion [2]. However, a number of their concluding statements deserve a comment."
 * I agree with Jakew that the reader can see the superscripts and the references section and that putting a larger number of references can give the reader an impression of weight, so we shouldn't put too many references for minority POVs in this summary article; more detailed info can go in the subarticles. Given the selection of refs for the mainstream viewpoint, listing 6 refs for the challenging viewpoint is way too many in my opinion. (involved editor) ☺ Coppertwig (talk) 17:37, 27 July 2009 (UTC)
 * Which is the minority POV? I have done a quick review of the first recent articles I could access via Scholar:


 * (supports thesis) Male circumcision reduced HIV incidence in men without behavioural disinhibition. Circumcision can be recommended for HIV prevention in men.


 * (supports thesis) Male circumcision significantly reduces the risk of HIV acquisition in young men in Africa. Where appropriate, voluntary, safe, and affordable circumcision services should be integrated with other HIV preventive interventions and provided as expeditiously as possible.


 * (supports thesis) Infections were 60% fewer in the treatment group, which seems to indicate that circumcised men are much less likely to become infected with HIV when having sex with infected women. In communities where HIV is common, circumcision may prove to be a valuable tool for reducing men's risk of getting infected. However, as with most studies, criticisms could be made of some aspects of the methods used, and more research is needed before we can be sure. We must also remember that circumcised men can still become infected, even though the risk might be lower. They should still take other steps to prevent themselves from getting HIV.


 * (no conclusion) No clear conclusion made - notes results look positive, but cannot be certain until mortality rates can be established.


 * (no conclusion) Modelling based on assumed outcomes of trials - discusses difficulties of circumcision programs.


 * (no conclusion) Discusses cost benefits of such programs


 * (no conclusions) Rveiews the 13 papers on issue, highlights the popularity of the practice in Africa, and details the limitations in existing studies as well as discusses the potential of poorly trained practitioners.


 * (refutes the thesis) Research did not find correlation in different population (not sub-Saharan Africa).


 * (no conclusions) Accepts the thesis, and looks at implications for treatment for people not part of the research group (e.g. MSM).


 * (no conclusions) Reviews research, which it acknowledges supports the thesis, but needing more evidence to extrapolate to other societies.


 * The papers about research on heterosexuals in sub-Saharan Africa support the thesis. The sole paper on a different cultural setting refutes the thesis.  The rest review the findings, find some problems with  sampling/methodology, and most suggest that the findings support the practice for heterosexual males in sub-Saharan Africa, but that more research would be needed in other cultures and amongst different populations (e.g. homosexual males) and that the findings cannot be generalised.  One UK paper suggests that for males coming to the UK from sub-Saharan Africa, the procedure would be supported, but not for indigenous gay males.


 * To report these findings as anything other than locational would be undue - and most of the reports that have not had corresponding results are from populations not covered by the research. The way evidence works, the refutation of a thesis should carry more weight than its denial - that is basic scientific method.  You cannot exclude research on the basis that there is less refutation than correlation, as it is the refutation shows the thesis is flawed. Once the thesis is re-stated in a more specific form, then it can be tested more widely, and so on.  All that can be said at present is that there appears to be a benefit for heterosexual males in sub-Saharan Africa.  Not that circumcision generally has any benefit.  So, the POV being expressed relates to a minority of the global population. Mish (talk) 18:41, 27 July 2009 (UTC)


 * Thanks for your input, Mish. Could you please clarify a couple of things: What is the "thesis" you're referring to?  What is the difference between "refutation" and "denial", or did you accidentally say "denial" when you meant something else? (I interacted with Mish a couple of days ago while I was responding to an RfC on another page.) Thanks again, ☺ Coppertwig (talk) 18:52, 27 July 2009 (UTC)


 * Apology - I was rushing due to other things I need(ed) to attend to. I meant to post this under the RfC. The 'thesis' is that male circumcision is an effective preventative measure against HIV infection (it is a thesis, as circumcision is a novel surgical procedure in this context - using genital surgery as a means of treating/preventing infection).  The research cited is in sub-Saharan Africa.  When somebody puts forward a thesis, other researchers test the hypothesis, to see whether the results are confirmed, if they don't, this refutes the thesis.  I think I may have used denial when I meant refutation - as I edited what I had started to write to make it more in line with how theories are proved/disproved.  Perhaps it would be helpful, rather than presenting positions as views, in that way.  The research shows that the hypothesis appears to hold in one context, but not in others, and those who review the research acknowledge that while this new form of treatment may be beneficial in that context, there is no evidence that this can be generalised. Mish (talk) 20:05, 27 July 2009 (UTC)
 * Most studies have assessed more specific questions than that, such as "is circumcision effective in reducing the risk of female-to-male transmission of HIV". The events that led to this RfC involved the statement that "Some have challenged the validity of the African randomized controlled trials [studying female-to-male transmission], prompting a number of researchers to question the effectiveness of circumcision as an HIV prevention strategy".  In any event, I'm very uncomfortable about trying to decide which viewpoint is correct &mdash; I don't think that's our role as Wikipedians; all we can do is to report on viewpoints in proportion to their prominence.  Jakew (talk) 20:36, 27 July 2009 (UTC)
 * Right. We're not testing which is correct, but how much prominence they have. The refs such as WHO don't state (as far as I was able to find) that circumcision is effective in sub-Saharan Africa; they simply state that it is effective.  So I'm not sure to what extent we can justify qualifying such statements here.  We might be able to state in a separate sentence that others have pointed out that it may not apply to the US or whatever, or we may be able to imply context: at least one of the refs stated in a separate paragraph that the report was focussed on sub-Saharan Africa. ☺ Coppertwig (talk) 20:43, 27 July 2009 (UTC)
 * Yes, reviewers either discuss the relevance for those outside the sample - homosexual males, people outside sub-Saharan Africa, women - and whether the sampling/methodology is adequate (especially as the research was not completed) - and how research in other situations has not produced the same findings. Because this is all discussed in relevant reviews, and released after the sub-Saharan reports, it is pertinent.  If you are looking for prominence, neither the original papers nor those that refute the findings are the most prominent, but reviews which acknowledge the significance of the findings, but point to the difficulties in extrapolating the findings beyond the context - so this has to be given as much weight as the sub-Saharan research.  The refutation should be acknowledged, but not accorded as much weight as these, as there have been  fewer papers that have reported contrary findings - but they are significant.  This has nothing to do with whether the research is correct, but about the weight the different perspectives should be given.  If anything, reviews and publication by WHO should receive more weight than the original papers, as these are effectively secondary sources, which guidelines suggest are preferable to primary sources (such as the original papers). Mish (talk) 23:11, 27 July 2009 (UTC)
 * I agree that reviews, policy statements, and meta-analyses are secondary sources and hence ought to receive the most weight. I can't speak for Coppertwig, but the purpose behind my sample of the literature was to try to ascertain, in a reasonably unbiased manner, the prevalence of various viewpoints in the literature.  The results show, as do Coppertwig's, that those who "[challenge] the validity of the African randomized controlled trials [studying female-to-male transmission]" are in a small minority.  I did not limit my results to reviews or meta-analyses; having done so (using PubMed's 'limits' tab), the first page of results contained:
 * 14 articles which were positive with respect to a protective effect against F->M HIV transmission:, , , , , , , , , , , , ,
 * 1 which was about female genital cutting, hence may be excluded:
 * 1 which was exclusively about MSM (hence not F->M), hence may be excluded:
 * 1 which was apparently negative, though perhaps not to the extent of challenging validity of RCTs:
 * 3 which were impossible to classify from the abstract (if present):, ,
 * For clarity, I should note that I'm saying that these articles are positive, but not necessarily gung ho. Many do list caveats.  Many do advocate some degree of caution.  But the more extreme viewpoint that circumcision is not protective against F->M HIV transmission, that the RCTs were invalid, etc., seems to be that of a small minority at best.  Jakew (talk) 09:23, 28 July 2009 (UTC)
 * Right! So we may be asking the wrong question. Instead of how much weight to put on each of two opposing viewpoints, we can ask how to represent the various nuances expressed by the articles, and try to avoid giving the impression that an unqualified positive viewpoint has any more weight (if any) than justified by the RS. ☺ Coppertwig (talk) 21:52, 28 July 2009 (UTC)
 * That seems a good suggestion Coppertwig. Maybe we should create a draft to try out ideas and see if we can satisfy everyone: Talk:Circumcision/HIVsectiondraft.  Maybe we can also agree on the purpose of the section.  Is it to just list studies or is it to discuss the social implications of the studies?  Mish has talked a bit about this above. Tremello22 (talk) 18:53, 29 July 2009 (UTC)

Confusion over versions of RfC
I suggest that in future, when we do RfCs, we discuss and agree on the wording of the question before the RfC starts. ☺ Coppertwig (talk) 18:32, 27 July 2009 (UTC)

Request for comment
Regarding circumcision for HIV prevention, what is the appropriate balance of critical and supportive references to balance WP:NPOV and WP:UNDUE? Please see above for details. -- Avi (talk) 14:47, 27 July 2009 (UTC)

Tremello's original RfC question was "The topic is circumcision for HIV prevention. One wikipedian thinks the addition of more critical references will create a NPOV problem ; another member thinks it won't cause a POV problem but will enhance the article. See above discussion for more info". Inserted after Mish had replied below) ☺ Coppertwig (talk) 18:26, 27 July 2009 (UTC)


 * From looking through what is on the article, it appears that there is no firm conclusion either way, so you ought to represent this, citing whatever sources are necessary, and mentioning any notable organisations in relation to their position(s) on the debate. You also need to draw a clearer distinction between individual benefits and epidemiological benefits in relation to the populations affected.
 * For example, what benefits there are (if any) of circumcision in the following situations:


 * HIV- circumcised man who penetrates HIV+ woman
 * HIV- uncircumcised man who penetrate HIV+ woman


 * HIV- woman penetrated by circumcised HIV+ man
 * HIV- woman penetrated by uncircumcised HIV+ man


 * HIV- circumcised man who penetrate HIV+ man
 * HIV- uncircumcised man who penetrate HIV+ man
 * HIV- man penetrated by circumcised HIV+ man
 * HIV- man penetrated by uncircumcised HIV+ man


 * And how that is seen as translating to epidemiology and the social benefits beyond individual protection (if any). That is what I would be interested in, as a reader. Mish (talk) 16:36, 27 July 2009 (UTC)

Clarification on the addition of 4 references
''It seems there was (still is?) confusion over what I see as a problem and the reason why I though the 4 references enhanced the article and didn't cause a POV problem. I have been discussing what the problem is on my talkpage with Avi and Coppertwig:User_talk:Tremello22. I'll continue the discussion here:

''I was discussing with Avi POV balance in general. I asked Avi why he thought the addition of 4 critical studies created a problem.''


 * Balancing sources and references so as not to unduly emphasize one side or the other outside of the existing penetration each has in the literature. -- Avi (talk) 19:25, 27 July 2009 (UTC)

Avi ,I go by common sense and the wikipedia guidelines.

PRO:

Over forty epidemiological studies have been conducted to investigate the relationship between circumcision and HIV infection.[143] Reviews of these studies have reached differing conclusions about whether circumcision could be used as a prevention method against HIV.[144][145][146][147]

Experimental evidence was needed to establish a causal relationship between lack of circumcision and HIV, so three randomized controlled trials were commissioned as a means to reduce the effect of any confounding factors.[147] Trials took place in South Africa,[148] Kenya[149] and Uganda.[150] All three trials were stopped early by their monitoring boards on ethical grounds, because those in the circumcised group had a lower rate of HIV contraction than the control group.[149] The results showed that circumcision reduced vaginal-to-penile transmission of HIV by 60%, 53%, and 51%, respectively. A meta-analysis of the African randomised controlled trials found that the risk in circumcised males was 0.44 times that in uncircumcised males, and that 72 circumcisions would need to be performed to prevent one HIV infection. The authors also stated that using circumcision as a means to reduce HIV infection would, on a national level, require consistently safe sexual practices to maintain the protective benefit.[151] An updated Cochrane review concluded: "Inclusion of male circumcision into current HIV prevention measures guidelines is warranted, with further research required to assess the feasibility, desirability, and cost-effectiveness of implementing the procedure within local contexts."[152]

As a result of these findings, the WHO and the Joint United Nations Programme on HIV/AIDS (UNAIDS) stated that male circumcision is an efficacious intervention for HIV prevention but should be carried out by well trained medical professionals and under conditions of informed consent.[8][13][153] Both the WHO and CDC indicate that circumcision may not reduce HIV transmission from men to women, and that data is lacking for the transmission rate of men who engage in anal sex with a female partner.[13][14] The joint WHO/UNAIDS recommendation also notes that circumcision only provides partial protection from HIV and should never replace known methods of HIV prevention.[153]

ANTI:

Some have challenged the validity of the African randomized controlled trials, prompting a number of researchers to question the effectiveness of circumcision as an HIV prevention strategy.[154][155]

Does this make you see my point a bit better? Tremello22 (talk) 19:33, 27 July 2009 (UTC)


 * I think the point you're making is that there is much more text (therefore much more weight) given in the article to the position that circumcision can reduce the risk of HIV, than to the position that there is still (after the RCTs) not enough evidence to conclude that circumcision can reduce the risk of HIV.
 * However, in order to argue for a change to the article, you would need to argue that this difference in weight doesn't correspond to the relative weights in the sources. See the three literature searches (by Jake, me and Mish) at Talk:Circumcision and the subsection following it; I think they show that in the sources, there is much more weight given to the former than to the latter, so it's not at all clear to me that any change is needed. ☺ Coppertwig (talk) 19:40, 27 July 2009 (UTC)


 * First, let me say that it is incorrect to place the following sentence in either the PRO or ANTI sections: "Over forty epidemiological studies have been conducted to investigate the relationship between circumcision and HIV infection.[143] Reviews of these studies have reached differing conclusions about whether circumcision could be used as a prevention method against HIV.[144][145][146][147]". That is an introduction. Second, WP:UNDUE requires weight and space in proportion with the penetration the positions have in the literature outside wikipedia, as I said above. The evidence being brought now indicates that the literature is not as balanced as I originally thought, and that proportionate weight should be given to the positions that are more supported by the literature, within bounds of balance. I've always said that I allow myself to be convinced by suitable and well-founded arguments. Jake and Coppertwig have brought evidence; by all means, please respond in kind. Another suggestion may be that perhaps you should post a "block" on the talk page of how you perceive a balanced section would be, and we can take it from there. -- Avi (talk) 20:02, 27 July 2009 (UTC)

Actually the introduction could be counted as procirc-biased because it doesn't say where they intended to use circ for HIV prevention. It is for places where the HIV rate is high; so primarily in Africa, not developed countries. Therefore it is wrong to assume that it was ever considered to use it in general, in all countries. It is subtle so I understand why you wouldn't spot it.

My basic point is that if people want to read the opposing view they have a choice of more than 2 studies to look at:

Studies included already: Additional reviews:
 * Cautious optimism for new HIV/AIDS prevention strategies. Siegfried 2006 Lancet - Comments on South African control trial saying it could over-estimate effect because trial stopped early.
 * Male circumcision and HIV prevention: is there really enough of the right kind of evidence?, Dowsett 2007 RHM
 * Neonatal circumcision does not reduce HIV/AIDS infection rates, Sidler 2008 S.AfMJ
 * Rolling out male circumcision as a mass HIV/AIDS intervention seems neither justified nor practicable, Myers 2009 S.AfMJ
 * Male Circumcision and HIV Control in Africa, Garenne 2006 PLoS
 * Male circumcision is not the vaccine we have been waiting for,Green 2008 FHIVT

These review studies discuss circumcision to prevent HIV in general as opposed to just criticising the RCT's. That was my main reason why I wanted all 6. The current HIV section creates the impression that circumcision is a worthwhile measure in combatting HIV in all countries. The question of whether circumcision is a good measure in tackling HIV in all countries is the question you should be searching pubmed and Google Scholar for, not just whether the 3 African RCT's are valid. If we are calculating the correct POV balance on this question (is circumcision is a good measure in tackling HIV in all countries) then it would be more 50/50, rather than balanced in favour of circumcision.

After my edit little has changed:

Some have challenged the validity of the African randomized controlled trials, prompting a number of researchers to question the effectiveness of circumcision as an HIV prevention strategy.[154][155][156][157][158][159]

It doesn't create an incorrect POV at all. What has happened is that in little space we have provided a lot more info for the reader to peruse if he/she so wishes. Tremello22 (talk) 21:05, 27 July 2009 (UTC)


 * I'm under the general impression that there is a fairly prominent viewpoint in RS that the results from Africa can't necessarily be assumed to apply in other countries. However, at the moment I don't know where to find any refs that state that, except one: the 18th hit (Millet et al) in my list above, but I think it's just a primary source.  Can you find any, Tremello? ☺ Coppertwig (talk) 21:22, 27 July 2009 (UTC)
 * Millet is an important study for the US because it looks at what effect, if any, circumcision will have upon a first-world country. Moreso than the African studies, which don't apply directly to first world countries because of differences in availability to healthcare and hygiene, as well as sexual practices. It is doubly important since it all of the authors are from the CDC and the study was funded by the CDC. For that reason, it needs to be given additional weight, if not here, then in the CDC discussion. Their lack of significant association from circumcision on HIV infection is in keeping with the common-sense observation that in the US, where 75% of men are already circumcised, the HIV rate is not lower than in, say, Europe, where most men are not circumcised. Frank Koehler (talk) 22:29, 22 September 2009 (UTC)
 * The CDC say in their "Considerations for the United States" section that there are certain differences between African and US settings. The main problem they outline, however, is not that the protective effect wouldn't exist, but that the absolute risk (and hence absolute risk reduction) is so much lower in the US that the risk:benefit balance is considerably different.  It's important to remember that there are two questions: whether it has a protective effect (which is a purely scientific question), and whether it should be done (sub-questions of this being 'where should it be done' and 'among whom').
 * On another point, none of the "additional reviews" Tremello22 identifies are actually reviews. Sidler is labelled as "OPINION". Myers is labelled as "EDITORIAL". Garenne is labelled as "Correspondence". And Green is identified as an "EDITORIAL" (as an aside, there was a fascinating response to Green et al. here). This isn't by itself an argument for or against (though in general we should prefer reviews over opinion pieces, and meta-analyses/systematic reviews over plain reviews), but accuracy is important. Jakew (talk) 09:45, 28 July 2009 (UTC)

What is the status of this discussion?
It seems to me nobody is adressing the question. We have had one outsider come in and recognise problems with the section. I expected that because I feel the same way and I trust my opinion. It seems to me that discussion has turned to major changes (or even a re-write?) to the section as opposed to just tagging on 4 additional references. I would have been happy with the additional references giving the opposing point of view. Maybe we can keep the additional references and someone can come up with a better sentence leading up to the references? Tremello22 (talk) 19:32, 28 July 2009 (UTC)
 * Do you understand the reasons that have been given why putting in those additional references would give undue weight to a minority POV? ☺ Coppertwig (talk) 01:17, 29 July 2009 (UTC)
 * (The above question has been recopied below by Tremello) ☺ Coppertwig (talk) 00:15, 5 September 2009 (UTC)

Summing up this discussion
Coppertwig, I'll answer your question and try to sum everything and list where I think the difficulties have arisen. I'll start a new section so everyone doesn't get lost.


 * Do you understand the reasons that have been given why putting in those additional references would give undue weight to a minority POV? ☺ Coppertwig (talk) 01:17, 29 July 2009 (UTC)


 * How we assign weight (i.e. text, and to a lesser extent superscript refs) is dependent on what the purpose of the section is.
 * A)Is it to just report the results of the RCT's? By this I mean figures.
 * B)Is it to list what major organisations say as a result of these trials?
 * C)Is it to i) mention the flaws some commentators found ii) Discuss whether the flaws effected the results.
 * D)Is it to talk about circumcision's role in combating HIV in Africa?
 * E)Is it to talk about the effect circumcision has on lowering HIV levels in places with low HIV prevalence (i.e developed countries - such as the USA or the UK)?
 * F)Is it to talk about how HIV compares with other prevention strategies? This would be dependent on place of course.
 * There are probably other questions too. Namely, should we talk about the progress in scaling up circumcision?  How it is going etc.


 * So that is why I don't think doing a search at pubmed or google scholar answered anything. In order to progress we need to answer the above questions first.
 * Let me say, I already knew that more people agreed with the conclusions of the RCT's, that circumcision lowers HIV. That is why major organisations such as the WHO,CDC etc adopted that position.  They wouldn't have done so otherwise. After most observational studies showing a protective effect (albeit with confounding factors) 3 experiments showing a protective effect is proof enough for the scientific community that it does reduce it.  What is done with this knowledge, there is more disagreement on that.
 * The argument here was specifically over whether the addition of 4 more references created an incorrect POV balance based on the literature ( as said above, whether the balance is correct is dependent on the purpose of the section - maybe people didn't take that into account?). It was not about how much of a minority view is the view that the RCT's were invalid/whether circ reduces female to male transmission of HIV; I am not arguing over that.
 * I separated the sections into PRO and ANTI above and I think that demonstrated that given that I feel that the section has more than one role, it clearly doesn't create an incorrect POV balance if we are assuming the section is trying to answer the questions I listed above (C,D,E,F). Why? Because the POV balance on those questions (C,D,E,F) is not weighted in favour of circumcision as much as it is with A and B.  Maybe the disagreement is simply over the anti-circ sentence (Some have questioned the validity...)? If people are serious about discussing any of this then Talk:Circumcision/HIVsectiondraft is probably the best place to start.  Alternatively we can take the hassle free route and add the 4 additional references as an easy way of giving an approximate balance (PRO vs. ANTI) on all the issues (not just the simple question of whether circ reduces females to male transmission)  Tremello22 (talk) 19:55, 29 July 2009 (UTC)
 * Tremello, please bear in mind that references do not exist in isolation, but are closely linked to the text. Increasing the weight of a given sentence can only be justified if the viewpoint expressed in that particular sentence has a corresponding weight in the literature.  Thus, for example, the sentence "some argue that the earth is flat" should have a weight roughly proportional to the prevalence of that viewpoint in reliable sources.  It would be inappropriate to base it on the degree to which it is believed that the flatness or sphericality of the earth is relevant for interplanetary travel.
 * In this particular case, they would follow the sentence: "Some have challenged the validity of the African randomized controlled trials, prompting a number of researchers to question the effectiveness of circumcision as an HIV prevention strategy." This sentence has nothing to do with your points D, E, F, so it makes no sense to use any perceived imbalance in terms of those to justify it.  Instead, the sentence is primarily about point C, but this brings us back to where we started because &mdash; as noted above &mdash; the viewpoint that the RCTs are flawed/invalid is that of a distinct minority.
 * The primary purpose of the section is to answer the question, "what is the relationship between circumcision and HIV". Everything else is secondary to that question, and must be because if there were no relationship then other questions would be moot.  There are several important sub-questions, involving M->F transmission, F->M transmission, and M->M transmission.  And there are secondary questions, too: how is this affecting prevention policies, etc.  But the weight given to the viewpoint of those who dispute the very existence of the relationship (by claiming that the evidence is invalid or that circumcision is ineffective) must reflect the prominence of that viewpoint. Jakew (talk) 21:24, 29 July 2009 (UTC)
 * This is not about anything nearly as clear-cut as whether the earth is flat - it is about a relatively recent matter, which is still debated and there is no clear scientific consensus. Until there is, it is simply not accurate to minimise the significance of a position that is based on evidence that refutes the other view.  That would be giving undue weight to what is at present a falsified hypothesis.  To exclude details about contrary views would present an inaccurate picture, and be inaccurate. Mish (talk) 22:26, 29 July 2009 (UTC)
 * Nobody has suggested, Mish, that the contrary views should be excluded altogether. However, reasonably objective searches of all PubMed articles, Google Scholar, and PubMed reviews/meta-analyses, have shown that these views represent those of a small minority.  Consequently, they should be given relatively little weight in the article. Jakew (talk) 07:58, 30 July 2009 (UTC)
 * I'd just like to add that I have seen your argument about "giving undue weight to what is at present a falsified hypothesis", but I do not consider it a valid argument because WP:NPOV policy is concerned with the prominence of viewpoints, not with whether they are correct. Describing it as a "falsified hypothesis" violates both NPOV and WP:NOR.  It violates NPOV because it is merely the opinion of those with contrary views that the hypothesis has been falsified.  It does not follow that everyone would agree; others might consider their arguments fallacious or their evidence an improper test of the hypothesis, for example.  (As noted above, there is ample evidence that most authors do not consider that it is a falsified hypothesis.)  It violates NOR because it is not our role to present our own analyses of sources or scientific hypotheses.  Again, what we need to do is to assess the prominence of the viewpoints.  It is not our role to assert that one is correct. Jakew (talk) 08:24, 30 July 2009 (UTC)
 * I take your point, and this is the problem, that there is no firm evidence upon which to base a conclusion - and the best that can be said is that certain organisations have recommended a course of action on the basis of the evidence there is, which is itself disputed. Africa has a history of coming up with cheap solutions to HIV/AIDS that avoid the costs of retrovirals (for example) because of the costs involved.  This could just be another one.  The efficacy of such programs will not be known for a generation, and it seems that the agencies that have commented on this argue that it could be one approach out of several that combined could reduce transmission and infection - not as a treatment in its own right.  That is about the best you can say, and that is what is in the sources - nothing to do with whether it is correct.  You have to refer to the most up-to-date research, and that is not a group of papers about some research that began but was cut-short, and presented the incomplete research as findings, but more recent research which failed to generate the same findings.  Falsifying something is not a claim - it is a scientific process.  If X causes Y is claimed, and somebody does X and Y does not result, then X causes Y is not true.  That leads to modifying the hypothesis in a way that it can be re-tested.  Nothing to do with what is correct. Mish (talk) 09:30, 30 July 2009 (UTC)
 * Hi Mish. I don't think we're trying to draw any conclusions, are we?  I know that the article is written very carefully to avoid asserting a point of view, hence for example we say that WHO, etc., "stated that male circumcision is an efficacious intervention for HIV prevention" rather than saying so ourselves.  I think we're just trying to summarise, in a limited space, the most significant viewpoints on the subject, and in rough proportion to their prominence.
 * Regarding referring to the most up-to-date research, ideally we should cite meta-analyses of randomised controlled trials as these are considered to be the highest quality evidence (see WP:MEDASSESS). After that, in descending order, come randomised controlled trials and then lastly observational studies (preferably meta-analyses of obs. studies rather than the primary sources themselves).
 * Regarding falsifying a hypothesis, it is not always the case that a given study is clearly, unambiguously, and non-controversially a proper and conclusive test of a given hypothesis. Sadly, human studies are not always as simple as "somebody does X and Y does not result".  Many studies are observational (as opposed to experimental studies such as RCTs), in which nobody actually does X at all.  Instead, X has been done previously, and the study involves looking for correlations.  If found, one may speculate that the correlation is due to a causative effect between X and Y (and if not found, one may speculate that it is because no causative effect exists), but this is not conclusive evidence.  The apparent relationship may be due to confounding factors, or for that matter these factors may counteract and hence obscure the effect of a true relationship.  Even if an RCT found no significant difference, we must remember that these are statistical studies, so one would have to consider the possibility of a statistical error, consequently it may not constitute proof that the hypothesis is false (though it is certainly important evidence).  I realise that this is technically OR; I hope that you'll find it interesting enough to forgive me for that, and will understand why I consider "falsifying a hypothesis" in this context to be a claim rather than an objective fact. Jakew (talk) 11:26, 30 July 2009 (UTC)
 * Unlike in female genital cutting, male circumcision has not been the subject of government campaigns. Some non-government organizations have sought to reduce or increase the number of circumcisions for medical, religious or social reasons. Male circumcision can now be partially reversed via a surgical technique, which gives back certain sensation to the genitalia.

There are two main anti-circumcision frameworks: the health model and the human rights-based model. The health model campaign offers conflicting definitions of circumcision as either harmful to men's health (physical and psychological trauma) or possibly reducing the transfer of HIV. The human rights model encompasses three important human rights discourses: violence against infants, rights of each person to determine their own sexual pleasure and rights to health and bodily integrity. —Preceding unsigned comment added by Gebrelu (talk • contribs) 20:15, 7 August 2009 (UTC)

Purpose of HIV section
(Discussion started on Jakew's talk page) I think this boils down to what we see as the purpose of the HIV section. So in order to progress I think it would be helpful if you gave me a good idea of what you think its remit should be. Tremello22 (talk) 20:06, 30 July 2009 (UTC)
 * Haven't I already done so, in the last paragraph of this edit? Jakew (talk) 20:11, 30 July 2009 (UTC)

Yes but I think it is the key question so I wanted to make sure you had considered it adequately. I'll presume you have. I'll separate your answer and reply separately to the points you made. "The primary purpose of the section is to answer the question, "what is the relationship between circumcision and HIV"."

It depends on what you mean by relationship. Relationship could mean how relevant circumcision is to different communities in battling HIV could it not? I'll take an educated guess and presume mean you want to answer the question: does circumcision lower HIV transmission at an individual level, and if so by how much? You would answer this by giving the results of the RCT's. The reader would get the answer : Yes. By 60%, 53%, and 51%. I guess the next question is, why is the primary purpose to give these figures without any context? Who decided this was the primary purpose? If it was you, how did you arrive at this assumption? The answer to the question by how much? would depend to a great degree on the individual's behaviour and his environment. These RCT's can only compare circed group with non-circed group. They can never know that circ reduces HIV by this amount.

"Everything else is secondary to that question, and must be because if there were no relationship then other questions would be moot." Not true. If there was no relationship, we would likely still document that studies had been done to investigate whether there was a relationship. We would then state there was no relationship.

"There are several important sub-questions, involving M->F transmission, F->M transmission, and M->M transmission." "secondary questions, too: how is this affecting prevention policies"

The question is how much weight do we assign to these "sub-/secondary" questions, if indeed we can describe them as such.

'''Change/expand the sentence? ''' The sentence in question:
 * "Some have challenged the validity of the African randomized controlled trials, prompting a number of researchers to question the effectiveness of circumcision as an HIV prevention strategy."

"But the weight given to the viewpoint of those who dispute the very existence of the relationship (by claiming that the evidence is invalid or that circumcision is ineffective) must reflect the prominence of that viewpoint."

I think this boils down to the sentence(s) that comes before these references. We could include the references elsewhere but in different contexts for example, commenting on: 1) How these results apply to countries with low HIV prevalence 2) How effective male circumcision is compared to other HIV prevention measures 3) etc.

Once we agree on the purpose of the section we can do this easily. Tremello22 (talk) 22:37, 30 July 2009 (UTC)


 * Regarding "does circumcision lower HIV transmission at an individual level, and if so by how much", your educated guess is essentially correct. You are not quite correct, however, when you suggest that I "would answer this by giving the results of the RCT's".  As I pointed out, the question can be divided into sub-questions: M->F transmission, F->M transmission, and M->M transmission.  The first is best answered by Wawer's RCT.  The second is best answered by meta-analyses and/or reviews of the South African, Ugandan, and Kenyan RCTs. The third is best answered by Millet's meta-analysis of observational studies.
 * You then ask, "I guess the next question is, why is the primary purpose to give these figures without any context?   Who decided this was the primary purpose?  If it was you, how did you arrive at this assumption?"  I think I've already explained why: "[it] must be because if there were no relationship then other questions would be moot."  In response to that, you say, "If there was no relationship, we would likely still document that studies had been done to investigate whether there was a relationship."  To put this another way, we would document the studies asking whether there was a relationship between circumcision and HIV, and then stop.  We wouldn't go on to discuss its use in prevention policies (for example) because nobody would suggest that it had a role in prevention programmes, and that would be because it had no effect.  As I said, other questions are secondary, and depend upon the primary question, being whether it has an effect.
 * You then ask, "The question is how much weight do we assign to these "sub-/secondary" questions..." The answer should be obvious: we weight them in proportion to the weight given to them in the literature.
 * Finally, I'm not opposed to changing the sentence, but I would expect that if the sentence is changed to represent a more mainstream viewpoint then other, higher quality, sources will be a better choice to support that sentence. Jakew (talk) 09:20, 31 July 2009 (UTC)
 * Maybe its observed effect is due to "irreducible confounding." Comments welcome.  Blackworm (talk) 03:43, 1 August 2009 (UTC)
 * Why don't you guys just write in the statistics from some of these studies or a meta-analysis? The results pretty much speak for themselves. The practice of circumcision is often an effective means of preventing the spread of HIV transmission, but - like all modes of protection - isn't a guaranteed mode of protection. Djinfidel (talk) 19:59, 12 August 2009 (UTC)djinfidel (PS) I'm knew to Wiki so bear with me.