Talk:Haplogroup R1a/Archive 7

A constructive proposal
PB666 wrote above:
 * Couple of other points:
 * The page has grown by 10 kb just in the last week, its now becoming a hard read. I fell asleep in the origins section.
 * Some areas are becoming increasingly difficult to read, in and of themselves.

Great! So now there is at least a consensus that communication is not functioning on this talkpage, and this is blocking constructive work on the article, which was by the way not too bad according to nearly everyone. (Many of the places PB666 points to in his latest load of rem notes are places he has recently had changed.) No other consensus is clear, because the discussions are confused and too long, and most editors are staying away. PB666 of course accepts no blame for this, and there's no point trying one more time to point out the obvious WP:TALK violations which, if stopped, would automatically improve everything. I would therefore like to propose something much more simple. PB666 and I should voluntarily limit ourselves to 200 words per day on this talkpage. This paragraph is a bit over 150 words. Do you accept PB666? I believe an answer can be written in less than 5.--Andrew Lancaster (talk) 08:15, 1 December 2009 (UTC)


 * Not talk page, Main page.PB666 yap 13:29, 1 December 2009 (UTC)


 * The proposal here is about the talk page, which everyone agrees has big problems. What do you think?--Andrew Lancaster (talk) 14:04, 1 December 2009 (UTC)


 * I think that if there are no more major critiques the conversation will die down now. Although, I would be happy to see more critiques.PB666 yap 14:07, 1 December 2009 (UTC)


 * The word count on this page is currently dominated by you, so it is all up to you. Are you saying you will reduce what you write here or not?--Andrew Lancaster (talk) 14:12, 1 December 2009 (UTC)


 * Takes two to tango Andrew.PB666 yap 14:31, 1 December 2009 (UTC)


 * My proposal involves two. What is your answer to the proposal?--Andrew Lancaster (talk) 14:37, 1 December 2009 (UTC)

Another constructive proposal
I would also like to kindly request that PB666 purge the comments boxes on this talkpage. They are currently very obviously filled with personal notes, mostly out of date, and mostly in a form which only he would be able to follow. (I think most of it is just copied from what he pasted in various other places previously including the GA review page when he accidentally made himself the reviewer.) Of course I have no problem that people would do such personal note keeping but this would surely be more appropriate on PB666's User space somewhere? I believe these comment boxes, and indeed the gradings this article has been given, should not normally be filled in by someone currently busy with it? For example short and simple points could eventually be extracted, once PB666 has got his point clear in his own mind, and discussed here?--Andrew Lancaster (talk) 12:44, 1 December 2009 (UTC)


 * OK, No problem.PB666 yap 13:14, 1 December 2009 (UTC)


 * Thanks. Both proposals?--Andrew Lancaster (talk) 13:17, 1 December 2009 (UTC)


 * All but three points.PB666 yap 13:21, 1 December 2009 (UTC)


 * Please explain.--Andrew Lancaster (talk) 14:00, 1 December 2009 (UTC)

OK, I see you reduced to three points only. Why not mention them here? These are actual understandable points now with real relevance to the article. Remaining comments about this page.
 * The map on the top of the page, and the text contradict each other, the origin map shows R1a and R1b coming from areas where neither have a strong presence. If this is true, then why is it not explained. For example Underhill said 'Origin of R1a remains elusive, Origin or R1a1 is probably close to R1a, but spread into South Asia also. What about R1* in Asia.
 * The map is a variation of the standard one someone made for Y haplogroups. Better to talk to the map maker, or else simply remove it or replace it. I think it is no big issue. On the other hand, it would be easy to defend this map:-
 * It is obviously a simple "artistic" abstract representation.
 * Nothing in genetics says that areas of low frequency can not be a point of origin.
 * Actually, though I am not sure the starting point is a low frequency point. The northern Caucasus has areas of high R1a and high R1b?
 * You say that the diagram contradicts the map, but actually this is also not clear. The starting point coming out of the Caucasus, is coming from the direction of "Asia".
 * But for R1a* the higher density areas is to the south. Its relatively broad, and as for spread, according to Klysovo, at least one branch must have spread to china or India, first.
 * I have no big problem with the map because I understand the complexities and its hard to really say anything specific, but other readers might note the inconsistency.PB666 yap
 * I am going to remove this issue from the comments, but we should be working toward generating a more representative picture, has anyone contact the editor who added this yet?PB666 yap 16:53, 1 December 2009 (UTC)
 * As far as I know this is something only you have raised any issue about.--Andrew Lancaster (talk) 18:02, 1 December 2009 (UTC)


 * Info box description in "defining mutations" should be reduced.
 * De gustibus non est disputandum.
 * The issue is not taste, the issue is to simplify and make more readable. Is it necessary to present all the branch-points in the section, most pages only identify the basal node. This is a minor point, however.PB666 yap 16:16, 1 December 2009 (UTC)
 * As discussed in the body, for this haplogroup we can quite literally say that "R1a" (the articles name) has different meanings out there in the real world, meaning people coming to this article (perhaps after having tried to read a genetics article for the first time, such as the ones Family Tree DNA has on its website for its customers) will need help to find an initial anchor point. This is a complexity out there in the real world which we need to handle somehow. The infobox is a place especially designed for finding those initial eye-brain-anchors?--Andrew Lancaster (talk) 16:31, 1 December 2009 (UTC)
 * Ah, that problem, how about then R1a and R1a1 only. Since most papers apply R1a as either R-M420 and or SRY1532.2.PB666 yap 16:47, 1 December 2009 (UTC) We do discuss R1a1a and R-M17 in the Lede.
 * How about deleting this: "3. M17 and M198 (equivalent to one another) define R1a1a, which is by far the most common type of R1a."?PB666 yap
 * Remember when you split the article and one of the problems was that a lot of what you thought was about R1a was about R1a1a? That's because this is how the article used to be - completely mixed up. And that in turn is because that is how a lot of reading matter out there is written. —Preceding unsigned comment added by User: (talk • contribs)
 * As a matter of fact, no, I knew it was mixed up, it was going to take time to separate the old from the new studies and old usages from new, I worked on that article for a few hours or so. Secondarily I though you wanted to limit talk page size. If you look at this page it is generally you always making the last comment in a hierarchy. The issue of discriminating R1a has nothing to do with R1a1a, Klyosov interchanges R1a with R1a1 several times in his paper. But that doesn't make a difference here, because most of the literature that is of relevance has R1a = R-M420 or R-SRY1532.2 no-one in recent literature calls R-M17 basal R1a, and if they do, that's a different issue. I think it looks cluttered, if I think it looks cluttered then at least 40% of the audiance will feel the same. So stop arguing about the issue.PB666 yap 20:37, 1 December 2009 (UTC)
 * Good on you for picking me up on that missing signature. In the above paragraph "most of the recent literature" really only concerns a couple of very recent articles, which are the ones that you know about this subject. And this is part of the reason you do not know what 40% of people will understand or want.--Andrew Lancaster (talk) 20:50, 1 December 2009 (UTC)


 * I am looking at the infobox and its presentability, I noticed that other info boxes use the class="infobox vcard" style="width: 22em; font-size: 88%; line-height: 1.5em; text-align:left" as the table format. However in our infobox the tabled information is text < /small> this is difficult to read as it is, but by adding alot of text you are creating an unreadable wall of words, which those reader are not likely to look at.PB666 yap 17:28, 1 December 2009 (UTC)
 * I do not really see what the problem is which is why I mentioned taste. It seems like just a matter of taste?--Andrew Lancaster (talk) 18:02, 1 December 2009 (UTC)
 * Improving quality also means improving appearance. I notice that the infobox page for this box has two unanswered critiques. Maybe it was an impromptu template that was never refined. In any case it what was meant to be a few loci, not 3 sentences, and it looks cluttered.PB666 yap 20:19, 1 December 2009 (UTC)
 * Of course we should improve appearance. But whose taste is the right taste? How do you think people handle questions like that normally on Wikipedia?--Andrew Lancaster (talk) 20:50, 1 December 2009 (UTC)


 * "Studies which have argued this case most strongly include Sengupta et al. (2005), Sahoo et al. (2006), and Sharma et al. (2009). Studies which have concluded that the data is at least consistent with this scenario include Kivisild et al. (2003), Mirabal et al. (2009) and Underhill et all. (2009)." cannot statements like these be simplified?
 * The sentences are grammatically very simple. So what I guess you mean is that this looks like excessive detail? I can say that nevertheless it was controversial that I stripped it down to this basic summary, leading to a certain editor User:Cosmos416 (who has since tried sock puppetry) being blocked. I think, based on the history of this article, that it is really necessary to make sure we are super neutral, and that we avoid creating edit wars.--Andrew Lancaster (talk) 14:59, 1 December 2009 (UTC)
 * Yes, but is there not a better way to do this that is more simple.PB666 yap
 * You mean without deleting those names of articles? I do not see it, but there might be a way. As mentioned, the grammatical structure of the sentences is actually very simple.--Andrew Lancaster (talk) 16:31, 1 December 2009 (UTC)

Confidence of molecular clocking using STR
[I added this section break as this section is oversized, I am limiting my responses to Andrew who has become WP:ARGUMENTATIVE and WP:IDIDNTHEARTHAT on this issue].PB666 yap 18:09, 2 December 2009 (UTC) Adding one more point
 * "With enough STR markers to compare, the chances of falsely identifying relative relatedness because of coincidentally similar haplotypes becomes trivial."' This sentence has recently appeared, I appreciate the effort to explain STR but the problem with the sentence is that it is opinion.
 * Reversions within a single STR cannot be detected without adequate branching. Reading Klyosov, about 2/3rds of STRs will have lost definition within 64,000 years. Errors at the surface, for example within 500 years and within very bushy trees will be 'trivial' however errors deeper in the tree and in regions were exclusion has occurred (exclusion = genetic drift and population size acting on and reducing allelic diversity) may have significant losses. For example Klyosov mentions that the first wave of Europeans were spread thinly from Bulgaria to Germany for 1000s of years, these are the types of situations were STR diversity can be lost. It is only through intense study of local clusters (around each R-M17 find) that this loss of STR can be detected. Ergo it may or may not be trivial. Without clear branching analysis correct estimation of diversity may be crippled. Reverse mutations at this level are rare, but when they occur, at this level they have profound effects. Klyosov specifically mentions the reverse mutation problem (not only reverse mutations however, also repeated homoplasies can cause the same problem).
 * In addition to this source of variation, Klyosov mentions and I concur, that basal clades and entry diversity is a problem, basal clades that were generated prior to migration but are not obviously entry clades may make younger populations look older, where as genetic drift may make older populations look younger. The problem with most studies is that there are sources of variation (such as random rates of change) that can be accounted for, but there are other sources of variation cannot be accounted for. Rich trees and large haplotypes are ideal situations, but most of the time, particularly intercomparing groups, this is not available, and therefore the confidence of assigning entry level diversity to a modal clads is also not trivial. I should also note that other authors (for example the website you referenced) suggest the rates have changed, looking at Klyosov he points out in the first paper a range of rates from 0.14 to 0.28 for different STR, it is unknown how much of this is STR specific variance, random variance, or clade specific variance.PB666 yap 16:46, 1 December 2009 (UTC)
 * Can you please consider my proposal in the above section about word limits? Will you tango with me? You still have not answered, and this post is again quite problematic. The first sentence shows that you did not even read the first words of the sentence you are commenting on. So this mass of words (360 words, significantly more than my proposed daily limit) is all irrelevant, all off on a tangent, discouraging and ruining all possible discussion. Holding yourself to shorter remarks would force us to read what you are responding to and what you are writing. The sentence above is not just opinion because "enough" is a relative word. It might be nice to find a citation, eventually.--Andrew Lancaster (talk) 18:02, 1 December 2009 (UTC)
 * Your writing is pretty much a distraction from the critique "the chances of falsely identifying relative relatedness" to me this "false" and "coincidentally" is translated into the consequence of either two things, reversions or homoplasies. So that in mentioning this, which is defined as a problem, not just by me, but by Soares et al, Klyosov, and a whole bunch of other folks who study rapidly evolving loci, that the issue is not trivial. In fact it is unpredictable, often invisible without adequate sampling, and its triviality is unattestable to. Second, since you are modifying the page, and inserting words that could be considered opinion, you have to also deal with the critique. Trivial is not applicable in this situation. Practically speaking, there is not going to be any other person critiquing this except me. I marked off this sentence in-situ some time ago and it was not reworded (see main edit page) and therefore the issue gets critiqued with all these words because it was unattended to. PB666 yap 20:19, 1 December 2009 (UTC)
 * Too many words, but I still did read. You are not correct in any meaningful sense. The sentence does not say how many that enough would be but we are sometimes now seeing papers with >60 STR markers. The chances of these problems, especially if we consider that we are here not talking about one individual's test, but a whole population, can be ignored. Not good enough for you, then let "enough" be 300 STR markers. The sentence only says that "enough is enough" and that is by definition correct, surely?--Andrew Lancaster (talk) 20:50, 1 December 2009 (UTC)
 * Unfortunately, this is not always true and the fact that you don't understand why makes the word "trivial" all the more problematic. You could use 1000 STR, however if you are selecting haplotypes from an area in which only one derivative lineage is detected, then a number of these STR could have undergone reversion undetected
 * The only thing that would improve with 60 versus 25 is the confidence interval would reduce because the Poisson distribution of reversions would collapse, the collapse = k / (#STR)^0.5 . But here again, no-one is doing 60 STR reliably yet, most of Klyosov is 17 to 25. Lets take an example. Suppose we have 25 sites all identical rates. Now suppose we have 10 mutations. We throw out the mutations 1 by 1, the state can be + or -. We seed the analysis by premising a branch a mutation, the other branch is very bushy and therefore can be resolved back to the seqMRCA. The first 1 thrown out will not revert anything, the second that is thrown out has a 1/50 chance of reverting the first, the third that is thrown out has a ~2/50 risk of the first or the second and so on in a Markov chain. There is good chance of a reversion that follows a Poisson distribution, and only by developing a Markov chain one can assess at the end of 10 selections the possible range of variants observed. In some cases you may observe 4 and in others 10. Then add a mixture of rate possibilities for each site, and repeat. At any point in the Markov chain, insert a branch(s), and continue on a 2nd Markov chain. In addition mid stream begin altering the mutation rate as observed. This is the way in which one determines the probability of missing a mutation. Adding to these sources of mutations, simply stated are binomial issues, rate errors, etc.
 * If you want to understand more about this read Endicott, P; Ho, SY; Metspalu, M; Stringer, C (September 2009), "Evaluating the mitochondrial timescale of human evolution", Trends Ecol. Evol. (Amst.) 24 (9): 515–21, doi:10.1016/j.tree.2009.04.006, . There are a few others.
 * This has nothing to do with the tiny probability of a COINCIDENCE of identical or near identical repeat results on STR markers, in a population. Surely you realize this? --Andrew Lancaster (talk) 23:39, 1 December 2009 (UTC)
 * P.S. 363 words. 50 would have done it?--Andrew Lancaster (talk) 23:41, 1 December 2009 (UTC)
 * The coincidence is immeasurable because these lines have reversed, Klysovo p.228 36 R1a1 covering 18 countries, that is 2 samples per country. 015 is a standalone chain, 012 has two basal branches and is stand-alone. Balkans has 4 standalone trees all of which are shorter. Klyosov claims there are no anomolies in the trees, just as Ingman et al 2000 stated he observed no signs of selection, 7 years later is a different story, people see what they look for, if they don't look for abnormalities in their work because it makes analysis easier, then, of course, they won't see it. You can argue with me all you want, but the precedences in molecular anthropology is that tight confidence windows are invariably overstated. Comments stay, you are not objective about Klyosov. Now go waste 2 hours counting letters.PB666 yap 02:47, 2 December 2009 (UTC)
 * The sentence you are complaining about has nothing to do with Klyosov. It just says that with enough markers, the chance of coincidence goes down a lot, and this is true. You have spent about a thousand words trying to argue against a tautology.--Andrew Lancaster (talk) 07:22, 2 December 2009 (UTC)

Correctly assessing variation is not "Trivial"
If man could build a rocket with enough rocket fuel we could easily go to alpha-centuari. Of to simplify to 16th century wisodm: If wishes were horses, beggars would ride. The problem is that no such rocket exists. Name one instance where 60 STR have been used in a global population survey.WP:CRYSTAL.PB666 yap 12:52, 2 December 2009 (UTC)


 * I believe there are now a few, and I just started to go looking but stopped myself. What am I doing? The number 60 has absolutely no relevance here. There is no discussion here about how much is enough, not in the article and not coming from you or me. The sentence only says "enough would be enough". By definition, this is true. "How long is a piece of string?" If you are claiming that researchers NEVER have enough STRs then the burden of proof is on your side because clearly the whole field disagrees with you. In any case, to make a general attack upon the use of STR markers in clustering, if that is your intention, would simply be WP:FRINGE (and just plain wrong), and in any case it is not something which should expounding upon here on this article talk page. You have used this url quite enough for your experimental "stream of consciousness" writing.--Andrew Lancaster (talk) 13:13, 2 December 2009 (UTC)
 * You could have responded in five words "I did not hear that", or "I am not going to change this no matter what evidence you present, and then I am going to grand stand for 200 words". The number of 60 is important because you brought it up, what I was reading they used 5, 6, 10, 17, and 25, back and forth. If you are not going to defend your statement you need to rewrite the sentence anyway it contained grammatica errors. Second I can site numerous paper after paper in molecular anthropology where one moron comes and says I attach this genetic change to this event +/- few percent error, and the next author comes along and blows his behind out of the water. I have blown a few of these out of the water, myself. Confidence is about what we know, what we want to know and what we want to assess. Independent sources variation are cross multiplied to form a product variation, if the distributions are normal (unskewed; almost never true in molecular anthropology) then variation (SD) = (RV12 + RV22 + ..... RVN2)0.5. where RV = relative variance (SDi/Averagei) created by a single source of variation. Relative confidence (95.4%) is a range centered about the mean and 4SD wide. In the case of most molecular studies data has to be coded as natural logs and handled differently. In the vast majority of cases where all sources of variation have been calculated the relative confidence ranges are typically -0.5 to +0.75 relative to the mean. With extremely bushy trees, and very large number of events covering many different evolving sites this improves, but in tree that are a single lineage, such as are common in certain subpopulations with R1a1 those optimal conditions are not provided.  Ergo it is misleading to single out a single source and say it is "trivial" particularly when you cannot step up to the plate and prove it, what really matters is the product of all sources of variation. And if you don't believe there are many sources of variation in estimating Y-DNA TMRCAs then place your references defending your point of view and I will bring forth mine and we shall see what is trivial and what is not.
 * This is particularly relevant since Underhill (the biggest review this year with a large number of authors) uses a different dating technique, no-one really knows within +/-60% relative range when the Y-chromosomal MRCA is, and so on, and here you are sitting here saying that "clearly the whole field disagrees with you", OK so put-up or . . . . .PB666 yap 18:09, 2 December 2009 (UTC)
 * You wanna know why the Y-DNA pages are in such bad shape, one simple reason, all the worshipers at the alter of the Y-chromosome (Male divinity) that edit these pages are so blinded by the latest saying of their favorite pundit that they never sit down to check the facts and the history. That is the whole reason of the Kurgan edit-war, outsiders to Y-DNA have a whole different take (as we have seen). If going by precedences of accuracy alone, we should be 99% skeptical of everything in the field that is published. And why this fixation to begin with, even under non-selective conditions Y correlates with 33% of the rest of the genomes evolution, with cultural selection (hegemony of male surnames) that can go down to 0%, if you were to look at HLA looking in Northern Ireland, with 5000 people typed for genetic evidence of Northwest/central Chinese origin you would find no evidence of migration at all. Idolic genuflectations, nothing more. PB666 yap 18:27, 2 December 2009 (UTC)


 * You seem to have a grudge.--Andrew Lancaster (talk) 19:20, 2 December 2009 (UTC)
 * I have 25 years experience in a pattern of mis-assumptions about statistics that never seem to stop repeating itself.PB666 yap 04:20, 3 December 2009 (UTC)
 * People with 25 years experience in a vaguely related field can still have grudges, and can still be disruptive editors, and they can even be plain wrong. It is pointless to try to pull rank on Wikipedia, as has been pointed out to you by admins and other editors in the past.--Andrew Lancaster (talk) 08:27, 3 December 2009 (UTC)
 * As I will show you below, that basic principles that assign confidence apply to all things that undergo change at a statistically definable rate. I repeat, there is nothing magical about Y-DNA that makes it immune to the same sources of variance that any other loci have. Again, I have focused on the whole of molecular anthropology, not just one little cravat niche, where cliche norms have evolved as a cover for really relaxed standards.PB666 yap 02:06, 4 December 2009 (UTC)

Rest of section
We have reached a point where agreement cannot be reached via consensus. Let's just leave the comments as they are for now and when objectively discussing these becomes a priority I will be happy to discuss them as always. There, that will clean up the talk page for the rest of the day. PB666 yap 20:39, 1 December 2009 (UTC)
 * I take it this means you do not want to "tango" (as per your words for my proposal above)? It is a shame you could not even bring yourself to answer my proposal in clear and honorable words. Honestly, I find these comments a bit mysterious. What do you mean? You do not get to say that you'll be ignoring consensus on Wikipedia, if that is what you mean. And I believe that if you want to put a line under failed conversations the best way is to let them be archived and try to move forward, for example as per my proposal.--Andrew Lancaster (talk) 20:50, 1 December 2009 (UTC)


 * No Andrew, I am really tired of these discussions where you ask rhetorical questions as a cover for WP:IDIDNTHEARTHAT issues. All you have to do is look at any other Infoboxes on wikipedia to know that there is a substantive quality difference. You don't need to argue with me. On this one issue I am going to open up a thread in HGH about remodeling the infobox, that is an area of my expertise, anyway, but what goes into these boxes is a garbage in garbage out issue. Apply TLDR ro your infobox.PB666 yap 23:27, 1 December 2009 (UTC)


 * It was not a rhetorical question. I made a very simple and concrete proposal on how to work here on this talkpage, which is based on a self-critical listening to the voices of the third parties. That's what you keep telling me to do right? Please answer to the proposal. Can you not live without these long postings which go off-topic and destroy communication? Are you telling the consensus concern to go jump in the lake after having lectured me so many times on its importance?--Andrew Lancaster (talk) 23:39, 1 December 2009 (UTC)


 * I would argue that you are destroying communication by the WP:IDIDNTHEARTHAT attitude you are simply disguising the communication strategy to promote POV edits. So enough with this, when you decide you want to be objective, I have a talk page, don't abuse it. I can totally be off this page until you decide to do away with the WP:OWN and WP:IDIDNTHEARTHAT issue. You asked me to amend the comments, based on the improvements I did. I didn't ask you to start a novel about the remaining three. I just simply gave you a reason why these critiques will persist. Save the page, look around at the diverse pages on Wikipedia, particularly featured articles, when your in the mood to discuss change, talk to me, otherwise . . . . . . . PB666 yap 02:22, 2 December 2009 (UTC)
 * Do I make myself clear? I am not limiting my edits, I am going to limit to whom I respond to. That is the better way and the comments stay.PB666 yap 02:22, 2 December 2009 (UTC)


 * As usual this is full of references to un-defined things (which POV edit am I supposed to be promoting? what novel did I write?), and it is also clearly very inappropriate to write, as you often do, in terms of "do this or else". By the way I asked your to remove the comments, not just amend them. What was the point of them being in a comment box? Who is meant to read them there? You are not using talk pages correctly, as has been pointed out to you by admins, and also in past.--Andrew Lancaster (talk) 07:22, 2 December 2009 (UTC)
 * There is nothing I need to further respond here.PB666 yap 12:52, 2 December 2009 (UTC)
 * It would at least be much more honest of you not to pretend to be responding to people. You are a monologue writer by nature. You do not read what you respond to. You notice a few words and then start writing about your latest thoughts on those words. However, if you are going to quit pretending to be in a discussion I think it is also time that you took my advice and moved your blog to a more appropriate webspace. As announced at the top of this page: This is the talk page for discussing improvements to the Haplogroup R1a (Y-DNA) article.--Andrew Lancaster (talk) 13:18, 2 December 2009 (UTC)

Andronovo associated w/ Indo-Iranian
I have corrected this from Indo-Aryan. Please look over Indo-Iranians, Indo-Iranian languages, Indo-Aryan migration, in addition to Andronovo culture again. DinDraithou (talk) 18:26, 2 December 2009 (UTC)

Thank you DD, I noticed the same thing.PB666 yap 18:29, 2 December 2009 (UTC)


 * I stand corrected. Thanks.--Andrew Lancaster (talk) 19:19, 2 December 2009 (UTC)

Balto-Slavic are not 'closely' related branches to Indo-Aryan. Apart from the satem shift, they are not a sub-branch of a common Indo-Aryan, but rather formed a continuum from Iranic-Balto-Slavic- Germanic. Only thing is, the satem innovation actually reached Slavic and Baltic areas, not Germanic, which is more peripheral. Hxseek (talk) 19:30, 2 December 2009 (UTC)


 * Germanic is essentially similar to Italo-Celtic in most ways, with some of its own peculiarities and a few things in common with Balto-Slavic. Like Italic and Celtic, it retains a bunch of Early PIE features. AFAIK the Germanic case system is the most conservative in all IE. DinDraithou (talk) 19:46, 2 December 2009 (UTC)


 * Hxseek, that sounds right to me, but hopefully I did not imply that Balto-Slavic developed from Indo-IRANIAN. :) Maybe the wording is a bit weird because I try to be accurate concerning what authors say, silly me, without making them sound dumb, naive me. Klyosov clearly makes at least a near equation between PROTO-Balto-Slavic and PROTO-Indo-Iranian and because that is not really in itself wrong, but it is a strange way to put it, I have tried to explain it in a way consistent with his way, but not using his very unorthodox, and potentially divisive, words. Of course how you and DinDraithou explain it is more conventional. But would a reader looking this up after reading Klyosov get that? Anyway, please change the wording if you see the path.--Andrew Lancaster (talk) 19:51, 2 December 2009 (UTC)


 * Balto-Slavic and Indo-Iranian are particularly close. Ancient texts like the Rig Veda can be easily translated into Russian but there are substantial difficulties involved getting them to make sense in Western European languages. DinDraithou (talk) 20:10, 2 December 2009 (UTC)


 * What is that sentence they do: God gave teeth, god gives bread? Looks even better with Baltic which is presumably close to what they all looked like once.--Andrew Lancaster (talk) 20:28, 2 December 2009 (UTC)

Folks this is an area of discussion that belongs on another page, DD is right about the phraseology, Indo-Iranian is the more appropriate, Indo-Aryan is more colloquial, that is what the argument should be over. The particular argument above is like saying English is not a Germanic language because is has alot of latin influences or that it is celtic because Gaelic was once lengua franca of the Ilses, systematics has assigned the languages and their names and we should use the proper versions there-of.PB666 yap 21:44, 2 December 2009 (UTC)


 * No there is no argument here, phraseology is also the wrong word, and Indo Aryan is a real technical word for a branch of Indo Iranian, not a colloquial term. Furthermore this is one of the first times for a while that several editors are using this tlk page to talk about the article: It is not for another page. Will you allow other editors to work here?--Andrew Lancaster (talk) 22:00, 2 December 2009 (UTC)


 * But it is also used colliquially to describe Eastern Indo-European languages, supposedly derived from the Aryan race. And I am talking about the article all the time, it simply you don't want to hear the critique. Ergo it was tagged and cn'ed, and your the one with WP:OWN problem.PB666 yap 04:18, 3 December 2009 (UTC)


 * Nonsense. It was not tagged and cn'ed. It was simply changed and everyone agreed to change it. There is no argument. Please stop posting fantasy accusations on this talk page.--Andrew Lancaster (talk) 08:25, 3 December 2009 (UTC)

Argumentative edits: please stop them
While these things are happening constantly, the future of this article remains a concern:-

1. PB666 removes a reference to Klyosov and then a little while later puts a  --Andrew Lancaster (talk) 08:53, 3 December 2009 (UTC)

2. PB666 tagging a section as POV, which means that he claims a Wikipedian is pushing his own opinions, when his talk page comments show clearly that he realizes that the POV being questioned is that of a published and notable author in the field: ,
 * WP:OVERTAGGING: If the tag doesn't say what you want it to say, don't use it!
 * Also see --Andrew Lancaster (talk) 08:53, 3 December 2009 (UTC)

3. Argumentative citation of wiki guidelines but clearly in opposition to the spirit of those guidelines: "Note:I challenged it. If it isn't referenced I get to remove it." . The guideline being cited is never meant to be used when the fact being questioned is one considered obvious, and the talkpage comments and indeed the edits of PB666 show that he realizes very well that the sentence he tagged is correct. He just preferred a different wording. There is no problem having opinions, and making a case for them like other Wikipedians, but the approach here is clearly aimed at avoiding any such discussion by citing "rules". This is referred to as WP:Wikilawyering, covering senses 2, 3, and 4 in the relevant WP page:
 * Wikilawyering (and the related legal term pettifogging) is a pejorative term which describes various questionable ways of judging other Wikipedians' actions. It may refer to certain quasi-legal practices, including:
 * 1. Using formal legal terms in an inappropriate way when discussing Wikipedia policy;
 * 2. Abiding by the letter of a policy or guideline while violating its spirit or underlying principles;
 * 3. Asserting that the technical interpretation of Wikipedia:Policies and guidelines should override the underlying principles they express;
 * 4. Misinterpreting policy or relying on technicalities to justify inappropriate actions.

Can it stop?--Andrew Lancaster (talk) 08:53, 3 December 2009 (UTC)


 * Something needs to stop, that's for sure. I'm sure both editors have the article's best interest at heart, however, look at this talk page ! its all from the past week. Seems like excessive, sometimes pointless edits arguing about minute details at best, sometimes arguements posted are tangential and cirumentative (?if that's a word) Hxseek (talk) 11:19, 3 December 2009 (UTC)


 * Yes, and interestingly enough Andrew is carrying on the war when my addition to the page has dropped markedly, He misses me, feels he needs a sense of closure. While he was working his Maelstorm, however, I have been gathering references for. Go to the ANI, you will see clearly they tell Andrew 1. Take a few days away from editing 2. He has ownership issues, which he argued with one editor about. OTOH, I left a 1 paragraph response. The thing is here, I am the one person he can't deal with. He can deal with Muntawandi, he has the upper hand, he can deal with Small Victory, obviously upper hand, but when it comes to me, I have access to far more literature, basic science literature and critiques than he does. Not only that I have been editing longer made more edits. And I know the difference between an encyclopedia than content like this:"Concerning E-M123* (tested and definitely without E-M34) Cruciani et al. (2004) located one individual in Bulgaria after testing 3401 individuals from five continents, and Underhill et al. (2000) located one individual in Central Asia. In a 568 person study in Iberia, Flores et al. (2004) found 2 E-M123* individuals, both in Northern Portugal out of 109 people tested there. In a 553 person study of Portugal, Gonçalves et al. (2005) also found 2 E-M123* individuals in Northern Portugal, out of 101 people, as well as 2 in Madeira out of 129 people tested there. Flores et al. (2005) found one individual out of 146 Jordanians. Cadenas et al. (2007) found none amongst the significant presence of E-M34 they found in their study of the UAE, Yemen and Qatar. Arredi et al. (2004) found 1 Tunisian in their study of 275 men in Northern Africa. Zalloua et al. (2008) found 26 E-M123 cases in Cyprus, out of 164 men tested; and 27 Palestinians out of 291 tested[28]".
 * He does not have the upper hand. And shortly I am going to bring several more references into the fray that really question why he carried on about by critique of 'trivial' errors. This is not a wall of words and what Andrew is going to do next is say I wasn't talking about this I was talking about that and your not on-topic. Watch.PB666 yap 01:16, 4 December 2009 (UTC)
 * Indeed it is completely off topic (you have given no explanation for the 3 edits I mentioned), and also ridiculously long, and also it is purely ad hominem i.e. not even vaguely anything to do with article quality, it contains accusations based upon distortions, and as usual it contains threats to edit war soon. Ticks all the boxes. I do not see you reducing your impact on this page?--Andrew Lancaster (talk) 07:18, 4 December 2009 (UTC)


 * Of course this talkpage should not be about individuals, but I think at this stage I need to summarize the problem in order to avoid misunderstandings. (I have no way of communicating with PB666 which works. See his talk page.) On the basis that there is a clear consensus on the problem you mention now, I have proposed (above ) to PB666 that he and I agree to limit our talk postings on this page. He has characteristically managed to take >1000 words to refuse, without any actual relevant explanation.
 * There are two reasons that just ignoring his posts is a questionable approach:-
 * 1. They are enormous, and their unnecessary size (they are filled with irrelevant chatter) is in itself very disruptive. This needs to stop in and of itself.
 * 2. They are filled with disruptive accusations and misstatements about what other editors are saying and doing, and he increasingly seems to be dedicated to deliberately trying to create disputes between other editors and me personally., , ,
 * Other editors might also want to keep in mind that PB666 has stated quite openly since he arrived at this article that putting me personally through a "painful" self improvement process as an editor is his main goal in what he is doing here. (Because he says then I can handle more Y haplogroup articles the same way as he is teaching me.)--Andrew Lancaster (talk) 11:57, 3 December 2009 (UTC)


 * You are doing that all by yourself. For example getting all peeved about the 3 remaining comments, and then going biserk about a new added comment. You could have just sat down and tried to see if there was a better way to word it, considered that to some trivial might be an opinionated word. Nope you pulled the red flag and tossed it in front of the bull, problem is you had movie props as swords. As for the rest of your diatripe below, I not taking your bait, you can wait, and I will defend my comments above and we will see if you actually have literature to back you up. "Friends, Romans, countrymen, lend me your ears;I come to bury Caesar, not to praise him. The evil that men do lives after them; The good is oft interred with their bones;So let it be with Caesar. The noble Brutus Hath told you Caesar was ambitious:....."PB666 yap 01:16, 4 December 2009 (UTC)


 * Although PB666's behavior here is clearly outside all norms (as several admins have confirmed to him on ANI)
 * , and on its own would normally get a block, admins tend not to like to get involved in anything looking like a content dispute between only two people. The very long postings of PB666 make it seem like there is a highly technical content dispute somewhere, but there is none. And he jumps on every third voice that appears, either to chase them away (attacking MarmadukePercy and Wapondaponda, for example, both of which cases spread to other talk pages) or to try to get them confused.--Andrew Lancaster (talk) 11:57, 3 December 2009 (UTC)

Response to AL
Andrew got some advice here on how to diffuse the issue from his side here. He has not listened to them very well. One editor pointed out that while my critique was taken to extreme Andrew did exhibit ownership issues. Having rather received a deaf ear he has come back here, again because of my recent edits to the main changing his 'heavily guarded' wording, he is playing his rather transparent Gaius Octavius Thurinus impersonation. Things aren't going to change for you Andrew until you realize that this is an encyclopedia. If at some point I start trying to trash the encyclopedia or make the page worse in that regard (all they have to do is look at my edits), and particularly when you insult the ANI people and infer they are too witless to understand the situation. DGG appears to understand the situation very clearly. As previously stated on the ANI board they see alot of notifications regarding pages similar to these. Rather than fueling their prejudices you should retract from you intransigent stances on various matters and begin looking at these matters from an 'improvement' point of view as in putting the goals of WIkipedia first. How can Andrew Lancaster help to make the pages Andrew Lancaster has worked on, like this page, like the E1b1b page, more encyclopedic and less like a dirty laundry list of factoids interdispersed between unreadable jargon. I leave it to ANI to look at these pages, and the mtDNA eve page you hang like an albatross around my neck, at least it is improving and at least it can be read.PB666 yap 01:59, 4 December 2009 (UTC)

Yes, this has been a difficult process, but unlike you who sees doom and gloom, sometimes it is the unbearable and adrenalin wrenching process by which one learns. Again, I will support you and Wikipedia in any way that I can as I always have, but I place the encyclopedia first. These talk pages will eventually scroll backwards in the archival oblivian, as talk pages often do, some talk pages have 30 archives, if we can get a decent page here done with only 3 then I would not be too complaining. What is important that the footprint of the Mainpage. Now I know you want to leverage me off the page, so that you can make your personal edits ad-libertum, but maybe thats not such a good thing, and maybe the folks on the ANI see that.PB666 yap 01:59, 4 December 2009 (UTC)

As for the guidelines other than the reference issue for the main page, did anyone on ANI correct my interpretation? So aren't you then trying to promote your point of view regarding guidelines, running a little R1a mutiny of Wikipedia. The goal is a cross purpose Andrew, the guidelines are there to help you promote you pages by making them more accessible to the general public. I am happy to pushed these, it has made the page much more readable, as stated by many. The problem however is not R1a, its all those other pages you have worked on that are unreadable, even for those with reasonable familiarity. I have compromised on the guides, the section heads, for example. However, there are other things that are obviously not to an encyclopedic standard, you might not see these today but in a year or so with more editing experience they might bug you. The one issue, the infobox is just one issue, you said gee we need this level of explanation for people who come here from the commercial organization FTDNA. Wikipedia is not an arm of FTDNA, we have not obligation to them, in fact WP is expressly the opposite. So one does wonder why you defend certain edits so fervously?PB666 yap 01:59, 4 December 2009 (UTC)

See the problem here, you have drawn your sword against your own shadow, your fighting your own deamons, so leave me out of it. Every chance I get to back off of this page you try to draw me back in. I have many things to do, the infobox is one, MtDNA is another, and these are not my interests, but unfortunately this section of WP is adrift without leadership and organization, we have become a callous on the buttocks of WP. You also have interest other than arguing with me, you have the R1b page which is a true mess. You have the E1b1b1c section, you have the E1b1b1a page, you have many many things to work on that are much more productive than arguing with me, so don't. As per the issue on confidence of relationships as per STR, I had to then go out and do more research, gather more references, and so, consequently I am going to present these here. When you are about a critique incessantly the the critic begins to think you need to learn more, consequently you drag out the issue.PB666 yap 01:59, 4 December 2009 (UTC)


 * You want people to work on articles? Then why do you continue posting 2000 word essays full of false accusations here, and continually making threats and edit warring such as in the above edits? Put your money where your mouth is and go and work on articles. Let other people do the same.--Andrew Lancaster (talk) 07:18, 4 December 2009 (UTC)


 * Despite your trolling Andrew, I have worked on other articles, I made a new template for Y-haplgroup frequencies, I worked on the Haplgroup A page, fixed its infobox, working on the cladogram, and off and on I have been working on mtEve, what have you done other than 16-18 hours a day hovering over this page. Not to mention I created a very nice table that can go into the Y-STR page, what about you?PB666 yap 02:50, 5 December 2009 (UTC)


 * If we must write about editors, then let's be correct. It is true that the day after the above posting of mine you started working on other things. For three days before you only worked on this one, and basically that has been your pattern for about a month. In one day, yesterday, you posted about 20,000 bytes to this talk page. You have now posted 406 times to this talk page, which is the most you have posted to any talk page, and is more than you have edited the article itself. And yet R1a is one of the articles you have worked on the most over your whole time in Wikipedia, despite your always trying to pull rank based on experience. For myself, although it should be absolutely irrelevant, and you should not have raised it here, I have actually I edit a wide range of articles when I have a period with enough spare time. Your work is mainly on articles that almost know one else sees, except for the controversial Mitochondrial Eve. As you know (because you mentioned some articles I was working on, on my talk page once, in order to harass me) I stopped worked on several jobs (especially Nicomachean Ethics) in early November because of some massive unilateral and very wrong and very obviously anti-consensus, edits happening here. (And this article had a history of edit wars.) During recent months I have indeed spent a lot of time on this article, triggered mainly by the new articles which have been published, which made it a good moment to try to get a more balanced and stable article. But of course it is has mainly taking up so much time, for example dropping work on other subjects, since early November and then only because of the need to respond to a constant flow of reactions to my editing, and sudden proposals or demands such as the demand to split the article before you even fully understood which parts were about which clades. Now, you yourself have described these actions as being mainly intended to put me personally through a "painful" experience. It seems a little odd therefore, that you would criticize me for spending time on reacting to you, when you clearly wanted this.--Andrew Lancaster (talk) 08:21, 5 December 2009 (UTC)

Here are some statistics concerning the last 500 edits on this page:--Andrew Lancaster (talk) 10:26, 5 December 2009 (UTC) The biggest posting was by you and was close to 19,000 bytes in one edit. This was RECENT, in the last couple of days, and was not in reply to anything being written here or being placed in the article. Nobody is pulling you back here. Your enormous essays of recent days for example are apparently simply about your objection to the word "trivial" which was resolved easily before you even posted your essay on the matter, by simply changing the wording. Concerning your main page edits, see the continuing edits I have listed in this section, above and below, aspects of which, such as the edits comments, are very difficult to interpret as anything other than trying to cause disruption. Yes, these are continuing. Again, although they pretend to be, they are clearly not in reaction to anything anyone else is doing. Your continuing insistence that you are aiming to cause discomfort to me personally does, please note, also raise serious issues that I have not attempted to address officially. WP:HA is however very clear. --Andrew Lancaster (talk) 10:47, 5 December 2009 (UTC)


 * Yes indeed, however, trust me i did not enjoy writing it no more than I enjoyed seeing an opionion on the page treating a source of variation as trivia. Capice. At some point I would hope you will cool your pipes and reassess what you have written cause you have already attracted confusion and people desiring to draw conclusions that you really wanted them not to draw. Its coming to the point very soon that problem #1 you will be having is with other editors on this page not me, and you have barely touched the R1b page, and 20 or so more pages to go. Better find a better strategy for dealing with conflict.PB666 yap 10:50, 6 December 2009 (UTC)
 * Oh And BTW regarding the Dongxiang and that group of people, the so-called 'mongolian-speaking chinese' there was an HLA paper written, I thought I recalled this so I investigated this, you might just want to check their putative origins before writing to much about Southern Siberian origins. Of course if I tell you, you will resent it, so its best you figure it out on your own. Chinese over the course of many thousands of years have engaged in forced migrations, others where willful migrations, without going into details much of the evidence has suggested that people made south to north migrations, some of the peoples who live in NW and N china once lived in other parts of Asia (and you will have to find out where) however it would make the Southern Siberian origin hypothesis much less likely. Not only that but the Dongxiang and others are admixed with people from Uzbekistan, Afganistan, because of recent events.PB666 yap 10:50, 6 December 2009 (UTC)


 * My usual question: are you saying the community has wrongly reported what is published, or are you saying that we should not have reported it?--Andrew Lancaster (talk) 19:38, 6 December 2009 (UTC)

A new example of an argumentative and pointless behavior
Please can we stop things such as these edit remarks by PB666:- The unfortunate thing about both edits is that PB666 is clearly trying to imply bad faith or errors on the part of other editors, when the record shows that he is the editor responsible for these and other link breaks, as well as other generally sloppy actions (even this new one left an extra "[[" needing clean up), on numerous occasions. The most relevant edit in this case is this one:, which is where the linked word [[haplogroup]] was deleted altogether by PB666. Please PB666, back away from the horse carcass and put down the stick?--Andrew Lancaster (talk) 08:44, 5 December 2009 (UTC)
 * add for some reason a wikilink that some one repeatedly eliminates. what's the point?.
 * Better link, lets hope it stays this time


 * Am I the only cause of sloppy edits, go check the log, how many grammatical errors of yours have I repaired, and I don't come here and tell you about every repair. Typos are no reason to delink wikilinks. Today alone I came across a section that had been messed with, I made edits to that and noted what caused it. I don't see what your problem is. And I am not harassing you. If I were harassing you I would be here writing pedantically about every edit you are making, as you are doing.PB666 yap 10:39, 6 December 2009 (UTC)


 * Maybe you missed the point. Why are you posting accusations against me in your edit summaries which are so easy to prove false? What is your intention? You have done it again today, as discussed below. And again with that one, when I point it out I get extremely indirect responses. Very strange. Have you reached the point of deciding that the infobozes are just so bad that they deserve actions outside the normal rules?--Andrew Lancaster (talk) 16:34, 6 December 2009 (UTC)

And another one
Please note that the changes were made by User:Zadeh79: Will PB666 admit he is wrong though? Of course the implication PB666 was trying to give was that I do this sort of thing all the time, and so now he is justified in his edit warring. So I doubt he will admit any error. It is too important to him.--Andrew Lancaster (talk) 09:14, 6 December 2009 (UTC)
 * Replace wikilinks for the 3rd time! Andrew you are clearly exhibiting WP:OWN issue when you exact these types of reversions.


 * Indeed it was. But as I claimed to Zadeh79 despite agreeing on the source of confusion, the argument is not as strong as one would prefer. So it is clear now that you have seen my critique, when are you going to modify the page so that we can avoid this new source of conflict?PB666 yap 10:36, 6 December 2009 (UTC)


 * PB666, this is very confusing:-
 * I asked if your edit comment accusing me of removing wikilinks was in error. I see no answer. It would very good for this talk page, and for your reputation, if you would give a clear answer. Was it an error? Good faith editors tend to admit errors quickly, as I have done with you on a couple of occasions.
 * Secondly, as quite often with you, after having made several different edits ABOUT this minor problem (some broken wikilinks), you ask why I have not fixed them. Didn't you fix them?
 * Thirdly, you have now reverted another edit which has nothing to do with this, with an edit comment pointing to an agreement on this talkpage. This has to do with you incorrectly re-categorizing the Caucausus under the Middle East and distinguishing Anatolia from the Middle East, which is simply a terminology error, as was mentioned on my original edit . I am going to assume this was an error also? Will you please correct this?
 * In the past I have defended you from the accusations of others that you are trying to bait me into things, or use other classic tricks of the trade. I have unfortunately had experience with more experienced edit warriors than you have had on your handful of technical articles and your templates, and this helps me see that this is not what you are (you are not an experienced edit warrior). However, it has to be obvious to a lot of people, and so I give you a friendly warning, that the above series of actions looks like a clumsy inexperienced attempt to try to get my revert count up. It would not work because the record of discussions is clear, but nevertheless it is very hard to find any good faith interpretation.--Andrew Lancaster (talk) 16:07, 6 December 2009 (UTC)


 * When you reverted my edit you also reverted several repairs. I am not sure your section head is the best section head in light of the critique. Anatolia is considered part of the near east. The traditional definition of the Middle East of the middle east does not included Anatolia. The eurocentric definition of Near East Does, West Asia includes the non-European part of the Caucasus. Therefore I think my section heads are more accurate than yours, as per the critique of the other editor and critiques made earlier. "Many have criticized the term Middle East due to its implicit Eurocentrism.[12][13] In contemporary Western academic & media venues, the term is often bandied about by both Europeans and non-Europeans; this stands in contrast to the similar designation, Mashriq, which is deployed primarily in Arabic-language contexts.[14]" [The rest of the above is diatribe].PB666 yap 16:32, 6 December 2009 (UTC)


 * So you think Anatolia is not in the Middle East and the Caucasus is? What on Earth is your point? Wikipedia uses standard geographical terminology. What is there to argue about. The only other "repairs" had to do with paragraph breaks? What is your point???????--Andrew Lancaster (talk) 16:36, 6 December 2009 (UTC)


 * The point is this is yet another one of your blind reversions, that is point enough.PB666 yap 16:43, 6 December 2009 (UTC)
 * Note to admins Andrew has once again blindly reverted my edit while at the same time leaving the paragraphing issues (also reverted) that I corrected, once again Andrew shows his disdain for the WP:MOS.PB666 yap 16:49, 6 December 2009 (UTC)


 * OK, now we are again writing in terms of third parties who are not really involved again (the reviewer etc)? If there is a problem, then I repeat that to begin with your initial edit request was either an error, or, given your refusal to admit this, fraudulent, and secondly I'll my requests for your explanation in case you have one:-
 * Do you think Anatolia is not in the Middle East and the Caucasus is?
 * Were the other "repairs" which you say that I blindly ignored, not just paragraphing changes which all went around the theme of treating the Caucasus as the Middle East?
 * I remain open to any explanation. But the ones given so far seem to be simple geographical errors or worse.--Andrew Lancaster (talk) 17:49, 6 December 2009 (UTC)

The evidence to support the origin of R1a1 is overwhelming....
I think has become clear that R1a1 formed in the regions between Iran, Turkey, and Armenia. Ice age maps, characterize these regions as having cold, but survivable landscapes. There is considerable evidence that Hap R originated in Iran, and there is the inarguable and long-standing, autosomal and craniometric evidence, showing these populations to pole highest in the Caucasoid corner. Additionally, while modern inhabitants of this region pair strongly (in the relative sense) with European populations, they are clearly genetic outliers in the Near East. In fact, all the data together suggests that the phenomena of loose pairing, in this case, is related to the strong genetic relationship of these populations to the theoretical population of 'root' (earliest) Caucasoids. —Preceding unsigned comment added by Zadeh79 (talk • contribs) 21:30, 3 December 2009 (UTC)


 * Hi. I think that in a very general way, saying that R1a has ancestry at some point in the Middle East is not controversial. But R1a itself is one specific male line. The question is not whether it was in the Middle East, and to be honest we can hardly say anything about where the mutation happened either, but what we can write about is where it actually spread from. In other words, how were the geographical patterns of today created? This is what the genetics papers write about, and these are the only sources we can use. We can not add our own ideas no matter how reasonable. BTW you deleted some things on the article without explaining why. People coming to Wikipedia want to know what published authors have written. They do not want to know what Wikipedians think. So we can not simply censor any well known and widely discussed ideas no matter what we think of them.--Andrew Lancaster (talk) 22:15, 3 December 2009 (UTC)

I gather you reverted his little unsubstantiated input, Andrew ? Hxseek (talk) 00:48, 4 December 2009 (UTC)


 * I wrote a slightly new version. Hope it is suitable.--Andrew Lancaster (talk) 07:01, 4 December 2009 (UTC)

People coming to Wikipedia, Lancaster, should be offered alternative theories of science, whether these theories are broadly supported, or not. I have not intended that people hear one side of the story, and that is exactly why I have included evidence to support the Middle eastern origin hypothesis (which should probably be changed to West Asian). I'm not sure what I've included that appears fabricated, but all of the necessary references have been provided. --Zadeh79 (talk) 19:24, 5 December 2009 (UTC)


 * Zadeh, rather than insult you by reverting your edits I will try to explain this. First, you should note I tried to get Andrew to change his wording of these sentences but he has refused, see below. Andrew has WP:OWN issues that he is not fessing up to. With regard to the issue at hand, we do not want yet to argue for your position for several reasons. Of the 7335 tested individuals elsewhere we have several problems.
 * The statistics are poor, we could probably argue that there is no or very trace levels of R1a in NW Europe, Northern Eurasia, etc, but actually defining levels around the caucasus, within Iran, Afganistan, Syria, Anatolia, Black Sea region ect, simply state 'absense of evidence is not evidence of absence'. There are statistical reasons that we should avoid drawing a conclusion at this point, after all M420 has only been used diagnostically for about 1 year. To state this more clearly, the levels of R-M420 are so low at its mode that other stronger modes could exist in surrounding regions and have gone undetected for statistical reasons or reasons of ascertainment bias. We need at least 10 counts of R1a from any one area to derive good statistics, and adequate sampling in adjacent areas to rule out the alternative possibilities.
 * There is a pool of R1* in India that remains untyped for M420, In the west invariably R1* is R-M420. This does not have to be the case in India, but it might. There has also been a cluster of R1a* R-SRY1532 and that is relatively significant in India.
 * The only way to assess diversity in this situation is to look for new defining mutations with the R1* clade, alternative one could look broadly at STR for slowly evolving haplotypes. This work may be in progress, as far as I know it has not been done.
 * I have argued to Andrew that we should not be using data points but roughly defining the frequencies over as broad a possible reason that good editorial judgement allows. He refuses, so I expect you will not be the only person to read this and jump to a premature conclusion, in fact the piece as it is written begs the issue.
 * I have argued to Andrew that we should at least mention the fact that there are untested R1* cases in India that may convert to R1a* with M420 evaluation and that this should be mentioned.
 * I warned Andrew of presenting the material in the way he did because it would invite critique and potential edit-wars, he does not see things this way.
 * I did change it to West Asia with the recommendation of another Editor. However I should point out that that hypothetical origins of R1a should probably include SE Europe (Eastern Mediterranean and Caucasus, SW Central Asia, And Southern Asia).
 * In terms of alternatives, we cannot offer every alternative theory, this page had 4 alternative origin theories last October, we have fought incessantly to reduce these, there are more or less proposals at this point, if enough references support a proposed origin hypothesis, then we can include it into the article. So that if you can garner enough information from the (hopefully recent and thorough) literature one can advocate addition. This particular feild of study has been highly subject to speculation in the literature, ballooning the R1a page into 'bad and popular science interpretations'. As a consequence some level of standards have to be kept.
 * Zadeh, please establish a talk-page so that other editors can exchange information with you when needed rather than on these talk pages.PB666 yap 07:46, 6 December 2009 (UTC)


 * Zadeh, apparently PB666 is talking about something else. I think there is a much simpler point concerning your edits. You present yourself as having tried to INCLUDE material about a hypothesis, and you present me as having deleted information about a hypothesis. The simple fact of the matter is that your simply deleted material about Iran. You edits do not match your talk page comment.--Andrew Lancaster (talk) 16:14, 6 December 2009 (UTC)

Sources of Variation
The first things that must be set forth are a few premises. Each of which will be substantiated with references throughout this. These premises are based in basic statistical and science principles and not in genealogy but forensic science studies.
 * 1. The measure of genetic relationship is best defined as the average temporal distance between two genetic variants and their most recent common ancestor. That time can be measured in years or generations. The most accurate is somewhere in between. According to the study "Mutation rates at Y chromosome Specific microsatillites" Gusmao et al. Human Mutation 2005:26:520, rates increase with fathers age, but not linear with age and not significantly. IOW there is ambiguity about which is more important.PB666 yap 07:03, 4 December 2009 (UTC)


 * 2. The second thing that needs to be established is that while there are statistical rules that apply here (WP:OR) forbids me from using these directly but it does not forbid me from explaining the basis which papers have published and which we all in this field use. Anyone actually can handle these complex statistics, they are basically inversion of the binomial probability distribution to obtain rates that explain specific values with a given level of confidence. For example, if the confidence range is 95% it means that if we accumulate probabilities scrolling through rates giving rise to n events in N trials probabilities from 0 to 2.5% and from 97.5 to 100%. These are the outside tips of the confidence range, rates that demarcate these two values (2.5 and 97.5%) define the confidence range. Table 4 In Gusmoa et al. (2005) they present in these rates. Note: that for DYS19 an observation of 12 mutational events in 7129 father-son pairs results in a rate (x 10-3) of 1.68 and a 95% CI of 0.87 to 2.938. These values are generated by the process above. The statistics have nothing to do with Y or DYS19 they have to do exactly with the range of rates that can generate 12 in 7129 tries at 95% confidence using a Markov chain process and the binomial probability distribution. The same technique is used to generate SSAP range for HLA alleles and haplotypes. I have created an excel program that can do this for extremely large N, for a non-wiki related work process as we can see below we are now dealing with 10,000s of sample points. The same process with 'exotic twists' can be applied to reversion or homoplasies.PB666 yap 07:03, 4 December 2009 (UTC)


 * 3.Any given ratio for example 12:7117 (24:14234, 48:28468, etc) the 95%CI will reduce at the rate of 50% per each 4 fold increase in N, as long as sample selection variance is the only source of variance. This goes by the basic idea the SE = SD/(N)^0.5 which you can find in any statistics book. Ratios based on smaller values have a relatively wider confidence range. Bizarre but 0 is not zero in statistics. Looking at DYS461 in the table 4 mentioned above please note that 0 detected mutation events corresponds to a rate confidence interval of "0-4.217". You might argue that this is not real because there is no median confidence, there actually is the median number in the distribution 0.692/N (e.g. 0.792), which is larger than some of the median values in table 4. Is this real, Black swan theory for some real life examples were failure to pay homage to the 0 gets one in trouble. What it basically means is that the SD ranges and 95%CI ranges are more important than the relative frequency, which have less value. In genetics this is most true as the observed number of mutations at sites approaches 0. The frequency placed in these various Y-DNA pages, they actually misrepresent reality. Reality is represented by that 95%CI. That is the reason that one study can show 0 evidence of E1b1b1b and then next study can show 10 instances. Many sciences, for example archaeology, paleoanthropology, etc. ignore zero, that why we see sayings like "Neandertals went extinct 31,000 years ago", and then 2 months later its "Neandertals went extinct 25,000 years ago". The problem is they are always chasing the last 1, instead of paying due homage to zero. If they did as such they would know that 0/N is a real number between 0 and ~4/N.


 * There is a philosophical point to be made here, what all the experts say (for example financial analyst working for the US Treasury Department in 2007 and 2008) is actually less important than how well the statistics have been done.PB666 yap 07:03, 4 December 2009 (UTC)


 * 4. For any two means and SD, a general rule of thumb is that common groups in normal distributions will overlap in 1 SD ranges. If not expect a significant different. If only means and 95%CI are available, if the 95%CI of one does not overlap with the mean of the other, its significant. How can we test this? Here is a simple test . For example in the paper Population and segregation date on 17 Y-STRS:.... Snachez-diz et al. For. Sic. Intl. Genetics. 2008:e31-e35, the authors found the following list of STRs in table 3. If we then take a look at these we see number like DYS458 (8 in 1258 = 8:1250) compared to DYS438 (2 in 4709 = 2:4707) and we FET (8, 1250, 2, 4709) for a p = 0.0001142. According to Zar, Biostastical analysis, if the two values significantly differ then it is best not to combine them directly and treat them as one (such as an common rate and common 96% CI). There are some exceptions, this is one such example, since we can make multiple comparisons in Table 1, we have to correct; however this only would increase the P-value about 5 fold, and a five fold increase in P is still highly significant indicating some values in this table should not be combined, as the authors did in the last row of their table (not shown on left).


 * If we compare DYS19 with DYS437 we see the mean is very close to the 96%CI of DYS19, if we place this in the Fisher exact Test the P value is 0.18, the reason it is not lower is because DYS437 confidence, like all lopsided ratios produce skewed upward confidence range, this is directly determined by the binomial probability distribution, these are not normal distributions. The 97.5% boundary of DYS437 is 3.085 which is higher than the mean relative rate frequency (2.318) of DYS19. Translation, use the rough guide in both directions, if both fail to overlap, then its significant. WP:OR forbids me from telling you to use the Fisher Exact Test, but if you are intelligent you can figure it out on your own :^).PB666 yap 07:03, 4 December 2009 (UTC)


 * 5. Rates are subject to 'clock' variation. And it turns out to have nothing to do with Y-chromosomal haplogroups. In the paper Mutation rate at Y-chromosome short tandem repeats in Texas populations. Ge et al: For. Sci. Intl. Gen. 2009:3:179-184 found that the rate of mutation increases with the number of repeats and the size of the repeats. IOW as the repeat complex grows in size it becomes more unstable, more likely to mutate, other papers have found that deletions occur more often as removal of 2 repeats. In calculating branch distances we don't actually know what the mutation rates thus we make assumptions based on the ranges, the rates may be corrected for repeat number or size, no such protocol has been written, the only thing we know is that they differ within ranges, and it will take 100s of father-son pairs mutations to figure it out.PB666 yap 07:03, 4 December 2009 (UTC)


 * 6. There is the largest source of variation in calculating branch lengths within local trees, because the number of events may be small, 1 or 2 mutations. This produces a large confidence interval of mutational breadth which has to factored against a variable mutation rate to create the genetic distance premised above as time. Independent sources of variation need to be cross multiplied, I have stated above in the Talk-page the roughest way of doing this. However the sources of reversion and homoplasies are dependent on the genetic distance, we call this a second order rate constant, this is because it goes through two different process, the second dependent on the first. Therefore as X--->Y then Y increases the Y can go to Z or back to X. Therefore the rate of X--->Y--->X is a hyperbolic with respect to time.


 * 7. Branching provides external reference for each relationship pair. The simplist way to present this is if X--->Y--->X and the process is hyperbolic, if another branch forms in the lineage before the probability of Y--->X begins to increase, essentially the new branch then traps the state of Y before Y disappears back to X. As a consequence trees that have lots of branches frequently are taller than trees with no branches in parsimonious trees, especially when such trees are rooted in outgroups. Examples of this are the Bonobo versus chimpanzee HVR, or the L0 mtDNA versus L1. IOW what is observed is very much dependent on how many observations one makes of the most similar but not-identical things, if such things are available to be seen. For example in dealing with mtDNA lineages, is a mutation in a branch in L1 is the result of a forward mutation in that branch or a forward mutation in mtMRCA-L1 branch and a reversion in a second branch of L1. We have to assume there are processes that are lost to observation. Any mutation we observe in one branch may be consequence of two reciprocating events in two other branches, the higher the mutation rate the more likely the double event.


 * 8. Some STR have undergone parallel duplications, DYS19 is one, there are other as of yet unknown black-swans that we also need to be particularly mindful of the newly used STR sites Klysovo is proposing since we are not yet aware of anomalies that might exist. Phylogenetic evidence for multiple independent duplication events at the DYS19 locus. Capeli et al:For. Sci. Intl. Gen. 2007:1:287 (comparable to the 9 bp deletion observed to have occurred twice with mtDNA).

So the basic principals of this form of rapid evolution. For DNA, simply, any site can go to 3 different states, but for instance in mtDNA there is generally two preferred states (transitions) two more restricted states (transversions) and two other states (insertion and deletion). STR have a large number of states, but in general they can flip up or down (sometimes down 2 or up 2, less frequently +/-3 or more). The issue is not altogether different than mutation in the HVR region of mtDNA. Rather than reinventing the wheel lets recapitulate a process that has worked, and lets see how it worked.

Vigilante et al used HVR to determine TMRCA and Exoafrican exit times, but she could not do so directly. HVR mutations were too fast and too unstable to measure between chimps and humans, so they used a two step process. They counted transversions used only the CHLCA variation of 4-6 million years, and they were heavily criticized for doing this. There is a binomial probability derived abstract rate (not applied) range for the 26 transversions they observed between chimps and humans. There is the risk of a reverse transversion and this is another source of variation (estimable but not applied). There is variation in the depth of the mtDNA tree that can be calculate based on coalescent theory (not applied). And there was variation in their own parsimonious estimate of human mtDNA tree depth, and there is the issue of clock rate variation. As we can see above, all the issues above applied to this study.

In 2000, Ingman et al. repeated the process with genomic mtDNA but decided not to deal with HVR, assumed a two rate model and excluded HVR rates. This study had many of the same problems as Vigilant. More recently Gonder et al, Soares et al, and Mishmar et al. have undergone a process of sorting mtDNA sites by rate class (same problem however in defining class rates above). These groups have produced TMRCA of 190-200Ka, however Endicott and Ho argue a TMRCA based on local clocking of ~110Ka. Compared to Y TMRCA estimates these are not too bad. The problem however Endicott and Ho is devoted to correct peripheral branch problems, not so much the TMRCA, but none the less the variation is not 'trivial'. MtDNA TMRCA is one of the most studied problems in molecular anthropology, but the TMRCA still varies over a 2 fold range.PB666 yap 07:03, 4 December 2009 (UTC)

Here is the critical issue with Y chromosomal studies: most of the publications on refined STR rates are within the last 4 years and are not that refined, counts are relatively low in some cases, no good papers for calibrating the mutation rates for 36, 60 or 100 haplotypes. There is at least one study that documented more dramatic changes than seen in these studies. For more haplotypes (25, 60, 100) one is dealing with STR which have poorer statistics (lower N, lower number of observed mutations, larger confidence ranges). And with these also come the black swans. When Ingman looked at mitogenomic selection with his 50 or so mitogenomes, no selection was obvious, but when the sample size was increased and certain assumptions were done away with (like a constant rate of evolution in coding region) then we see examples if selection are abundant, as expected. It became a massive problem that need to be dealt with (Endicott and Ho, Soares et al, Mishmar et al. as examples) in estimating branch depths in local trees. Soares spent a considerable effort trying to deal with homoplasies and reverse mutations, some sites evolved so fast CRS16519, CRS16182-16183 that they needed to be masked (or otherwise TMRCA is biblical eve in age). As each year has increased since 2000 the story of selection has grown, grown, grown. Very few things are trivial, and those that have argued so, generally, are proven wrong. Branch placements in the lower tree, and other places in the tree are still argued over, etc. PB666 yap 07:03, 4 December 2009 (UTC)

Andrew is going to gripe about the size of this, but since we are not dealing with 1 Y-DNA page but many, and since this issue is one discussed by many papers (I presented just a few), and because we need to understand that what is published (e.g. Algerian = 4.4% i.e. 1 in 22 individuals is not 4.4% but a fraction within a confidence interval which contains a SSHP of 0.044) can be a gross oversimplification of reality, we as editors can avoid conflicts with other editors over conclusions, often oversimplified, often based on bad statistical analysis often based on ascertainment bias. We can avoid these issues by not taking on opinionated stances of others, but by creating a better stance of sticking to the facts we can observe ourselves. If we do not see an SD range or confidence range with a statistic (like 4.4%) why should we trust it, if 1 in 22 are counted what does it mean, should we be adding SD ranges or confidence ranges. As we look at the problem, one method of presentation is dishonest, the other method will not be understood by most reader, and the third method is WP:OR. So why are we placing all these percentages in the Y-DNA pages, flooding the infoboxes with 100s of %s when in fact they don't mean that much? Simplify, simplify, simplify these pages should not be presenting 100s of percentages but relative frequency ranges. We should not be taking opinionated stances on such a rapidly evolving field of studies were wide ranges of variation in quantities (divergence, TMRCAs) are frequent.

When we quote Klyosov let us consider 17-STR disclosed in figure legends, those are real sites, with real mutation rate ranges, with other processes (black swans) we do not yet fully understood.PB666 yap 07:03, 4 December 2009 (UTC)


 * So in summary you are saying that all existing clustering and age calculations for Y haplogroups are estimates, which can be wrong. Correct? And so the article should use terminology which makes it clear that they are estimates which could be wrong. Correct? Which it already does. Correct? And nobody has ever argued against you about this. Correct? (4 questions. Each can be answered by yes/no.)--Andrew Lancaster (talk) 08:56, 4 December 2009 (UTC)
 * One answer, as consumers of the literature, we as Wikipedians should never put ourselves in a position to overstate confidence, overstated confidence is a POV, is a reason for propagation of critiques within the literature, and will also be a point of propagation of edit conflicts here. If we avoid the position of overstating confidence (as per the origin of R1a in the above section) then we can potentially avoid future edit conflicts. And this is the point to build concensus here we need to avoid opinions that are not warranted by quantitative confidence.PB666 yap 17:01, 4 December 2009 (UTC)
 * I do not see any answer in this. Maybe I am just not up to it, but please do help other editors by using simpler words like yes or no. After writing so many words, it would seem such a waste if no one would even be able to understand what your point is. Obviously if you write so many words here on a communal webpage dedicated to discussing only what should be in the R1a article, it must be because you really want people to really understand some message about what should be in the article right? Please try answering the questions with yes or no.--Andrew Lancaster (talk) 08:00, 5 December 2009 (UTC)

I only scimmed throught that thesis, however, the content would be better discussed on another talk page for a different article, ie the specifics of estimation, etc in genetics studies. THis hardly directly applies to an article on a specific haplogroup, nor is it the best place to be discussed ad nauseaum Hxseek (talk) 15:05, 4 December 2009 (UTC)
 * Absolutely correct, this is not the place, it is not a thesis either (mine was couple hundred pages) unfortunately we got here because one editor refuses to understand that variance is a problem that has to be handled delicately and is never trivial. To cut to the chase, as shown about, several STR variances do not overlap in what we call rate probabilities, the 96%CI range shows. As a consequence, genetic relationship (time to common ancestor) should be handled separately, although in many of these papers they are not. For the reason why see Soares et al. 2009. To be brief, each rate has its own phenomena. Fast evolving STRs can grow, but then become unstable, they can revert and undergo homoplasy unseen in all but the bushiest trees. Slow evolving things are least prone to these processes but their confidence ranges may be much larger (because counts as observed in the right hand table above, and BTW this is the source of mutation rates in other studies). In complex with both of these are other sources of variation which need to be crossmultiplied into the analysis. Finally they can be combined by weighted averaging (and crossmultiply those resultant variances). The product when best described is a confidence range of times to most recent putative ancestor within a branch. If we look at Klosovo's circular cladograms we see that many branches have very uneven lengths, the is very typical for all trees of this type (nothing special about STR except the rates). Some regions of the trees are bushy (reliable) and other regions are not- trees which are generally shorter, and very difficult to correct (ergo they are generally left uncorrected). Again if you want to move this to the STR-Page thats fine, but Andrew and his misguided statements above force me into a situation of proving how 'experts' have weighed in on the sources of STR variation. I would have you focus on the line in the table DYS448  0.19, 1.6, 5.7 : 2 of 1258  That is a 20 fold range of rates, and these are the types of ranges that are being used in the 17 STR, 25 STR, and 60 STR comparisons because few comprehensive studies have been done. As Andrew argues correct or incorrect, its an issue of corrrect handling and fair statement of variance.PB666 yap 16:15, 4 December 2009 (UTC)


 * Now that your essay is here, can you answer yes or no to the 4 questions I posed above? If you read them, you'll see that these 4 questions are written by trying to take your essay seriously AND THEN apply it to the R1a article which we are supposedly discussing. I believe that answering those questions (or not) will help show your intentions, which could help create better understanding. --Andrew Lancaster (talk) 17:38, 4 December 2009 (UTC)


 * Whats your problem now, Andrew? Article estimates that are biased often conflicted. The article could use better phraseology (avoid words like trivial when describing variance). The article is much better than it was 2 months ago, but you were recently arguing against changing some of this troubling wording. Despite the increasing STR decrease the confidence limits, there are pitfalls in the use of the new STRs, not as well characterized.PB666 yap 02:43, 5 December 2009 (UTC)


 * I am not arguing anything. They are yes or no questions. Why can't you answer? They are very basic questions to help understand the relevance of your essay. Don't you want anyone to understand you? To restate the questions:-
 * Are you basically saying that all existing clustering and age calculations for Y haplogroups are estimates, which can be wrong? Yes or no?
 * Is this so difficult Andrew. Do I have to spell it out for you? You can only combine groups in which there is no statistical difference in the rate for reaching a rate average. So to take an argument of best fit 6 of the 17 could not fit into a single group statistically, which would call for at least 2 preferable 3 rate groups. STRs within the rate group could be combined without violating statistical rules for a single rate class with a lower variance. However that would mean that STR counting would have to be subdivided, time estimates determined independently and then re-merged together. As Soares et al. 2009 did for mtDNA.PB666 yap 22:41, 5 December 2009 (UTC)
 * In addition slower evolving rate classes take precedence in forming parsimonious trees relative to fast evolving, in these cases because faster evolving mutations are more likely to revert. In addition, to add onto the problem these newly discovered STR may have phenomena associated with them that only large region sequencing would reveal, Such as with DYS19.
 * So in practice this would mean that the article should use terminology which makes it clear that they are estimates which could be wrong. Yes or no?
 * The best means of being honest to the data is to use confidence intervals, with a very careful consideration that for STR that are very recently described and not well characterized that there could 'black swans' lurking within the STR assessments. PB666 yap 22:41, 5 December 2009 (UTC)
 * Do you not agree that it certainly does this already? Yes or no.
 * As I have stated I prefer data to be encoded into percent ranges, broad ranges like 10 to 30% or 5 to 20 or 1 to 5%, reflecting sampling issues. As I have said with regard to the SNPs, we should not be presenting a list of numbers but a geographical range along with a percent range, while keeping the data (such as in notes, tables or list pages). The material in the text should reflect that statistical nature of the data without using complex statistics in the text nor creating a tally board of data, and many Y-DNA pages are. This more fairly reflects that level of uncertainty that is actually present in the data to the reader, it is simple to read, to understand and it does not make a grand pretense of something that otherwise is not that well defined. Here are areas where this article begs trouble with readers like the above section demonstrates'
 * "apparently mostly in the Middle East and Caucasus: 1/121 Omanis, 2/150 Iranians, 1/164 in the United Arab Emirates, 3/612 in Turkey. Testing of 7224 more males in 73 other Eurasian populations showed no sign of this category so far." The trouble here is that 7224 also contains samples close to this region that might also have M-420 that are either too low in frequency, ascertainment bias, or random factors that miss equally high SSHPs. Also what about the R1*s that Sharma detected, are they possibly M-420s if not what does that tell us about R1 and India?
 * "1/51 in Norway, 3/305 in Sweden, 1/57 Greek Macedonians, 1/150 Iranians, 2/734 Ethnic Armenians, 1/141 Kabardians. Sharma et al. (2009) also found 13/57 people tested from the Saharia tribe of Madhya Pradesh, and 2/51 amongst Kashmir Pandits," here is another example. Someone looking at this might conclude that SRY1532.2 evolved in India and spread assymetrically to the middle east, the Madya Pradesh is a very small region that could have a recent expansion as a consequence of a founder, this may exist in any other region, including regions with 0/N observations. As discussed about 0/N is not zero, its a confidence range which at best is <4/N when N is very large and assumes no ascertainment bias. When one deals with low frequencies one has to handle data rather delicately as not to promote a theory or confuse the reader. Again not to deny the raw data, that can go into a note, but to create a readable passage that conveys both a possibility but not a conclusion in as simple, clear a manner as possible.PB666 yap 22:41, 5 December 2009 (UTC)


 * Has any other editor ever argued that such things should not be in the article? Yes or no.
 * "Trivial", the presentation of data points (see Subclade definition section). This level of detail is unneccesary and it begs to reader to be a statistician, to understand what the binomial probabilities are. To make a more readable document we need to present the soft, amenable to future revision version of frequencies rather than precise data points and all that these infer. IOW if the editors do their work properly, the readers do not have to. Take a look at the R1a1a7 map, how much of this map is actually punctate frequency variation within the population (Single selection haplotype probabilities per geographic position) and how much of the color waffling is good ole, plain and simple issue of undersamping and interpretative sample variance?PB666 yap 22:41, 5 December 2009 (UTC)
 * Please help everyone understand your point.--Andrew Lancaster (talk) 07:55, 5 December 2009 (UTC)

I really do not see any answers. Indeed it is hard to see anything which is even concerning the same subject as your 19,000 byte esay. Just to repeat them one more time:- All of your above remarks, apart from being extremely confused and unclear, appear to be about a different subject, i.e. about frequency estimation in populations, something you have never raised before today as a point of concern. (It is certainly one more problem in this field, but I would ask almost the same questions: for example, who disagreed with you about this?) Your essay was not about this at all I thought, but rather about clustering based upon STR haplotypes, used both in order to estimate relative relatedness, and TMRCA age. Am I wrong?--Andrew Lancaster (talk) 16:20, 6 December 2009 (UTC)
 * Are you basically saying that all existing clustering and age calculations for Y haplogroups are estimates, which can be wrong? Yes or no?
 * So in practice this would mean that the article should use terminology which makes it clear that they are estimates which could be wrong. Yes or no?
 * Do you not agree that it certainly does this already? Yes or no.
 * Has any other editor ever argued that such things should not be in the article? Yes or no.


 * Here you go with the bolding again. I have answered your question fully and honestly. The post was about two problems. One larger problem concerned sampling statistics, the version of the problem addressed by opinion you added to the page addresses the STR problem. Both issues have been answered, it is hoped that you will fix the sections in which you added an unnecessary level of detail to the Main text, and hope that you refrain from adding opinion in the future. Other than responding to this responding to the rest of your talk-page expanding diatribe will accomplish nothing. As for credibility, don't blame me because you got in an edit war with yet another person on this page, engaging in reversions over several edits makes it difficult for people to follow the page history, please refrain from edit warring with people over content that begs critique.PB666 yap 16:39, 6 December 2009 (UTC)


 * No, there were 4 simple yes/no questions. We already had essays, and the aims it to try to get to the bottom of what your actual point is with regards to the R1a article. For example, is the point just that you think your original research is better than what is published? Where did your write your answers? --Andrew Lancaster (talk) 17:38, 6 December 2009 (UTC)

Further Balto-Slavic-Indo-Iranian links
Perhaps there should be an article covering all of this in an Indo-European context. Beyond DNA and language there are frequently remarked upon similarities between Indo-Iranian religion and (Balto-)Slavic mythology, as far as the latter has come down to us. The cycle of Perun and Veles is asserted to be close to that of Indra and Vala and/or Vrtra. And of course there are Svarog and Svarga. I'm sure if we had a picture of ancient Slavic religion we could find many more. DinDraithou (talk) 20:06, 5 December 2009 (UTC)
 * That there was a homogeneous proto-Slavic mythology is debatable. However, cetainly, there is some evidence of cultural flow between Balto-Slavic and Iranic Hxseek (talk) 22:08, 5 December 2009 (UTC)


 * Aunt Sally? Mythology is a scholarly term. DinDraithou (talk) 22:31, 5 December 2009 (UTC)


 * These are called Oral or Artistic Traditions, which more broadly focuses on traditions other than classical mythology.PB666 yap 07:50, 6 December 2009 (UTC)

Original research noticeboard
I have posted a thread that involves this article here. Wapondaponda (talk) 08:17, 6 December 2009 (UTC)


 * I'm not taking any sides but this has no place here. You should move this to Pdeitiker's talk page. It may not be your intention, but this looks like an attempt to drive him away from the article. Is it? You have to keep in mind what newcomers will think. DinDraithou (talk) 16:18, 6 December 2009 (UTC)


 * Hah! I Invite everyone to come and help participate on the improvement of mitochondrial Eve page and haplogroup A (Y-DNA). Clearly MW does not want to work with me, that's fine.PB666 yap 16:23, 6 December 2009 (UTC)


 * Umm. I think newcomers will already be driven away? Do you want diffs of PB666 driving editors away from this article if they disagree with him? Wapondaponda is posting here correctly because this article is mentioned in a broader discussion, and by the way, PB666 himself has made clear on several talkpages around Wikipedia that there is a big action going on, with him against others. Do not get things round the wrong way. You might want to compare the talkpage of Mitochondrial Eve to this one, and also see "Without any other facts, based on 10,000s of edits here on a wide variety of pages, I conclude that I have the better representation of reality. Therefore I am going to be disgruntled and gripe about the issue as I prepare mentally before going through 100 poorly written pages based on awful molecular anthropology to fix an issue of carelessness. If someone out there doesn't want me (self-admitted Y-DNA skeptic) doing resectioning of their precious Y-DNA pages I suggest they take the initiative." . PB666 wants to make an impact.--Andrew Lancaster (talk) 16:30, 6 December 2009 (UTC)


 * New comers are not apparently being driven away, you should be aware of this because you are currently trying to eat newcomers. lol. PB666 yap 16:41, 6 December 2009 (UTC) [eating being an exaggerated form of biting].


 * As usual, when making accusations, give diffs and make the accusations clear. Otherwise people will ignore you.--Andrew Lancaster (talk) 17:20, 6 December 2009 (UTC) By the way, which newcomers are you talking about? Are you talking about a single use account who came and sounded familiar? Where is MarmadukePercy? Did you not fairly openly discourage Hxseek and Wapondaponda from posting here?--Andrew Lancaster (talk) 17:23, 6 December 2009 (UTC)

Undue weight being given to Klyosov
I have been reading through the literature with regard errors and critiques. I think given the responses to Klyosov that undue weight is being given throughout the article to klyosov. Its OK to use his critiques but using his conclusions and ignoring the primary authors conclusions exhibits a POV. Couple of papers. Shi et al. 2009 MBE "A worldwide survey of human male demographic . . .". This group recognizes the "Zhivotosky problem". Zipping to the discussion first "The choice of mutation rate was perhape the most difficult issue. [as it should be given the history] While it influences only the temporal scale of the conclusions, this scale is crucial for comparison with other genetic and archaeological or climatic data. The OMR ["observed" mutation rate] is based on a large number of observed transmissions from some loci ..., but far fewer for others, so the uncertainties differ between loci. [This is almost exactly what I told Andrew above, the confidence ranges are reliable for 17 STR sites or less, but not the entire 67]. But I should note that their 95% CI includes the time range consistent (140 Kya) with the divide between SAK and non-SAK africans of Behar, so that at least in this case statistically they are in a ball park that appears to be settling out.PB666 yap

To get where the clocking problem comes from an how they solved it "The final methodological issue that we needed to address was the choice of mutation rate." [They derived their rates from genealogical and general evolutionary rate (Zhivotovsky), and as an aside my critiques resides in the meshing between then two, when internodal branch distance increases and the certainty of placement decreases] "We used both of these, but were concerned that while the OMR was appropriate for Y-STRs with similar level of variability to the eight loci used be Zhivotovsky et al. to estimate it, it would not be appropriate for significantly more or less variable Y-STRs which are expected have a different mutation rates." [They go onto to describe a method of separating Y STRs into 5 classes one of which was comparable to Zhivotovsky]. These rate classes ranged from 10-3 to 10-6 (i.e more than a 1000 fold range, much higher than I showed above). This exactly iterates my point, since statistically the rates are different they should not be averaged together or treated as one.PB666 yap

So that is the theory part, and Klyosov does not separate his STR studies by rate. Separating them by rate makes it easier to correct for the effects of saturation (observed in the high rate classes).PB666 yap

The second paper that disagrees with Klyosovo is Keyser et al. These people looked at 10 Y DNA samples from the Bronze age to the Iron Age, in contrast to what Klyosov showed this group has Y-DNA R1A1 matched population living to the east but also central and northern Europe. Absent in these was matches with Ukrainians or Russian Steppe groups. There were scattered matches across Asia, but these come simply be homoplasies. There were more close matches in Turkey and the Caucasus than in the Ukraine or Ural-Volga region. They noted that no exact matches were found, but when they lowered their criteria a high density of matches were found with high frequencies in the Altai republic.PB666 yap 03:48, 23 December 2009 (UTC)

The CRS - mtDNA was found twice, along with many instances of European mtDNA, including mtDNA of the U5 group. However as a whole it would appear that mtDNA were far more of East and Central asian ancestry than Y indicating the Y-invasion bias. Of these mtDNA only one had a high density in Ural-Volga-Ukrainians, and one was elevated in Ural-Volga region. However 1 was only found in Sardinia, 2 were found in groups with little putative Ukraine contact in Europe. One was spread from Turkey to China (which correlates with a path of spread of HLA from west to east). The mtDNA in only 4/26 individuals was consistent with a Ural-Volga-Ukraine origin. Many instances of the Mediterranean/Europeans H, HV, CRS found, H6 (corsica) indicating a SE mediterranean influence. All and all the evidence for a specific Ukrainian/ Ural Volgan influence is present but not strong. The influence of Eastern Asia is at least as strong. I strongly suggest that Keyser et al be reread and used as a primary source and use of Klyosovs controversial corrections be removed.PB666 yap 03:48, 23 December 2009 (UTC)

In addition In Keyser et al. 10 were attribute to the Andronovo culture, 4 were attributed to the Karsuk culture, 12 to the Tagar cultur and 6 to the Tachtyk 1 culture. Only 3 Y of these were Andronovo 2 R1a1 and the other one belonged to an unknown C-haplgroup. The two that were R1a1 were also found in Turkey, Altains, Hugarians, Polish and Cretans. So there goes that theory. Were is the Ural-Volga-Ukraine origin?PB666 yap 04:00, 23 December 2009 (UTC) [When in doubt cut it out].

Summary: Keysers data is entirely consistent with cultural influences from the "Kurgan" culture but only with mtDNA, but the Y-DNA results are consistent with a wide number of possibilities, since the Sample of R1a1 is small and exact matches were seen with the Anatolian regions with some dropped sites, this is consistent with a model of gene-flow (Invasion) of males from Bronze age early IE speaking cultures near the fertile crescent.PB666 yap 16:39, 23 December 2009 (UTC)


 * Your making up your own theories. Ie. not only have you proposed your own genetic hypothesis, but none of the major linguistic frameworks actually posit a spread of IE languages from the Fertile Crescent during the Bronze Age Hxseek (talk) 23:10, 26 December 2009 (UTC)


 * Sentences removed from Main, until Keyser is reread and represented correctly they should stay off the main.PB666 yap 05:02, 28 December 2009 (UTC)


 * Your main point is that you personally do not agree with Klyosov or these steppe theories. Good for you, but that is not a satisfactory reason for removing big chunks of texts, or reference to a well known article? One thing this article has always had problems with are people wanting to filter what is presented so that it reflects their favorite theories.--Andrew Lancaster (talk) 13:19, 28 December 2009 (UTC)


 * No, the point is that either you read Keyser and ignored it and inserted interpretations and WP:OR based on what you thought it meant, or simply you did not read it. Keyser makes it clear that the R1a matches were not close and those that were, were over a broad area, the map they show makes no indication that there were hits within the Ukraine or Western Steppe region. Instead you went out of your way to draw that opinion in the text, that is POV editing and until you can find a way of writing it without drafting only Klyosov's point of view, I will continue to erase it.PB666 yap 16:31, 28 December 2009 (UTC)
 * The age of mythological Y-DNA is over, from now on these pages will rely on the cold-hard facts or the info will be removed. Its amazing that in this process I have had to check every reference with a fine toothed comb, you claim to be well read on Y-DNA, why is it so difficult for you to get you facts strait?PB666 yap 16:31, 28 December 2009 (UTC)
 * Just to make this clear, I am not advocating the placement of an Anatolian origin on the page, what I am advocating is the removal of speculation and misinterpretation from the page. Correct representations of the literature are allowed, the is an exercise in getting NPOV editing.PB666 yap 17:31, 28 December 2009 (UTC)
 * What you are responding to expresses no opinion of mine. I simply point out that arguing against a published author, saying why you personally think them wrong, is totally irrelevant on Wikipedia. Arguing with me about that is pointless. Concerning what Keyser wrote, the footnote now includes a direct quote, and this published and peer reviewed article seems quite clear in its claims, as is Klyosov. I make no argument for or against them, but I see no reason to forbid mention of Klyosov or Keyser. Both articles express opinions found in numerous places. Whether they are wrong or right, they are certainly not "fringe".--Andrew Lancaster (talk) 19:29, 28 December 2009 (UTC)
 * OK lets take a look at the word fringe
 * "However, we are concerend that this haste may not be in the best interest of sound scientific process, as evidence by the heavy reliance on the non-refereed publications, unpublishted work and works not accessible to the scientific reading audience through usual scientific channels (categories that cover all five citations to Dr. Klyosov's previous work). Reference to the non-peer reviewed and poorly accessible data and formulations renders the constructive critique process problematic.- Hum Genet (2009) 126:725–726"
 * "Dr. Klyosov's Comment makes the erroneous claim that the haplotypes found in a population (and representing a sample) form a genealogy, and that 'familial' rates are therefore appropriate. For mutation rate estimates, Dr. Klyosov (Comment, p. 13) refers to Chandler (2006); however, this author suggests a method of estimating relative mutation rates at Y-chromosomal STRs based on haplotypic distributions in populations. For estimating absolute mutation rates, Chandler (2006) used estimates of overall Y-STR mutation rate from data on father-son pairs obtained by Gusmão et al. (2005). Hum Genet (2009) 126:725–726"
 * Paraphrasing - His critiques are fringe science. But shall we use this to judge our conclusion. Here is what Keyser says: ::::::"Haplogroup (hg) assignment, based on Y-chromsomeomal SNP-typing, revealed that except for S07, which was fiund to belong to C(xC3) all specimens were affiliated to hg R1a1. This finding is in agreement with the relative similarity of the [STR] haplotypes mentioned above. - Human Genetics 126:401"
 * ",R1a1 is widely spread across Eurasia. It is found among western Eurasia, southern Asian, central Asian and Siberian populations. This haplotype is thought to trance the migration patterns of the early Indo-Europeans, perhaps stemming from the Kurgan culture (Zerjal et al. 1999; Seminoe et al. 2000). The additional analysis performed on the Xiongnu specimens revealed that whereas none of the specimens from the Egyin Gol valley bore this haplogroup, the Scytho-Siberian skeleton from the Sebystrei stie exhibited R1a1 haplogroup Human Genetics 126:401"
 * "A search of the YHRD database as well as in our own databank revealed that none of the Y-STR haplotypes obtained from the south Siberian samples perfectly matched (at 17 loci) those included in the databases. Neveretheless, when not all loci were scored, matches were found for all samples except two, (S07 and S32) Human Genetics 126:401"
 * Noting the Andronovo R1a1 Y-DNA samples are S10, and S16 have the same haplotype (Table 3).
 * "S10/S16 haplotype matched the most frequent R1a1 haplotype (12 loci) seen in the souther siberian population of Derenko et al. (2006) This haplotype is also found in the Altaians [. . . .] The S10/S16 haplotype is also found in eastern Europe (Humgary, Slovenia, Poland) as well as in Asia (Central Anatolia). Human Genetics 126:401"


 * Here is what Klyosov says:
 * "However as Figure 9 shows, all of the haplotypes nicely fit to the 17-marker hapltyep tree of 252 Russian R1a1 haplotypes with common ancestor of 4750 +/- 490 ybp.JoGG 5:217:256"
 * As I stated to you previously Klyosov did not do an impartial analysis, such an analysis would have also included samples from other regions for control references including those "also found in eastern Europe (Hungary, Slovenia, Poland) as well as in Asia (Central Anatolia).", and the figure he shows has no scale and so the relevance of the distinction is unclear. Figure 9 has no scale and different deep branches in the Russians may have come from multiple regions outside of ethnic Russian areas.
 * Combining the fact that other researchers are regarding this as fringe science, that it is written in a non-standard scientific style, and that he is not running the proper controls, I would argue that his conclusions are fringe regarding this issue.
 * Once again, your refusal to properly read and understand the literature has forced me to make an unnecessarily long reply to rebutt your claims - if had you not carefully read and compared the literature. You are well aware of the Klyosov controversy in Hum Genetics. and you are also well aware that he fragmented populations based on recent ethnological and political boundaries without concern for longer range relationships. I suggest you carefully read the critiques again and work on removing the majority of Klyosov's claims until they can be established by a more deliberative scientific process.PB666 yap 02:23, 29 December 2009 (UTC)


 * In all your long reply though, as usual, there is no actual demonstration of what you are claiming. Frankly, you are either unwilling or unable to stick to any subject. By the time you get to the end of one of them you are writing about a new subject.
 * The fact that Hammer and Zhivitovsky disagree with Klysov's critique of their theory is obvious. So what? It does not mean they call Klyosov a fringe theorist. And indeed you are on record several times as saying you agree that Klyosov has a point concerning that particular subject. And you know very well that Klyosov is not the only geneticist who feels big doubts about the Zhivitovsky method. On the other hand, and this is kind of important, what does any of this particular exchange have to do with the particular material you want to remove from this particular article? Here is the passage you keep deleting:-


 * This modest passage has nothing to do with age estimation controversies, but is all about looking at STR clustering. Klyosov's STR clustering is not all that controversial. His novel conclusions, by his own account, come from looking at different data sets. It is also not something that Underhill or anyone else have criticized or even spent much time looking at. OTOH, your own vague and confused opinions, fascinating as they are, are simply not relevant. Please do not waste time pretending you do not know that this is the major issue with your edit warring on this subject. Nothing you have said anywhere justifies deleting mention of this theory as "fringe".--Andrew Lancaster (talk) 11:19, 29 December 2009 (UTC)
 * This is nothing more than a diversion, it is clear that some authors believe that Klyosov is presenting fringe work. They are not going to call him fringe; however, what they have said is almost the same. The analysis you presented is intentionally incomplete and misleading. And your statements like "In all your long reply though, as usual, there is no actual demonstration of what you are claiming. Frankly, you are either unwilling or unable to stick to any subject.", "OTOH, your own vague and confused opinions, fascinating as they are, are simply not relevant." demonstate that you do not want to stick to the subject. The long reply was mostly quotes of the relevant material, precisely pertinent to the material being questioned. Since you cannot dispute the quotes you attack the messenger.  The above is not my opinion but quotations from the literature. Unless you can come up with independent support for Klyosov's work all material relating to Klyosov should be removed. Wikipedia is not a newspaper. You have no basis or reasoning that is valid to keep the material on the page. I consider you inability to address the specific issue here as evidence that the material should be removed, and I should note to admins, Andrew once again is n not working under good faith, rather than try to correct clear identified problems he is simply reverting the deletes.PB666 yap 16:06, 29 December 2009 (UTC)

(outdent) PB666,

1. You write that "it is clear that some authors believe that". Just give a quote. Give a reference. No problem. Then we can discuss.

2. This is the latest of many occasions where you have demonstrated a lack of understanding, feigned or real, of the difference between:-

Please familiarize yourself with the difference between the three. The third one is the one which is allowed, and which you keep mixing up with the others.
 * Personal attacks. (example: you are stupid and you have a big nose) Note: personal attacks are not the only type of ad hominem diversion from a subject, and I think it is technically possible to write a personal attack which is not a diversion from the subject matter at all.
 * Criticisms of a person, which divert away from the real subject matter and are therefore "ad hominem". (example: this reminds me of how brilliant you were in that discussion we had about another article) Note: ad hominem does not always imply personal attack and attack does not always imply ad hominem diversion of an argument
 * Criticisms of an argument. (example: this argument you present shows the following problems...)

3. Concerning reverting edits, you have been doing the same thing in this case. The critical thing is that you know very well that you are reverting against consensus.

4. Back to the real subject, just to make sure we do not loose it in the smokescreen. You want to censor mention of a published article because it does not agree with YOU personally. That is not on.--Andrew Lancaster (talk) 18:16, 29 December 2009 (UTC)


 * First, off the above is a personal attack
 * Second, I have repeatedly tried to steer this back to the discussion with real quotes and you have derailed it with ad-hominem attacks.
 * Third, you are using this tactic clearly cover your original research and POV edits.
 * In the above you presented no facts to support your case, clearly indicating you are trying to hide the fact that it is you that have added the POV edits.
 * Where are the facts in this case? show were Klyosov compared within his cladogram STRs from R1a1 outside his clusters. Clearly what you have done is washed away Keysers interpretation and backfilled in with Klyosov's. Clearly Klyosov is treated as a fringe experct. Clearly you cannot defend your edits, stop grandstanding and support your case.PB666 yap 03:34, 30 December 2009 (UTC)


 * You presented nothing apart from your own opinion which had any relevance to the material you keep deleting. And you repeat this again now. Also what I wrote above is NOT a personal attack, nor ad hominem. It responded to what you wrote. Deliberately pretending that others are writing personal attacks or ad hominem when they are not is text book smoke-screening which you have a long record of doing. You accusation of OR is a bit weird isn't it? You want to delete mention of something in a published article and I have never defended the theories in that article. If I am promoting OR which does not comes from an RS then which OR? Why can you not explain?--Andrew Lancaster (talk) 08:39, 30 December 2009 (UTC)


 * The statements you presented from Keyser are taken out of context, they were meant to contrast their observation, that observations was they observed was a wide spread. The rest of your statement is nonsense.PB666 yap 04:13, 31 December 2009 (UTC)


 * Do you want to re-word that so it means something? (BTW turning into 2000 words of gobbledygook would not make it better). What is your accusation actually? After so many posts you still can not be clear whether I am publishing OR or else I am faithfully reproducing something which is published, but which you consider "fringe". What is the OR? BTW, your latest edit is obviously almost as unreadable as your above posting, so obviously it is not going to stay.--Andrew Lancaster (talk) 11:02, 31 December 2009 (UTC)

Other wikipedians please look at what PB666 is proposing:. Keep in mind that as usual the edit summary is highly distorted. Obviously the other version does not only mention Klyosov! And there has been discussion on the talk page.--Andrew Lancaster (talk) 09:04, 30 December 2009 (UTC)


 * By a rough count, PB666 has now deleted all mention of Klyosov about 5 times in his most recent burst of revert editing?--Andrew Lancaster (talk) 11:02, 31 December 2009 (UTC)


 * You seem to keep pandering to other people, you should spend more time pandering to the literature you are using as a source - and get the usage sources as correct as possible. The question you should ask first is 'has anyone else here read Keyser et al. 2009 and are there any other interpretations or suggestions'.
 * Klyosov, in retrospect is a very weak source. He has an interesting critique of the evolutionarily applied mutation rate, but that's about it. The other work he has recently done is arguably controversial, and pointedly fringe and I expect most of his critiques will be debunked in short order. Summarily we should avoid putting fringe, transiently applicable sources that in 6 months people will be fighting to remove from the page. If you want to use a better critique, a critique that has more notable authorship "A worldwide survey of human male demographic history based on Y-SNP and Y-STR date from the HGDP-CEPH population. Shi et al . Moiecular biology and Evolution October 12th 2009. This particular paper places the TMRCA between 83- and 168 kya) which is the only recent paper which places a chronology within the context of a 2:1 effective breeding ratio and the mtDNA TMRCA timeframe of 160 to 200 Ka. Read the discussion, paragraph 3, the decisions making process they use is markedly more complex than klyosov and reflects that fact that a pair-analysis TMRCA deviate from Zero time the ability to estimate the TMRCA begins to deviate from its actual value.
 * What I found astounding is that you claim to have expert review of the literature, and yet you shamefully pander to a fringe author and virtually ignored the presentation of what the primary author wrote, you ignored the fact that there were only 2 Andronovo R1a1 samples, that these were virtually identical, and (12 markers) had no match from the Ukraine, Volga-Ural or even regions proximal east of the Urals. The fact that Klyosov did not include any groups outside ethnic Russians in his spread and given the deep branches in his phylogenetic diagram would have alerted any serious molecular anthropologist that he is representing a biased view, ethnic relationships of 2009 have very little bearing on genetic relationships 4600 years ago in most places. You need to get your act together on fairly and unbiasly representing work, this is not a review for an unrefereed science journal.PB666 yap 16:30, 2 January 2010 (UTC)


 * A discussion which you largely ignored.PB666 yap 16:30, 2 January 2010 (UTC)

Amongst the garble you maybe have hit upon something above. Yes, my editing on Wikipedia panders to other people (i.e. consensus) and it panders to whatever has managed to get published in reliable sources. I am not even trying to defend Klyosov. I do not even necessarily disagree with your concerns. Please learn to "pander" or else go away from Wikipedia. Wikipedia is not meant for any other type of editing. As I have said many times to you, Usenet and blogs are always out there beckoning for what you want to be able to do. What you want to do is in violation of WP:OR, WP:Neutral and WP:SYNTH.--Andrew Lancaster (talk) 21:22, 2 January 2010 (UTC)


 * Not trying to defend Klyosov? and not pandering, no-one twisted your arm and forced you to add Klyosov's interpretations to 7 sections. I violated of nothing, I want the mythology removed. I have only added what was presented in the literature. You are the one who took quotations out of context and dressed them up as agreement. It is only because you want to continue to add one extremist point of view that other points of view need to be added. I am against as many folks on the notification board were against adding connections between language, genetics and the Kurgan culture. And you are the one who has violated WP:Neutral. Not only are you defending mythology on this page but you are aiding Muntawandi and his afrocentrist agenda on the mtDNA page, little hint it will not fly. BTW, the claim of pandering to wikifolk is not excuse for you using Klyosov as a source for Keyser, rather than carefully reading Keyser first, for if you had read Keyser you would have known there were only 2 R1a1 from Andronovo and you would have known that they did not closely match any Ukraine, Volgo-Ural samples.PB666 yap 04:05, 3 January 2010 (UTC)


 * What do you mean by mythology? You mean you do not like it? And what exactly is wrong with the wording about Keyser? I think you are saying we must include interpretive comments about what Keyser did not say, right?--Andrew Lancaster (talk) 11:42, 3 January 2010 (UTC)
 * I need to point out that this series of edits is clearly a case of WP:POINT making:, , , . Let's just summarize the facts of PB666's history on this:-
 * In this current round of edits he is now removing all mention of Klyosov, although most of what he is removing has been in the article for months, and PB666 is on record arguing that much of it is correct.
 * Klyosov is most controversial when it comes to his criticism of the Zhivitovsky method for estimating clade ages. The one source PB666 has come up with to try show Klyosov can be treated as a fringe author is a letter on this subject he sent to a Nature publication, and the reply from the authors, defending themselves. But many respected authors do not use this method, and PB666 is on record sympathizing with this criticism.
 * PB666's edit warring all comes from a dispute about one quite uncontroversial claim of Klyosov, i.e. that he thinks there is a component of Indian R1a which is similar to "Indo European" R1a, and another component which is not. I know of no clear reason to call this a fringe claim or even a controversial claim. PB666 has also not mentioned any. He is instead trying the more desperate measure of trying to get all mention of Klyosov erased.
 * The reason PB666 is so concerned with this section is because he has a personal fringe theory that R1a and Indo European languages come from the Middle East. He has filled this talk page already with hundreds of words about it. But he has no source for this. It is a theory based on very limited knowledge and should certainly be ignored here on Wikipedia.
 * If anyone thinks I am making any error in reversing the above-mentioned edits, or if people agree with what I have written, please do mention it. It is important to get the voices of others. I'll just state my assumption is that I am basically acting on behalf of what all editors of this page want apart from PB666.--Andrew Lancaster (talk) 12:09, 3 January 2010 (UTC)


 * Certainly if another editor can read both Klyosov and these other sources and can add judiciously elements back I don't have a major problem with that, the critical problem here is based on how you cherry-picked information out of context from Keyser and how you cherry picked information from Underhill I don't think you are neutral or enough to add it back. If Genie or Dindraithou desire to read these papers and carefully compare them, I have no problem, but at the moment you have lost all appearance of objectivity. If however I find edits in which the editors appear not to have read multiple sources and borrow, as you did, from a single source, I will delete it.PB666 yap 16:19, 3 January 2010 (UTC)


 * PB666, you name no error in my above description and you still do not supply the missing explanation for treating Klyosov as a fringe author writing "mythology". So your censorship is clearly not justified by any Wikipedia rules. You also know very well that the sections you deleted were already long being edited by other editors including you, so this is nothing to do with any doubts about consensus. Indeed, you have made clear many times that you think my big mistake concerning this passage is to "pander" to other people and not to accept your superior authority. Let's face it you only deleted the Klyosov mentions as one new step in an on-going edit war aimed at leaving space for your fringe theories and your grudges. Your real aim is to remove one thing only: mention of a proposed connection between India and the Steppes, though you know this is entirely mainstream. You only now started to propose deleting all mention of this author because you can not get rid of this, even though it is an entirely mainstream proposal. Also your latest edit summary is ridiculous:- "Undid Andrew's reversions, The use of Klyosov's speculative research in this article is not justified, we can wait until other papers are able to confirm his methods." This means you are inventing a rule especially for this one source. If you applied to other sources in this field you would have to delete or make wrong nearly everything in every haplogroup article! There is also in fact nothing particularly new or special or controversial about looking at STR clusters in this way (concerning India, which is the bit you really want to delete). So who is going to write anything about it? Such things have been published in mainstream articles for years.--Andrew Lancaster (talk) 16:37, 3 January 2010 (UTC)


 * It is clearly quoted above, or is this yet another WP:IDIDNTHEARTHAT situation.


 * Unsigned comment above is by PB666 as usual. And as usual, it can not be understood. What is "it"? You refer to something you wrote previously, as usual. But there is never any substance to "hear".--Andrew Lancaster (talk) 07:36, 4 January 2010 (UTC)

Just to make sure it does not slip past people, this new more aggressive approach of "if I can not get what I want then watch this" is by his own account in retaliation for the recent pruning which has finally begun of PB666's work on Mitochondrial Eve. Here is justification for my remark, something PB666 never gives for his... But of course if anyone at all believes they see good reasons to say that my understanding of the consensus on this article, or on Mitochondrial Eve, was wrong, please say it. I am assuming I speak for all other editors on both (a) that it is wrong to delete mention of Klyosov entirely, or concerning his relatively uncontroversial assertion that he sees Indo European like STR haplotypes in Indian data; and (b) that the massive extensions to the Mitochondrial Eve article were way off topic and inappropriately written for Wikipedia.--Andrew Lancaster (talk) 17:44, 3 January 2010 (UTC)
 * Here and here are diffs he posted to show admin User:Dbachmann.
 * Here is PB666 on the Mitochondrial Eve talk page saying that he will be "carving pertinent research" from pages including R1a in consensus retaliation for the removal of material on Mitochondrial Eve, proposed many times in the past and most recently here. (As usual PB666 shamelessly uses accusations which do not make sense, accusing User:Muntuwandi of inserting "afrocentrism" with this deletion.)
 * Here is a threat posted on this page, but referring to recent editing on Mitochondrial Eve, and giving a "little hint it will not fly".

Good grief. Let me be crystal clear. I personally disagree with a number of conclusions by Klyosov. That is no reason to remove all mention of a published author from a Wikipedia article. Of course publication in itself does not guarantee a high level of scholarship. We should give preference to work published in referred journals or academic texts. That includes The Journal of Genetic Genealogy. Even that is no guarantee that the author(s) will be proved right by further studies. The pursuit of knowledge is a never-ending journey. We can only present the current state of knowledge. Klyosov has made unique contributions to the current body of work on R1a. --Genie (talk) 18:22, 3 January 2010 (UTC)

I disagree with many of the conclusions of Klyosov. However, in an encyclopedia article it is important to include all viewpoints on the origins of this haplogroup, and whether or not I agree with Klyosov should have no bearing on inclusion of his take on things. MarmadukePercy (talk) 19:44, 4 January 2010 (UTC)


 * I agree with Genie's statement. The hypotheses stated by Klyosov may or may not be correct, but since his article has been published by a peer reviewed journal, the material is eligible for inclusion in the article. Though the argument has been "undue weight being given to Klysov", the editing is more akin to censoring Klyosov. If one disagrees with Klyosov, then the appropriate course of action is to demonstrate weaknesses in Klyosov's arguments that have been published in reliable sources. Wapondaponda (talk) 18:45, 3 January 2010 (UTC)


 * Would you like to rewrite the passages, if so I suggest you very carefully read Klyosov, look at where he got his data from and how he has compared sequences relative to other authors, read for instance what other authors who have looked at STR relationships beyond national borders. I am not going to converse here with people who have not read this literature, and certainly editors who do reversions without having read the literature, that will also go backwards. I did not remove all mention, but even the material I left behind, there are now better sources for the critique than klyosov. To state this clearly and simply, some of klyosov's dating techniques may only be applicable for hundreds of years between branch-points. Of course the Zhivotosky technique is up in the air because of the ever fluctuating Y-DNA molecular clock, but that is another issue. I suggest anyone who wishes to objectively address read the reference I provide above, the make note in the first paragraph that the most difficult problem in dating major branch-points is the inconsistent molecular clock. This is something that Muntawandi is desparately trying to cover up in the mtEve page.PB666 yap 20:51, 3 January 2010 (UTC)


 * PB666, the material you want to delete has nothing to do with anything you write here. Klyosov's dating methods are not the subject of extensive discussion in any version of the article. The material you wanted to delete was concerning STR haplotypes in India, and how some might come from the Steppes. Such STR analysis is very orthodox, and Klyosov goes through the whole range of R1a inhabited regions, in I think more detail than anyone previously. In fact, this is not even a Klyosov theory as such but simply one he is the latest person to defend (and the only person since all the data made a simple one wave model of R1a spread from the Steppes less tenable). You are also doing your case a lot of damage by throwing in wayward personal accusations where your substance should be. Genie obviously claims to have read Klyosov and to disagree with some of what he says. Your suggestion of Muntawandi's "cover up" is not only not referenced or explained here, but also not on the Mitochondrial Eve talk page. It is just one of your typical throw away accusations which you use instead of rational discussion. You do not get to say that you are going to ignore other Wikipedians.--Andrew Lancaster (talk) 07:36, 4 January 2010 (UTC)
 * The contention that Klysokov's article is correct because it does not contradict the Indo-European homeland being outside of Indo-Iranian territory is not nearly sufficient enough. If there are significant problems with Klysokov's method then it is worth stating in the article. GSMR (talk) 19:37, 6 January 2010 (UTC)


 * (a) We here on Wikipedia are not in the right place to be arguing whether Klyosov is correct, and no one is making the argument you mention. He is published in a reliable source, and that is why he should be mentioned.
 * (b) If we wish to mention concerns with his method, then there are some really simple conditions:-
 * First of all it would need to be aspects of his work which are actually discussed in this R1a article, and relevant to the conclusions being mentioned. This is obviously not an article about Klyosov.
 * Second this criticism needs to be the criticism not of us, but of someone published in a reliable source we can refer to.
 * With these two conditions in mind, I can not think of any controversial aspects of Klysov's work which are essential to anything in the present R1a article and not being flagged as controversial? --Andrew Lancaster (talk) 20:00, 6 January 2010 (UTC)

Note. This edit which is a repeat edit (a "slow edit war" edit) has an edit summary once again clearly implying this is tit-for-tat editing because of deleted materials on Mitochondrial Eve. (MW=Muntuwandi.) Discussions about that article should be on that article's talk page. PB666 is requested to partake in consensus building discussion (he did not even reply to the pre-discussions which took place between other editors on that article) or else at least not work against consensus. For those wondering, I post messages for PB666 on article talk pages because discussion on his personal talkpage is more impossible. (It is "his" talkpage, so he feels no compulsion to write with any clarity, focus, control or formatting there.)--Andrew Lancaster (talk) 12:56, 11 January 2010 (UTC)

splitting out Keyser discussion
PS, if PB666 has some sort of point about Keyser's paper it is absolutely unclear to everyone who has looked at his edits, which appear to be synthesis, and not in any way more accurate. If he wants to try to explain his point he should start a new talkpage section. I am going to look at it again also. I and everyone else have been reading this section as being about Klyosov.--Andrew Lancaster (talk) 16:08, 4 January 2010 (UTC)


 * The point is quite clear Andrew and if you have read the paper, which I still doubt you have the figure shows no relationship between two andronovo or any other South Siberian aDNA any Volga-Ural, or Eastern Ukrainian peoples. Either admit you have not read the paper and stop trying to mislead people. The clear contrast between Klyosov and Keyser is that Keyser compared samples over a much broader region, and Klyosov cherry picked comparisons, which means by emphasizing Klyosov you are clearly misrepresenting the literature.PB666 yap 04:49, 11 January 2010 (UTC)

PPS, just looking at PB666's comments about me cherry picking the Keyser article, I should note that this is misleading. The passage was not written by me. PB666 has played as much of a role in the current passage as anyone. My biggest affect on that passage has been to remove material which is incomprehensible (so yes, a lot of what what PB666 writes comes under this category) or obviously off topic or obviously synth (PB666's insertion of commentary about what haplotypes did not match where). If I can find a way to actually make a better passage I will.--Andrew Lancaster (talk) 16:21, 4 January 2010 (UTC)


 * You are cherry picking. The concensus here is that Klyosov is debatable (problematic) and yet there are 5 paragraphs solely representing his 'reworkings', there is a disproportionately large presentation of klyosov relative to the other literature. Clearly you presentation of Klyosov's work on this page is representative of you believe, your bias and your non-neutral POV. It was such overrepresentation of mythology that screwed up this page to begin with.PB666 yap 04:49, 11 January 2010 (UTC)


 * Actually, it is not representative of what I believe. I think most of your own doubts and theories about R1a can all be traced back to remarks made by me to you at different times. Does this claim shock you?--Andrew Lancaster (talk) 12:35, 11 January 2010 (UTC)


 * Klyosov is overrerpesented particularly giving his controversial conclusions.PB666 yap 03:39, 12 January 2010 (UTC)


 * I appears you are deliberately trying to mix discussion of Klyosov and Keyser. Anyway it is very simple:-
 * The consensus could not be more clear. Look how many people have reverted you. Look at the comments above.
 * Being controversial as a person does not get all mention of you removed from Wikipedia articles. You need to show that the particular Klyosov ideas or articles mentioned in the article are really considered a fringe and (see the consensus above) this very obviously is not the case.
 * Presumably if you had a case you'd have made it by now?--Andrew Lancaster (talk) 08:17, 12 January 2010 (UTC)