Talk:Mass killings under communist regimes/Archive 57

Undue weight given to source
This paper is cited no less than three times in the article, as in the following: "Another common criticism, as articulated by anthropologist and specialist in former European communist regimes Kristen Ghodsee and other scholars, is that the body-counting reflects an anti-communist point of view..." The article describes Ghodsee as a specialist in former communist regimes, however a look at her published work shows a focus on gender studies and Bulgaria, with nothing on the USSR, China, or Cambodia. Given the clear ideological bias of this academic, and the extent to which we are running the magnifying glass over cherry picked sources and domain expertise of actual, widely cited historians, is it really fair that this academic is sourced at all for this article, let alone 11 times? AShalhoub (talk) 22:05, 21 December 2021 (UTC)
 * I like your approach. However, it is valid only when applied consistently. We can limit ourselves with sources that say "A" and the sources that criticise "A" only in one case: if "A" is a majority viewpoint, and no other significant viewpoints on the subject exists.
 * The problem with this article is, however, quite different. Thus, Rummel was criticized only by a couple of authors, and none of them performed a global analysis of his "democide" data set. Normally that implies wide acceptance, but... Take a look at such a topic as Soviet Russia history, or at Great Chinese famine. Imagine that you have no preliminary knowledge about that, but you know that there were some political repressions in the Soviet Union under Stalin, and the Great Leap caused a terrible famine in China. With this knowledge in mind, try to find most reputable sources that discuss history of these two events. After that, try to find how the authors estimate the number of victims, what terms they use, and what causes they propose. Finally, try to check how frequently each of them cite Rummel.
 * All of that would be quite instrumental for understanding of who is cherry-picking and who gives undue weight to which sources. Paul Siebert (talk) 22:19, 21 December 2021 (UTC)
 * A hint: if you will do a neutral and unbiased source analysis you will find that most country experts essentially ignore Rummel, so both those who made estimates of "global Communist killings" and many of those who criticize them do not express a majority viewpoint, whereas majority scholars just ignore this topic, and they prefer to focus on study of separate countries. --Paul Siebert (talk) 22:26, 21 December 2021 (UTC)
 * I disagree with mentioning anyone's opinions in text unless their opinions are significant to the topic. For example, I would mention Raphael Lemkin's opinions on genocide because he came up with the concept and is frequently cited. Is this "common criticism" the consensus view in the literature, is it a left-wing view or is it the view of communist apologists? Policy requires us to explain the weight of different opinions. Incidentally saying something is a "common criticism" is a violation of Unsupported attributions. Although we can use that term if it is taken from reliable sources, we cannot make that conclusion based on our understanding of the literature. TFD (talk) 23:10, 21 December 2021 (UTC)
 * What you say is indisputable but trivial. Now answer the following question: is the opinion of Rudolph Rummel significant to such topic as "Stalinist repressions in the USSR"? If "yes", how do you explain a nearly total lack of support, criticism and even mentioning of his works by major experts in this topic?--Paul Siebert (talk) 23:18, 21 December 2021 (UTC)
 * No, it is not significant. Furthermore, since he is not an expert on this topic, his writings on it published outside academic publishing are not reliable for facts. TFD (talk) 23:43, 21 December 2021 (UTC)
 * Rummel’s books Death by government (cited by 1577), Power kills: Democracy as a method of nonviolence (cited by 540), Lethal Politics: Soviet Genocide and Mass Murder Since 1917 (cited by 238) and China's Bloody Century: Genocide and Mass Murder Since 1900 (cited by 190] were all originally published by the academic publishing house Transaction Publishers, with new editions by Francis Taylor. --Nug (talk) 01:51, 22 December 2021 (UTC)
 * There has been previous discussion of Transaction Publishers. See for example Reliable sources/Noticeboard/Archive 54. While their main focus was reprints of post war era social sciences classics, they were a publisher for modern extreme right books, that reputable publishers would avoid, such as books defending racism. So if you are an academic who wants to argue that the black race is inherently violent, the wrong side won the U.S. Civil War or the Communists killed 100 million people, that's the publisher you went to. The company was bought by Taylor and Francis in 1997. After it disappeared as a separate imprint, Taylor and Francis did not publish and new works from Rummel, who turned to self-publication. TFD (talk) 02:40, 22 December 2021 (UTC)
 * Oddly enough the only people seeing an issue with Transaction Publishers (and certainly no mention of anything untoward in the article) at Reliable sources/Noticeboard/Archive 54 are involved editors who have perviously voted to delete MKuCR, all other uninvolved people say it is perfectly fine. Strange coincidence that. As for Taylor and Francis they certainly have published Rummel since acquiring Transaction Publishers. --Nug (talk) 03:38, 22 December 2021 (UTC)
 * I think this discussion is senseless. We already had two (or even more) RSN discussion, and we have a consensus that Rummel is generally reliable, his democide theory is seen as a serious contribution, but it contradicts to later works by second generation of genocide scholars. Valentino and especially Mann explicitly disagree with his conclusions. And, more importantly, Rummel is not reliable for figures (except Cambodia). Therefore, yes, you are right about the source in general, but you are not right about some aspects of this source. TFD is not right about the source in general, but we still cannot use Rummel as a source for figures, and we need to discuss Rummel's theory in a context of latter works by other scholars. Paul Siebert (talk) 04:42, 22 December 2021 (UTC)
 * Why are you continuing to discuss Rummel, when this section is about Ghodsee? --Nug (talk) 05:06, 22 December 2021 (UTC)
 * Because too much weight is given in the article to the views of Rummel and few other authors that advocate similar ideas. Actually, the whole article is written from their perspective. Therefore, we need to fix this issue first, and then focus on Ghodsee and similar sources, which are on the opposite side of the opinia spectrum. It doesn't matter (now) whether Ghodsee is a good or bad source, it is not a major article's problem. Later, when we fix the main problem, we will probably think about putting Ghodsee in a proper content (and I cannot rule out a possibility that this source may be completely removed; I have no strong opinion yet, because that is not urgent). Paul Siebert (talk) 06:13, 22 December 2021 (UTC)
 * Dismissing User:AShalhoub's legitimate concern regarding WP:UNDUE usage of Ghodsee and continuing to focus on Rummel after making this commitment: "If you agree to apply policy "even handedly", let's do that. First, White should be deleted as a poor quality tertiary source authored by a non-expert, which duplicates figures from the sources that are already presented in this article. After you self-revert, I am ready to discuss other issues." raises a question about your sincerity. --Nug (talk) 06:44, 22 December 2021 (UTC)
 * And your post raises a question about your abilities to understand what I write.
 * I never dismissed any legitimate concern, I just wrote that it is not my priority, because, WP:UNDUE should be primarily applied to Rummel himself. I cannot rule out a possibility that when we change the article's structure to something more neutral, and put Rummel, Courtois etc into a proper context, there will be no need in such sources as Ghodsee at all. Paul Siebert (talk) 14:15, 22 December 2021 (UTC)
 * That's because the Transaction Publishers was at the time so small that it had very few original volumnes. In fact most if not all of the discussions at RSN were about Rummel. While you may believe theories advanced in their books that blacks have substandard intelligence, that goes against mainstream thought. Basically, it was a a publisher for extremists who couldn't get published anywhere else. But as you say, Nug, you don't see an issue with it. TFD (talk) 04:46, 22 December 2021 (UTC)
 * No, I said uninvolved editors like User:Jayjg, User:Crum375, User:Jayen466 and User:DGG had no issue with the publisher. --Nug (talk) 05:06, 22 December 2021 (UTC)
 * Transaction also published a book by Ivan Van Sertima who is best known for the obviously fake theory that Amerindian culture was created by Africans. Calling it discredited would give it too much credit. Serious writers cannot be bothered to reply to it, which is similar to Rummel's theories. Usually people who hold minority views don't argue their views are mainstream, they argue that the mainstream is suppressing their views because they are controlled by whomever. Do you think that reputable publishing houses publish racism and pseudohistory? TFD (talk) 06:15, 22 December 2021 (UTC)
 * This is nonsense, Francis and Taylor still publishes Rummel but not Ivan Van Sertima, so clearly they still find Rummel's work relevant. Do I detect a personal attack in your "Usually people like you...", do you want to refactor that? --Nug (talk) 06:44, 22 December 2021 (UTC)
 * I don't know what the policy is when they acquire another publishing company. They still have books by Joseph Scotchie. See Neo-Confederacy: A Critical Introduction, p. 40: "Scotchie is a neo-Confederate sympathizer, has been a League of the South member....He has published essays in Southern Review that are hostile toward imigrants." No mainstream academic publishers would publish these books if they were written today. The League of the South incidentally is listed by the SPLC as a neo-confederate hate group. TFD (talk) 14:42, 22 December 2021 (UTC)


 * I agree with that usage of Ghodsee is WP:UNDU and saying something is "common criticism" is a violation of WP:WEASEL. Unfortunately Paul's justifications for ignoring policy with: "However, it is valid only when applied consistently" and "What you say is indisputable but trivial" followed by a digression to Rummel raises questions on whether we can trust Paul Siebert to apply policy even handedly rather than selectively. --Nug (talk) 01:28, 22 December 2021 (UTC)
 * If you agree to apply policy "even handedly", let's do that. First, White should be deleted as a poor quality tertiary source authored by a non-expert, which duplicates figures from the sources that are already presented in this article. After you self-revert, I am ready to discuss other issues. Paul Siebert (talk) 03:29, 22 December 2021 (UTC)
 * I don't think Wikipedia policy is meant to be transactional but rather an absolute we all have to adhere to. In any case I already reverted nine hours ago after I read this. --Nug (talk) 03:44, 22 December 2021 (UTC)
 * Ops, didn't see it, sorry. I agree that our policy is absolute.--Paul Siebert (talk) 04:20, 22 December 2021 (UTC)
 * I agree this academic seems to be given undue weight in the article. With regards to their 'bias', that doesn't make them an unreliable source although in-text attribution of their views may be appropriate. To be honest, I think on such a contentious article as this on such a politically polarised subject in-text attribution is going to be needed for almost all sources. I also agree that we should make changes on individual merit, not taking an 'all or nothing' approach. Vanteloop (talk) 13:18, 22 December 2021 (UTC)
 * This section should be called Undue weight given to source[s] because I fail to see how one can read the article, and not see any problem and think Ghodsee (of all sources) is the issue. By the way, everything is attributed (the article is a bunch of "he said, she said", which is not a good approach in writing a good article) because pretty much everything has been recognized as a minority view, and is why I think we should refocus and restructure the topic to something for which we have tertiary sources and a clear weight of sources, but we make no distinction between significant ones and fringe ones. Are not Stephen Hicks, George Watson, and many others also not only undue but even non-experts? If it can be argued that people raising concerns about Transaction Publishers' history of publishing fringe authors and dismiss them because they are critical of this article, it is telling that any source dissenting or criticizing the topic is dismissed as undue (first it was Karlsson for being dismissive of Courtois and Rummel, two core sources), even though it is attributed like everything else and at least has some expertise and publications on Communism. Davide King (talk) 14:11, 22 December 2021 (UTC)

Undue weight
Ok, we achieved a consensus with that NPOV should be applied even handedly. I take Nug's words not just as mere a declaration, because he demonstrated his good faith by self-reverting his (re)addition of a highly questionable White's book.

Now it is my turn.


 * An important note. Anticipating possible accusations of original research, I am reminding you that what the Wikipedia community understands as neutrality means carefully and critically analyzing a variety of reliable sources and then attempting to convey to the reader the information contained in them fairly, proportionately, and as far as possible without editorial bias. The rest of this post is an attempt to do such an analysis. Therefore, any accusations of OR will be treated as a bad faith attempt to derail the discussion.

I decided to apply WP:DUE to Rummel first.
 * Step 1. Search for mass deaths. Rummel (with impressive 236 citations appears on the first page. Ok, he passes a preliminary step 1.
 * Step 2. Let's see who cites him. To this end, check other authors from the top of the list.
 * Ellman (#1 in the list): doesn't cite Rummel
 * Wheatcroft (#2 in the list): no citations
 * Mertelsmann and Rahi-Tamm (next in the list) no citations,no citations
 * Rummel himself (just to test the system): 12 self-citations
 * Wemheuer: this author does not cite Rummel
 * Rosefielde: (next in the list): cited Rummel 3 times. I was not able to see in which context he cited Rummel, because although his "Lethal politics" was in the list of references, there were no inline references in the text, and no mention of Rummel. It looks like Rosefielde added this source just as one of books on the subject.
 * Kay: (next in the list) cited Rummel 0 times
 * Krischel: (next in the list): cited Rummel 0 times
 * Harris (this source is irrelevant) (and it doesn't cite Rummel
 * Eberstadt: no citations
 * Dyadkin: no citations
 * Ginger - irrelevant
 * Norman - a very specific source about one late Soviet incident. Irrelevant.
 * Snyder: frankly, I expected Timothy Snyder to cite Rummel, but... he doesn't

These are all sources that I found at the first two pages of the google scholar search on this topic. I didn't cherry-pick them, and even didn't know the results of my check: I am typing in a real time, I check one source in the list, then I type what I found, and then I move to the next one. I am totally transparent. My preliminary conclusion: Rummel and other sources exist in "parallel universes": they almost do not interfere.

I didn't do the opposite check for an obvious reason: Rummel's books are old, and most sources in this list are more recent, so it is not a surprise that he doesn't cite them.--Paul Siebert (talk) 18:47, 22 December 2021 (UTC)


 * Step 3. Let's check who cites the authors from this list.
 * Ellman (his second article): this list includes such authors as Cormac O Grada, Mark Harrison, Steven Rosefielde, Mike Haynes, Oleg Khlevniuk, Barnett, Stephen Wheatcroft - most of them are experts in Soviet history. It is easy to check that most of those authors cite Ellman's works., so all those scholars are interconnected by a dense networks of cross-references, which demonstrates that they exist "in the same universe" (incontrast to Rummel).
 * Wheatcroft: actually, everybody is welcome to analyse this list using the same scheme. I randomly checked some sources, and they confirm my initial expectations: they cite Ellman, they cite other experts in Soviet Russia, but they do not cite Rummel.
 * (I'll add my analysis of other authors from this list if needed)
 * Step 4. What those authors tell about deaths in the USSR, and what sources do they use? To answer this question, I first looked at this list of sources, and one of them, On Sources: A Note. Author(s): Michael Ellman. Source: Soviet Studies, 1992, Vol. 44, No. 5 (1992), pp. 913-915 seems especially relevant. Although it is old, it is interesting for two reasons. First, the author is acting as an arbiter in the dispute among two prominent scholars, Wheatcroft and Rosefielde, so this source provides us with information about opinia of three authors. Second, Ellman analyses the the relative merits of 'statistical' and 'literary' sources for the study of Soviet history.
 * In this article, Ellman concludes that literary sources, and, especially, the book by Antonov-Ovseenko (in the latter book he reproduces some figures that he heard when he was a GULAG prisoner) are unreliable sources, and that early estimates of human life losses in the USSR (which relied mostly to literary sources) are unreliable.
 * Another article by the same author, [Soviet Repression Statistics: Some Comments. Author(s): Michael Ellman. Source: Europe-Asia Studies, Nov., 2002, Vol. 54, No. 7 (Nov., 2002), pp. 1151-1172 https://www.jstor.org/stable/pdf/826310.pdf] is a more recent source, and it performs a detailed analysis of human losses in Stalin's era. To the best of my knowledge, these figures (at least, all that relates to GULAG) represent scholarly consensus. I make this conclusion based on the results of my work on the GULAG article. The edits that I made many years ago, and the sources that I added, are still there, and they have not been seriously challenged by other users. Therefore, I may conclude that the sources that I found during the above search procedure are good, and they reflect scholarly consensus. Note, I found those sources "in real time", I didn't cherry-pick them, and this procedure lead me to essentially the same facts and sources that currently represent a consensus view on the history of repressions in the Soviet Union. That means the above search procedure is quite adequate.


 * Step 5. What about Rummel's sources?
 * Analysis of Ellman's "Soviet Repression Statistics..." shows that the author used a large amount primary and secondary sources about different aspects of Soviet repressive machinery. The full list of sources and endnotes is 7 pages long, so it is impossible to reproduce it here. A significant part of those sources are books, articles, and documents that became available after 1990 ("archival revolution"). Now it is a good time to compare it with Rummel's sources.
 * Rummel's sources for the USSR can be found here, here, here, here, here, here, here, and here
 * The problem with these lists is that they provide no full references. I can look only by names, but this, or this is not informative at all. I recognize some names, for example, Nikolay Tolstoy, a post-WWII British author whom Dulic dismissed as extremely unreliable. Next, the name of Antonov-Ovseenko already appeared in this discussion, it is the very same source that is considered by Ellman as non-reiiable (that is actually a hearsay). Similarly, Rummel uses Bukovsky's memoirs, Solzhenitsyn's figures etc as a source for statistics, and all of that is not considered as reliable data by modern authors, such as Ellman and other scholars from his pool. That my claim is totally verifiable: just open the sources that are listed in step ##2,3 (and other sources in that list) and check that by yourself.
 * Not only Rummel's sources are unreliable, they are outdated. Rummel cites Dyadkin, but that book was published in 1983. Rummel cites Dallin&Nikolaevsky, and I perfectly know that source: it is "''David J. Dallin and B. I. Nikolaevsky, Forced Labor in Soviet Russia (New Haven, 1947), p. 13". Just think: it is the source published in 1947!!!
 * Furthermore, Rummel cites some "census data". Obviously, he cites some old and incomplete data, because more recent data, published by E.M. Andreev, L.E. Darskii & T.L. Kharkova (this source is extensively used by Whaeatcroft, Ellman and others) totally debunk his "democide" estimates: ADK say that demographic losses in the USSR from 1927 to 1946 were about 56 million people, and these losses included 26 million death during WWII, more than ten million birth deficite, migration etc}}. How can all of that be consistent with Rummel's data in table 4a (11,440,000 deaths in 1929-35), table 5 (4,245,000 deaths in 1937-38), table 6 (5,104,000 in 1940-41), table 7 (13,053,000 deaths in 1941-45: I assume he concludes all civilian deaths, including the Holocaust deaths were "Communust democide"?), and  table 8 (15,613,000 deaths in 1956-53). It is inconsistent with all available modern data.
 * I can tell you what was the main source of Rummel's error. If you look at the tables you will see that Rummel (i) dramatically exaggerated the population of GULAG, and (ii) made a totally arbitrary assumption about the death rate. These two blatant errors are responsible for a significant part of his exaggerated figures. I am familiar with this topic, because creation of this table was essentially made by me. I know that some early estimates of GULAG population were unreasonably high, and that modern authors do not consider these figures are reliable. Obviously, the word "modern" cannot be applied to Rummel in that sense.
 * It is now a scholarly consensus that the number of people who passed through GULAG was 14 million + 4 million in colonies (two years term), and the mortality rate was really high only during short periods of its history. These two blatant errors are sufficient to totally remove Rummel's data and leave it in a footnote as a historical estimate. I would support keeping Rummel because he is popular in blogosphere, so its complete removal may lead to accusations of "bias".
 * Conclusion
 * Why should we use such sources as Ghodsee for critique of the views of the authors who share Rummel's ideas? The answer is simple: serious authors ignore Rummel. It is as hard to find serious sources that criticise Rummel as to find serious sources that debunk flat Earth theory.
 * Your comments are welcome. Participants of the DRN may make their comments at WP:MKUCRSA too, whereas other users are welcome to comment here.--Paul Siebert (talk) 19:09, 22 December 2021 (UTC)
 * It seems to be your logic that criticism of Rummel's numbers indicates that they aren't to be relied on, but also, lack of criticism of Rummel's numbers shows that his numbers can't be relied on. Do you see the problem? Regarding flat earth theory, to rewrite the laws of spacetime is not the same thing as to have death estimates that are 10-20% too high; that is an analogical fallacy AShalhoub (talk) 09:14, 23 December 2021 (UTC)

Paul's flawed methodology applied to other authors gets the same results

Paul's methodology is gravely flawed and can be used to "prove" any author is "undue". Let's do the same for Ellman. So by your methodology everyone ignores Ellman his use is "undue" any where in Wikipedia. --Nug (talk) 21:29, 22 December 2021 (UTC)
 * Step 1. Search for mass deaths. Ellman (with impressive 121 citations appears on the first page. Ok, he passes a preliminary step 1.
 * Step 2. Let's see who cites him. To this end, check other authors from the top of the list.
 * Ellman: cites himself twice
 * Wheatcroft: one citation
 * Mertelsmann and Rahi-Tamm no citations,no citations
 * Wemheuer: no citations
 * Rosefielde: cited Ellman once.
 * Kay: cited Ellman 0 times
 * Krischel: cited Ellman 0 times
 * Harris (this source is irrelevant) (and it doesn't cite Ellman
 * Eberstadt: no citations
 * Dyadkin: no citations
 * Ginger - irrelevant
 * Norman - a very specific source about one late Soviet incident. Irrelevant.
 * Snyder: he doesn't cite


 * Compare this and that.
 * Compare this and that
 * Compare this and that
 * In addition, this is just a preliminary step. The authors whom I site are in a permanent interaction: they take data and arguments from each other, they comment on each other, they belong to some common pool. In contrast, I saw no attempt to discuss Rummel by any of those authors. Paul Siebert (talk) 21:47, 22 December 2021 (UTC)
 * Interesting, thanks to you, I found Rosefielde's review on Rummel's "Lethal politics" among the above search results. Rosefielde finds 10 million collectivization death a quite plausible figure, but that was 1990 Rosefielde! He reconsidered his estimates after "archival revolution" (which Rummel refused to do). In addition, Rosefielde notes serious flaws:
 * "Specialists of various persuasions however will be dissatisfied with numerous aspects of the work. Rummel relies entirely on English language sources, some of which are of dubious worth. In one instance he cites a Newsweek report of a classified CIA study to support his estimate of post-Stalinist democide. Unfortunately the reporter either misread the primary document, or was misbriefed. As a consequence, Rummel greatly overstates the likely magnitude of democide 1954-87. Similarly, he ignores most of the specialized demographic, jurisprudential and Sovietological literature on the subject which makes his calculations extraneous to the on-going debate. His estimates are not so much erroneous as they are unresponsive to a welter of serious counter-evidence."
 * Actually, even the 1990 version of Rosefielde confirms my conclusions that I made. Rosefielde was not aware of many facts that we currently know, but he correctly identified the problem. Another important problem is Rummel's indiscriminated approach:
 * "Sovietologists will also be disappointed that Rummel does not carefully explore the applicability of his thesis to the current situation. Is democide an artifact of immature authoritarian Marxism, or should we anticipate a new wave of mass murder if and when Gorbachev finally capitulates to the right? Are glasnost' (openness), demokratzatsiia (democracy), novoe myshlenie (new thinking), and perestroika (radical economic reform) antidotes, or will democide persist even under a liberalized Soviet Marxist regime? These are profound issues which should not have been overlooked."
 * Again, Rosefielde correctly notes that Rummel's obsession with numbers and formal aspects makes his "theory" worthless: it has virtually no predictive power.
 * I think it is obvious from this review why Rummel is essentially ignored by serious historians. Paul Siebert (talk) 22:10, 22 December 2021 (UTC)
 * Actually, that is just a start of the work. If we continue, we will find that there is a huge domain of sources connected by cress-references, where authors discuss each other's findings or use them. I just started this work, but I have a strong impression that Rummel is not a part of the domain that combines Ellman Wheatcroft, Davies, Getty, Alexopoulos, Werth, Graziosi, Khlevniuk etc.--Paul Siebert (talk) 21:55, 22 December 2021 (UTC)


 * So far, only and  presented a reasonable criticism. Nug pointed at some inconsistencies in my analysis, which, as I demonstrated, were just apparent. I am still waiting for his comments on other aspects of my analysis of Rummel.
 * AShalhoub made some comment that I find superficial. It should be clear from my analysis that I didn't speak about merely a lack of criticism. The point was different: (i) Rummel uses different sources that all other (modern) scholars do, and these sources are considered by them outdated and unreliable, (ii) Rummel does not use the sources modern scholars rely upon, (iii) Rummel is not cited by modern experts in Russian and Soviet history, and (i)&(ii) is a possible explanation. In this situation, lack of criticism by no means can be seen as an argument in support of his reliability. For experts in Soviet Russia, Rummel is a fringe author.
 * WRT "10-20% too high", that is a total ignorance. Rummel's figures for the USSR are 6-10 times higher, and that follows from the information presented by me. Please, read my posts carefully before commenting.--Paul Siebert (talk) 15:52, 23 December 2021 (UTC)
 * @Paul Siebert In your argument above, I gather that you want to strike Rummel's estimates from the article. I think the approach of throwing every link and argument against the source without context is not the correct one. In the first place, I don't believe any of us are domain experts and we're not in a position to pick apart whether a researcher as cited as Rummel is using the correct methods, because this is a type of original research. What we have to do is cite other people that are domain experts, that have made this criticism of Rummel (like Karlsson): "While Jerry Hough suggested Stalin’s terror claimed tens of thousands of victims, R.J. Rummel puts the death toll of Soviet communist terror between 1917 and 1987 at 61,911,000. 121 In both cases, these figures are based on an ideological preunderstanding and speculative and sweeping calculations." That is a fair criticism. I would ask you, with your claim "Rummel's figures for the USSR are 6-10 times higher", do you mean the death toll of Soviet Communism is actually 6 to 10 million people, or are you cherry picking what you see as the weaker estimates of Rummel to discredit his estimates entirely? Also, the tables you posted (without context, which is never a good idea) don't seem to show what you described. For one example, table 8 does not seem to show death estimates from 1953-1956, but aparently from 1930 (again, the context to this table is lacking, so it's hard to tell). In any case, I definitely think it's hyperbolic to compare Rummel's estimates to flat earth theory, if he indeed placed them at 60 million. Is this really far enough from "the consensus" to make that kind of comparison?AShalhoub (talk) 16:41, 25 December 2021 (UTC)
 * I always prefer a step-wise approach. First, we must come to agreement about one obvious thing: Rummel's figures (especially for the USSR) are dramatically inflated and outdated, and Rummel's theory is nearly totally ignored by historians who study Russian, Chinese and Cambodian history. IMO, that is absolutely obvious from my analysis, from the quotes that I provided, and from the results of two RSN discussions.
 * Second, after we all agreed upon that, we should, keeping in mind that Rummel's ideas are popular in blogosphee and were reproduced by many politicians and journalists, should discuss a proper context and a way this information should be presented in this article. I will oppose to a complete removal of this information from this article, and I have some ideas how to present these figures. However, I prefer to discuss that after we come to consensus on teh first issue.
 * WRT your concrete question about figures, a consensus figure for GULAG deaths is 1.6-1.7 million (some authors include deportation deaths and get 2.5 million). Compare this figure with Rummel's GULAG mortality data (he obtained several million of GULAG death only for the period of a complete GULAG dissolution, and the deaths in GULAG proper were tens of millions, according to him.) Paul Siebert (talk) 17:12, 25 December 2021 (UTC)
 * Most of Rummel's estimates were actually ranges with a probable mean, for example with the Soviet Union, Rummel estimated a range from 28 million to 127 million deaths with a probable mean of 62 million people. Rummel did also note that he would be amazed if future research did not come up with figures that deviated significantly from his own. In any case, Valentino estimates between 10 to 20 million died under Stalin's rule alone and up to 2.5 million pre-Stalin deaths, so the claim of Rummel's estimate being 6 to 10 times higher is certainly an exaggeration. Again you repeat "Rummel's theory is nearly totally ignored by historians who study Russian, Chinese and Cambodian history", but that is a flawed because global experts rely upon and cite country experts for the original death toll numbers, but the inverse is not necessarily true. --Nug (talk) 19:19, 25 December 2021 (UTC)
 * (The talk page becomes so convoluted that I may overlook your posts. Please, ping me next time if you want me to respond)
 * And why do we need to know his range if modern figures are much more accurate and trustworthy? In addition, this range is meaningful only if each estimate is equally trustworthy, and the distribution is not skewed.
 * You can see that by yourself. Take the figures from this table and calculate a range, and the most probable estimate. And then compare it with modern consensus data (1.5-2.5 million depending on a year). That will allow you to see how "accurate" Rummel's approach is.
 * WRT Valentino, I believe both you and I already agreed that genocide scholars are not good source for figures, because they are not experts, and they, as Harff conceded, do not need to know accurate numbers.
 * WRT the rest, I responded in details in my post in the very last section. Paul Siebert (talk) 23:09, 25 December 2021 (UTC)
 * And, by the way, if you provide a range from 28 million to 127 million, how can you claim the most probable figure has so many significant figures? He says "61,911,000". It should be clear to any educated person, that that precision is simply illiterate! Paul Siebert (talk) 23:21, 25 December 2021 (UTC)

I continue with Rummel. In my previous posts, I presented the results of just one google scholar search. To rule out a possibility of a bias in keyword selection, I've done other searches with different keyword sets. In all those lists, I see familiar names, or names of co-authors of the scholars from the previous search, or names of those whom the scholars from the previous search cite. However, I couldn't find Rummel at first 5 pages of each of those lists. I conclude that my initial search result was more an exception than a rule. You can easily check this my claim by doing an independent analysis of the above lists.
 * soviet repressions collectivization
 * soviet Volga famine
 * soviet stalinist repressions

I think, if there will be no additional arguments or considerations, a conclusion from this discussion should be: an undue weight was given to Rummel's figures and Rummel's "democide" in this article. The scholars like Ghodsee hardly deserve mention in this article, but Rummel must be moved to the very bottom into some section that should be named "Attempted generalisations" (or something like that), and mentioned only briefly as a minority POV.--Paul Siebert (talk) 16:12, 23 December 2021 (UTC)


 * Paul’s “interaction” analysis:
 * Compare this and that.
 * Compare this and that
 * Compare this and that
 * is also flawed, as it is looking for interaction across different disciplines, between single country experts and systematic global study experts. Global study experts like Rummel rely upon the data from single country experts to compile data from each country into a global dataset. But the inverse does not hold, single country experts do not rely upon data from systematic global empirical study experts, that would be circular. So Paul’s claim that single country experts ignore Rummel doesn’t indicate anything other than these people work in a different domains, one involving global study of data complied from single country experts, the other involving single country studies by single country experts.
 * Barbara Harff is a renowned scholar in the global study of mass killings, and yet according to Paul’s flawed methodology, there is little interaction between Harff and Wheatcroft or Ellman and Harff, therefore single country experts are ignoring Harff because her data is unreliable. Paul analysis doesn’t really prove anything. --Nug (talk) 20:42, 23 December 2021 (UTC)
 * With due respect, I do not accept your criticism. First, you discuss just a minor part of my initial post, but you make a claim about some general flaw in my approach. That is an unjustified generalisation. Please comment on other aspects of my original post, and if you see any problems or flaws, please, explain that. If no criticism will follows in next few days, I have a right to interpret that can implicit acceptance of my arguments and conclusions, which I will do.
 * Second. I agree with your argument about " different disciplines, (...) single country experts and systematic global study experts". Yes, these are two different groups of authors.
 * Third, I find the rest unconvincing. To acknowledge the existence of two groups is just a first step. The next step is to establish interrelations between these two groups, and relative weight of their views. That is what I am doing, and that is what you failed to do. And, after we identified the first group, we need to establish a relative weight of Rummel within the first group, for even many authors in the first group disagree with him.
 * Fourth. With regard to global data, as I demonstrates, Rummel's data are of poor quality, and country experts do not rely upon them not because they belong to different domains, but because they are garbage.
 * Your argument about Harff is valid. It demonstrates that Harff also works in isolation from country experts and she belongs to the "genocide scholars" domain. I see no reason why that may be an indication of any flaws in my approach, because that is a conclusion made in one peer-reviewed publication that I cited on this talk page: yes, most "genocide scholars" work in isolation from country experts, who ignore their works. And that requites us to decide, what viewpoint should be present as a majority view and what is a minority/fringe. I proposed to discuss it, and that is a reason why I started this source analysis.
 * This work has already lead to one positive conclusion: both and I acknowledge existence of two distinct domains (school of thought), country experts and "genocide scholars". The next step is to develop a common approach to determine relative weight of their viewpoints.--Paul Siebert (talk) 21:06, 23 December 2021 (UTC)
 * P.S. You are not right about Harff: . Actually, she is recognised by country experts, but these experts specialise not on Russia, but on Cambodia. Harff performed a comparative analysis of Cambodian and Indonesian genocide, and this work is recognised by country experts.--Paul Siebert (talk) 21:10, 23 December 2021 (UTC)
 * All of Paul’s arguments against Rummel are based on the premise that the basic goal of all academic research is to determine an accurate number of deaths, and thus if he shows Rummel’s numbers are inaccurate then that somehow makes Rummel unreliable. But that premise is flawed because accuracy, while important, isn’t the core goal here, as Harff states:
 * ”Compiling global data is hazardous and will inevitably invite chagrin and criticism from country experts. Case study people have a problem with systematic data because they often think they know better what happened in one particular country. I have sympathized with this view, because my area expertise was the Middle East. But when empiricists focus on global data, we have to consider 190 countries and must rely on country experts selectively. When we look for patterns and test explanations, we cannot expect absolute precision, in fact we do not require it. ”
 * Rummel’s numbers were in fact estimated ranges with a likely minimum, maximum and median numbers, and to focus on accuracy of a fixed number as some kind of measure of "fringe" really is disingenuous. As Harff also states:
 * "As someone who had tried to identify all cases of geno/politicide since World War II, I fully understand what it takes to collect reliable, unimpeachable global data. It is impossible. The best we can do is to seek advice about additional episodes, and to report ranges of fatality estimates, as Rummel does."
 * Your comment "Fourth. With regard to global data, as I demonstrates, Rummel's data are of poor quality, and country experts do not rely upon them not because they belong to different domains, but because they are garbage." is flawed. Global experts rely data from country experts, not the other way round. --Nug (talk) 21:24, 23 December 2021 (UTC)
 * See my response to
 * A core of Rummel's approach is finding correlations, more concretely, the number of "democide victims" vs "regime type". If his figures are 10 times greater than in reality, that dramatically affects eigenvalues and eigenvectors, and he may see correlations that are not observed if other data sets are used.
 * However, Harff's point is valid: "genocide scholars" do not need to have very precise data. However, that means they should not be used as a source of data: if we know (and they themselves recognise that) that their data are not precise, what is a reason to show them? Maybe, it makes much more sense to show much more preceise and fresh data obtained by country experts?
 * With regard to Rummel's words about "impossibility" to collect some data, I doubt that opinion of a non-expert is relevant. Rummel is not an expert in source criticism, he is not an expert in demography, he is not an expert is Soviet history, he is not an expert in Chinese history, and I would prefer to see the opinion of some country expert or demographer on that.
 * And, yes, global experts rely on data from country experts, but Rummel relied on OLD data from UNRELIABLE country experts, and these data are considered as unreliable by MODERN country experts. That means, Rummel relied on BAD and OLD data from QUESTIONABLE country experts. Paul Siebert (talk) 21:37, 23 December 2021 (UTC)
 * That may be a valid and true criticism. However I think the best way to proceed would be to find reliable sources that espouse that criticism and we can evaluate them. As it stands this is OR that we cannot put in the article. Vanteloop (talk) 22:59, 23 December 2021 (UTC)
 * At the top of this section I posted an explanation that careful and critical analysis of sources is what NPOV requires us to do. I also reminded all participants of this discussion that a reference to OR will be interpreted as disruption. Please, stop disruption.--Paul Siebert (talk) 00:34, 24 December 2021 (UTC)
 * Ah yes, the old 'It's not OR because I said so and anyone who disagrees with me is being disruptive'. Not very convincing Vanteloop (talk) 00:38, 24 December 2021 (UTC)
 * It is not OR because that is what policy requires us to do. You accused me of OR, please, demonstrate what exactly I am proposing to add to the article that was not explicitly stated by some reliable source. You should either present some evidence or apologise. Paul Siebert (talk) 00:41, 24 December 2021 (UTC)
 * I literally just asked you to provide some sources for your claims, you then claimed that was disruptive behaviour. I ask you to strike that through as a gesture of good faith Vanteloop (talk) 00:44, 24 December 2021 (UTC)
 * I cannot provide sources that say "there is no sources on that". And it would be ridiculous to expect somebody else could do that.
 * I explained everyone that references to OR is a disruption, but if you agree not to mention OR in that context then the incident may be considered resolved. Paul Siebert (talk) 00:50, 24 December 2021 (UTC)
 * And I explained to you that you do not get to dismiss concerns about policy violations just because you post a message saying that raising those concerns is disruptive. Your opinions on Rummel are not given special weight just because they are on a talk page, any more than they would on a random blog. They have to have citations so that we can verify your claims, else please publish your research in a journal and then come back Vanteloop (talk) 00:58, 24 December 2021 (UTC)
 * There is a big difference between raising a legitimate concern about a policy violations and wikilawyering. To claim that careful and critical analysis of a variety of reliable sources is an original research is hardly a legitimate concern about policy violations. Paul Siebert (talk) 16:42, 24 December 2021 (UTC)
 * Furthermore, to be clear your accusation my request for ciations was 'disruptive' and then refusal to strike constitutes a clear assumption of bad faith, I suggest you take this second opportunity to clarify otherwise Vanteloop (talk) 01:08, 24 December 2021 (UTC)
 * Can you please explain me this your post in different words? I simply don't understand it. " your accusation my request for ciations" - what does it mean? Whom I accused and of what? Which "citations"? Paul Siebert (talk) 16:38, 24 December 2021 (UTC)
 * I note that this is not the first time you have said you don't understand when called out on bad faith accusations. I will assume good faith and remind you of the comment just a few above this one. I literally just asked you to provide some sources for your claims, you then claimed that was disruptive behaviour. I ask you to strike that through as a gesture of good faith. I also note you didn't respond until you accused me of personal attacks and I reminded you that this comment went unaddressed. I don't believe any editor can reasonably be asked to engage when requests for evidence are immediately met with accusations of bad faith. Vanteloop (talk) 16:59, 24 December 2021 (UTC)
 * Well, your original post (22:59, 23 December 2021 (UTC)) contained two main theses:
 * "the best way to proceed would be to find reliable sources that espouse that criticism and we can evaluate them."
 * "As it stands this is OR that we cannot put in the article."
 * I responded that "careful and critical analysis of sources is what NPOV requires us to do". If it was not clear from that my post, I am explaining it again: the whole section where I presented my analysis was aimed not to put some new information in the article, but to rearrange/remove some information. It should be clear to any user who understands our policy that this my activity has no relation to NOR, and it is in a full accordance with NPOV. I believed that was obvious to everybody, but if it was not obvious for you, then I am sure it is obvious now.
 * Furthermore, I explained in my another post, that it is ridiculous to expect a source that confirms that Rummel should not be used as a source for some claims in Wikipedia articles. And that should be also obvious to anybody who is familiar with our policy. By writing that, I already addressed your first claim. Moreover, I presented some sources in my other post (01:09, 24 December 2021 (UTC)), which you de facto ignored (I do not consider your acrimonious comment as an adequate responce).
 * Therefore, I owe no apology to you, because I assumed you understand our policy.
 * My previous warning is still in force. Paul Siebert (talk) 17:17, 24 December 2021 (UTC)
 * I will assume good faith that you misunderstood again, because your comment completely fails to address my argument that immediately accusing a user of 'disruptive behaviour' for requesting evidence is not acceptable, especially when asked multiple times to withdraw the accusation. Please withdraw your baseless accusation to 'stop disruption' instead of misdirecting onto some other matter. My previous warnings are still in force. Vanteloop (talk) 17:36, 24 December 2021 (UTC)
 * You probably haven't noticed, but the whole huge post that I made at the beginning of this section was an analysis of tons of sources and other evidences. This post was supplemented with a reservation that that it is not a proposal to add some statement to the article, and therefore it is cannot be considered as a piece of OR.
 * In responce, you requested for some evidences and claimed that that may be OR. That may mean one of the following:
 * 1. You haven't bothered to read my post, or
 * 2. You didn't understand it, or
 * 3. The post was not a good faith action.
 * I would prefer if 1 or 2 were true reasons. If that is the case, please, read my explanation and stop this nonsense Paul Siebert (talk) 17:50, 24 December 2021 (UTC)
 * And yet you immediately accused another user of bad faith. I have extended good faith in that your misunderstandings are genuine, however it is clear at this point you will not apologise for for this baseless accusation of bad faith (a sanctionable offence), despite being given more than enough opportunities and explanations. Therefore this conversation may be carried on in a more appropriate venue Vanteloop (talk) 18:05, 24 December 2021 (UTC)
 * Here and elsewhere Siebert backed up their statements, so what is the problem exactly? Your opinions on Rummel are not given special weight just because they are on a talk page, any more than they would on a random blog. Yet, that applies to everyone involved, so why dismiss Siebert when they say Rummel is given more weight than is warranted but not doing the same to those who disagree, even though they did not provide any scholarly source explicitily saying that either? And if we are wrong, it should be easy to provide, and we have actually provided them time and time again (one of this article's core source Karlsson saying Rummel's estimates are ideologically biased, see also both Harff's summary and Mann's criticism, and RSN's discussions). Siebert's source analysis has actually been positively reviewed in an academic journal, so while this does not mean they are always right, I find criticism of them deeply unfair and unjustified. You also should not ask us to prove a negative (e.g. that country experts do not rely on Rummel for the events or estimates, even though I think both Siebert and Harff 2017 have already proved and supported this claim). As I said many times, Rummel is unreliable or undue for estimates and events, especially about Communism, but he is reliable, though not without criticism and other genocide scholars hold more weight, ot at least due for the theories and links of Communism and mass killings. Davide King (talk) 16:56, 24 December 2021 (UTC)
 * Yet, that applies to everyone involved, so why dismiss Siebert when they say Rummel is given more weight than is warranted but not doing the same to those who disagree, even though they did not provide any scholarly source explicitily saying that either? I couldn't agree more. Every editor should be able to provide citations for their claims, and no editor should make accusations of bad faith in response to a single comment asking for evidence. Vanteloop (talk) 17:05, 24 December 2021 (UTC)
 * I am summarising, one time, specially for you.
 * Rummel was criticised by Dulic (I provided links on this page) for flawed approach in general and for usage of poor sources for Yugoslavia.
 * Rummel was criticised by Rosefielde (I already provided links on this page) for usage of questionable sources and for his approach that ignore important nuances
 * Rummel was criticised by Nathan (I already provided links on this page) for poor sources for China
 * Rummel was criticised by Karlssen (the work is cited in the article) for inflation of figures and for ideological bias, which makes him a bad source
 * Rummel's "democratic peace" theory was criticised by Mann (p.22 of his book)
 * Rummel uses very old sources that are not considered as serious sources by all modern experts in Soviet Russia (see the analysis on this page)
 * Rummel uses a very ridiculous procedure for calculation GULAG "democide" (see Rummel himself): he takes absolutely unreasonable data on GULAG population (which are currently considered as a nonsense) and calculated the number of deaths based on his own assumptions about the prisoners mortality rate (absolutely amateurish approach, which is fare less reliable than detailed and nuanced estimates made by modern authors)
 * Rummel's figures are ignored by experts in Soviet Russia, and his works are ignored too (see my analysis).
 * Rummel was concluded to be unreliable for figures twice at RSN.
 * Now please answer, why Rummel is still in this article?--Paul Siebert (talk) 01:09, 24 December 2021 (UTC)
 * As Nug rightfully pointed out. Siebert's Special Analysis tm could be used to say any author is 'undue'. How about this, if you are so confident in your patented analysis, post it as an online blog and then we can take your blog to RSN. If it shown as reliable I will personally apologise to you. Furthermore you have twice been asked to rectify an assumption of bad faith and it is noted you have not done so Vanteloop (talk) 01:21, 24 December 2021 (UTC)
 * You may apologise right now. This author in this article said that I, Wikipedia user Paul Siebert use a good approach for identification of reliable sources. He says:
 * "It is Paul Siebert, champion of the more traditional information routine, who digs deeper into the sources than anyone else."
 * (To better understand a context, keep in mind that Luyt consider "traditional information routine" as something good.) Since 2013, my approach became even more advanced.
 * Actually that is not an argument. My approach is good simply because you guys haven't proposed anything. No alternative approaches, no fresh sources, no reasonable criticism - nothing. Are you going to collaborate? Paul Siebert (talk) 01:41, 24 December 2021 (UTC)
 * Brilliant! I assume that means you are taking up my suggestion. Be sure to include that paper in your blog so we can discuss it at RSN. I'm sure such an even more advanced pioneer of historical analysis will have no trouble getting consensus as reliable. I await your blog post. While you write, it may be worth considering that if the wheels fall off your entire system when one users asks for evidence it may not go very far Vanteloop (talk) 01:48, 24 December 2021 (UTC)
 * It looks like an attempt to derail a serious discussion. Please, stop. Paul Siebert (talk) 01:54, 24 December 2021 (UTC)
 * Paul Siebert, champion of the more traditional information routine, Let me remind you of my first comment which started your digressions and which you have still not answered. That may be a valid and true criticism. However I think the best way to proceed would be to find reliable sources that espouse that criticism and we can evaluate them. As it stands this is OR that we cannot put in the article. In fact you are the one who derailed the discussion because you were unable to answer this simple constructive criticism. Vanteloop (talk) 01:58, 24 December 2021 (UTC)
 * My post that I made on 01:09, 24 December 2021 (UTC) was a detailed answer to your question. My editor automatically added it to a wrong place. Read it and give me reasonable comments. Paul Siebert (talk) 02:02, 24 December 2021 (UTC)
 * Nug, do you understand that what you quoted proves that the topic should be B? Genocide scholars are good sources for B, not for A, for which they must rely on country experts. Country scholars are the best sources for A, so the only way to have a NPOV C article is to rely on country experts for A (e.g. summarizing and describing the mass killing events, rather than treating them as death toll events) and genocide scholars for B. Davide King (talk) 21:55, 23 December 2021 (UTC)
 * No it doesn't. Please don't interrupt discussions if you have nothing relevant to add. --Nug (talk) 22:28, 23 December 2021 (UTC)
 * Actually, DK is partially right, and you are somewhat rude. Paul Siebert (talk) 03:14, 24 December 2021 (UTC)
 * Some genocide scholars may be good sources for "A", and some country experts may be good sources for "A". A relative contribution of teh former and the latter can be determined based on source analysis. However, the only user who proposed any approach was I and . 's criticism focuses on petty nuances, and we cannot move forward. Are you guys going to work together? Can you propose your own ideas how to solve the problem?Paul Siebert (talk) 00:46, 24 December 2021 (UTC)
 * , I do not necessarily disagree but that is the point — they have continually refused to do that, and if they truly think your approach is flawed, to semi-quote Valentoop, "get [yourself] published in an academic journal"; as things stand, an academic journal positively reviewed your approach, which does not mean you are always right, but we clearly need better criticism than that. In regards to this, if one does not see a difference between lower than 10 million (low) and 20 million (high), and 28 million (low) and 127 million (high), I do not know what to tell them.
 * But perhaps they outlined the problem there — they unwittingly confirmed that country experts are majority sources and genocide scholars are the minority because the former are relied on by the latter. If the reverse was true, and they too were a majority, surely genocide scholars would be widely cited in the context of Communism? And yes, NPOV remains the major issue (again, I ask Nug that they stop single outing me in regards to OR/SYNTH when I am clearly not the only one to have brought it up and the AfD's closure gave it enough weight, along with NPOV issue, to not be able to result in consensus for 'Keep' despite the obvious numerical advantages, which makes it an even bigger result because our arguments were able to hold enough weight to not warrant an unambiguous 'Keep'), so can you tell us how we can solve this when your comment shows that country experts represent the majority views and genocide scholars are the minority? Davide King (talk) 19:39, 25 December 2021 (UTC)
 * And before I get strawmanned ("If the reverse was true, and they too were a majority, surely genocide scholars would be widely cited in the context of Communism?"), while I would not find it absurd if a country expert cited Harff of a genocide scholar (just because the latter write in a global context, it does not mean country experts cannot find useful insight from particular events and countries), what I am referring to is the lack of references to Harff, Rummel, Valentino et al. in works about Communism as a whole (e.g. The Cambridge History of Communism and Oxford Handbook of Communism). They discuss sport, religion, and link with violence and terror (B) but not this topic (A or C). Michael David-Fox actually contributed to The Cambridge History of Communism, so he is a mainstream figure and his criticism of Malia cannot be dismissed as "revisionist." Davide King (talk) 19:55, 25 December 2021 (UTC)

A general discussion

 * This is precisely what we need to write a NPOV article without OR/SYNTH; sources need to be connected and cross-referenced, and we should use those who do, and not those who operate in a vacuum like Harff and Rummel, even though I personally respect them both and appreciate their scholarly work. The problem is that none of those sources (country specialists) appear to write within the context of this article, so I think TFD is right that we cannot use them,1 which is why I had to rely on Ghodsee and is absurd to consider her undue as things stand. But the problem is not Ghodsee, the problem is that mainstream scholarly sources ignore the sources who write within that context, and genocide scholars are distorted to falsely imply they support Courtois, Malia, and Rummel's thesis, which is the really notable topic but currently fails NPOV and OR/SYNTH because we got the structure totally wrong. As asked: "Is this 'common criticism' the consensus view in the literature, is it a left-wing view or is it the view of communist apologists?" It is, in fact, the consensus or mainstream view (e.g. scholarly reviews of The Black Book of Communism).2
 * Notes
 * 1. Again, here comes the contradiction. To solve NPOV issues, we must rely on country experts and mainstream, majority views of each mass killing events (e.g. structure A — using Ó Gráda and Wheatcroft rather than Conquest and Dikötter), and on genocide scholars for the theories (e.g. structure B — some genocide scholars hold more weight than other); however, scholarly sources about A write within the context of each country, scholarly sources B write within the context of global Communism (though majority of them limit themselves to Stalin, Mao, and Pol Pot). By merging the two, we are committing OR/SYNTH, falsely implying they rely on, and support from, each other.
 * 2. The problem is that there is no literature for what we currently do (as noted by, "such an amalgamation (as we have currently) effectively uses the list from A as synthesis to support the arguments presented in B; it's an inherently non-neutral article", which is also what TFD and Siebert have been saying for years, e.g. this article tries to prove Courtois', Malia's, and Rummel's thesis, hence Ghodsee appears as criticism, when it is simply a mainstream view (in regards to body-counting, especially Communism) but there are no better sources because they ignore the topic as currently structure. As noted by Siebert, this article does not even present a NPOV summary of the mass killing events according to majority scholarly sources and specialists, it treats them as death toll events.3
 * 3. Rather than trying to explain the events when we discuss them, what happened, why (which does not necessarily imply communism), and what is to be learned from it, we put death tolls. This is why I think TFD suggested to name it Victims of Communism because that is what we do; we do not summarize the events, we treat them as death toll events/communism as the main cause, and since we put an overemphasis on ideology and generalize and oversimplify too much (this is not surprising considering the level of sources currently used, all representing minority views), we are essentially following the Victims of Communism Memorial Foundation's narrative as mainstream and majority view, rather than disputed (e.g. on scientific and methodological grounds, not that many people were indeed victims of the system, which is not disputed or denied) and controversial (e.g. "Advocating the cause of the victims of Communism" in The Criminalisation of Communism in the European Political Space after the Cold War).
 * Davide King (talk) 22:44, 22 December 2021 (UTC)
 * If it was renamed "Victims of Communism," then we could concentrate more on Rummel because he was a member of the National Advisory Council of the Victims of Communism Memorial Foundation, along with Robert Conquest, Bob Dole, Jack Kemp, Richard Pipes and others. Rummel was possibly the first to come up with the the over 100 million victims estimate in his 1993 unpublished essay "HOW MANY DID COMMUNIST REGIMES MURDER?" The Black Book was not published until 1997. And then there would be no need to cite authors who don't connect mass killings to Communism. TFD (talk) 23:31, 22 December 2021 (UTC)
 * The "Victims of Communism" article will become a loose collection of facts or events that were described as "victims of Communism" by one reliable source. That article will be an NPOV nightmare. And we don't need to concentrate on Rummel, for he and his democide already have their own articles.
 * Actually, I propose to stay focused. This section must answer the question: "What should we do with Rummel as a source?". --Paul Siebert (talk) 00:02, 23 December 2021 (UTC)
 * In regards to "a loose collection of facts or events that were described as 'victims of Communism' by one reliable source", that would be a clear violation of both our policies and guidelines, and the structure, since TFD and I advocate such renaming only for B, which would not entail summarizing or describing the events, so I am not too worried about that. We may distinguish between the popular press views (e.g. Courtois, Malia, and Rummel's thesis popular there — Communism as main cause) and the scholarly views (e.g. Valentino and other scholars who discuss Communism and mass killings as correlations and theories, not as events, which are more nuanced — it is more complicated than simply blaming communism, or communism as a whole rather than authoritarianism or one particular current). In regards to Rummel, he is due only for B but we should cite secondary coverage of him. Davide King (talk) 01:13, 23 December 2021 (UTC)
 * No, because Wikipedia is not a dictionary, it is an encylcopedia. Whereas dicitonaries are about words and their various meanings, dictionary entries are about topics. While words can have different meanings, topics can share the same name. For example, the article "social liberalism" is about the version of liberalism that emphasizes the welfare state. In the United States, people use the term to refer to cultural liberalism, which has its own article. If you define the topic, then the article will stay on focus. We don't have it to call it "the version of liberalism that emphasizes the welfare state." In any case, the current approach hasn't worked despite over a decade of discussion and never will. TFD (talk) 03:46, 23 December 2021 (UTC)


 * I think you guys get it wrong. There is actually a consensus of mainstream academic sources (many of which you just cited above) that all these deaths (in Gulag, during the Holodomor, Red Terror, Cultural revolution, etc.) were a result of the existing political systems. This is something articulated in the Black Book of Communism and in a lot of other books and articles by mainstream researches in this field. Are you saying these deaths were NOT a results of the communist political system? Which sources make such claim? Yes, perhaps one can find a bunch of revisionist historians like Zhukov, Grover Furr, Arch Getty etc., but they would be in minority here. Do not push WP:GEVAL. My very best wishes (talk) 02:40, 23 December 2021 (UTC)
 * If this consensus is so strong then why is the closed circle of the BBoC (and other works by the same authors) and Rummel all anyone can point too? If this consensus was as strong as you claim it is then we wouldn't be here. If everyone agrees on this, then go ahead and prove it. BSMRD (talk) 02:45, 23 December 2021 (UTC)
 * If you look at the thread above, it includes maybe 20 historians (cited on the page), all of which support this mainstream position. There are many many more. Almost all sources currently cited on this page support such position. Including even former top communist party officials, such as Alexander Yakovlev. My very best wishes (talk) 02:54, 23 December 2021 (UTC)
 * No, they don't. That is just a list of authors who have mentioned "soviet mass deaths". You can't just say "this is the mainstream position" and not back it up. That mass death happened under (some!) Communist regimes is not questioned by anybody, but that it was somehow the "fault" of Communism, or that Communism is somehow innately to blame is very much not an undisputed mainstream position. That almost every author cited on this page supports the assertion that "Communism" is responsible for crimes commited under leadders who believed in it is not supported by what they have actually written. Indeed, a minority of the suthors cited on this page are even used to support that claim, being cited for completely unrelated statements. Several authors have been misquoted, and others have been given vastly more credence than they recive in scholarly literature. Frankly, the fact that we are still pretending that Courtois et al. are some respected scholars of Communism or that the travesty that is the BBoC should be used for anything other than ridiculing it's authors is just tiresome at this point. BSMRD (talk) 03:07, 23 December 2021 (UTC)
 * Did you actually read their books? No, most of them do make such connection more or less directly. For example, Yakovlev do makes such connection explicitly. So is Snyder. But it is enough to look at citations currently provided on this page. I should say this page is informative, and in a better shape than a lot of other pages in WP. My very best wishes (talk) 03:21, 23 December 2021 (UTC)
 * Did you really just compare Getty to Furr and Zhukov? You appear to not understand "revisionism" as a paradigm in the context of Soviet and Communist studies (Getty), and historical revisionism (Zhukov), and Furr who is not a historian. I think you are the one who is pushing GEVAL. If you actually read The Black Book of Communism you would know that Werth draw a link between Stalin and Lenin through Nechayev (who was criticized by Marx as "barracks communism"); it is really mainly Courtois, Malia, Rummel, and a few authors who go back to Marx himself; majority of country specialists focus on societal context (e.g. most Communist regimes arising in countries with no long-lasting democratic traditions, civil rights, or private enterprise, etc.). I also have to agree to what BSMRD wrote here. Davide King (talk) 03:12, 23 December 2021 (UTC)
 * Yes, I did read the book. Yes, Nechayev is a quite important and symbolic figure (that's why I edited page about him). But The Black Book of Communism is not anything outstanding. It is based entirely on a lot of other sources (as typical for many books by academics), and does not really contradict anything I read in other books, like the book by Yakovlev, for example. My very best wishes (talk) 03:45, 23 December 2021 (UTC)
 * Ignoring comments about Poland, here Newimpartial (I do not know whether I should ping them or not) gave a good summary of recent scholarship on The Black Book of Communism, which has since been supersed by the Oxford Handbook History of Communism and the Cambridge History of Communism as more accurate and nuanced references.1 In addition, I understand the "political system" to not necessarily refer to communism (just like there is a difference between an ideology justifying and causing something — many Communist leaders indeed justified their policies ideologically but whether Marx or communist ideology actually justified that is disputed) because scholars refer to authoritarianism or totalitarianism (e.g. Rosefielde's "immature authoritarian Marxism") rather than communism itself or as a whole. If this is what you mean by "political system", I do not disagree with you in this regard and I certainly do not condone the awful and tragic stuff they caused.
 * The problem is that this is not really notable, what is notable is communism as main cause (Courtois', Malia's, and Rummel's thesis) but whether this is presented as a mainstream view, it is a controversial, disputed, and minority view in academia but very popular among the non-academic press. To quote Neumayer 2018, "the Black Book of Communism contributed to legitimising the equivalence of Nazi and Communist crimes. The book figures prominently in the 'spaces of the anti-communist cause' comparably structured in the former satellite countries, which are a major source of the discourse criminalising the Socialist period."
 * Notes
 * 1.
 * The quoted part is, in fact, what I believe TFD, Siebert, and I have been saying this whole time and what I just pointed out above, and that this article, on which your thoughts I can respect but are clearly not reflected by DRN and AfD's conclusions, is deeply lacking. Davide King (talk) 04:25, 23 December 2021 (UTC)
 * What you cite are valid sources. They can be used if they say anything about "Mass killings under communist regimes" (I am not sure). Yes, this is really a good question to what degree Lenin and Stalin were driven by the communist ideology. I do not think any serious authors (including authors of "Black Book") ever claimed that Stalin was driven exclusively by the communist ideology; some say that he was not driven by the ideology at all. But this is not the point. What made these crimes possible was not just the ideology (although it did play a role), but the entire political system in the countries under discussion. The "communist regimes" is simply an established terminology, although as always in humanities, one can easily criticize terminology as poor. My very best wishes (talk) 06:19, 23 December 2021 (UTC)
 * Still, the source would have to say that. In the Vietnam War for example, both sides killed civilians. Did North Vietnam kill civilians because the entire political system made this possible or because in any conflict of this nature, both sides kill civilians? TFD (talk) 06:24, 23 December 2021 (UTC)
 * According to my understanding of this subject, the answer is "yes", i.e. the war and people being killed was a product of political systems in all countries involved in this conflict, politically and militarily. One can not blame just the war. But I am not an expert. My very best wishes (talk) 15:31, 23 December 2021 (UTC)
 * TFD&Davide King! That is why I proposed to move the discussion of sources to WP:MKUCRSA. On this page, it immediately becomes so messy that it is not possible to follow. Please, either focus on the analysis of sources or use this section for a general discussion.--Paul Siebert (talk) 15:41, 23 December 2021 (UTC)
 * It's unclear what aspect of Rummel you consider undue? Rummel numbers or Rummel/Horowitz? The source analysis seems to be looking at the former, and general discussion focused on the latter. For the discussion in this section, why not proceed from literature reviews such as this "Other more recent studies identify the communist state or regime as perhaps the most structurally predisposed toward genocide and mass killing, owing either to the magnitude, speed, and scope of the revolutionary goals (Valentino 2004) or to perceived factionalism and dissent that tend to be framed as categorical betrayals of an organic concept of the nation (Mann 2005)."fiveby(zero) 19:03, 23 December 2021 (UTC)
 * I think the issue is that Siebert see all options as neutral and in respect of our policies, and on this we disagree, but I think we agree that Rummel is undue if we focus on the events, on which we must rely on country experts and specialists. For me, and I think them too, Rummel is acceptable for B but not for A. For B, I prefer using secondary coverage of him (e.g. the source you cited, which seems to be perfectly fine for B) rather than the primary work of Rummel himself, and this for other authors too; what we need to rely on is precisely such reviews of the literature to establish WEIGHT and write a NPOV article about it rather than cherry pick from their own primary works. To conclude, my main issue with Rummel is his estimates (A) and treating his theories (B) as universal and uncontroversial, or even as proof in support of C, which is how Rummel is currently used, rather than notable but not universally accepted or without criticism. Davide King (talk) 19:56, 23 December 2021 (UTC)
 * Rummel is clearly undue for figures. In that sense, he (and all old sources) has just a historical value, and they should be combined together and moved to a footnote. With regard to the rest, one has to keep in mind the following:
 * The core of Rummel's approach is the search for correlation (number of killed vs regime traits). If his figures are wrong (and for the USSR they are dramatically wrong), that directly affects eigenvaluses and eigenverstors that he obtains. Therefore, his conclusions should be treated accordingly.
 * Wayman&Tago state Rummel's database is a good framework for studying mass killings during the 1900-1987 period. --Nug (talk) 22:12, 23 December 2021 (UTC)
 * Rummel did not discover "democide", he invented it. He did not conclude that his results lead to a discovery of some new phenomenon, he just said: "Look, many states kill their people. I call that "democide"". He defined a new term that combines some events in one new group. This grouping is not accepted by majority of authors. Thus, Harff (another genocide scholar) does not include majority of "democide deaths" into her database: according to her, only four out of 18 Communist states were genopoliticidal. Therefore, Rummel is undue for "mass killings", because his "democide" is not what most experts see as "mass killings"
 * Wayman&Tago state that Harff's database (politicide + genocide) is a subset of Rummel's database (politicide + genocide + mass murder) and Rummel's 13 out of the 18 Communist states was equally valid given the broader superset that Rummel's database represents. You have wrongly equated "mass killings" with (politicide + genocide), but in fact it has a broader definition. --Nug (talk) 22:12, 23 December 2021 (UTC)
 * As Rosefielde notes, Rummel ignores regime evolution, and he implicitly considers Communist regimes as equally totalitarian from the very beginning to very end. As a result, his approach totally overlooks the fact that most Communist regimes that were killing their people did that only during a very short parts of their history. The same was noted by Wayman&Tago. That also make Rummel's theorising undue.
 * Where did Wayman&Tago say this. I fail to see how regime evolution is relevant here. --Nug (talk) 22:12, 23 December 2021 (UTC)
 * Finally, Rummel's conclusions were rejected by Valentinio, and, what is more important, by such a prominent sociologist as Michael Mann. That, again, makes Rummel outdated and UNDUE. Paul Siebert (talk) 20:05, 23 December 2021 (UTC)
 * Where does Valentino or Mann explicitly reject Rummel's conclusions? --Nug (talk) 22:12, 23 December 2021 (UTC)
 * Mann. "The Dark Side of Democracy", p.22. Paul Siebert (talk) 00:28, 24 December 2021 (UTC)
 * I'll answer to your other questions after you move them from my post, otherwise it will be a mess. In future, please refrain from typing inside my posts. Thanks. Paul Siebert (talk) 00:30, 24 December 2021 (UTC)
 * I don't think it is fair to say that Rummel's conclusions were rejected by Valentinio:"Rummel and others have marshaled strong evidence to demonstrate that democratic forms of government are associated with lower levels of mass killing and human rights abuses than other governmental systems, especially totalitarian and communist regimes. The fact that democracies engage in less mass killing of their own citizens than other forms of government is one of the most carefully documented findings of the theoretical literature on mass killing. Nevertheless, two problems limit the utility of this insight."(p. 27) and citing
 * One major idea to take from that literature review i pointed to is just how complex and difficult genocide studies are, and the perils of WP:OR by editors in this area. fiveby(zero) 11:45, 24 December 2021 (UTC)
 * You are correct, it is possible but we have to be extra careful in assessing claims by editors and ensuring OR does not make its way to the article from the talk page. Vanteloop (talk) 12:05, 24 December 2021 (UTC)
 * Also, while Mann is certainly not convinced by the democratic peace theory, outright rejection of Rummel is not correct:"Like scholars of ethnic cleansing, most observers of leftist atrocities have adhered to a statist view of the perpetrators: mass killings were the top-down work of a dictator or political elite or of totalitarian regimes (e.g., Conquest, 1990; Courtois et al., 1999; Locard, 1996: 131; Rummel, 1992). They emphasize the coherence, premeditation, and planning of the killing. Indeed, these were highly statist regimes, not remotely democratic. They had abandoned their original minimalist view of the state to embrace a decidedly statist (and militarist) view of social transformation, dictatorial top-down planning backed by military-police repression...Yet the process by which this eventuated was complex..."(pp. 320-1) fiveby(zero) 12:35, 24 December 2021 (UTC)
 * It was a pleasure to read your detailed responce. Yes, I am pretty familiar with all these sources, and I agree that it was some exaggeration to claim these author totally reject Rummel. However, you must keep in mind a context. I wrote that during the discussion of a criticism of Rummel, who is currently one of the core sources this article is based upon: It is easy to see that the article's structure is organised in such a way that it implies that Rummel's views are mainstream, and other authors fit into this scheme. In that situation, some exaggeration during a talk page discussion was completely justified.
 * I fully agree that Rummel cannot be totally dismissed, simply because it is correct that some totalitarian regimes were murderous. However, as your quotes correctly say, his theory has a limited applicability, and we must establish these limits: it is definitely a minority view, but we must agree about the amount of space that we can devote to it in this article, and about a proper context.
 * I concluded that Valentino disagrees with Rummel, because that is how secondary sources describe his views (you must agree that Valentino's book is a primary source for his own views). Wayman&Tago say that Valentino "disagrees" with "Rummel's finding that authoritarian and totalitarian government explains mass murder", and "Valentino argues that regime type does not matter". These authors do not use the word "reject", but they use the word "disagree", but the difference is not that dramatic. The same point makes Strauss (World Politics, Vol. 59, No. 3 (Apr., 2007), pp. 476-501). Therefore, although "reject" was incorrect, to say "disagree" would be absolutely correct.
 * WRT Mann, he says that Rummel is "tautologically correct" (P.22), for any regime that kills its people cannot be considered as truly democratic. However, he points an many, many exceptions that undermine a value of the "democratic peace theory" as a whole. And he point at the fact that many totalitarian regimes were more efficient in stopping ethnic cleansing than democratic ones.
 * The quote from Mann was taken from the beginning of his chapter about Stalin, Mao and Pol Pot. In this quote, he summarises the views of others, and his own conclusion is different. His own analysis of Stalin's, Mao's, and Pol Pot's regimes leads him to a conclusion (p.350 - ) that "leftist" mass killings share significant common features with democratic (etnonationalist) mass killings, and, whereas the latter were a results of perversion of democracy, the former were a result of perversion of socialism. And he demonstrates that many mass killings in Communist states were not "top-down", but "bottom-up" (i.e. they had a broad popular support, i.e. were democratic).
 * In general, you must agree that the theories of those authors are more in agreement with what we see: many democracies at the early stage of their development committed mass murders and mass killings, and the same can be said about immature Communist societies. But, as we all know, both mature democratic and mature Communist societies commit much less mass killings, in a full agreement with Mann.--Paul Siebert (talk) 16:28, 24 December 2021 (UTC)
 * And, if you want me to respond to your next post, please, ping me (the talk page is becoming too convoluted) Paul Siebert (talk) 17:00, 24 December 2021 (UTC)
 * In regard to Valentino, while Wayman&Tago says "Disagreeing with Rummel's finding that authoritarian and totalitarian government explains mass murder, Valentino argues that regime the does not matter", further on they also say "A complication is his work is that one of Valentino's three main categories of killing is 'communist' mass killing, so he brings in regime type, after all, and so ends up a bit closer to Rummel than would have expected at the outset of the book". So Wayman&Tago are essentially arguing that Valentino's disagreement with Rummel is weakened by Valentino adopting the category of 'communist' mass killing. Therefore you can not claim Valentino rejects Rummel's regime classification because Valentino actually uses it in his study despite his conclusions. The main conclusion from Wayman&Tago was to confirm that "autocratic regimes, especially communist, are prone to mass killings generically, but not so strongly inclined toward gene-politicide.". In other words, while communist regimes cannot be considered genocidal, they certainly had a greater tendency towards indiscriminate mass killings. --Nug (talk) 10:57, 25 December 2021 (UTC)
 * This is a controversial and ideologically tense subject area. Basically, almost everyone in this field criticized almost everyone. Merely the fact that a scholar X was criticized (in certain aspects) by scholars A and B does not disqualify his work as RS per WP:RS. Just provide the entire range of views per WP:NPOV please. My very best wishes (talk) 23:34, 23 December 2021 (UTC)
 * WRT Mann, he says that Rummel is "tautologically correct" (P.22), for any regime that kills its people cannot be considered as truly democratic. However, he points an many, many exceptions that undermine a value of the "democratic peace theory" as a whole. And he point at the fact that many totalitarian regimes were more efficient in stopping ethnic cleansing than democratic ones.
 * The quote from Mann was taken from the beginning of his chapter about Stalin, Mao and Pol Pot. In this quote, he summarises the views of others, and his own conclusion is different. His own analysis of Stalin's, Mao's, and Pol Pot's regimes leads him to a conclusion (p.350 - ) that "leftist" mass killings share significant common features with democratic (etnonationalist) mass killings, and, whereas the latter were a results of perversion of democracy, the former were a result of perversion of socialism. And he demonstrates that many mass killings in Communist states were not "top-down", but "bottom-up" (i.e. they had a broad popular support, i.e. were democratic).
 * In general, you must agree that the theories of those authors are more in agreement with what we see: many democracies at the early stage of their development committed mass murders and mass killings, and the same can be said about immature Communist societies. But, as we all know, both mature democratic and mature Communist societies commit much less mass killings, in a full agreement with Mann.--Paul Siebert (talk) 16:28, 24 December 2021 (UTC)
 * And, if you want me to respond to your next post, please, ping me (the talk page is becoming too convoluted) Paul Siebert (talk) 17:00, 24 December 2021 (UTC)
 * In regard to Valentino, while Wayman&Tago says "Disagreeing with Rummel's finding that authoritarian and totalitarian government explains mass murder, Valentino argues that regime the does not matter", further on they also say "A complication is his work is that one of Valentino's three main categories of killing is 'communist' mass killing, so he brings in regime type, after all, and so ends up a bit closer to Rummel than would have expected at the outset of the book". So Wayman&Tago are essentially arguing that Valentino's disagreement with Rummel is weakened by Valentino adopting the category of 'communist' mass killing. Therefore you can not claim Valentino rejects Rummel's regime classification because Valentino actually uses it in his study despite his conclusions. The main conclusion from Wayman&Tago was to confirm that "autocratic regimes, especially communist, are prone to mass killings generically, but not so strongly inclined toward gene-politicide.". In other words, while communist regimes cannot be considered genocidal, they certainly had a greater tendency towards indiscriminate mass killings. --Nug (talk) 10:57, 25 December 2021 (UTC)
 * This is a controversial and ideologically tense subject area. Basically, almost everyone in this field criticized almost everyone. Merely the fact that a scholar X was criticized (in certain aspects) by scholars A and B does not disqualify his work as RS per WP:RS. Just provide the entire range of views per WP:NPOV please. My very best wishes (talk) 23:34, 23 December 2021 (UTC)

"Communist holocaust" listed at Redirects for discussion
An editor has identified a potential problem with the redirect Communist holocaust and has thus listed it for discussion. This discussion will occur at Redirects for discussion/Log/2021 December 27 until a consensus is reached, and readers of this page are welcome to contribute to the discussion. K.e.coffman (talk) 19:28, 27 December 2021 (UTC)

Synthesis?
Some people still make the claim that grouping communist countries together is OR/SYNTH, but Valentino has published such a grouping. If we already accept Valentino's definition of "mass kiliings" as 50,000 deaths over five years we should also accept his grouping. Below is a table from page 75 of his book:

Valentino describes the events where the mass killings occurred in the table, so how is it SYNTH to include summaries of those events like Red Terror and Great Leap Forward in the article? --Nug (talk) 11:53, 25 December 2021 (UTC)
 * As I said many times, the grouping is perfectly fine for B because if they theorise about the grouping, it is perfectly acceptable. The problem is discussing the events without connection. Majority of scholars do not make connections, or discuss them individually, or at best make a comparative analysis of Stalin, Mao, and Pol Pot, not every other Communist regime, or Communism as whole or in general, as we currently do, and this is not just because they are country experts as you say; as can be seen in the scholarly criticism of Communism, Dallin, David-Fox, and others directly criticize such grouping and how the authors gave no explanation or connection. We need sources doing that for us, but they mostly limit to Stalin, Mao, and Pol Pot, which is a scope I can accept; yet you want to give sections to all those other Communist regimes, when Valentino could not verify them (therefore, we can only write a very short paragraph about it, as we do at the introduction to "Other states"). Finally, Valentino is not an expert of Communism or about A — I support Valentino for what is his expertise (B). Since you complained about me, I suggest you that you further discuss this with . I do not enjoy talking to walls and repeating stuff either. Also ping, , , and — discuss this with them. It is certainly not just me. Davide King (talk) 13:19, 25 December 2021 (UTC) [Edited] Davide King (talk) 13:52, 25 December 2021 (UTC)
 * This your post is one of the reason why this talk age discussion is so not and convoluted. I already explained, many times, that
 * There is NO synthesis to group Communist regimes together. At least, because two authors, Rummel and Courtois, did that;
 * There is a potentially huge problem with NPOV, because we have a strong reason to believe that the authors who group Communist regimes together and draw some general conclusions about a linkage between Communism and mass murder are minority of insignificant minority views.
 * To answer this question, I proposed to develop a joint and neutral approach to source evaluation.
 * I provided an example of application of that analysis and asked everybody to present their reasonable criticism and/or alternative approaches.
 * You responded with some petty critique, and then magically disappeared from the discussion (as you usually do when we start speaking seriously). After that, we re-appeared with your straw man arguments and proposed the thesis with which noone can disagree, but which is totally irrelevant to the major problem of this article.
 * I am sorry, but if you will not return to the discussion of the major issue, I will try to minimise my interaction with you. Paul Siebert (talk) 16:44, 25 December 2021 (UTC)
 * I explained to you that your methodology of finding cites of global experts by country experts as a measure of due weight is flawed because global experts generally cite country experts, not the other way around; that some country expert does not cite a global expert means nothing. Can you then atleast devote some time to convincing David Kinge to stop using "OR/SYNTH", and use DUE instead. Surely he understands the difference and is not purposely trying to confuse the discussion. --Nug (talk) 18:57, 25 December 2021 (UTC)
 * I do not accept your explanation because it is a logical fallacy. Consider this example Anatole Klyosov published a series of works on "DNA genealogy". In his works (you can easily find tghem by yourself, but I cannot provide the link, because it is in a Wikipedia blacklist), he is citing works of reputable experts in population genetics, but his own works are ignored by them (they are cited either by his colleagues or by non-experts). Does the fact that he cites true experts makes "DNA genealogy" a true scientific discipline? No, DNA genealogy is considered a pseudoscience. Therefore, your argument, which is equally applicable to Rummel and Klyosov, is not working: it cannot reveal real pseudoscientists, and, therefore, it does not prove that Rummel is not a fringe author. Disclaimer. I fully realise that the analogy with Klyosov is not a proof that Rummel is fringe. However, it proves that your criticism is superficial and fallacious.
 * I expect to see your serious criticism, otherwise I propose to use my approach to resolve UNDUE issues. I am glad that you agree that UNDUE is a main article's problem.
 * WRT SYNTH, I think you guys can develop a common vision of the situation if you discuss among you (in some subsection or at the DRN page). It seems there are some elements of synthesis in your interpretation of Valentino (and I think I will be able to explain you what I mean). In that sense, DK is, to some degree, right. However, you are right that Valentino groups Communist regimes together, and it is not a synthesis to claim that. In other words, the truth is somewhere in between your and DK's position, but (if you want to know my opinion) I need to dive deeper in your and DK's rationale to understand whose view is closer to the real picture. Paul Siebert (talk) 22:58, 25 December 2021 (UTC)
 * I never denied that Valentino or others did the grouping, or at least that is not what I meant. But Valentino, and others, is used as a SYNTH to justify writing about Communism as a whole, when proven Communist mass killings include Stalin, Mao, and Pol Pot (I do not think we should consider other cases not yet verified, especially if we must lower the most used threshold to do so, and certainly they are undue to have separate sections), and you are right that the fact he considers Afghanistan to be counter-guerilla mass killing prove that this is not as easy as Nug make it appear to be. What I always said was that the grouping is controversial or disputed by mainstream scholars (Dallin, David-Fox, and others), therefore we cannot treat it as fact or monolithic, the latter of which is criticized even by serious works that do a Communist grouping (e.g. Cambridge and Oxford — interestingly enough, they do it for anything, including terror/violence links, but mass killings).
 * By doing the grouping and treating it essentially as fact and uncontroversial, we are giving undue and unjustified weight to Courtois and Rummel, who are the ones who discuss Communism as a whole (Chirot, Jones, Mann, and Valentino limit themselves to three, proven Communist leaders who engaged in mass killings, while Tago & Wayman 2010 discuss 18 Communist regimes, not any nominally Communist regime). Such OR/SYNTH issues were seriously considered in the closure and are clearly connected to NPOV issues because we are giving massively undue weight to those who do the grouping, and treating the topic as representing majority views rather than minority views, as it is in fact the case. Another thing to consider is that it is dispossessive mass killings, not Communist mass killings, that is a major category; the latter is a subtype of the former, and the fact Valentino has published nothing else about Communist mass killings (other than passing mentions but not sure about that too) means that we have been cherry picking and acting as though Communism is Valentino's main focus.
 * You may also find useful what I wrote here. Davide King (talk) 23:26, 25 December 2021 (UTC)
 * As I wrote here, and echoing, the fact that we are still discussing the same authors and sources after all those years should be telling. If A and C are such notable topics with the correct structure, surely there would be new scholarly sources coming out every few years or so? Yet, we all go back to Bellamy, Chirot, Jones, Mann, Rummel and Valentino, with only Rummel having a work fully devoted to Communism, while everyone else's is chapters about, and clearly focused on, Stalin, Mao, and Pol Pot, three Communist leaders, not Communist regimes as a whole. Davide King (talk) 23:38, 25 December 2021 (UTC)


 * Also still acting as those OR/SYNTH claims are crazy or debunked, when if that was the case, the latest AfD would have resulted in 'Keep' rather than 'No consensus.' It literally said that "the Wikipedia editing community has been unable to come to a consensus as to whether "mass killings under communist regimes" is a suitable encyclopaedic topic", and took OR/SYNTH issues seriously enough. So please, stop acting as though this has been debunked or is no issue. Davide King (talk) 14:27, 25 December 2021 (UTC)
 * Just to clarify, policy distinguishes between synthesis in reliable sources and synthesis by editors. We are of course allowed to report the synthesis made by experts. But that is governed by other considerations:
 * Weight. You need to establish the weight that Valentino's interpretation has in the body of reliable sources on the topic.
 * You can't say Valentino says these events are connected and then provide additional information that Valentino omits, per no synthesis.
 * Take for exmaple the theory that Michigan, Ohio, Pennsylvania, Southern Ontario and other regions constitute a nation called "The Foundry" with Detroit as its capital. (The Nine Nations of North America, 1981.) Your article has to explain why some social scientists believe this. You can't just accept the existance of this nation as a fact and flesh it out with information about the various cities and states making up the nation. If you do, you are not objectively reporting the theory, you are trying to persuade the reader that the nation exists.
 * You also cannot say that Valentino's theories are similar to Rummel's and synthesize the two into a general theory: you need a secondary source that does that.
 * TFD (talk) 14:24, 25 December 2021 (UTC)
 * Valentino's theory is not similar to Rummel's simply because reliable secondary sources say they are not. That had already been exhaustively discussed, with references and quotes, on this talk page. Paul Siebert (talk) 16:46, 25 December 2021 (UTC)
 * In few words:
 * Valentino discusses several states that belong to the group of Communist regimes, and he notes that some of them engaged in mass killings, whereas others weren't, and after a comparative analysis he concludes that regime type is not a good predictor of mass killing's onset.
 * Rummel, using his lousy database and tautological terminology, concludes that these is a strong correlation between what he calls "democide" and totalitarianism/Communism.
 * Clearly, these are totally different theories. Paul Siebert (talk) 16:55, 25 December 2021 (UTC)
 * Can you name a source that compares Valentino and Rummel? TFD (talk) 16:59, 25 December 2021 (UTC)
 * "Disagreeing with Rummel's finding that authoritarian and totalitarian government explains mass murder, Valentino (2004) argues that regime type does not matter; to Valentino the crucial thing is the motive For mass killing (Valentino, 2004: 70)." {Journal of Peace Research, Vol. 47, No. 1 (january 2010), pp. 3-13}
 * Strauss, World Politics, Vol. 59, No. 3 (Apr., 2007), p. 482 Paul Siebert (talk) 18:03, 25 December 2021 (UTC)
 * Regarding the first source, Wayman&Tago also say "A complication is his work is that one of Valentino's three main categories of killing is 'communist' mass killing, so he brings in regime type, after all, and so ends up a bit closer to Rummel than would have expected at the outset of the book". So Wayman&Tago are essentially arguing that Valentino's disagreement with Rummel is weakened by Valentino adopting the category of 'communist' mass killing. Therefore while Valentino concludes that regime type "communist" is not a predictor of mass killings, Valentino actually uses mass killing type "communist" as part of his topology of mass killings. In other words, Valentino has identified a correlation between communist regimes and mass killings (as did Rummel) but differs from Rummel in asserting that this correlation does not imply causation, which is what I have been saying all along. The main conclusion from Wayman&Tago was to confirm that "autocratic regimes, especially communist, are prone to mass killings generically, but not so strongly inclined toward gene-politicide.". In other words, while communist regimes cannot be considered genocidal, they certainly had a greater tendency towards indiscriminate mass killings. --Nug (talk) 18:34, 25 December 2021 (UTC)
 * Regarding the second source, there is no direct comparison between Rummel and Valentino, Strauss is actually focuses on second generation genocide scholarship which includes Valentino, and places Rummel in context of first generation scholarship, in terms of defining a predictor of the onset of genocide. --Nug (talk) 22:00, 25 December 2021 (UTC)
 * Yes, W&T say, that Valentino disagree with Rummel, but some aspects of his views are closer to Rummel's views that he initially declared. When I say that I disagree with you and after that I partially accept some of your points, does it mean I agreed or not?
 * WTR Strauss, he says:
 * "Rudolph Rummel claims "absolute power is the key factor" (p. 481)
 * According to Strauss, that is one of the core idea of "first generation genocide scholars" including Rummel. And Stauss says, quite clearly and unequivocally
 * "The second-generation scholarship also consistently reject arguments about a link between authoritarian regime type and genocide" (ibid.)
 * Clearly, since this statement was made in a context of the Rummel's mantra about absolute power, Strauss uses the term "genocide" in its colloquial meaning, which covers mass killings, democide etc.
 * Therefore, I absolutely cannot understand how could Nug overlook this clear and unequivocal statement. Paul Siebert (talk) 23:59, 25 December 2021 (UTC)
 * I just explained to you that Valentino, while seeing a correlation between mass killings and communist regimes, says that correlation does not equate to causation. You are saying the opposite, that because there is no causation then there is no correlation, but that doesn't follow. The path to resolution is for you to honestly admit that such a correlation does in fact exist. --Nug (talk) 01:13, 26 December 2021 (UTC)
 * "Correla..." is found only three times in his book. This scholar does not discuss correlation neither in this, nor in his other works. That is my last post on this topic until the RfC ends. You are more then welcome to continue the discussion of source selection procedure. Paul Siebert (talk) 01:28, 26 December 2021 (UTC)

I would expect something more substantial. A couple of lines in an article isn't very useful, especially when they give conflicting accounts, viz., did Valentino attribute the killings to Communism or didn't he. TFD (talk) 21:43, 26 December 2021 (UTC)
 * Do you not comprehend the difference between correlation and causation? Given that Valentino devotes an entire chapter to "Communist mass killings" he clearly sees there is a correlation, otherwise why did he group mass killings under communist regimes into one chapter in the first place? Valentino's conclusion that the causation isn't attributed to "communism" directly: "I contend that mass killings occur when powerful groups come to believe it is the best available means to achieve certain radical goals...". Valentino argues that ideology can shape why leaders believe that genocide and mass killings is the right course of action. So he isn't attributing the killings to Communism, but to the leadership who see mass killings as the best way to achieve Communism. That's why we have Mass killings under communist regimes and not Mass killings under Communism, there is a distinction. --Nug (talk) 00:32, 27 December 2021 (UTC)
 * Saying that because Valentino sees a correlation between mass killings and communism because he has a chapter about Communist mass killing is synthesis. Suppose there were an article "Literature under communist regimes." That does not mean the author saw a correlation between literature and communism, since non-communist states are just as likely to have literature as communist ones. TFD (talk) 01:43, 27 December 2021 (UTC)
 * No, if an author is writing a chapter titled "Communist Literature" he has correlated literature written by communists together, Nazi's are just as likely to produce literature but we wouldn't see any mention of "Mein Kampf" in such a chapter. --Nug (talk) 01:58, 27 December 2021 (UTC)
 * Maybe. But then an article about Commonwealth literature doesn't necessarily imply the author has correlated literature written by Commonwealth writers together. Commonwealth literature incidently is a frequent grouping, see for example The Journal of Commonwealth Literature. That's why Wikipedia has a policy against edtiro synthesis. TFD (talk) 02:22, 27 December 2021 (UTC)
 * What synthesis? The Journal of Commonwealth Literature says it is a leading source for "literature written and published within the Commonwealth", obviously it wouldn't include literature written and published outside the Commonwealth, would it? --Nug (talk) 03:46, 27 December 2021 (UTC)
 * You wrote, "Given that Valentino devotes an entire chapter to "Communist mass killings" he clearly sees there is a correlation, otherwise why did he group mass killings under communist regimes into one chapter in the first place?" [00:32, 27 December 2021]
 * I mentioned that there is a publication called, The Journal of Commonwealth Literature. There is no correlation between Commonwealth literature - different languages, cultures, genres, etc.
 * I agree with you that Valentino's title implies a correlation or even causation, but it does not explicitly say one exists. (That's why I have always objected to the title of this article.) But if you conclude that because he used that title he saw a correlation, you are engaging in classic synthesis.
 * Synthesis btw does not mean you are wrong. Experts whose works are used for articles use synthesis. The difference is that synthesis by editors is not permitted.
 * TFD (talk) 06:41, 27 December 2021 (UTC)
 * Nope. The criteria for inclusion is clearly "literature written and published within the Commonwealth", different languages, cultures or genres is irrelevant here. --Nug (talk) 20:52, 27 December 2021 (UTC)

We were not discussing the criteria for inclusion, we were discusxing synthesis. Presumably that's why you titled this discussion thread "Synthesis." Below an editor asked why these discussions are so long. Your posting helps to explain it. TFD (talk) 21:44, 27 December 2021 (UTC)

1.6% of editors responsible for more than 60% of text
Since the 22nd November, when the AfD for this page was opened, 123 unique editors have contributed to this talk page. 2 of these (Paul Siebert and Davide King) have contributed 62.4% of the text, and 52% of the edits to this talk page. at time of writing In comparison,the 3rd largest contributor (Nug) has contributed less than 10% of the text to this page. We appreciate your contributions and insights but when such a small number of editors are dominating the discussion in such an active page it leads to diminishing returns for further discussion. While you may have some very valid points, they get lost due to the dominant behaviour and others are less likely to consider your viewpoints because of it. I humbly request these editors and any new editors arriving on this page allow discussions room to breathe to avoid any perception of accidentally impeding the process.

Let me emphasise, I am not asking these or other editors to stop participating here. I am simply asking them to reconsider that their manner of doing so, in the spirit of collaboration. I know I, and I'm sure a fair few other editors would greatly appreciate if you could commit to this. Kind Regards, Vanteloop (talk) 23:14, 24 December 2021 (UTC)


 * Interesting. I shall sit back & watch. GoodDay (talk) 01:12, 25 December 2021 (UTC)
 * I'm just trying to point out the flaws in arguments of Paul Siebert and Davide King (who essentially just repeats and amplifies what Paul says anyway). Are you planning on expressing a position in the RFC? --Nug (talk) 11:08, 25 December 2021 (UTC)
 * Can you please stop make such generalizations? "... who essentially just repeats and amplifies what Paul says anyway." This is not true — I agree with them on the article's long-standing issues but I would say my position is closer to TFD's, and I actually agree with you that we cannot use country experts as long as the article's structure is Communism in general but I agree with them that this structure is wrong. To remain on topic, while such statistics may be true and I apologize for taking so much space, and I indeed welcome many other users to participate, it is also highly misleading, for I made many edits to correct typos, indents, and fix missing signature, and Siebert's and mine posts have been the biggest because we have analyzed in-depth sources, scope, and topics.12 Davide King (talk) 17:09, 25 December 2021 (UTC)
 * I was using you as an example to contrast to the top 2 contributors, rather than include you in that category. My request was the other 2 take a step back more than anything else, I don't think your level of contribution is necessarily a large cause of the difficulty of this talk page. Apologies if I was unclear Vanteloop (talk) 22:38, 27 December 2021 (UTC)
 * This is an outrageous and appalling edit which should be withdrawn by the user. It seems like some are determined to pursue the same McCarthyite tactics on the talk page as their intellectual idols quoted in the article itself. In particular, since we’re being personal, the suggestion that the user Nug’s behaviour on this page has been anything other than politicised, polemical and disruptive is absurd. Withdraw and delete, or this should be taken further. DublinDilettante (talk) 16:38, 27 December 2021 (UTC)
 * The purpose of this post is to comment on one of the possible reasons this talk page has been so dense and difficult to keep up with, and humbly ask all editors to help in improving it. This Is the aim simply to attribute mass killings to any and all “communist regimes” for political purposes? Don’t bother answering, it’s obvious that that’s the purpose of this entire article and other such comments to not help to do that in my opinion. A good faith editor can't take much from that Vanteloop (talk) 22:38, 27 December 2021 (UTC)
 * Isn’t this the dreaded “whataboutism?” You have called out, by handle, two users whose perspective you disagree with, in an abusive and targeted manner. You know that other users, whose perspectives you presumably agree with, have behaved abominably and disruptively in this section, and you have chosen to ignore this. This is an appalling attempt to tar two reputable editors for a transparently political purpose, and should, I repeat be immediately withdrawn or escalated further. DublinDilettante (talk) 23:22, 27 December 2021 (UTC)
 * “whataboutism?” Not everything you disagree with is whataboutism. I am arguing that your perspective of those who disagree with you is fundamentally against the purpose of this post, and is why you misinterpret it. an abusive and targeted manner No. for a transparently political purpose No.Vanteloop (talk) 00:34, 28 December 2021 (UTC)
 * Dublin, your own behavior is transparently abusive and disruptive. If vanteloop has a grudge against Paul and Davide, this would be more constructive to contact them on their talkpage first and foremost rather than here, and by using a tone that is appropriate. If Vanteloop would hypothetically continue to hound Paul and Davide, they then can be taken to the abitration committee. I digress-
 * Vanteloop's post, as it appears, was made to highlight that the article could have POV issues which often does occur with articles edited by the same few people on a frequent manner. Wikipedia was last seen on the news because a single person edited all by themselves most of the articles on the Scottish Wikipedia, and I highly doubt anyone at Wikipedia is pleased by this. MarioSuperstar77 (talk) 12:54, 28 December 2021 (UTC)

"often paled"
Coming from Signpost to take a look what all this fuss was about, I noticed several sloppy things. I fixed the lede a bit (I think), but I noticed a "micro-edit war" in the history. Generally I would frown upon random comparisons who killed more, but I agree that is a notable expert makeess such comparison, it deserves to be mentioned. That said, the phrasing "Red" terror often paled in comparison to "White" terror doesnt strike me as particularly enlightening/encyclopedic: what is "often", what is "paled"? what exactly was compared? under which criteria? etc. fortunately this page in the source is available, and in the broader context this statement does make sense. Also, since this is a qualitative opinion, I would suggest to put a direct quote:
 * "..when both sides engaged in terror, the "red" terror usually paled in comparison with the "white"...

I.e. the author was talking about "mutual destruction" and not a random comparison of, say, Thermidore with Holodomor (sorry for the pun; unintended), which is reasonable. Loew Galitz (talk) 03:23, 29 December 2021 (UTC)
 * I've changed the sentence to include the quote. X-Editor (talk) 07:03, 29 December 2021 (UTC)

Cuba
What is the basis for including Cuba on this list? A single source attributes a maximum of 8,335 killings, allegedly of political opponents, over the course of eleven years following the revolution.

For reference, judging by established rates, it is likely that police repression and violence killed at least 10,000 people in the United States during the same period. On what basis does the 8,000 figure, arising from a period of counter-revolution, belong in this article? Is the aim simply to attribute mass killings to any and all “communist regimes” for political purposes? Don’t bother answering, it’s obvious that that’s the purpose of this entire article. DublinDilettante (talk) 19:56, 27 December 2021 (UTC)
 * Police repression and violence in the United States has literally nothing to do with the topic of this article. Whataboutism is not an argument against the inclusion of these killings. There is also already an article on Police brutality in the United States. X-Editor (talk) 20:41, 27 December 2021 (UTC)
 * The silly term “Whataboutism” is the last refuge of the apologist. The point, as you know, is whether a badly-sourced claim of a number of deaths from multiple causes across the span of a decade constitutes a “mass killing”. Does the police brutality article define the actions of US police as a “mass killing?” If it does not, why (except for the obvious political reasons) are we applying it here? You have made no argument for retaining it, and I’d say the Cuba section is on shaky ground. DublinDilettante (talk) 23:15, 27 December 2021 (UTC)
 * I'm not apologizing for anyone and it is bad faith to assume so. Both police brutality in the US and the killings in Cuba are bad. What I was trying to say is that your argument is poor because it is a logical fallacy. X-Editor (talk) 02:33, 28 December 2021 (UTC)
 * While violence in the US is wholly irrelevant, a good point is raised re: number of death and period of time. If we operate under the "5,000+ over 5 years or less" definition of a mass killing, Cuba fails and should be removed from the article. BSMRD (talk) 23:31, 27 December 2021 (UTC)
 * , the most accepted definition or criteria is actually 50,000 over five years, not 5,000; only Stalin's, Mao's, and Pol Pot's regimes, plus the Red Terror within the context of civil war, fit this criteria and the only famine to be actually debated as a mass killing is the Holodomor. Davide King (talk) 23:44, 27 December 2021 (UTC)
 * Bulgaria, East Germany, Romania, North Korea and Vietnam also fit that criteria according to Valentino. If we accept Valentino's criteria of 50,000 over five years, we should also accept his mass killing categories of "Communist", "Ethnic", "Territorial", "Counterguerrilla", "Terrorist" and "Imperialist". You can't just cherry pick only the criteria you like. --Nug (talk) 23:53, 27 December 2021 (UTC)
 * They are possible cases, and as such they can be summarized in a single paragraph, there is no need to give each of them a section on par on universal and verified mass killings. Only Stalin's, Mao's, and Pol Pot's regimes fit the category of Communist mass killings according to Valentino. Davide King (talk) 23:58, 27 December 2021 (UTC)
 * This is also in line with the literature in discussing Stalin's, Mao's, and Pol Pot regimes (Jones, Mann, etc.), which Valentino also explicitily used as his bases for the chapter's name. Davide King (talk) 00:00, 28 December 2021 (UTC)


 * It does not matter how many people were killed in specific country. Those killings in Cuba are described in sources that consider such political repressions as a phenomenon typical for communist countries in general, for example the Black Book of Communism. My very best wishes (talk) 03:26, 28 December 2021 (UTC)
 * So's Afghanistan, but Nug's source says, "we exclude borderline cases such as Afghanistan." (See below.) TFD (talk) 04:09, 28 December 2021 (UTC)
 * This is strange because Afghanistan was definitely the place of mass murder by the communist regime of Babrak Karmal who was installed as a puppet by the USSR. My very best wishes (talk) 05:48, 29 December 2021 (UTC)
 * Cuba is included in Wayman & Tago's dataset. --Nug (talk) 06:49, 28 December 2021 (UTC)
 * In the dataset? Maybe. Please cite relevant page numbers, so we can see what exactly they say about Cuba. And whether they actually express any opinion on what any data on Cuba actually has to say on that specific country. AndyTheGrump (talk) 07:06, 28 December 2021 (UTC)
 * Yes, in the data files mentioned in the section below. It shows Cuba as having perpetrated democide but not geno-politicide. --Nug (talk) 09:41, 28 December 2021 (UTC)
 * So, WP:OR of a 'dataset'. What a surprise... AndyTheGrump (talk) 13:43, 28 December 2021 (UTC)
 * Not OR at all. The paper presents data in tables as well as provides a link to the published dataset, so it forms a part of the discussion where it states "Of these 18 cases, Harff identifies four as having geno-politicide during the communist period, which means a geno-politicide occurred in 22% of cases. Rummel identifies these four plus nine other communist regimes as having democide, meaning 72% of communist regimes.", so it is perfectly valid to look at the dataset published in conjunction with the paper to identify those nine other communist regimes the authors refer to. --Nug (talk) 18:11, 28 December 2021 (UTC)
 * Yeah, I'm just gonna remove Cuba because no-one has been able to cite an encyclopaedically valid reason for retaining it. DublinDilettante (talk) 15:08, 28 December 2021 (UTC)
 * You should revert your removal, despite your edit comment that there is no credible source, it does in fact cite Ulfelder & Valentino. It is also mentioned on page 648 in Pascal Fontain's chapter Communism in Latin America in BBoC. The fact you edit war other text claiming RS, yet remove this text which is also backed by RS indicates a level of WP:TENDENTIOUS editing. --Nug (talk) 18:11, 28 December 2021 (UTC)

Off topic
As we already have Democide for that, it makes sense to limit this article to geno-politicide (Harff is still the dataset most used by scholars). Or hear me out, we actually create Geno-politicide (we may merge Political cleansing of population and Politicide to it) and we discuss all events listed there, rather than cherry pick Communism as a separate topic. Davide King (talk) 16:04, 28 December 2021 (UTC)
 * It makes no sense to limit this article to the narrower geno-politicide, it isn't called Geno-politicide under communist regimes. --Nug (talk) 18:20, 28 December 2021 (UTC)
 * And it should not be called like that — the bottom line is that only Courtois and Rummel have discussed Communism as a whole as a separate topic, Mann, Valentino, and others discuss it as a subtopic, which means it should be a subsection of mass killings in general, and limit it to Stalin, Mao, and Pol Pot (Cuba is not discussed at all), therefore Communist regimes is very misleading when majority of scholarly sources emphasis three Communist leaders. Davide King (talk) 19:01, 28 December 2021 (UTC)
 * MKuCR is already a sub-topic of many other articles. --Nug (talk) 19:10, 28 December 2021 (UTC)
 * Are you really using Wikipedia to support this? Of course it is going to be linked as long as this article, in the current structure, exists. The fact Communist mass killings (Stalin, Mao, and Pol Pot) are discussed in chapters in general works about genocide and mass killing makes it abundantly clear that this is a subtopic, not a main topic. A main topic may be the theories about them. Davide King (talk) 19:31, 28 December 2021 (UTC)
 * Nug, it's not a subtopic, your link only shows that a word or phrase used in each of those articles is linked to this one. For example, I just linked your user page, it doesn't me you are a subtopic of this page. Furthermore, I just checked half a dozen of those pages and the links no longer exist. TFD (talk) 13:06, 29 December 2021 (UTC)

This talk page is totally unreadable
It is becoming increasingly difficult for me to participate in this talk page discussions, because too many discussions take place concurrently. Actually, most of them are not necessary or premature, because they will not lead to any change in the article's content until the RfC gives us a final answer about the article's topic. Therefore, I am going to stop my participation in all discussions until the RfC comes to some logical end. I make one excepttion: development of the mutually acceptable procedure of source evaluation. We need that, because, independently on the RfC results we need to come to an agreement which viewpoint is a majority and which is (are) minority view (or views). In connection to that, I am asking, , , , , and  if they are interested in that discussion. Please, let me know if you have any ideas, or if you have any objections to the approach that I propose. If you need me to re-explain my approach, I will gladly do that at WP:MKUCRSA. I may forget to mention someone's name, so I apologise in advance if I forgot someone.--Paul Siebert (talk) 00:21, 26 December 2021 (UTC)
 * I long ago stopped trying to keep up with this talk page. For my part I'm not planning to participate in the RFC because I do not have an understanding of what the major sources are for the article in any form. I don't feel that I can form an opinion about the proper scope of the topic until I know what sources we're summarizing. Yes, I'd be interested in a discussion about sources (and procedure or criteria for source evaluation is the place to start, IMO), but I won't have much time until after New Year. Levivich 06:20, 26 December 2021 (UTC)
 * Can someone who knows how adjust the duration of the auto archiver from 21 days to, maybe, 7? I think that would improve talk page readability. schetm (talk) 14:02, 26 December 2021 (UTC)


 * One thing that might help, skimming recent discussions, is to set up a subpage where people can list high-quality sources, with key pull-quotes, that they feel either show that aspects of the central thesis being argued here are mainstream and widely accepted, or that it is controversial. That could help establish the overall acceptance of various aspects of it in the sources (the broad connection, specific numbers, etc) and how it is generally described, while producing a list of sources that could possibly be used in the article.  Editors could also reply to specific sources there with discussion of whether it's being summarized accurately, how mainstream / prominent / reliable the source and author are, and so on. Look at the discussion there's a ton of "the sources totally say X", "no, they largely say Y", which makes it tricky for people to contribute to the discussion or even figure out where it is - it's not reasonable to expect every editor to read 20-odd sources in-depth!  Having a page of key pull-quotes for each one and a place to hash out general summaries of what they say could help us get a sense of the scholarship and would be something we could build on going forwards. I'd break it up into sections based on key aspects of the debate or something, especially the basic question of "is there a causal connection between Communism and mass killing?" and the 100 million figure, which seem to be major points of contention. --Aquillion (talk) 06:53, 27 December 2021 (UTC)
 * I agree on the initiative, that would weed out improper sources for this article which would be a good thing in long term. MarioSuperstar77 (talk) 13:15, 27 December 2021 (UTC)
 * WP:MKUCRSA? Levivich 13:58, 27 December 2021 (UTC)
 * Yes. Davide King (talk) 14:38, 27 December 2021 (UTC)
 * I added a table to properly categorize good sources from bad sources. MarioSuperstar77 (talk) 15:22, 27 December 2021 (UTC)


 * Part of it is 's consistent misrepresentation of what other readers have posted, distortion of policies and guidelines and novel interpretations of sources. See for example the discussion thread he set up at {{Talk:Mass killings under communist regimes#Synthesis]]. TFD (talk) 21:56, 27 December 2021 (UTC)
 * Another unsubstantiated personal attack for my diff list. Thanks. --Nug (talk) 23:57, 27 December 2021 (UTC)

It's been roughly a week, since I've given up on following the multiple discussions. GoodDay (talk) 18:57, 29 December 2021 (UTC)

Why is the section under mass killing not properly sourced??
Why is that section citing another Wikipedia article section that doesn't exist. I think a primary source claiming 8k died would be good. In my opinion that section seems very biased and without citation. 2607:FB91:1989:106D:91F8:38EE:24DD:BAC4 (talk) 15:12, 29 December 2021 (UTC)
 * Oh jeez, what's this now. 10 new discussions begun, within the last two or so days? GoodDay (talk) 19:00, 29 December 2021 (UTC)
 * They failed to delete the whole thing, so now they're trying to make article revisionism. It's clear they don't even know how Wikipedia and this discussion is a prime example of it, not even citing where the ""problem"" is. 191.136.192.156 (talk) 15:52, 31 December 2021 (UTC)
 * The problems are neutrality and synthesis. It literally says so at the top of the article. You could also read all of the discussions above this one. X-Editor (talk) 23:03, 31 December 2021 (UTC)

This would be the most constructive thing to do
This has gone on for a while, so let me say what would be the best course of action. Since the information therein this article exists elsewhere, which is Criticism of Communist party rule, I think we can safely remove "Estimates", "Proposed causes" and "Debates over famines" which are all equally contentious in this article here specifically as they constitute opinion pieces whereas Mass killings under Communist regimes is supposed to be focused on facts.

Picking A, B or C will not fix WP:TOOBIG, which is a crystal clear cut violation at that which was raised by nobody, except me; WP:SYNTH; and last, but possibly not least WP:UNDUE. This is not about the quality of the sources themselves, this is about the presentation. The issues that I have noted previously will never be fixed if we continue to try to repair the article using any previously mentioned methodology, whether it would be Paul's source analysis, replacing some paragraphs with anything else or adding counter points for the sake of balance. Analogically, this would be the equivalent of tearing out some food from a dish because it tastes odd and then give it a different aroma, this is not how this should be dealt with because at the end of the day it will still taste weird. In that same analogy, Mass killings under Communist regimes would be a dessert, a chocolate cake more precisely and Criticism of Communist party rule would be a veggie salad with some beef on the side - everyone here thinks the chocolate cake does not taste good, therefore, they attempt to add a different kind of pepper into the dessert in the vain attempt to make it taste differently, maybe better, but it will not because it is a chocolate cake for crying out loud; that pepper would better serve its purpose over the veggie salad and the beef. And, as it did turn out, the reason for the chocolate cake's odd taste was due to someone that had previously added lemon slices into the recipe. Therefore, the best course of action would be to remove the lemon slices to get the best possible chocolate cake available.

In itself, it is surprising to see so many people complain about POVFORK when there is already a noncontroversial opinion-focused FORK of this very article: Criticism of Communist party rule which I encourage everyone to improve instead of focusing on this article here and its misused sources. Every single one of them should be purged from MKUCR with no exception. As for each section that is not contentious and does deserve to stay there, they should be expanded with reliable tertiary sources focused on data gathering rather than gossip from the likes of Rummel, and every other scholar, journalist and historian, to give proper context to each of the countries' mass killings. More facts, less hear-say I say!

This is the correct solution, and I would be bummed if this is not the route taken to clean up this article proper. MarioSuperstar77 (talk) 21:30, 23 December 2021 (UTC)
 * Do you realise that "Terminology", "Estimates" and "Causes" are the tree section that are the core of this article, and that are the main NPOV problem? If we remove these sections and slightly rewrite the rest, we will get a summary style ("option A").
 * Of course,, that does not mean I object. I support it, of course.--Paul Siebert (talk) 21:43, 23 December 2021 (UTC)
 * Which is especially because they are the main NPOV problem that this article draws so much ire. If your computer's processor fails, you change it as soon as you can. This is clearly time for this article's main points to be entirely re-written. MarioSuperstar77 (talk) 21:58, 23 December 2021 (UTC)
 * Yes. But deleting these section is almost tantamount to the attempt to delete the whole article (which caused the larges AfD in Wikipedia history). I am saying that because deletion will cause a storm of totally emotional and absolutely irrational opposition. But we can try. Paul Siebert (talk) 22:08, 23 December 2021 (UTC)
 * I agree, the problem is that there is no tertiary source about the topic (the closest one is Karlsson & Schoenhals 2008, who are dismissive towards Courtois and Rummel, and do not mention Valentino, yet it has been disputed due to being old and not cited), there are no academic books fully dedicated to Communist regimes in general (The Black Book of Communism is controversial, ideologically, charged, and chapters do not necessarily focus on mass killings, and Red Holocaust is mainly about excess deaths), only chapters limited to Stalin's, Mao's, and Pol Pot's regimes (even when they say 'Communist regimes' or 'Communist mass killings', contrary to what we do here by broading the scope, the scope is limited to universally recognized mass killing events under Stalin, Mao, and Pol Pot, three very specific Communist leaders), three Communist leaders who are universally recognized to have engaged in mass killings (the Red Terror can be considered another mass killing event but is placed within a totally different context). The global Communist death toll is a minority view that is ignored or criticized by majority of scholars; there are not much tertiary sources about it but there are plenty about its narrative.
 * A more accurate and descriptive title would be Mass killings under Joseph Stalin, Mao Zedong, and Pol Pot; however, we already discuss each event individually and together in each country's history, what would such an article add without any theories or connection about them?1 If we actually had such tertiary sources, the article would have been fixed by now. B is the only topic that has such tertiary sources, and also the only topic where we can actually discuss more than those three Communist regimes (e.g. Tago & Wayman 2010's discussion of 18 Communist regimes and the link with mass killings). D may be a solution but B should be the goal.
 * Notes
 * 1. Imagine this same article with removal of "Terminology", "Estimates", "Proposed causes", "Debate over famines", and reducing "Other states" to the opening paragraph and removing merge of excess death events with mass killings. It will solve many issues but what will remain that is not new? Davide King (talk) 22:18, 23 December 2021 (UTC)


 * No, I disagree. This article is mostly focused on facts about the mass killings, but it also includes interpretations of these facts, which is only natural. In contrast, the Criticism of Communist party rule is a different and a wider subject, which is mostly about interpretations. My very best wishes (talk) 23:30, 23 December 2021 (UTC)
 * but it also includes interpretations of these facts Yes we know, that's what we call WP:SYNTH. MarioSuperstar77 (talk) 00:41, 24 December 2021 (UTC)
 * It's only synth if a wiki editor does the interpretation. And other words for interpretation (by non-editors) can be "scholarly study", "analysis" etc. North8000 (talk) 00:44, 24 December 2021 (UTC)
 * "Criticism of Communist party rule" should probably be deleted per Criticism. While that article is not a policy or guideline, it makes sense that an article about what is bad about something is inherently not neutral. The other inherent problem is that it assumes there is a commonality of various Communist party rules. And, as other editors have pointed out, mass killings were only one of a number of criticisms. TFD (talk) 01:45, 24 December 2021 (UTC)
 * I completely disagree. The article you are referring to literally says "Dedicated "Criticism of ..." articles are sometimes appropriate for organizations, businesses, philosophies, religions, or political outlooks, provided the sources justify it" and there are many sources that provide criticism of communist party rule. Since it is not an official policy or guideline, which you yourself admit to, it cannot be used as justification for deletion anyway. X-Editor (talk) 05:45, 24 December 2021 (UTC)
 * Actually, yes. If the article's topic is "criticism", that implies a discussion of this criticism, not the subject of that criticism. That includes a criticism of that criticism, and a discussion of its place in the opinia spectrum. Paul Siebert (talk) 15:38, 24 December 2021 (UTC)
 * Indeed, that is why I have also once proposed to turn those two articles into a general scholarly analysis of Communism, which is of course going to include what would be considered as criticism, except it would be of higher weight. If we can get a good article out of it, it may be the cornerstone to use for all relevant Communist-related articles and significantly improve them, as the article should give us good indication of what views are mainstream, majority, minority (significant), and fringe. Indeed, I am interested to actually find this out (apart from being interested in preventing mass killing, which means we should be accurate when we represent scholars, e.g. Valentino and leaders, and societal background and context, not a narrow focus with generalizations and oversimplifications about ideology and Marx, which is why I want to greatly improve this article) because if we can do it, we may avoid so many discussions and controversies about their weight in Communist-related articles. This would also be an article where discussion of Communism as a whole can be warranted (e.g. as is done by the Cambridge History of Communism and the Oxford Handbook of Communism) because it would also include criticism of generalizations and discussion about the grouping itself (e.g. communisms vs. Communism, and other more nuanced and middle views), and not presented as facts as we do here. Davide King (talk) 16:22, 24 December 2021 (UTC)
 * Since communist states have pursued various policies in different countries at different times, most criticism will not have application to all of them. Hong Kong for example has different economic policies from Kampuchea. The various ideoligal opponents of communism - liberals, fascists, authoritarian conservatives and Trotskyists - all criticized Communist party rule for different reasons. Furthermore, Communist states criticized each other, particularly China and the Soviet Union. The article only works if one sees Communism as a monolithic system and there being only one valid ideology from which to criticize it. One example of that is Jewish Bolshevism.
 * Anyway, I do not think there are sources written about criticism of Communism, there are books about Communism that criticize it.
 * TFD (talk) 17:13, 24 December 2021 (UTC)
 * Based on that logic, any page pertaining to a group of 'things' which have correlation with each other, but are also different in other ways, are not encyclopedic because they are treated as a monolithic entity. Fine by me, then any article focusing on Autism, gender and Schizophrenia, just to name a few, are not encyclopedic because they treat each respective subject as a monolith, whereas they each work on a spectrum. Communism works on a spectrum too, but all stem from Marxist theory with the added Totalitarianism of a crazy leader atop of it. That argument could also be used in regards to Capitalism because the Capitalism of China is not the same as the Capitalism of the US which is not the same as the Capitalism of the EU. MarioSuperstar77 (talk) 13:04, 26 December 2021 (UTC)

In hindsight, this article is not the only one to have the WP:TOOBIG issue, COVID-19 pandemic in mainland China has it too. MarioSuperstar77 (talk) 20:34, 31 December 2021 (UTC)
 * There are 3,390 articles longer than this one. Levivich 20:47, 31 December 2021 (UTC)
 * If you read WP:TOOBIG, it actually refers to "readable prose size", which is 69 kB (10935 words) according to the Page size tool. --Nug (talk) 21:36, 1 January 2022 (UTC)