Wikipedia:Dispute resolution noticeboard/MKUCR Source Analysis

Ground Rules
Here are the ground rules for this source analysis subpage:
 * 1)  Be civil and concise.
 * 2)  Civility is required everywhere in Wikipedia and is essential in dispute resolution.  Uncivil statements may be collapsed.
 * 3)  Overly long statements do not clarify issues.  (They may make the author feel better, but the objective is to discuss the article constructively.)  Overly long statements may be collapsed, and the party may be told to summarize them.  Read Too Long, Didn't Read, and don't write anything that is too long for other editors to read.  If the moderator says to write one paragraph, that means one paragraph of reasonable length.
 * 4)  Do not report any issues about the article or the editing of the article at any other noticeboards, such as WP:ANI or Arbitration Enforcement.  Reporting any issue about the article at any other location than the article talk page, DRN, the DRNMKUCR subpage, or this source analysis subpage is forum shopping, which is strongly discouraged.  Any old discussions at any other noticeboards must be closed or suspended.  If any new discussions are opened elsewhere while discussion is pending at DRN, the mediation at DRN will be failed.
 * 5)  Comment on content, not contributors.
 * 6)  The purpose of discussion is to improve the article, not to complain about other editors.  (There may be a combination of content issues and conduct issues, but resolving the content issue often mitigates the conduct issue or permits it to subside.)  Uncivil comments or comments about other editors may be suppressed.
 * 7)  "Comment on content, not contributors" means that if you are asked to summarize what you want changed in the article, or left the same, it is not necessary or useful to name the other editors, but it may be important to identify the paragraphs or locations in the article.  It isn't necessary to identify the other editors with whom you disagree.
 * 8)  Discuss edits, not editors.  This means the same as "Comment on content, not contributors".  It is repeated because it needs repeating.

Robert McClenon (talk) 01:01, 15 December 2021 (UTC)

As a reminder to editors from the main DRN page (including myself), or an introduction to new editors from the article talk page, I have pasted the following paragraph regarding this page Vanteloop (talk) 18:23, 15 December 2021 (UTC)

Paul Siebert
For the discussion, we must operate with a common set of sources, for, per Aumann's Agreement Theorem, creation of a common knowledge domain is important for achieving consensus. Obviously, we are incapable for analyzing all sources on the topic, so I propose to agree how to create a representative set of them for each subtopic of our discussion. The procedure must be neutral, which means each of us cannot just pick the source they like: our procedure must be transparent, reproducible, and independent on a personality of those who perform a search. Does everybody agree with these principles?--Paul Siebert (talk) 15:20, 15 December 2021 (UTC) You rules say:
 * "Reporting any issue about the article at any other location is forum shopping, which is strongly discouraged."

That de facto eliminates us from the process of working on the article. This situation cannot last long, so I insist on strict deadlines and prompt responses. We either finish that process quickly, or I will not consider this rule as mandatory.--Paul Siebert (talk) 19:04, 15 December 2021 (UTC)
 * User:Paul Siebert - I have added an explanatory phrase to that rule. Do not report any issue about the article at any conduct forum, such as WP:ANI or Arbitration Enforcement, at least not if you want this process to continue.  Do not report any issue about the article at the neutral point of view noticeboard; we know that there are neutral point of view issues.  Any source questions that are taken to the reliable source noticeboard should be coordinated.  Does that address your concerns?  Robert McClenon (talk) 22:34, 15 December 2021 (UTC)
 * Actually, I would prefer if some strict schedule was proposed that each party agreed to stick with. I am ready even not to edit the article at all during this discussion, provided that it will move fast. Paul Siebert (talk) 22:45, 15 December 2021 (UTC)

I would also like the source analysis to move quickly. What does User:Paul Siebert want the other editors to submit by 18 December 2021? Robert McClenon (talk) 04:41, 16 December 2021 (UTC)


 * as I explained, we need to agree about the common search procedure for RS identification and selection. I would like to avoid a situation like: "My source says X", "But my source says Y" - "Your source is an obsolete garbage" - "No, yours source is a trash!"
 * That may end at RSN, and the process may take years, because in the worst case scenario, each source may require its own RSN discussion.
 * To avoid that sxcenario, I propose to select (jointly) a representative set of sources on each concrete subtopic. "Representative set" does not mean "exhaustive", it means that the sources from this set adequately represent all important views, so addition or removal of few sources to that set does not change the overall picture appreciably. I have an idea how organise the process of sources selection, but if I propose it right now, I may be accused (again) of attempts to take the process under my control. Therefore, I would like to hear the opinion (or opinia) of other users first.
 * This discussion may take some time, but as soon as we achieve some agreement about the procedure, it may be applied to each topic.
 * Actually, I am already waiting for ideas. We may start right now. Paul Siebert (talk) 04:57, 16 December 2021 (UTC)

16 December 2021
In addition to anything else that is agreed on, I will ask each editor who is taking part in the source analysis to identify between two and four sources no later than 18 December 2021, and then other editors can comment on whether they are acceptable. Robert McClenon (talk) 07:53, 16 December 2021 (UTC)


 * I am afraid that is a little bit too broad.
 * First, we must agree on some concrete subtopic. I propose every user to name a subtopic, and then we select the subtopic by voting.
 * Second, anticipating possible disputes about neutrality in source selection, I propose to agree that the sources must be not cherry-picked. In other words, a user who proposes a source must explain how the source was identified. The procedure must be neutral, which means that any user with no preliminary knowledge of the topic would find the same (or similar) source using the same procedure.
 * Third, there should be some commonly accepted criteria of source's quality. They must be formal, and they should not depend on one's taste.
 * I think that will allow us to minimise possible misunderstanding and conflicts. Paul Siebert (talk) 15:52, 16 December 2021 (UTC)
 * To avoid possible accusations of taking the discussion under my control, I will make my next post after other three active participants expressed their opinia. Paul Siebert (talk) 15:54, 16 December 2021 (UTC)

Levivich
I have no idea what the formatting is for this page (e.g. separate section headers for each editor?) so if I'm doing this wrong, somebody please refactor or let me know.

I agree step #1 is to identify the criteria for source inclusion, before talking about specific sources.

The sources should be:
 * 1) Published in the 21st century (per WP:RSAGE) - for this topic, about which much scholarship is written, anything over 20 years is obsolete, too old to be a good source for this topic
 * 2) *Just to give everyone a sense of scale, Google Scholar has over 1,900 results for "mass killing" "communism" since 2017, and 5,780 since 2000 . There are plenty of 21st-century sources to draw from.
 * 3) Scholarship (per WP:SCHOLARSHIP), meaning published by
 * 4) a university press (e.g. Cambridge University Press)
 * 5) an academic publisher (e.g. Brill Publishers, see Category:Academic publishing companies) (but not mainstream non-academic publishers, like Penguin Books)
 * 6) an academic journal with an impact factor greater than 1.0 (an arbitrary cut-off designed to focus on mainstream scholarship instead of obscure or fringe journals)

Two obvious top-notch examples: Cambridge History of Communism and Oxford History of Communism. Levivich 16:27, 16 December 2021 (UTC)

Robert McClenon
User:Levivich - I think it will be second-level headings for dates, and third-level headings for editors. As long as the back-and-forth does not get out of hand, there is not a prohibition on back-and-forth at this time. Robert McClenon (talk) 20:01, 16 December 2021 (UTC)

Davide King
I believe that Siebert gave a good intro and I hope we can agree on (i) subtopic, (ii) neutral research, and (iii) source quality, for which Levivich gave good suggestion. For the topic about the link and theories, "mass killing" "communism" is fine as a good Google Scholar search, though we may get too much results of works about mass killings in general; for the topic about the events, I suggest "communist mass killings" (also used in the AfD's closure) or "mass killings" "communist regimes". One thing that must be noted is that "anti-communist mass killings" gave me 50 results in total, while "communist mass killings" gave me 81 results in total; however, that is a bit misleading, since the latter includes many sources about anti-communist mass killings, which gave us more results than communist mass killings, which is reduced to 31 results in total.

The Cambridge History of Communism and Oxford History of Communism are certainly good and balanced sources. It is interesting that the latter has chapters about "Religion under Communism" and "Sport under Communism" but no "Mass Killings under Communism" or "Communist Death Toll" chapters. Cambridge's History of Communism includes "Communism, Violence and Terror" by Hiroaki Kuromiya, which may be used to discuss links and theories. Davide King (talk) 01:54, 17 December 2021 (UTC)

Paul Siebert
I find this very useful. I also encourage all participants to read WP:RSVETTING I respectfully request and  to confirm their intentions to participate in this discussion, otherwise I see no reason to continue our conversation here, for there seems to be no significant disagreement among other participants.--Paul Siebert (talk) 20:30, 17 December 2021 (UTC)

Paul Siebert
I presented some source analysis at the MKUCR talk page, but if DRN participants prefer to discuss it in a quite place, I am ready to continue here. In general, it may be useful to make long posts with sources and quotes that the article's talk page, and to discuss them here. What do you guys think?--Paul Siebert (talk) 20:41, 22 December 2021 (UTC)

Davide King
I think it is a good idea to "make long posts with sources and quotes that the article's talk page, and to discuss them here." I think Siebert delivered in their first analysis, and I do not think Nug gave a good rebuttal here. I believe Siebert provided plenty of evidence that Rummel's estimates are unreliable; if someone is unreliable about something (in this case estimates and Communism), there is no point discussing whether it is due or undue. Next, I would like to see Siebert perform an analysis of Bellamy, Chirot, Jones, Mann, Sémelin, and Valentino (e.g. the sources used to support the status quo and listed here), and what they actually say and support, and what weight they hold in regards to genocide studies and Communist studies, what they have published about Communism in particular, and in general how accepted they are. Davide King (talk) 21:45, 22 December 2021 (UTC)

Davide King
It is Final Solutions: Mass Killing and Genocide in the 20th Century, not Final Solutions: Mass Killing and Genocide under Communist Regimes; The Dark Side of Democracy: Explaining Ethnic Cleansing, not The Dark Side of Communism: Explaining Ethnic Cleansing; Purify and Destroy: The Political Uses of Massacre and Genocide, not Purify and Destroy: The Communist Uses of Massacre and Genocide; and Why Not Kill Them All?: The Logic and Prevention of Mass Political Murder, not Why Not Kill Them All?: The Logic and Prevention of Communist Mass Murders. Can one truly look at their own publisher's synopsis and tell me that Communism is their main focus? All of those sources may be acceptable for B, not for A or C as has been done, and most important they are not supposed to be synthesized to broaden the scope to Communism as a whole just because they say Communist regimes or Communist mass killings, when what they mean and discuss are Stalin, Mao, and Pol Pot, three very specific Communist leaders who are universally recognized to have engaged in mass killings but communism was not the main cause, not Communist regimes as a whole. Davide King (talk) 14:58, 27 December 2021 (UTC)

List your sources and tell us which should be included
Per WP:CCC, you may change the reliability of each source if the consensus changes. MarioSuperstar77 (talk) 15:08, 27 December 2021 (UTC)
 * I would propose a little bit more focused approach.
 * Let's remove the RS column: we don't need to discuss the sources that are obviously unreliable;
 * Let's focus on some concrete topic. Keeping in mind the possible outcome of the RfC (the option "B" is clearly winning, to my big surprise), the most relevant subtopic is causative factors. Therefore, I would propose to focus only on those sources that discuss causes of mass killings in Communist states, and the sources that discuss these sources.
 * To facilitate a discussion, I propose to summarise (i) which category of mass killings each source discusses, and what terminology it uses, (ii) what each source tells about causes, (iii) does it link mass killings with Communism as a key factor, (iv) does in discuss these mass killings in a broader context. Paul Siebert (talk) 19:27, 27 December 2021 (UTC)

To demonstrate my point, I created another version of the table. If you like it, feel free to edit/expamd.--Paul Siebert (talk) 19:50, 27 December 2021 (UTC)
 * That is indeed more informative than my table, but I urge you to keep the unreliable listed sources with a header that clearly states that much. It will help editors remember what sources should not be used, and it will warn newer editors to not use those sources. As such, the sources will not be reintroduced into the article as a good-faith edit. MarioSuperstar77 (talk) 21:42, 27 December 2021 (UTC)
 * I have updated it in this same edit. Apart from the introduction to the Black Book and the VoCMF, which are either unreliable or not reliable (I prefer 'not reliable' for Courtois and Malia because there is consensus they are not 'green' but whether they are 'yellow' or 'red' depends on what we take from them), the real issue is that of NPOV and WEIGHT. Do they discuss Communist mass killings as a separate topic or as a subtopic? Is their focus on the link and the theories, or on the events? Do they represent a majority, mainstream view that is supported by country specialists and historians, or are they an isolated minority? Davide King (talk) 22:14, 27 December 2021 (UTC)

Regarding the table above, Malia wasn't even discussed in the two RSN discussions, yet it is listed as "not reliable" in the table. The first RSN discussion was about the reliability of Courtois' estimate and I don't see any consensus, and the second about the whole book BBoC, with most !votes saying use with attribution for the introduction. If we can't honestly represent RSN discussions, it doesn't bode well when it comes to interpreting published sources. Regarding the table below, the two columns "Cause(s)" and "Links to Communism as a primary causative factor?" can be merged into a single column "Cause(s)". The "Is there any controversy?" is subjective and should be removed, and the column "Broader context?" needs to be renamed "Context". --Nug (talk) 23:43, 27 December 2021 (UTC)


 * The first table is a subject of a separate discussion. I think "reliability" should be used only in some concrete context, and the claim about unreliability should be supported by a link(s) to previous discussions. That means we need one more column in the above table.
 * With regard to the bottom page, no. "Cause" and "Link to Communism" must be separate columns, because many authors (e.g. Valentino or Harff) discuss causes, but they do not discuss the link to Communism; others mention Communism in passing (Mann). We can merge these two columns only if we want to write an absolutely biased article.
 * "Controbersy" is by no means subjective. Thus, a number of sources exist that call Courtois introduction "highly controversial".
 * "Broader context" means that mass killings are discussed in some context that is broader than the article's topic. For example, in a context of politicide, or in a context of Asian genocides etc. Paul Siebert (talk) 00:31, 28 December 2021 (UTC)

Additional analysis on the Communist democides by Nug, from talk page. MarioSuperstar77 (talk) 20:42, 31 December 2021 (UTC)

Communist democides can be, and already are, discussed at Democide. Rather than take Rummel or Valentino's theories as fact, it would be good to present their weight in the literature. Valentino says nothing know and what I have repeatedly told about, namely that universally recognized mass killings were done under Stalin's, Mao's, and Pol Pot's leaderships, and his emphasis on leaders is in fact very grounded in reality; whether those other possible cases are relevant, we cannot base it from Valentino alone but from its weight in the literature, especially if some of them fit another category than Communist mass killings (e.g. counterguerilla mass killings, as is done for Afghanistan, which means that being a nominally Communist regime is not as important, and Siebert's point about Afghanistan is still correct, and is indeed also supported by Tago & Wayman 2010 — we need not to confuse Valentino's Communist mass killings category with any mass killing that happened under a nominally Communist regime); the literature (Chirot, Jones, Mann, and others) emphasizes Stalin, Mao, and Pol Pot, and unless those other possible cases become universally recognized as Stalin's, Mao's, and Pol Pot's, we should not give them more weight than universally recognized events warrant, other than a short paragraph saying this. Davide King (talk) 21:21, 31 December 2021 (UTC)

Kuper 1981

 * pp. 98-9 Any suggestion of ideological (idealistic) determinism must certainly be rejected. However, one can hardly deny the significance of ideological factors, even though they were not determinant. The revolutionary changes in Russian society were carried out in the name of Marxism-Leninism, and key concepts of the theory were applied in the framing and implementation of policies.
 * pp. 99-100...returning to the more general problem of the influence of Marxist theory, there are elements of this theory which have an affinity for highly destructive conflict. Acceptance of the theory of increasing polarization between proletariat and bourgeoisie, between the emancipatory forces of revolution and the enslaving forces of reaction, readily yields a Manichean vision of a world torn apart in deadly conflict between good and evil. The preoccupation with classes encourages an indifference to the individual and to individual suffering. The belief that being determines consciousness, a cardinal belief in traditional Marxism, readily yields a conception of guilt by social origin; people are guilty for what they are, rather than for what they do. Given the movement to a higher level of civilization, social categories, whose origins or function cast them in a reactionary role, are excluded from, or rendered marginal to, the moral community. In the light of a Messianic faith in a final and definitive upheaval which will lay the foundations for a utopian society, whole groups become expendable for the realization of an ultimate good, and a whole armoury of antitheses and lethal malediction becomes available&mdash;enemies of the people, bourgeois, nationalists, revisionists, rightists, reactionaries, doomed classes, counter-revolutionaries.
 * p. 100 It seems paradoxial to suggest that Marxism incorporates a dehumanizing ideology, conducive to most lethal and merciless conflict. Marxism is, after all, one of the great humanist doctrines of our day, with its vision of a society free from inequality and conflict, in which men may live in full creativity. Yet this combination of the highest ideals with massive slaughter is all too common. Indeed the high ideals often serve as warrant for massacre.

Fein 1993

 * As expected, unfree, authoritarian, and on-party communist states (in ascending order) are most likely to use genocide
 * One-party communist states are 4.5 times more likely to have used genocide than are authoritarian states
 * ...the discrimination between totalitarian and authoritarian states reflects a historical dimension of utmost relevance to the history of genocide in the 20th century...Although there has been much dispute about the value and implications of this division and of totalitarianism as a concept, especially given the disintegration of the communist bloc and wide variance in practice from the original ideal types, it is still a useful guide to states' ideological models for the period under study. Totalitarian ideologies which designate groups as enemies&mdash;either because they are from the wrong class or the wrong group&mdash;are ideologies which imply justifications for annihilation.
 * Harff, surveying both genocides, politicides, and mixed cases since 1945, concludes that "revolutionary one-party states are most likely to commit geno/politicides...closely followed by praetorian states [military-dominated authoritarian]" However, Harff's study, based just on the characteristics of perpetrator-states, did not take into account the number of states of each type extant. Her conclusion corresponds to that of much research on totalitarianism and violence.
 * Communist states were 4.5 times more likely to have committed genocide than all unfree states (36%: 8%) in these regions. This finding supports theories that explain why totalitarian states have greater needs and justifications to systematically slaughter groups of their citizens than authoritarian states. Theories reviewed by Schmid (1991) suggest that both state structure based on bureaucratic absolutism and Marxist-Leninist ideology interact to account for state murders and other gross violations of human rights in the former Soviet Union.

Fiveby (talk) 15:22, 28 December 2021 (UTC)


 * Thank you for adding new sources, but I would prefer if we follow the format that was set in the above table. Concretely, before discussing Kuoer, it would be necessary to place him into the table and fill all fields (does he propose/use some specific terminology, what is the focus of his study, etc). Without that, I cannot discuss this source. The same can be said about Fein. Can you please do that, and then we continue. Paul Siebert (talk) 20:52, 21 January 2022 (UTC)
 * well i tried. I think it just goes to show how we began working at cross-purposes here, don't think Kuper has any use in defining scope for the article. More towards a bibliographic narrative or possible narrow use. Feel free to remove that row. Or we could leave it there as a counter to 's AfD observations on human nature. fiveby(zero) 23:42, 21 January 2022 (UTC)

Strauss 2014
"The convergence of terror and revolution has been noted from the time of the French and Russian revolutions. This article lays out the ways in which the concept of extreme political violence in communist revolution from Russia to Cambodia continues to be both politicized and resistant to agreement about such basics as definition, scale, and numbers of victims. There is particular disagreement about whether to count as victims of terror those who were the collateral damage from poorly conceived and brutally implemented policies such as collectivization and the Great Leap Forward alongside terror deliberately inflicted upon particularly targeted individuals and groups. The article suggests that whether deliberate or incidental, terror in communist systems is best understood as proactive and reactive campaigns to ensure regime security by mobilizing the bureaucracy and engaging in a display of communicative theatre with mass populations through such forms as mass trials."

Cowe 2014 (review)
"In his introductory essay, S. A. Smith acknowledges the basic contradiction within the conditions needed to propagate Communism, as outlined by Marx, and the reality of those states which actually adopted it practically. With certain notable exceptions, he shows that Communism often took root either as a direct result of war/colonial insurrection and/or within countries with authoritarian systems already in place 'changes of borders, the devastation caused by war, genocide and forced migration as a consequence of the imperial politics' that beleaguered Eastern Europe and that 'played an essential role in the establishment of communist regimes' (p. 204). Thus the basic premise is that Communism took root in countries which were unprepared economically and as a result, the implementation of it at a state level was flawed from the beginning."

Kuromiya 2017
Abstract needed. may be able to get full access and summarize.

Both of those are much better, and more importantly mainstream, sources on the topic. Davide King (talk) 15:43, 28 December 2021 (UTC)
 * Please, add a description of those source to the table 2. Without that, we cannot discuss them seriously. Paul Siebert (talk) 20:54, 21 January 2022 (UTC)
 * , I am not sure how to describe it — it is mainly about the Soviet Union, China, and Cambodia but it does mention the Eastern European countries, and the scope appears to be the Red Terror and collectivization campaigns, though I am not sure about the terminology. I am also not sure about the causes but it does not appear to see generic communism as the primary cause, rather Marxist–Leninist vanguard party system. I hope that you can verify whether this reading is accurate and if you have access to it (I got it through The Wikipedia Library), so that you may help to verify and give a definite description for the table. If it is not a problem to use a quote, "Conclusion" says: "'There were repeated stimuli that pushed communist regimes to pursue campaigns of terror against their citizens, and ultimately against themselves. Part of this was situational: communist parties came to power in wartime contexts rife with competitive militarization, brutality, and the disintegration of even the most basic elements of physical security. Communist parties were in part successful because in many environments they were better organized, better disciplined, and more ruthless than their competitors. Once established, the dictatorship of the proletariat required that all manner of defined enemies be either exterminated or cowed into submission, and the main way in which this was accomplished was through the institution of the bureaucratic campaign. The problem that many communist regimes ran into was that freed of legal or social constraints, the vanguard party, convinced of its access to absolute truth, was also free to make catastrophic errors. Whether it was the pursuit of disastrous policies to (p. 368) realize economies of scale in grain production or the collateral damage from a campaign against putative internal enemies, there were no institutional restraints on the vanguard party. Social groups that might have blunted excesses were eliminated, fractured, or cowed into silence. In a system in which the vanguard party was only accountable to its own fluctuating notions about how to best pursue the revolution, the methods that had enabled it to come to power and consolidate its rule against real or likely enemies became naturalized as methods of rule that, by vanquishing imaginary and fantasy enemies, would ensure the maintenance of the power of the party and the ultimate victory of socialism.'" Subsections include:
 * "What Is This Thing Called Terror? Concepts, Claims, and Numbers"
 * "The Security, Social and Political Context(s) of Terror: Militarization, State Building, and Uncertainty"
 * "Terror as Communicative Theatre"
 * "Escalation of Terror: Ideologically Charged Responses to Elite Contention and Social Uncertainty"
 * "Conclusion"
 * An interesting paragraph and a note:
 * It uses "excess deaths" just like Siebert explained many times and I believe it confirms what they have been saying the whole time, writing:
 * "And contemporary demographers who reconstruct numbers of excess deaths in the notoriously bloody twentieth century—i.e. deaths in excess of the statistically expected number of deaths in a population over a specific length of time—feature communist regimes prominently, although not exclusively. A quick perusal of websites on the topic shows that many of the activities associated with terror (in lower-case letters) and 'The Terror' of the Stalinist Soviet Union and its imitators feature on lists of what is often called 'democide': purges, executions, forced deportations, and famine-induced excess deaths. It is natural to slide between terror and democide, since all forms of mass killings clearly involve a great deal of terror for the victims. The more careful of such websites acknowledge the wide variation in estimates of excess deaths, with annotated citations, suggesting that these differences are down to a left–right bias on the part of the individual researcher or else are due to incomplete or inaccessible data. However this conflation of terror and excess deaths in general conceals more than it illuminates."
 * "The literature on what Lenin was and was not responsible for in terms of establishing the basic template for terror in the Soviet Union is considerable. Recent books that identify Lenin with terror and callous regard for human life are Christopher Andrew and Vasili Mitrokhin, The Sword and the Shield: The Mitrokhin Archive and the Secret History of the KGB (New York: Basic Books, 2000) and Stéphane Courtois, Mark Kramer, et al., The Black Book of Communism (Cambridge, MA: Harvard University Press, 1999). Both, while cataloguing undoubted horrors, lump together different causal dynamics. The subject matter of both, and the lack of comparative and contextual framing, skews the resulting analysis in favour of a very high number of deaths."
 * Davide King (talk) 22:32, 22 January 2022 (UTC)


 * I have not been able to fully read Kuromiya 2017 but I found this from the American Historical Review (2019): "'The second half of the first volume presents eleven essays summarizing (mainly) the state of Soviet historiography on Stalin's consolidation of state power on the eve of World War II, from 1924 to 1941, and four more essays on the broader Eurasian context. In the opening essay of part 2, Hiroaki Kuromiya brilliantly summarizes the relationship between Communism, violence, and terror in Stalin's rise to power: 'Terror and violence are endemic to dictatorships. In the history of political violence, the Soviet practice marked a new stage. Political violence was ideologically justified and exercised by the first communist state in world history. Backed by Moscow, communists elsewhere also resorted to violence as a political weapon' (1:279). Here Kuromiya takes a bow to The Black Book of Communism (1997) and its lead essay by Nicolas Werth. In this way, Kuromiya seems to embrace the controversial Black Book and its conclusions. But a close reading of Kuromiya's essay substantially reduces the absurdly exaggerated data of the Black Book, presenting numbers that are a fraction of those in the Black Book—chronicling evil, yes, but on a much smaller scale than the scales of atrocities depicted in the Soviet-bashing typical of Cold War and post–Cold War Western historiography.'"
 * I found very interesting the part about the "close reading"—that is precisely why we must rely on secondary academic coverage rather than our own views of each work like has been done for Valentino.Davide King (talk) 22:45, 22 January 2022 (UTC)
 * It seem this book is not available from my university's library. However, from other articles by Kuromiya I conclude that he is either "revisionist" or "post-revisionist" (according to Karlsson's classification). Paul Siebert (talk) 00:10, 23 January 2022 (UTC)

Davide King
As expected, majority of sources cited are cherry picked from passages about Communism and Communism as a whole but are about genocide and mass killings in general, therefore they are in fact in support of my Mass killing expansion and/or Mass killings in history article proposal. Even those that are focused on Communism, and as a result have 'Communist' in the title, focus mainly (if not only) on Stalin, Mao, and Pol Pot, not Communism as a whole, because it is those three leaders who are universally recognized to have engaged in mass killings. In addition, the major issue is NPOV and WEIGHT, not VERIFY; of course there are going to be sources discussing Communist regimes but in what context? Majority of given sources do not fit 's suggested criteria here, which is in fact very acceptable. Finally, several of them still rely on totalitarian theory, which is clearly a minority view at best in academia. If possible, I would like to comment on Levivich's suggested criteria and whether the points I raised here are warranted.


 * References

Davide King (talk) 15:20, 28 December 2021 (UTC) [Edited to add Guilhot 2005] Davide King (talk) 15:50, 28 December 2021 (UTC)
 * Not sure what you'd like to see here, and everyone seems to be heading off on the own directions with the analysis. I was proceeding from the reviews located so far, and identifying references to earlier works. If, for instance Harff&Gurr say that Kuper, Fein and Charny "are of particular interest" then why should we not add those sources to the analysis? I would point out that the totalitarianism sources will naturally appear before the revisionist sources both chronologically and when authors review prior works. There is quite a bit of material to get through. If this is the character of this dispute resolution and an indication of the arguments that will be happening during the source analysis then it will certainly make no progress. fiveby(zero) 16:21, 28 December 2021 (UTC)
 * I would have expected that we follow Levivich's suggested criteria because the major issues are NPOV and WEIGHT, therefore we must identify mainstream sources and more recent sources that summarize the literature for us to determine their weight. Davide King (talk) 16:41, 28 December 2021 (UTC)

Paul Siebert
I am not sure the discussion develops in a right direction. It is premature to discuss any quotes until we came to an agreement about reliability of each source, its scope, terminology, main thesis and controversy. The reason is obvious, because: Therefore, I would be grateful if, instead of filling the space with large quotes, we all sick with the scheme shown in the second table. Please, add new sources to that table and fill all fields. Furthermore, as soon as we started to discuss Harff, let me point out that this source is very old. It was published before the "archival revolution" in the USSR, and the data used by the authors are obsolete, Cold war era data. Thus, I checked her Table 1, and the first line is 1943-47 politicide in the USSR that killed 0.5 to 1.1 million repatriated Soviet nationals. I saw no mention of those deaths in modern sources that discuss victims of Soviet repressions. Wheatcroft, Ellman, Rosefielde, and other authors never tell about that. I decided to check if my conclusion was correct, and I found this source. It is the first in the list, it is peer-reviewed, it is cited by peers, it is telling specifically about repatriation of Soviet citizens, and it is recent. All of that makes it much more trustworthy. This source says:
 * We must clearly understand what concrete events each author means. Thus, when some author means a linkage between Communism and genocide, they may mean, e.g., only Cambodia or Cambodia and China, etc.
 * We must be sure the authors use the same term in the same or different context. It is especially important for such terms as "mass killing" and "genocide".
 * We must make sure we do not misinterpret the author's opinion by taking their view out of context. Thus, if we use Mann, Valentino, of Harff we must clearly explain that he didn't see a linkage between mass killings and Communism
 * We must clearly separate controversial and non-controversial sources.
 * "Although repatriation was unmistakably compulsory and arrest was frequent, declassified Soviet archives have demonstrated that Cold War-era works vastly exaggerated the scale of repression."

Moreover, the Gulag article cites a source (Zemskov V.N. On repatriation of Soviet citizens. Istoriya SSSR., 1990, No.4) that provides a detailed statistics of a fate of repatriated civilians: out of 4.1 million of repatriated citizens, only ~7% of repatriated citizens were imprisoned in Gulag, and others were sent home, conscripted etc. There is no information about a million of killed in some "politicide": teh first line in Harff's table 1 tells about a politicide that never was, which adds no cfredibility withg other lines in her table. That means by using Harff as a source, we create a POV-fork: we tell a story of a politicide that never occurred, according to new sources. This is a very complex situation, because we cannot combine Harff and Zemskov in one narrative without a danger of OR. We have a very unusual situation when different groups of sources tell different stories and present different facts, but there is no dispute between them. We need to develop a general approach to this situation.--Paul Siebert (talk) 21:34, 28 December 2021 (UTC)

Nug - Additional sources
See Dispute resolution noticeboard/MKUCR Quotes from Sources.

(I have moved the 69 sources because, with separate headings, they make it nearly impossible to navigate the Table of Contents. Robert McClenon (talk) 01:05, 29 December 2021 (UTC)}}

Davide King
Sorry but what Nug did above is not only unhelpful but even disruptive in light of Siebert's comment before, and the moderator's rules about posting here. To avoid escalation of disputes, I am not going to do it but I kindly ask to revert them — Nug are free to link the same list of sources from a sandbox, there is no need to bloat the page like this, and is not unhelpful at all. The issue is not VERIFY, but NPOV and WEIGHT, therefore providing cherry-picked sources, most of which are already used in the article as it stands, and we all know what the AfD said about it, does not lead us forward but backward. Davide King (talk) 23:39, 28 December 2021 (UTC)