Talk:Mass killings under communist regimes/Archive 55

Archive 50

←

Archive 53

Archive 54

Archive 55

Archive 56

Archive 57

→

Archive 60

100 million figure in the lead

The claim that the 100 million figure is the most common-cited number has been reverted back into the lead, even though there is no support whatsoever for that wording in the body. The closest things in the body are a cite to the blog of the Victims of Communism Memorial Foundation - an obviously WP:BIASED source funded by a think tank - and a quote by Engel-Di Mauro that that is the most common figure used by a specific, narrow surge of anti-communists in that particular time period, which is devoted to exhaustively debunking it. Neither of these are anywhere near sufficient to support the argument that that is the most common figure in general, which is what the proposed text implies. I feel it's undue for the lead in any form, but it is flatly factually inaccurate and unsupported by both the article and its sources in the version people keep attempting to add. In fact, the one source that was being used for anything remotely resembling it in the text specifically did not support the idea that it was a common claim in general - it specifically said that it was an opinion that originated solely in the introduction to The Black Book of Communism and was otherwise unsupported. The lead rewrite is already WP:BOLD; this addition is clearly contentious and needs discussion per WP:BRD (though I see absolutely no possibility that it could be supported in its current form - it is a misreading of the article, nothing more.) The only sources in the article that actually endorse it (as opposed to bringing it up to criticize it) are the intro from The Black Book of Communism and the blog of the Victims of Communism Memorial Foundation, which are some of the lowest-quality and most WP:BIASED sources in the article and which certainly cannot be used to define the lead. This is a highly-controversial article; when an WP:EXCEPTIONAL new addition to the lead is disputed, discuss it, don't just put it back with an edit summary saying it's fine. Do not restore it without consensus. --Aquillion (talk) 02:58, 6 December 2021 (UTC)

It has been strongly suggested, or I am now (again), that is the synthetic topic the article is based on. I'm starting to favour the title change to "Victims of Communism", the canard that mention 100 million. ~ cygnis insignis 03:16, 6 December 2021 (UTC)

@Aquillion: that is in my to-do list. However, first, I would prefer to focus on the Terminology section, which creates an absolutely false impression that MKuCR is some well defined scholarly topic. After we resolve this problem, I propose to totally rewrite the "Estimate" section: in reality, not only the figure by itself, but even the very idea to come up with some single figure and link it to Communism is seen as deeply politicised and controversial idea. That means the discussion of any global estimates must be done only in a context of political implications of that figure. It is deeply incorrect to use a pseudo-neutral scholarly tone for that description. Instead, we must explain why the idea to combine all figures into a singe aggregate figure was proposed, who links it with Communism, what are political purposes, who supports this approach, who criticises, etc. After all of that will be done, the lead will be updated.

I propose you to join my work on "Terminology", and after that we will jointly work with "Estimates". I have a plan, I have a lot of sources, and we will be able to convert the section into a nice and neutral story. Paul Siebert (talk) 03:21, 6 December 2021 (UTC)

That is fine, but this is a new addition to the lead. It's easier to object when it is first added, especially given that I feel that the body simply doesn't support it; the article has gone without this line for years, it can go without it for a while longer until / unless we reach a consensus on some alternative version. --Aquillion (talk) 03:23, 6 December 2021 (UTC)

Usually, the lead is supposed to reflect what the article's body says, so the best way is to start from the article's body. Since the latter says "100 million", there is no strong reason not to include it into the lede. It may be reasonable to remove it, but that is not a urgent need. And keep in mind that the article is under 1RR, so you may be reported for technically exceeding the 1RR limit. I am warning you, because the only block that I got during my Wikilife was exactly for that reason and when I was editing this article. Paul Siebert (talk) 03:33, 6 December 2021 (UTC)

As soon as we started talking about estimates, take a look at this:

"In 2017, historian Stephen Kotkin wrote in The Wall Street Journal that communist regimes killed at least 65 million people between 1917 and 2017, commenting: "Though communism has killed huge numbers of people intentionally, even more of its victims have died from starvation as a result of its cruel projects of social engineering."

From this quote, a reader may conclude that Communists killed 65 million, and even more died from starvation. That is a direct lie. Kotkin said that 65 million deaths is the demographic estimated. ...in the Soviet Union, China, Mongolia, Eastern Europe, Indochina, Africa, Afghanistan and parts of Latin America—communism has claimed at least 65 million lives, according to the painstaking research of demographers. That means, (i) Kotkinn cites someone other's figures (not his own data), (ii) he speaks about demographic losses. The current text of this "well written and well sourced" article contains tons of misinterpretations or a direct lies. It need a thorough analysis and, probably, major rewrite. However, let's finish with Terminology first. Paul Siebert (talk) 04:01, 6 December 2021 (UTC)

Paul Siebert, with respect, you are not parsing Kotkin's statement accurately, probably because English is not your first language. Kotkin's statement means that, while communists did in fact kill people deliberately, the majority of the 65 million+ deaths they caused were the result of unintended famines. There is no other way to parse Kotkin's statement that would be internally consistent or logical. When Kotkin states that there were at least 65 million victims of communism, some of whom were purposefully killed, and that "even more of its victims have died from starvation," [emphasis added] the reference to "victims" refers back to the larger whole of 65 million. In English, "Though communism has killed huge numbers of people intentionally, even more of its [65 million] victims have died from starvation" does not mean that "communism has killed 65 million people intentionally, and even more victims have died from starvation," nor could anyone fluent in the language reasonably construe it to mean such a thing.TheTimesAreAChanging (talk) 04:46, 6 December 2021 (UTC)

It seems you parse my own statement incorrectly, for I am saying exactly what you say.

The full paragraph says:

"But a century of communism in power—with holdouts even now in Cuba, North Korea and China—has made clear the human cost of a political program bent on overthrowing capitalism. Again and again, the effort to eliminate markets and private property has brought about the deaths of an astounding number of people. Since 1917—in the Soviet Union, China, Mongolia, Eastern Europe, Indochina, Africa, Afghanistan and parts of Latin America—communism has claimed at least 65 million lives, according to the painstaking research of demographers.

Communism’s tools of destruction have included mass deportations, forced labor camps and police-state terror—a model established by Lenin and especially by his successor Joseph Stalin. It has been widely imitated. Though communism has killed huge numbers of people intentionally, even more of its victims have died from starvation as a result of its cruel projects of social engineering"

This paragraph does not allow double interpretation: Kotkin says that, (i) according to demographers, at least 65 million people died prematurely, and (ii) these deaths included such categories as mass deportations, state terror, starvation etc.

Demography is intrinsically incapable of separating death by categories. It is a pure statistical discipline that deals with deaths (from all causes), births, migration etc. The figure of 65 million may mean either "excess mortality" (all premature deaths) or "population losses" (unborn infants, emigration AND premature deaths). Period. I totally rule out an possibility that Kotkin, a professor in Princeton, use "population losses" figures (that would be ridiculous), so he definitely means "excess mortality". And, obviously, he is too educated to refer to demographic data as an estimate for the number of execution deportation etc. Demography cannot provide such information. To claim the opposite would be as ridiculous as to claim that by using a multi-meter it is possible to tell if the electricity in your home was produced at nuclear or gas power plant.

That means, Kotkin's statement is in agreement with other sources: yes, "excess mortality" (all premature deaths) in Communist states (excluding WWII deaths in teh USSR) amounted to 65 million at least, and lion's share of those deaths were Great Chinese famine deaths. That is what he says.

And what we see in the MKuCR article?

• It falsely ascribe to Kotkin a claim that Communists killed 65 million (that is a double lie: that figure was obtained not by Kotkin, but by unnamed demographers, and Kotkin never said 65 million were killed by Communists)
• It claims that famine deaths and other categories are not included in those 65 million (that is a direct lie, for Kotkin says that, according to demographers, Communist rule lead to a loss of 65 million, and that included executions, deportations, starvation etc).

Just an example of one (out of many) falsifications that I found in this "pretty well sourced and well written article". Paul Siebert (talk) 23:37, 7 December 2021 (UTC)

How does one deal with such a proliferation of misstatements as that uttered by Paul Siebert above?

• "It seems you parse my own statement incorrectly, for I am saying exactly what you say." Right, which is that our text (consisting almost entirely of a direct quote from Kotkin) is very clear and cannot reasonably be misunderstood as double-counting the 65 million excess deaths that were caused by communist governments in the twentieth century.
• "It claims that famine deaths and other categories are not included in those 65 million (that is a direct lie ... " Our Wikipedia article, which includes a footnote with the entire excerpt for readers to examine for themselves, very obviously says nothing of the kind (nor has any other editor supported your peculiar interpretation), so your statement could indeed be considered a "direct lie."
• "It falsely ascribe to Kotkin a claim that Communists killed ... " To the contrary, our article merely quotes Kotkin accurately as stating that "communism has claimed at least 65 million lives ... Though communism has killed huge numbers of people intentionally, even more of its victims have died from starvation as a result of its cruel projects of social engineering." [emphasis added] Note that Kotkin's use of "intentionally" implies that communist governments also unintentionally killed people.
• "And, obviously, [Kotkin] is too educated to refer to demographic data as an estimate for the number of execution deportation etc." Based on this comment, it seems like you are suggesting that even though the source (Kotkin) uses demographic data to estimate excess deaths, which he himself refers to as both intentional and unintentional "killings," the source is wrong and should know better (based on his education) that only direct executions (which demography cannot separate from other causes of death) qualify as "killings" (or "mass killings"). If your original research were accepted, this would set a drastic precedent for the rest of the article, but, alas, we cannot use Paul Siebert as a reliable source. We have to stick to what the source says, and cannot change it based on editor opinion alone.
• "It falsely ascribe to Kotkin ... that figure was obtained not by Kotkin, but by unnamed demographers ... " For the record, the current version of the article states: "In 2017, historian Stephen Kotkin wrote in The Wall Street Journal that ... 'communism has claimed at least 65 million lives, according to the painstaking research of demographers.'" Paul Siebert's suggestion that we are somehow misrepresenting or distorting Kotkin, in some way, or that "wrote" isn't the correct form of attribution, and hence that the only solution is to remove Kotkin entirely as a source, seems like an astonishing (and disappointing) case study in bad-faith wikilawyering.TheTimesAreAChanging (talk) 23:29, 9 December 2021 (UTC)

• The body certainly does not say, in the article voice, that the most common estimate overall is 100 million (that was my specific objection.) It doesn't say anything remotely close to that. It only mentions it a handful of times, most of them criticizing the number or attributing that view specifically to a handful of fierce anti-Communists. I assume you are not going to argue that we can use the blog of the Victims of Communism Memorial Foundation unattributed in the lead; and even attributed I would argue it is undue for the lead (there is currently a reasonable argument above over whether they should even be used in the body, but citing their blog in particular seems dubious even in the body - personally I would never support putting them in the lead. Their blog could only be cited as WP:RSOPINION at best and they're just not noteworthy enough for their opinion to be leadworthy. "Think tank / advocacy group believes thing in line with their views", with no secondary sources treating it seriously, isn't really noteworthy.) Even before I changed it, the bit cited to Engel-Di Mauro still (accurately) stated that the figure is used only by anti-communist scholars, making it misleading for the lead to state that the figure is common overall. If we mention it in the lead at all, we would have to word it very differently - one thing the WP:RSN seems to be leaning towards is that we definitely cannot use Courtois without attribution, which this would be doing (since the sources near-universally agree that that number originates exclusively from him.) Regarding the 1RR, my last edit was two days ago - unless you meant my two recent edits; the second one wasn't a revert, but even if it were, multiple consecutive edits with no intervening edits are counted as one for 3RR / 1RR purposes. --Aquillion (talk) 04:58, 6 December 2021 (UTC)

The 100 Million claim is obviously very contentious, and as far as I can tell only used by the fiercest partisans of this topic, i.e. Courtois, the VOCMF etc. To say that it is the most common number is not supported by the body or the sources provided, and if it is presented in the lead or even really the article at all it must be presented as what it is, a (WP:FRINGE at worst, minority at best) view held by people who have demonstrated clear ideological bias with regards to this topic. BSMRD (talk) 03:42, 6 December 2021 (UTC)

Please, go to the AfD talk page and read the quotes collected by me. That representative sample of sources gives an impression of how this approach is treated by scholars. It is especially interesting to read Werth's opinion. IMO, all those sources must be used in the re-written "Estimates" section to provide a proper context. And the section should be renamed to something like "Attempts to propose global estimates of Communist death toll and its political implications". Paul Siebert (talk) 04:05, 6 December 2021 (UTC)

• Howdy hello folks. I added the 100 million figure based on my reading of the article. I have no connection to the article, I just like writing leads a lot (its my favorite part of Wikipedia). When I saw this article only had like three sentences for a lead, that felt just criminal! So I spruced it up a bit from what I got out of the article. The 100 million figure stood out because it was mentioned repeatedly. If it isn't reliable, that should probably be made more clear in the body. Or perhaps there could be a note in the lead saying something like "100 million is mentioned commonly in the popular press, but is likely inaccurate as well." But I think some number should exist in the lead. Folks don't want to sift through 20 different estimates. That's why I included the high and low values, and what appeared to be the mode (100 million). I have no attachment to that number, aside from that some numbers should be mentioned in the lead. Remember, the majority of readers don't ever make it past the lead. CaptainEek ^{Edits Ho Cap'n!}⚓ 05:52, 6 December 2021 (UTC)

• CaptainEek, this goes back to what is the main topic supposed to be — Communist mass killings, in which case the events and what happened is more important (e.g. we would say what happened according to majority scholarly sources, and limit ourselves to Stalin, Mao, Pol Pot, and the Red Terror with the Russian Civil War and White Terror in mind, e.g. mass killing events according to the most accepted definition of 50,000 within five years, which makes it pointless to mention death tolls, as that is out of the scope), or Communist death toll, in which case the events are to be seen as death toll events (e.g. we would say number of people died under event A, and we may mention all Communist regimes, in which the case the scope is much broader). The problem of the article is that it attempts to do both things, and treat all events as generic mass killings, plus ascribing a number of causes that are too ideologically-centered and not as nuanced and context-minded as majority scholarly sources and country experts do.
• Until we actually agree on the main topic, its structure, and core sources, we are always going to have this problem. I would not have a problem mentioning estimates, perhaps just changing the wording that lower estimates are more reliable, higher estimates are from the popular press and/or include famines and other events which scholarly sources say should not be counted, and of course the criticism of the body-counting itself as a useless exercise to score a political point and push the view that Communism was worse than Nazism because it killed 100 million, or more people in general. Davide King (talk) 06:17, 6 December 2021 (UTC)

• CaptainEek is obviously correct in that the lead needs some sort of summary. I agree with Davide King: I would not have a problem mentioning estimates.XavierItzm (talk) 13:03, 6 December 2021 (UTC)

It should be pointed that Valentino (Cornell University Press, 2005) says: "Communist regimes have been responsible for this century's most deadly episodes of mass killing. Estimates of the total number of people killed by communist regimes range as high as 110 million." This puts the lie to the canard that only Courtois (Harvard University Press, 1999) or Dissident (2016) have arrived at 100 million.XavierItzm (talk) 13:28, 6 December 2021 (UTC)

Can’t believe I have to point this out, but saying “estimates include” does not constitute making an estimate. If you want examples of people quoting the 100m figure, any fascist meme page on Facebook will have plenty. The issue is where the estimate comes from. DublinDilettante (talk) 14:49, 6 December 2021 (UTC)

• No, it does not support your contention that someone other than Courtois reached 100 million - note, the blog you mentioned as your other source was vague as to who it was summarizing, which is to be expected for a mere blog post, but, as far as I can tell, was merely referencing Courtois as well. "Estimates include" means it surveyed all other estimates and that, presumably, Courtois was one of the ones they included (obviously, it doesn't support saying that it was the most common estimate, but I think we're past that particular dispute - I don't see anyone still arguing that 100 million is the most common or that we can say that in the article voice.) If you want to show that someone other than Courtois reached that figure, you need to show another source reaching the figure, or saying that someone other than Courtois reached the figure, not a source vaguely acknowledging that that is a figure that has been reached (we know it has been reached, by Courtois.) Also, could you stop referencing the blog as Dissident (2016) as if it were some sort of academic source? It's a blog; we cannot use it in the article. --Aquillion (talk) 23:22, 7 December 2021 (UTC)

A distinction without a difference. Dissident (2016) is the Victims of Communism Memorial Foundation putting out its own number. What you are trying to say is that a Tweet by The Coca-Cola Company is not a tweet by Coca-Cola Company because obviously Coca-Cola Company has no fingers so it can't type a tweet. Duh! Obviously the person who writes Coca-Cola tweets is an employee of the company and whatever is published by the official tweet account is considered as spoken by the company. Ergo, what Dissident put out in 2016 is evidently what the Memorial Foundation is saying. XavierItzm (talk) 04:32, 8 December 2021 (UTC)

If you believe that someone's opinions or speculation on a random blog is an appropriate cite for a figure in this article, you're free to take it to WP:RSN (though the last discussion near-unanimously found this exact source unreliable for this exact statement; I'm baffled as to how it somehow creeped back in.) But I think your time and energy would be better spent finding better sources - if, as you say, this figure is actually significant, then it should be easy to find actual high-quality WP:RSes covering it, with a weight comparable to the other sources in the section, rather than being reduced to arguing that we can include random opinions from blogs. You might feel that it makes no difference where an opinion is published, but WP:BLOGS and WP:RSOPINION disagree - RSOPINION covers only some sources; it does not simply allow us to use any source we please as long as we have an in-text citation, while BLOGS flat-out says such sources are largely not acceptable. Again, seriously - you are arguing that this is a vital figure of paramount importance that belongs in the lead, and you are citing this to a rando think-tank blog! Come on. If you're convinced its so central to the topic, just find a better source. --Aquillion (talk) 21:12, 10 December 2021 (UTC)

@Aquillion: let's be realists. For many years various users were trying to fill the "Estimates" section with reliable data, and that lead to the current (quite a pathetic) state. Just in case if you haven't seen it, take a look at my post that I made in the "Blog..." section. It seems there lousy sources are the only sources on this topic. The section must be completely rewritten (followed by the lede). Taking into account, that the DRN is going to be very slow, I have a feeling I should not wait for its end. Paul Siebert (talk) 21:32, 10 December 2021 (UTC)

@XavierItzm:

Valentino cites Rummel, who is recognised as an unreliable source for figures (see RSN). Dulic provided a nice analysis of all flaws in Rummel's approach, and, according to Harff, Rummel's response to Dulich's criticism didn't satisfy Dulich.

Valentino's own study do not involve any analysis of figures. He just cited this figure without any critical analysis. He is neither demographer, nor historian, he is a genocide scholar, and his expertise is not figures. The main goal of his study was to identify causes of mass killings and find the methods for their prevention. By the way, do you know what is his main conclusion? A conclusion is: regime type does not matter (and that is equally applicable to Communism too). In connection to that, don't you find it amazing that those users who extensively cite Valentino, simultaneously resist to bringing the article into accordance with his main conclusion? Paul Siebert (talk) 23:54, 7 December 2021 (UTC)

Paul, stop repeating this canard that Rummel's data unreliable. Wayman and Tago did a comprehensive analysis of his data and found Rummel numbers are consistent and his database is a good framework for studying mass killings during the 1900-1987 period. --Nug (talk) 00:32, 8 December 2021 (UTC)

@Nug: Thanks. I saved this diff.

Please, explain (with quotes) what concrete analysis of Rummel's figures was performed by Wayman&Tago.

Assuming that your assertion is correct (in reality, it is not), explain, how this W&T's assertion can be reconciled with the outcome of Rummel vs Dilic discussion, and with Harff's summary of it?

Assuming that your assertion is correct, how it can be reconciled with Karlsson's conclusion?

Assuming that this your assertion is correct, how can it be interpreted in a context of the recent RSN discussion of Rummel

Please, explain that, otherwise I have a serious reason to accuse you of personal attacks (accusation in repeating a "canard") and cherry-picking. In addition, repetition of already debunked arguments may be considered as a sign of WP:IDHT. Paul Siebert (talk) 01:12, 8 December 2021 (UTC)

I've explained this to you previously. Dulić's analysis was criticised for extrapolating his findings on Yugoslavia to Rummel's complete dataset, moreover his analysis only covered a portion of the time period covered by Rummel (Rummel, RJ One-thirteenth of a data point does not a generalization make: A Response to Dulić. Journal of Peace Research 41(1): 103–104.). Secondly, Karlsson's conclusions on Rummel is grossly WP:UNDUE and is not cited by anyone apart from you here on Wikipedia, it is confined to a couple of sentences in his paper with no further elaboration, whereas Wayman&Tago devote an entire 11 page paper showing that after examining the 18 consensus cases of communist regimes, Rummel's identification that 72% of communist regimes have engaged in mass killings compared to Harff's identification of only 22% is due to the fact that Harff's dataset is a narrower dataset strictly confined to geno-politicide and excludes other mass killings outside of that definition. Thus Wayman&Tago state: Most communist regimes, one can say, based upon the literature as we have reviewed it, are 'guilty' of democide but not of geno-politicide. --Nug (talk) 03:59, 8 December 2021 (UTC)

@Nug: sometimes, I have a feeling that you are acting in bad faith, and all what you write is a pure demagogy. However, your other posts (including your last post) create an impression that you are a quite reasonable person, who sincerely wants to achieve consensus. I sincerely want to resolve this puzzle. Let's agree about the following: I answer to your arguments in full, and I will try to address all your major points. If you disagree, in your responce, please explain me why exactly you find my arguments unconvincing. If you don't answer, I will conclude that my arguments satisfied you. In the latter case, please, do not raise the same arguments again, neither during a discussion with me, not in your responses to other users; if you do that, I will conclude you are acting in a bad faith, and will act accordingly. The same is equally applicable to me: if you demonstrate that my arguments are flawed, I will never repeat them at any Wikipedia page. Deal?

If you agree, please, let me know, and I will provide a detailed responce about Rummel, Dulic, Karlssen, Harff etc. It is a really interesting issue, which I will gladly discuss with you. Paul Siebert (talk) 18:33, 8 December 2021 (UTC)

First I'd like your input on what you think Wayman&Tago2010 is discussing and what you believe their conclusions are with respect to Rummel. --Nug (talk) 21:27, 9 December 2021 (UTC)

Ok.

First, you must understand that the author's goal was to demonstrate "that differences in the two major datasets can significantly affect the results of predictions of mass political killing". They do not examine validity of datasets. They just assume that both Rummel and Harff use reliable data, and W&T do not question their validity.

They acknowledge that Rummel's figures and Harff's figures are different, but W&T attribute this difference to different inclusion criteria used by these two authors. In other words, if Rummel's "democide" in, some state was 40 million, and Harff's "politicide" was 2 million, that is not because Rummel's or Harff's data are wrong, but because these two authors include different categories of deaths into their database.

However, they noted that

• "...most scholars except Rummel have worked with Harffs data." (W&T, page 7). That directly contradicts to your previous assertion that Rummel's database is at least as popular as Harff's. In connection, your allegation about my ostensible POV-pushing in Mass killing look somewhat groundless. I am waiting for your comments on that. What about some form of apology?
• that Harff's data are systematically lower (more than an order of magnitude for most periods) (ibid)
• that application of a lower threshold eliminates a large case of democides from the analysis, but it does not eliminate or even significantly reduce the basic difference between the Rummel's and Harff's data. (W&T, p 10)
• the only plausible explanation is that Rummel includes non-targeted (a.k.a unintentional) mass killings, whereas Harff doesn't (famine, the most deadly events, are not included into politicide database).

However, I haven't seen any indication of even a minimal attempt to verify validity of raw data in Rummel's and Harff's database.

W&T just say: "Ok, these two authors, using two different approaches and two different data sets, came to different conclusions. Leaving the question of validity of their datasets beyond the scope, what may be a reason of such a discrepancy?

Since W&T do not conclude that Rummel'd data are correct, but they assume that (and leave that beyond the scope of their analysis), this article cannot serve neither as an evidence of validity of Rummel's figures nor as a demonstration of its non-validity.

Do you have any objections to this my interpretation? If you do, please, explain. Paul Siebert (talk) 22:29, 9 December 2021 (UTC)

I never said "Rummel's database is at least as popular as Harff's", I pointed out that it is at least as significant as Harff's (PITF), as demonstrated by its inclusion on Oxford University affiliated website[1] If Rummel's dataset was such trash as you claim it to be, why would W&T even bother in writing a 12 page paper on comparing the two datasets? Clearly in the minds of W&T the validity of Rummel's database is taken as granted, and they do say it is a good framework for studying mass killings during the 1900-1987 period. --Nug (talk) 22:18, 10 December 2021 (UTC)

• Ahoy hoy cap'n! We should all try to remember that, thanks for getting that lead reader ready. /s ~ cygnis insignis 15:41, 6 December 2021 (UTC)

@Nug: What is a measure of significance? That it was included in a list of four datasets? What if they just listed all available datasets (which is highly likely)?

W&T do not focus on comparison of datasets, they are trying to understand the difference between conclusions made by Harff and Rummel. And they note that their datasets were dramatically different. They do not say whose dataset is correct, they leave that question beyond the scope.

In connection to that, I do not understand why you are carefully ignoring Dulic, who devoted two articles specifically to the analysis of Rummel's data. He clearly says:

"In any case, it seems the discrepancy between Rummel's figures and what can be reasonably be estimated is simply much too high to be taken lightly. However, these deficiencies cannot be eliminated by mere adjustments if Rummel uses similar sources in other case studies. On the contrary, I argue that the problematic presuppositions regarding the variation principle warrant a thorough revision of the method. Whether such a revision would change Rummel's overall results, particularly when it comes to the ranking of 'democidal' regimes, remains undecided."

This is a clear and severe criticism, which, according to Gleditsch, Rummel failed to adequately address. Why are you ignoring it? And why the opinion of Karlsson, who is notable enough to use his article as a core article for CAHuCR, but who is not notable for his opinion about Rummel? Paul Siebert (talk) 04:39, 11 December 2021 (UTC)

Mann, Marx and "classicide"

I am currently reading reviews on Mann's "The Dark Side of Democracy", and several reviewers mention Mann's name in a context of "the work of the great comparativists of classic sociology - Karl Marx and Max Weber." It is interesting to see how the MKuCR article uses Mann as a reliable source and simultaneously describes another great comparativist, Karl Marx, as a founder of a murderous ideology. What sources are you reading, colleagues? Paul Siebert (talk) 06:33, 11 December 2021 (UTC)

It is possible to believe both that 1) Marx was an influential and incisive thinker 2) the views of historians who criticise his writings as legitimising violence should be included in this article. Vanteloop (talk) 08:33, 11 December 2021 (UTC)

Again, this is a prime example of VERIFY vs. NPOV/WEIGHT. That there are some historians who say that (VERIFY) is not sufficient if they fail WEIGHT; it is the same thing for the VoC estimate — if they do not get secondary coverage, they fail WEIGHT, and if they do not get significant scholarly coverage, they fail NPOV. Without a tertiary source, we must assume that all those views are minority — but are they significant enough? They all appear either one-sided or decontextualized, and generalizations which lacks the full, necessary context (hence POV forks accuses), and fails NPOV in presenting all significant views in proportion to their weight in the literature. Simply asking us to add stuff is not a good approach either, if those sources do not write within the context and structure of the current article. If we cannot identify majority views, and if this article is a collection of minority views, some of which significant, some of it not, some of it fringe, then what is the point? Davide King (talk) 12:38, 11 December 2021 (UTC)

@Vanteloop: Marx, without any doubts, legitimised violence of a certain kind. However, there is a big difference between the statement: "Marx ligitimised violence" and "Marx's theory was a justification of mass killings".

Legitimisation of violence was a mainstream trend during XIX and early XX century, for revolutions (note, mostly bourgeois revolutions and national revolutions, which is essentially the same) created at least a half of all currently existing states in Europe, including Italy, France, Germany, Hungary, etc. Are you going to seriously claim they were not violent? A revolutionary leader or activist was one of role models. Moreover, violence is justified by, e.g. US constitution: its Second Amendment stipulates the right of people to bear arm, for organised militia serves as a "moral check" against both usurpation and the arbitrary use of power. The very spirit of the Second Amendment is to keep people prepared for violent actions when it is needed.

Therefore, the question is not if Marx justified violence (he obviously did), but if he stayed apart from a mainstream trend (which is highly unlikely, for violence was universally seen as a quite acceptable tool both in a political struggle and for solving territorial disputes between states).

The transition between "he theoretically justified violence" and "his theory justified mass killings" is something that does not follow from the sources cited in this article. It is (i) a piece of original research, and (ii) an obvious minor POV-pushing, for I provided several mainstream peer-reviewed publications that clearly say otherwise. They can be found on this talk page and the AfD's talk page. Paul Siebert (talk) 21:06, 11 December 2021 (UTC)

Paul Siebert, there is clearly a dispute between WP:VERIFY and WP:NPOV/WP:WEIGHT in this article, where that the events indeed happened, that events under Stalin's, Mao's, and Pol Pot's regimes universally fit the category of mass killings, that there are sources that mentions Communist regimes (by which they really mean Stalin, Mao, and Pol Pot — that is what Chirot, Jones, Mann, and Valentino really discuss), and that this is sufficient to have an article about it but it is not, in fact, sufficient due to NPOV and WEIGHT concerns. I would say that also in light of the AfD's closure, this is no longer sufficient because, to quote from the closure, 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.

The current structure simply cannot be improved or fixed to no longer have NPOW and WEIGHT issues because the only possible tertiary source (Karlsson) is dismissive towards Courtois and Rummel, does not mention Valentino at all, etc. and the article is a collection of minority views, and we cannot add more sources because majority of sources do not write within the context of this article, do not write in a commonality Communist grouping, and do not engage in a global Communist death toll. By attempting to fix NPOV and WEIGHT, we too engage in OR/SYNTH because the article's structure is reversed and wrong — it is, in itself, OR/SYNTH; if it was not, the result would have been 'Keep.'

As noted by the AfD, sources are the main issues, and their issues are in regards to both WEIGHT and VERIFY (e.g. Chirot, Mann, and Valentino are clearly respected academics but are they discussed fairly and correctly in the article or are they misrepresented? — since we are not citing them to secondary coverage but to their own works, I am not surprised that this is the case. Do they separate Communism in a special category as a new topic? If so, how accepted this approach is and do they represent majority or minority views?). The simplest way to fix this article is to find a topic for which we have tertiary sources (Neumayer et al.), and that is what TFD, others, and I propose. If you want to discuss Communist mass killings, we have Mass killing for that, if you want to have its own article it must only discuss Stalin's, Mao's, and Pol Pot's regimes, as Chirot, Jones, Mann, and Valentino do, or we may have a disambiguation page linking to events themselves. Anything would be better than merging events with theories, merging Communist mass killings with global Communist death toll and theories by non-experts on Communism. In fact, doing so may be disruptive and it amounts to ignoring both the closure and moderator acknowledgment of issues, which are the result of this merging approach, hence no consensus as to whether MKuCR is a suitable encyclopedic topic.

Davide King (talk) 11:57, 11 December 2021 (UTC)

Kotkin

Paul Siebert, please respond in this newly-created thread in order to help make this talk page easier to navigate, as we have moved on to a distinct topic with our discussion of Kotkin.TheTimesAreAChanging (talk) 23:38, 9 December 2021 (UTC)

As soon as we started talking about estimates, take a look at this:

"In 2017, historian Stephen Kotkin wrote in The Wall Street Journal that communist regimes killed at least 65 million people between 1917 and 2017, commenting: "Though communism has killed huge numbers of people intentionally, even more of its victims have died from starvation as a result of its cruel projects of social engineering."

From this quote, a reader may conclude that Communists killed 65 million, and even more died from starvation. That is a direct lie. Kotkin said that 65 million deaths is the demographic estimated. ...in the Soviet Union, China, Mongolia, Eastern Europe, Indochina, Africa, Afghanistan and parts of Latin America—communism has claimed at least 65 million lives, according to the painstaking research of demographers. That means, (i) Kotkinn cites someone other's figures (not his own data), (ii) he speaks about demographic losses. The current text of this "well written and well sourced" article contains tons of misinterpretations or a direct lies. It need a thorough analysis and, probably, major rewrite. However, let's finish with Terminology first. Paul Siebert (talk) 04:01, 6 December 2021 (UTC)

Paul Siebert, with respect, you are not parsing Kotkin's statement accurately, probably because English is not your first language. Kotkin's statement means that, while communists did in fact kill people deliberately, the majority of the 65 million+ deaths they caused were the result of unintended famines. There is no other way to parse Kotkin's statement that would be internally consistent or logical. When Kotkin states that there were at least 65 million victims of communism, some of whom were purposefully killed, and that "even more of its victims have died from starvation," [emphasis added] the reference to "victims" refers back to the larger whole of 65 million. In English, "Though communism has killed huge numbers of people intentionally, even more of its [65 million] victims have died from starvation" does not mean that "communism has killed 65 million people intentionally, and even more victims have died from starvation," nor could anyone fluent in the language reasonably construe it to mean such a thing.TheTimesAreAChanging (talk) 04:46, 6 December 2021 (UTC)

It seems you parse my own statement incorrectly, for I am saying exactly what you say.

The full paragraph says:

"But a century of communism in power—with holdouts even now in Cuba, North Korea and China—has made clear the human cost of a political program bent on overthrowing capitalism. Again and again, the effort to eliminate markets and private property has brought about the deaths of an astounding number of people. Since 1917—in the Soviet Union, China, Mongolia, Eastern Europe, Indochina, Africa, Afghanistan and parts of Latin America—communism has claimed at least 65 million lives, according to the painstaking research of demographers.

Communism’s tools of destruction have included mass deportations, forced labor camps and police-state terror—a model established by Lenin and especially by his successor Joseph Stalin. It has been widely imitated. Though communism has killed huge numbers of people intentionally, even more of its victims have died from starvation as a result of its cruel projects of social engineering"

This paragraph does not allow double interpretation: Kotkin says that, (i) according to demographers, at least 65 million people died prematurely, and (ii) these deaths included such categories as mass deportations, state terror, starvation etc.

Demography is intrinsically incapable of separating death by categories. It is a pure statistical discipline that deals with deaths (from all causes), births, migration etc. The figure of 65 million may mean either "excess mortality" (all premature deaths) or "population losses" (unborn infants, emigration AND premature deaths). Period. I totally rule out an possibility that Kotkin, a professor in Princeton, use "population losses" figures (that would be ridiculous), so he definitely means "excess mortality". And, obviously, he is too educated to refer to demographic data as an estimate for the number of execution deportation etc. Demography cannot provide such information. To claim the opposite would be as ridiculous as to claim that by using a multi-meter it is possible to tell if the electricity in your home was produced at nuclear or gas power plant.

That means, Kotkin's statement is in agreement with other sources: yes, "excess mortality" (all premature deaths) in Communist states (excluding WWII deaths in teh USSR) amounted to 65 million at least, and lion's share of those deaths were Great Chinese famine deaths. That is what he says.

And what we see in the MKuCR article?

• It falsely ascribe to Kotkin a claim that Communists killed 65 million (that is a double lie: that figure was obtained not by Kotkin, but by unnamed demographers, and Kotkin never said 65 million were killed by Communists)
• It claims that famine deaths and other categories are not included in those 65 million (that is a direct lie, for Kotkin says that, according to demographers, Communist rule lead to a loss of 65 million, and that included executions, deportations, starvation etc).

Just an example of one (out of many) falsifications that I found in this "pretty well sourced and well written article". Paul Siebert (talk) 23:37, 7 December 2021 (UTC)

How does one deal with such a proliferation of misstatements as that uttered by Paul Siebert above?

• "It seems you parse my own statement incorrectly, for I am saying exactly what you say." Right, which is that our text (consisting almost entirely of a direct quote from Kotkin) is very clear and cannot reasonably be misunderstood as double-counting the 65 million excess deaths that were caused by communist governments in the twentieth century.
• "It claims that famine deaths and other categories are not included in those 65 million (that is a direct lie ... " Our Wikipedia article, which includes a footnote with the entire excerpt for readers to examine for themselves, very obviously says nothing of the kind (nor has any other editor supported your peculiar interpretation), so your statement could indeed be considered a "direct lie."
• "It falsely ascribe to Kotkin a claim that Communists killed ... " To the contrary, our article merely quotes Kotkin accurately as stating that "communism has claimed at least 65 million lives ... Though communism has killed huge numbers of people intentionally, even more of its victims have died from starvation as a result of its cruel projects of social engineering." [emphasis added] Note that Kotkin's use of "intentionally" implies that communist governments also unintentionally killed people.
• "And, obviously, [Kotkin] is too educated to refer to demographic data as an estimate for the number of execution deportation etc." Based on this comment, it seems like you are suggesting that even though the source (Kotkin) uses demographic data to estimate excess deaths, which he himself refers to as both intentional and unintentional "killings," the source is wrong and should know better (based on his education) that only direct executions (which demography cannot separate from other causes of death) qualify as "killings" (or "mass killings"). If your original research were accepted, this would set a drastic precedent for the rest of the article, but, alas, we cannot use Paul Siebert as a reliable source. We have to stick to what the source says, and cannot change it based on editor opinion alone.
• "It falsely ascribe to Kotkin ... that figure was obtained not by Kotkin, but by unnamed demographers ... " For the record, the current version of the article states: "In 2017, historian Stephen Kotkin wrote in The Wall Street Journal that ... 'communism has claimed at least 65 million lives, according to the painstaking research of demographers.'" Paul Siebert's suggestion that we are somehow misrepresenting or distorting Kotkin, in some way, or that "wrote" isn't the correct form of attribution, and hence that the only solution is to remove Kotkin entirely as a source, seems like an astonishing (and disappointing) case study in bad-faith wikilawyering.TheTimesAreAChanging (talk) 23:29, 9 December 2021 (UTC)

  The source says that according to demographers, 65 million died prematurely. That is what Kotkin says when he cites demographers. He cannot say that Communism killed 65 million, according to demographers, because demographic figures contain no information about causes of deaths. Therefore, there is no other way to interpret Kotkin: 65 million, or maybe more, died prematurely under Communists.

  He continues, that those deaths may be a result of " mass deportations, forced labor camps and police-state terror", but also "even more of its victims have died from starvation as a result of its cruel projects of social engineering".

  In other words Kotkin says that, according to demographers, 65 million people died prematurely, and those deaths were a result of " mass deportations, forced labor camps and police-state terror" but mostly "from starvation as a result of its cruel projects of social engineering". All those categories are included in 65 million.

  The article says a quite different thing. It says that communist killed 65 million, and even more were starved to deaths. That is a lie, and I cannot understand why you are defending that.

  And, finally, citing op-ed by Kotkin as a source is hardly an indication of a good job. If he cites demographers, find them. But demographic figures are hardly relevant to this section, for demographic consequences of the Communist rule is a separate topic (the main consequences was not "excess deaths", but "excess lives": both in USSR and PRC, Communist rule lead to an explosive growth of life expectancy, and, as soon as demography is concerned, that also must be discussed.

  One way or the another, you failed to address my arguments. Paul Siebert (talk) 23:51, 9 December 2021 (UTC)

"He cannot say that Communism killed 65 million, according to demographers ... " Unfortunately for you, Kotkin does use the word "killed," including in reference to those who were unintentionally killed in famines. WP:V and WP:TRUTH apply here. "The article says a quite different thing. It says that communist killed 65 million, and even more were starved to deaths." No, our article does not say that, at all, but if you think that our summary is unclear or misrepresents Kotkin in some way, then you are free to propose drafting improvements to make Kotkin's intended meaning clearer for readers, rather than deleting this sourced content wholesale. "And, finally, citing op-ed by Kotkin as a source is hardly an indication of a good job." This is a new argument that you have not made previously. Yes, op-eds can never be used for factual claims, but they can be used for attributed opinions from subject matter experts. This is partially a matter of editor discretion, but Stephen Kotkin is an academic expert on Soviet communism, so I don't think that the source is obviously WP:UNDUE. Either way, the long-standing content should remain in place until there is a consensus to remove it. I'm not going to engage with you any further on this matter as I feel that you are bludgeoning the discussion (and, no, silence does not mean that I am conceding you are correct, as you've suggested previously), but I certainly welcome other editor's views on our use of Kotkin as a source.TheTimesAreAChanging (talk) 03:14, 10 December 2021 (UTC)

I do not care what exact word he use, but the source of his information is demographic data. Demographic data are the source for "excess deaths" or "population losses", not for "killings", and if Kotkin is inaccurate in terminology (which is normal for non-peer-reviewed publications), that is not an excuse for us. Kotkin may use the word "killed", but you must understand that he meant "excess deaths from all causes", and if you don't understand that, keep in mind WP:CIR. Kotkin cannot cite demographic data in a context of deaths from some specific cause, he definitely meant deaths from all causes.

The article says:

In 2017, historian Stephen Kotkin wrote in The Wall Street Journal that communist regimes killed at least 65 million people between 1917 and 2017, commenting: "Though communism has killed huge numbers of people intentionally, even more of its victims have died from starvation as a result of its cruel projects of social engineering." The words in quotation marks may be understood as a comment on the figure of 65 million, but they can also be understood as an additional remark implying that "even more of its victims" are those who were not included in the number of "65 million killed", "have died from starvation as a result of its cruel projects of social engineering." The latter is the most plausible interpretation of this text. That is a problem, amd this problem is not only my concern. You re-added this ambiguous text, and I expect you to propose an idea how to fix this problem.

I am supporting the idea to discuss demographic consequences of the Communist rule, but Kotkin's op-ed is not the best source for that. Paul Siebert (talk) 03:42, 10 December 2021 (UTC)

"The words in quotation marks may be understood as a comment on the figure of 65 million, but they can also be understood as an additional remark implying that 'even more of its victims' are those who were not included ... " I don't think that interpretation is even remotely plausible, but, if it will resolve your concern, I would support changing the text to: "In 2017, historian Stephen Kotkin wrote in The Wall Street Journal that communist regimes killed at least 65 million people between 1917 and 2017, primarily by starvation."TheTimesAreAChanging (talk) 03:54, 10 December 2021 (UTC)

Do you understand what the words "demographers data" mean? Kotkin's words actually mean: "(i) I claim that Communists killed many people, and many more died from starvation, and (ii) according to demographers data, there were at least 65 million excess deaths in Communist states". That is a precise meaning of their words. Do you understand the difference? Paul Siebert (talk) 04:04, 10 December 2021 (UTC)

Yes, I understand the field of demography and that demographers calculate excess deaths. But I'm trying to stay focused on specific article text improvements, so I really don't have time for lengthy metaphysical digressions.TheTimesAreAChanging (talk) 04:11, 10 December 2021 (UTC)

If we limited ourselves to the most accepted definition of 50,000 killings within five years and discussion of Stalin's, Mao's, and Pol Pot's regimes (while noting that other Communist regimes killed people on a smaller threshold but not discussing them as we do because they are out of scope, unless we finally move on from mass killings), which is, ironically enough, exactly what core sources do (Chirot, Jones, Mann, and Valentino), we would not have such issues about terminology and global death tolls. If this article uses mass killings but then in practice uses it in a very liberal way, with no defined criteria, conflating universally recognized mass killing events for excess deaths and demographic losses, or direct killings not fitting the most accepted criteria, merging mass killings with a global Communist death toll (much broader than mass killing), acting as though there is a common terminology, then we are bordering OR/SYNTH and have no clear topic, or it is problematic. Davide King (talk) 07:08, 10 December 2021 (UTC)

I think Davide King's re-write of the section is a good compromise which represents the source well. The only change I would suggest is re-adding 'at least' before 65 million. Either way, the full quote is there for readers to assess themselves. Vanteloop (talk) 23:41, 10 December 2021 (UTC)

The source used is "Editorial: The legacy of 100 years of communism: 65 million deaths." See what News organizations says, "Editorial commentary, analysis and opinion pieces, whether written by the editors of the publication (editorials) or outside authors (invited op-eds and letters to the editor from notable figures) are reliable primary sources for statements attributed to that editor or author, but are rarely reliable for statements of fact." That means we cannot use this article to report what Kotkin said. We can't use Kotkin's article either, because he is not an expert on mass killings. TFD (talk) 04:06, 11 December 2021 (UTC)

Yes, as I already said, Kotkin's opinion is that "Communists killed many people" (which is non-controversial, but trivial), and that "some demographers calculated the number of excess deaths under Communists, and it is 65 million (which is hardly controversial, because the Great Chinese famine alone caused nearly 40 million deaths). The first part of this statement hardly belongs to the "Estimates" section. The second part refers to a totally different category, namely, "excess deaths". Again, if we decide to discuss demographic consequences of the Communist rule, I can provide many interesting sources, but that is a totally different topic. Paul Siebert (talk) 04:46, 11 December 2021 (UTC)

And, if we want to discuss demographic data, maybe, it would be better to look for demographic sources, not their summary in some op-ed? I am not sure Kotkin is a big expert in demography. Paul Siebert (talk) 04:48, 11 December 2021 (UTC)

You sure you want to go down TFD's suggested path? I suspect that most of the sources you suggest that do not discuss those mass killings in a context of Communism are not experts on mass killings or experts in demography either. --Nug (talk) 10:18, 11 December 2021 (UTC)

That is a circular reasoning. You arbitrary select one term, and you declare that the authors who use this term are experts in the subject, whereas the authors who use other terms are not. However, the subject is not "mass killing", but some concrete event that lead to a loss of human lives (no matter how different authors call that). If Valentino call Great Purge "mass killing", and Mann "classicide", who is a better expert in Great Purge? Paul Siebert (talk) 21:38, 11 December 2021 (UTC)

Playing terminology games again? So a "classicide" isn't a "mass killing" according to you? --Nug (talk) 02:06, 12 December 2021 (UTC)

@Nug: First, in contrast to you, I've read The Dark Side of Democracy by Mann, and I know what exactly he means under "classicide" (a spoiler: Mann's "classicide" is just a small subset of Valentino's "mass killing", and it is a "dark side of democracy" rather than a product of a murderous ideology). Second, I find your habit to stop responding to arguments in one section, and then starting a dispute in another section somewhat suspicious. I am getting impression that your goal is not to find some productive solution, but to filibuster the process of the article's improvement.

By the way, if you are going to respond to my previous arguments (and it would be impolite not do do so), please, ping me, because the talk page becomes so long that it is becoming increasingly difficult to see new posts in the middle of the page. Paul Siebert (talk) 02:55, 12 December 2021 (UTC)

@Nug: Upon reflection, I came to a conclusion that you totally misunderstood my argument. It does not matter if "classicide" and "mass killing" are the same (a spoiler: they aren't, but that is not important in this context). Let's assume they are totally the same thing. However, if you read Mann, you probably noticed that he discusses "classicide" mostly not in a context of Communism. Actually, Valentino also discussed his "mass killings" not in a context of Communism. That does not necessarily make them non-experts. Paul Siebert (talk) 06:20, 12 December 2021 (UTC)

Abysmally terrible sources and citations

Just one peak at this wiki and I see why so many people were calling for its deletion. Death toles are cited using works by people such as Robert Conquest, a former propagandist for the British government's secret Information Research Department, and whose works have been beaten into the dirt by history professors such as Wheatcroft, Suny, Davies and Manning.

Within several areas of the wiki the "Victims of Communism Memorial Foundation is used as a source, despite the fact the organisation has such a terrible history of fudging numbers to make higher death toles that it counts every single death of every person (edited: during the Covid pandemic), no matter which country they are from as a "victim of communism". They're so desperate to artificially inflate deaths that they count people who died in capitalist regimes because they could not afford privatised healthcare as victims of communism. Again a look at their funding shows them to be a propaganda front for the US government and their board has included former war criminals such as George W Bush.

As if the sources couldn't get even worse, the infamous The Black Book of Communism is used as a source, a work so terribly researched that many of its own contributors denounced it, and its own wiki page includes an eight paragraph long list of professional historians attacking the book for its oversimplifications, sloppy research, and quest to achieve as higher and higher death toll numbers.

This is just what I've spotted in an hour but I'm sure it will get worse the closer I look at the citations. I could be wrong but I think we should also take a closer look at the sources for some of the photographs cited. A brief look at the source for one of them, the "Ukrainian American Youth Association", appears to have some dodgy links with the one-time Nazi collaborators of the Organisation of Ukrainian Nationalists, nad includes some glorification of fascists like Stepan Bandera.

This is just terrible, I've rarely seen wiki pages which are this much of a mess. I don't know how we can possibly fix all this because anytime somebody tries you get an activist who just undoes everything. BulgeUwU (talk) 21:18, 5 December 2021 (UTC)

See WP:RSE. Those sources are deemed by the community at large reliable and presumably appear on other Wikipedia articles talking about Communism. If you remove a source that is deemed reliable of course you will be reverted that is why I have not done it myself. The first paragraph must include the title written in bold, and currently, there is a debate ongoing to rename the article to something more appropriate. MarioSuperstar77 (talk) 21:23, 5 December 2021 (UTC)

You might want to make your case over WP:RSN to get those sources removed. That was suggested to another editor a week ago. MarioSuperstar77 (talk) 21:28, 5 December 2021 (UTC)

Books published by Oxford University Press and Harvard University Press are considered reliable, gold-standard sources on Wikipedia. The misleading presentation above has no basis in Wikipedia policy.TheTimesAreAChanging (talk) 21:36, 5 December 2021 (UTC)

Is the Black Book of Communism a reliable, gold-standard source, yes or no? Yes or no? That specific book. Yes or no? DublinDilettante (talk) 21:56, 5 December 2021 (UTC)

Conquest's The Great Terror is certainly a gold-standard reliable source, and, yes, at least Werth's and Margolin's contributions to The Black Book of Communism without question meet that standard as well.TheTimesAreAChanging (talk) 22:52, 5 December 2021 (UTC)

There’s a very old joke about a diffident curate who gets served a rotten egg at the vicar’s table, and when the vicar points it out, the curate says, “oh no, sir, I assure you, parts of it are excellent!” It’s meant to be a joke and not a guide for editing Wikipedia. A crank publication edited by a crank for the purposes of right-wing crankery is still that, even if it came ex libris God himself. DublinDilettante (talk) 00:02, 6 December 2021 (UTC)

Take a look at the RSN discussion about the Black Book. There is no binary answer to your question. Paul Siebert (talk) 03:10, 6 December 2021 (UTC)

Books by acclaimed university presses are generally reliable, but no publisher is perfect. Even Homer nods; I can think of at least one philosophy book from Oxford UP that is absolute dreck. XOR'easter (talk) 21:58, 5 December 2021 (UTC)

In response to a recent edit by BulgeUwU, I have reverted the change partially removing the reference to the Victims of Communism Memorial Foundation. I think there's might be enough disagreement here that we could have a discussion about that source alone. IDontHaveAnAccountYet (talk) 23:22, 5 December 2021 (UTC)

I note that user BulgeUwU, who instigated this valuable section and accurately characterised the nature of this article, has now been blocked from editing, apparently due to concerns over their user name (which means nothing to me). My own account was blocked for 31 hours due to a user following me from this page to edit war on another. There is absolutely no way any progress can be made, or any credible RfC undertaken, in this total absence of good faith and with this rampant degree of disruption. DublinDilettante (talk) 17:17, 6 December 2021 (UTC)

I used to be "BulgeUwU, and I frequently edited on wikipedia for over a year with no problem. Then the very moment I made edits to this wikipedia page somebody falsly flagged my username as being inappropriate, which then led to me being blocked and forced to change it, and even continued to block me even days after I requested a change in username. I'm certain somebody who didn't like my perspective on this wiki page (which I believe should be deleted) decided a quick way to shut me up for a week would be to falsly flag my account. All it's really done is make me more determined to fix the countless problems this article faces. The History Wizard of Cambridge (talk) 02:46, 10 December 2021 (UTC)

If that is true that you edited on Wikipedia for an entire year with that username, then that is surprising, but also possibly explainable. When I saw your username I thought it was weird, but I did not think it was offensive enough to warrant reporting you, so because I deemed that was ultimately trivial and demanded more effort than just simply going on with my business I did not report you and I presume you survived 1 year with that username because you happened to come across editors like me who err at your username, but would rather not bother with you because there was inherently nothing disruptive about your username. Additionally, I know for a fact that people on Twitter have had similar usernames to yours, but much more objectionable. I also presume it was Captain Eek who reported you based on what they said on your talkpage, and in fact, they participate in the discussion below, therefore, you are completely correct when you say that your participation on this talkpage is what caused your username to finally be reported to the admins. MarioSuperstar77 (talk) 12:59, 12 December 2021 (UTC)

Further nomination should be considered disruptive behavior

I saw the media coverage of this article's deletion discussion.

https://www.telegraph.co.uk/news/2021/11/27/wikipedia-may-delete-entry-mass-killings-communism-due-claims/

https://www.foxnews.com/politics/wikipedia-page-mass-killings-communist-regimes-deletion-bias

https://www.mrctv.org/blog/mass-killings-under-communist-regimes-wikipedia-page-being-considered-deletion

https://www.rebelnews.com/wikipedia_wants_to_delete_its_mass_killings_under_communist_regimes_article

Hungarian- https://precedens.mandiner.hu/cikk/20211130_wikipedia_kommunizmus_cenzura_cancel

Slovak- https://www.postoj.sk/93610/wikipedia-chce-vymazat-stranku-o-vrazdach-spachanych-komunistickymi-rezimami

With time, an article's notability is likely to increase, not decrease. If anyone makes another fifth nomination, it should be closed as speedy keep. And if the nominator includes anyone from the past nominators or past delete voters in the last four discussions, then he should be blocked for one month. Knight Skywalker (talk) 06:28, 8 December 2021 (UTC)

Thanks for letting us know you think extremist right-wing media should have a veto over Wikipedia content. This is obvious trolling from the user above, but if one thing has been made clear in the “debate”, it’s that a future AfD is inevitable, and should happen sooner rather than later. DublinDilettante (talk) 09:47, 8 December 2021 (UTC)

I don't know who is an extreme wing of right or left. I know only about Fox News. And I did a google news search of, Mass killings under communist regimes. These websites showed in search results. If you have doubt, then search recent media coverage.

Adding :

https://www.washingtonpost.com/news/volokh-conspiracy/wp/2017/11/07/lessons-from-a-century-of-communism/

https://www.levandehistoria.se/sites/default/files/wysiwyg_media/crimes-against-humanity-under-communist-regimes-research-review.pdf

https://www.degruyter.com/document/doi/10.7591/9780801467172-006/html

https://www.ceeol.com/search/article-detail?id=86279

https://www.cambridge.org/core/books/abs/dark-side-of-democracy/communist-cleansing-stalin-mao-pol-pot/5BC0D5F39EF9C5A1F6BA171572F419E9

https://www.jstor.org/stable/25654524

Knight Skywalker (talk) 11:09, 8 December 2021 (UTC)

The AfD's media coverage listed is all from one side (several of which even very far to the right), and only The Telegraph can be considered generally reliable. The closure was very clear that this was not an issue of notability but of NPOV/OR/SYNTH. If you were right, it would have resulted in a speedy 'Keep' closure already but it did not.

Google is not a good way to make such a research (WP:GSNR), Google Scholar is much better for this, especially because "Mass killings under communist regimes" is considered to be a descriptive title. You should be searching "communist mass killings" (the name also used in the AfD closure) on Google Scholar, and the painted picture is not as clear, with many sources being about mass killings of communists (Indonesia). By the way, have you noticed that two scholarly sources you linked are about genocide and mass killings in general? Of course, Communist regimes are discussed because the events indeed did happen and many, many people have died — but does that warrant Communism as a separate category or new topic, rather than expansion of Mass killing? Mann's main thesis is that democratic transformation can result in genocide, as it has happened in Rwanda, hence the title, and many similar general works have chapters for capitalist regimes and other regime types, but that by itself does not mean that a new topic is created or it is to be considered a separate or special category by majority of scholarly sources. Davide King (talk) 12:45, 8 December 2021 (UTC) [Edited to copy edit and fix typos] Davide King (talk) 13:26, 8 December 2021 (UTC)

Adding to Davide King's comment. The result of the AfD was 'no consensus to delete'. That is a significant distinction to 'keep' that should be noted. Of course using common sense we should not re-nominate this for a while following that exhausting AfD but renomination is not off the table in the future if editors feel the issues with the article have not been resolved Vanteloop (talk) 13:02, 8 December 2021 (UTC)

A newb mistake dear, the spectrum is Consensus => delete -> keep => No Consensus. ~ cygnis insignis 13:10, 8 December 2021 (UTC)

@Knight Skywalker: The Empire did nothing wrong, how dare you name check that murderous rebel! ~ cygnis insignis 13:04, 8 December 2021 (UTC)

Media coverage was not about the subject of this article (whatever that might be), media coverage was about the AfD process, two different things. FWIW, no consensus = status quo ante. Regards, --Goldsztajn (talk) 02:53, 9 December 2021 (UTC)

Fox News needs no introduction and "rebel media" is infamous for giving fascist activists a voice and a paycheck. No wonder you're so deep into anti-communist propaganda if this is where you get your news. The History Wizard of Cambridge (talk) 18:44, 11 December 2021 (UTC)

I mostly focus on BBC, CNN, NYTimes. And when you say anti-communist propaganda; there is no need for propaganda, they have done enough killings; and as we can see China, a developed country, trying to dominate Hong Kong, Taiwan, Manipur. Knight Skywalker (talk) 14:36, 12 December 2021 (UTC)

Agree. In fact a study showed that people who watch Fox News are less informed than people who read or watch no news at all. TFD (talk) 19:59, 11 December 2021 (UTC)

Blog source in the estimates section

I mentioned this above, but since it keeps getting brought up in other contexts I feel it deserves its own section. The paragraph I tagged here is cited solely to a blog; that is obviously not an WP:RS or an acceptable source on an article of this nature, especially given the wide range of academic sources available. It needs to be removed entirely - if people think the figure is important, perhaps a better source can be found to replace it (though of course we would have to adjust the wording to whatever the new source said), but obviously someone's personal opinions and speculations on a blog, with no indication of any process to ensure fact-checking and accuracy, don't qualify as meaningful estimates or as a valid way to assess the literature. --Aquillion (talk) 23:29, 7 December 2021 (UTC)

Actually, the whole section must be rewritten. It contains several sources that were found unreliable during previous RSN discussions, it misinterpets other sources, it takes the data out of context, etc. However, since I resumed my participation in the DRN discussion, I am not going to edit the article for a while. Paul Siebert (talk) 23:42, 7 December 2021 (UTC)

It does, but this one seems like low-hanging fruit to me. Determining the best academic sources to use and how to use them is difficult and will require a lot of time and discussion; determining that we should not use a blog seems like it ought to be easy. --Aquillion (talk) 23:46, 7 December 2021 (UTC)

Actually, I recall there was a discussion about White at RSN, and a consensus was "not reliable" I am not sure why it is still in the article. The same story is Rummel. If I remember correctly, WOC was also discussed at RSN a couple of years ago. Yes, it makes sense to start purging it, but I am not ready for that work. Paul Siebert (talk) 00:01, 8 December 2021 (UTC)

By the way, the current state of the Estimate section is a brilliant demonstration of a desperate lack of good sources for this topic... Paul Siebert (talk) 00:03, 8 December 2021 (UTC)

A couple of cites in a section are dubious, hence the whole section is bad and therefore demonstrates the whole topic lacks good sources. Nice fallacy of composition argument Paul. --Nug (talk) 00:39, 8 December 2021 (UTC)

@Nug: As I already explained elsewhere, the section is bad not because a couple cites are dubious. It is bad because (i) ALL sources are either dubious (like Rummel) or misinterpreted (like Kotkin), and (ii) the very concept of "estimates of a global death toll of Communism" is highly poloticised, and it cannot be represented as a neutral scholarly topic: it should be discussed only in a context of its meaning, support and criticism. Paul Siebert (talk) 00:56, 8 December 2021 (UTC)

@Aquillion: Look at This and that, it may be helpful. Paul Siebert (talk) 01:13, 8 December 2021 (UTC)

• I'd agree entirely with Aquillon; that paragraph needs to go. It's a serious due weight problem. Vanamonde (Talk) 01:06, 8 December 2021 (UTC)

  Not exactly. The figure of 100+ million can be frequently found in many popular web sites. We should not remove it, we should explain who made these estimates, what these estimates mean, what concrete political idea is conveyed by the author(s), who agrees with that, who ctiticises that concept, and why. If we just remove the whole section, Wikipedia may be accused of a "leftist bias" (we have seen it during the recent AfD). Paul Siebert (talk) 01:18, 8 December 2021 (UTC)

And? The claim that Jews run the world or that the earth is flat can be found on "many websites". Let people accuse us of "leftist bias", the info is undue and the article isn't about these claims. If the scope changes we can discuss them properly, but right now there is no need. BSMRD (talk) 02:39, 8 December 2021 (UTC)

Your analogy is wrong. A Jewish conspiracy is not discussed at WSJ or NYT, and is not published by Harvard University Press. The information that Communists killed 100+ is unreliable, but the claim is very popular. Therefore, instead of just deleting it, we need to discuss it, and to explain why Courtois made this claim, how this figure was obtained (including explanations of all manipulations), is this very popular claim supported by other authors, and who criticises it and why, and who supports it. I think that would be a nice and detailed story that will explain all nuances of that minority viewpoint. Paul Siebert (talk) 03:14, 8 December 2021 (UTC)

Paul Siebert is quite correct in all respects and we should keep the section and keep the various claims (including the one made by the Victims of Communism Memorial Foundation) and explain the various claims and their origins exactly as Paul Siebert rather cogently explains.XavierItzm (talk) 04:12, 8 December 2021 (UTC)

That risks inserting a WP:SYNTHy POV narrative as to the origin of the 100+ million claim. The mind boggles at how upset some people get over this, even if the WP:True number was half or even a quarter of 100 million, does it really put communist regimes in a better light? — Preceding unsigned comment added by Nug (talk • contribs) 04:27, 8 December 2021 (UTC)

I think whether a claim is merely detached from reality or detached so far from reality as to belong to another universe is an important consideration for something that aspires to be an encyclopaedia (and sign your edits, we all know who this is). DublinDilettante (talk) 09:53, 8 December 2021 (UTC)

XavierItzm totally misinterpreted my words. I didn't mean to keep all those ridiculous figures. In particular, WOC Memorial is unreliable source per RSN, and should be removed without any reservations. Paul Siebert (talk) 18:16, 8 December 2021 (UTC)

Are you truly saying that the article shouldn't cite a source because the source might be biased? Consider the following case: the current Wikipedia article on the Lebensraum extensively quotes the Zweites Buch (i.e., Mein Kampf part II); will people come out of the woodwork to say it should not be cited, because it is evidently biased because it was penned by Adolf himself?
Not to mention that by definition all sources contain their own biases; this is why we survey, present them, and attribute them as required. Apologies for the reductio ad hitlerium.XavierItzm (talk) 05:37, 9 December 2021 (UTC)

Sorry but how did you come to the conclusion that Siebert said that? Siebert has not mentioned biased at all but rather unreliable, which is a different thing — a biased source can still be generally reliable. The issue is of weight — if there is no secondary coverage in reliable sources citing, discussing, or mentioning the Dissident article, it is undue. That is a problem with the whole article because it is a bunch of "he said, she said", and rather than cite it to secondary sources, which may contextualize what the authors said (e.g. if they are writing within the context of general mass killings or if they are discussing Communism as a separate or special category that warrants a new topic, which is the point of our dispute), we cite them to the author themselves; this is not necessarily an issue because we are not disputing their reliability (e.g. they are academics) but whether they are correctly paraphrased and contextualized. If we cannot distinguish from majority, minority, and fringe views, how can we write a NPOV article, which explicitily requires that we give weight to each views proportionally to their representation in reliable sources? Davide King (talk) 07:17, 9 December 2021 (UTC)

The Victims of Communism Memorial Foundation is clearly a reliable source for its own views. Certainly the data point presented by such a prominent organisation merits inclusion, with proper attribution and contextualization, of course. So, fear not your ability to distinguish majority and minority views: let policy be your guide.XavierItzm (talk) 16:20, 9 December 2021 (UTC)

Firstly, your comment can be summed up as a meaningless word salad that does not disprove any of the points made here, nor does this rebuke the over all belief that the source is partisan. Secondly, the consensus is very clearly tipping in your defavor. In the page linked above, not a single person marked this source as being reliable, and nobody, besides you, supports the inclusion of this source on this talkpage right here. Additionally, I would suggest not to speak in such a snobbish condescending tone, this does nothing to help your argument. MarioSuperstar77 (talk) 19:43, 9 December 2021 (UTC)

@XavierItzm: You misinterpreted my words twice. I think, the best apology would be if you carefully read the text below, and answer if you agree or disagree. If you disagree, then provide a detailed answer. If you agree, please, do not repeat your arguments again in any discussion with me or other users.

I never said biased sources could not be used. That can, and they should be used if their bias is properly explained. However, these sources must be reliable. Second, as Rummel correctly noted (that is a part of his approach, and, in that aspect, he is totally right), the sources must be independent on each other. That means, each of them must present figures that were obtained independently: if a source "A" says that 100 million were killed, and the source "B" says 100 million were killed, we must be sure "A" does not just cite "B", otherwise, only "A" or "B" should be used. Third, all sources are supposed to discuss the same topic, and that topic must be well defined and non-controversial. If it is poorly defined and/or controversial, all of that must be clearly and exhaustively explained.

I can give you an example of a good presentation of that type data, the GULAG article: some modern consensus figures of GULAG mortality are presented, and historical estimates are given in a separate section to show how human knowledge of the scale of that tragedy was developing (actually, the "Historical estimates.." table was written mostly by me). As a result, a reader can clearly see, what was a GULAG death toll according to modern consensus figures, and, simultaneously, they can understand why other figures frequently appear in mass-media (usually, they come from sole old source).

Note, almost every source cited in this and that sections is a research article or a book that is specifically devoted to the same, well defined topic (i.e. a number of people who died in camps or few years after their release). Each source provides independently obtained data, and this is not just a repetition of the same information taken elsewhere (actually, not every source, I've just noticed some fresh additions, so the article may need some cleanup, but it will not lead to any significant changes of the main idea).

And what do we have here?

• Old and new data are mixed together. That confuses a reader, who wants to know real facts, not a history of the topic.
• Most sources are not peer-reviewed publications or university level books, but blogs, newspaper op-eds, or self-published sources
• It is almost no evidence that these sources are independent (i.e., they do not provide figures taken from each other or from other sources)
• The very topic is poorly defined, and it is not clear what some concrete author means (most importantly, some authors speak about executions and similar coercive deaths, other authors include famine, others speak about "demographic losses" or "excess mortality"). All of that are different categories and different topics.
• And, finally, the most important thing: this section totally ignores the fact that the very idea to come up with some aggregate estimate of "Communusm death toll" is extremely controversial, and majority of authors either ignore it or openly criticize. That means, that the sources presented in this section is an extremely biased sample that expresses a minority POV.

I remind you (and all other users), that a discussion of those problems cannot be superficial. Keep in mind that WP:CIR, so superficial objections and unsubstantiated refusal to accept facts and sources may elevate to conduct issues. Please, approach this discussion seriously.

Now let's take a look at concrete sources:

• Culbertson. A desperately outdated source that was not cited and has no review on Jstor.org. What is the reason to add this garbage?
• Lenczowski. The article by John Lenczowski in The Christian Science Monitor about the events in Nicaragua. Just think: a US Department of State person, a professional propagandist, published an op-ed in a conservative newspaper 36 years ago, were he wrote about the events in Nicaragua. What is a probability that this article presents his own data? What is a reason to believe a reader may be interested to know an opinion of the US propagandist about the events that he mentions just in passing? I see no idea who added this ideologically motivated and outdated garbage, which cites some figures from unknown sources>
• Brzezinski. Another former US official and propagandist. In contrast to the two previous sources, his book is well cited, but the death toll is not a focus of his study, for the book discusses a totally different subject. A snippet view provided as a reference does not allow us to understand the context, and we even don't know whom Brzezinski cites. However, it is obvious that 60,000,000 is not his own figure, he took it from other sources, and those sources are "desperately outdated", and we even don't know what category of deaths they discuss. Conclusion: remove this garbage.
• US Congress. I do not understand why US official document is used as a source of data. Maybe I missed something, and Wikipedia is now affiliated with US government?
• Rummel. As I explained elsewhere, Rummel did no his own research. His approach is as follows: to collect ALL publications that discuss all cases of mass killings and obtain an average. As Dulich noted, Rummel was not doing any source criticism, so he treated all sources as equally trustworthy. Furthermore, his statistical approach (when higher figure have no upper limit, whereas lower figure have a natural limit, a zero) inevitably leaded to inflated figures, because too high data points are not compensated by too low ones. That was recognized by Harff and by Karlsson. Obviously, Rummel's estimates work fine when high quality data sets with a normal distribution are available, and that is why he provided pretty good figures for Cambodia. However, the major part of his global Communist figure is USSR, and the data in his "Lethal politics" are awful. He uses mostly indirect estimates obtained in the West by the end of 1970s, and he even never tried to update his data set. Currently, much better data are available, thus, Gulag mortality is, according to scholarly consensus, c 1.5 million, whereas Rummel says it was tens of million. Rummel sources for USSR are worse than the historical sources cited here. All other scholars, even Conquest, even Rosefielde, recognised their earlier errors and reconsidered their estimates, but Rummel refused to do so. That his mistake undermines any credibility of his figure of total deaths Rummel total figures are garbage, and they represent only historical interest.
• Courtois. I (and many other users) already explained that numerous sources criticize Courtois for the very idea to combine all deaths and attribute them to Communism. We already had several RSN discussion, and this source was not recognised as reliable for this concrete figure. Why Courtois is still here?
• Valentino. If you read his book (I did), you know that the total figure is not a focus of his study. He just cited data presented by others (mostly by Rummel). He is not an expert in that, and he never pretended to be an expert. Incidentally, the core conclusion of his work is that there is no significant connection between a regime type and mass killings. He never did any research of that type, and these data do not belong to him. Valentino was taken out of context, he is not an independent source, there is absolutely no need to present the figure from his book, which discussed a totally different aspect of mass killings
• Matthew White. Who he is? Why he is here? Maybe, the next step would be to use Guiness book as a source? This is obviously a tertiary source of a questionable quality. By using White, we discredit this article and demonstrate a desperate lack of good sources on that topic.
• Bellamy. I looked through this essay, it is a well written and a serious source with a large number of inline citations. The problem is that the figure of 6.7 and 15.5 million has no inline citation. It is clear from the context that Bellamy took it from some other source (the essay's topic is not figures), but, due to the lack of inline citation, we cannot figure out the source. Bellamy, although it is a high quality source in general, cannot be seen as an independent source for those figures, because the author does not explain where he took those numbers.
• Julia Strauss. Not bad. The problem is, however, that Strauss provides no total figures. She gave two figures for two countries. If we want to include data by country experts, what is a reason to include only Strauss, why tens of other authors are ignored? But, if we include other authors, that may be SYNTH, because most country experts do not support the very idea that different events in different Communist states had significant similarities, and importantly, they even do not consider most of them as "mass killings". We should either include all most important country experts (which requires a total rewrite), or remove Strauss.
• VOC is considered unreliable per a recent RSN discussion (I already provided a link in this section). Why it is still here?
• Kotkin. As I already explained above, we have a direct falsification here. Kotkin never wrote that Communists killed 65 million, he said that, according to some (unnamed) demographers, population losses (or excess deaths from all causes, it is not completely clear: his words allow a double interpretation) in Communist states amounted to 65 million. And they included, among other categories, execution, famine, etc. In contrast, the Wikipedia article falsely says that, per Kotkin (a first lie: that is not Kotkin's figure, he obtained it from some unknown source), Communists killed 65 million, and even more were starved to death (a second lie: a figure of 65 million includes all categories, maybe, even a birth deficit)

In summary, I don't know who wrote that (I've just finished my analysis, I didn't look at that section before), we are dealing either with a direct attempt to introduce a false information (a conduct issue), or there is a WP:CIR problem. There is no sources in this section that are both of a reasonable quality and are used in a correct way. That demonstrates a desperate lack of high quality sources for this topic.

Believe me, I know what I am talking about, because I am speaking based on my experience that I obtained during my work with GULAG article, which now has (partially due to my efforts) a large number of high quality sources.

I propose to delete this piece of ****. Currently, I am participating in the DRN, and I am not going to edit the article until the discussion ends. However, this section is a shame, it needs a complete rewrite (which may depend in the result of the discussion about the article's topic). In any event, it cannot exist in this form, that discredits Wikipedia. Paul Siebert (talk) 21:07, 9 December 2021 (UTC)

I removed a blatantly false statement, I am not going to edit the rest for a while.--Paul Siebert (talk) 21:16, 9 December 2021 (UTC)

I'll review each of your points and respond in due course. This fixation on numerical precision is misplaced, as Harff writes: "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. ... Over time I have become more critical of country experts who challenge systematic empirical studies. Case studies are scarce, of dubious accuracy, or non-existent for some episodes of mass death, and estimates vary greatly.". --Nug (talk) 22:21, 9 December 2021 (UTC)

Thank you for quoting the source that I quoted on this talk page for the first time several years ago (and which is being cited by others since then). However, that demonstrates my point: what is the reason to cite a scholar who had never been interested in obtaining precise numbers, and who is known to provide inflated figures? And, whose estimates for the USSR (61,911,000 million) is totally inconsistent with modern data? By the way, when I was a student, my professors told me that a too large amount of significant figures is the first sign of a poor education. Paul Siebert (talk) 22:38, 9 December 2021 (UTC)

WRT "Over time I have become more critical of country experts who challenge systematic empirical studies", the problem is not only in what Harff think about country experts, but what country experts think about her. And guess what they think? Nothing. They just ignore her "theorising". Paul Siebert (talk) 22:46, 9 December 2021 (UTC)

I think the discussion has gone on long enough, I'm taking the initiative of removing the controversial text from the article and its associated citations. Just in case, I will not be touching the museum and memorial section of the article, I will exclusively change the contentious information marked as WP:UNDUE. MarioSuperstar77 (talk) 19:54, 9 December 2021 (UTC)

It should be pointed out that a discussion is ongoing and that you have jumped the gun. XavierItzm (talk) 17:55, 10 December 2021 (UTC)

@XavierItzm: It seems you have a time to comment on another user under a pretext that "a discussion is ongoing", but you ignore my responce to you, where I explain what is fundamentally wrong with all sources. In connection to that, your reference to some "ongoing discussion" looks somewhat odd.

In addition, what "ongoing discussion" is possible about the source that was recognized unreliable at RSN? I saw no fresh arguments, except "It is reliable for VOC position". However, that is not an argument, because it is equally applicable to almost every source. I am still waiting for your comments on my post. Paul Siebert (talk) 20:18, 10 December 2021 (UTC)

@Paul Siebert:, please observe WP:NODEADLINE. All good things come to those who wait. On the other hand, due to some people jumping the gun, it was important to highlight that the discussion is ongoing and that MarioSuperstar77 had no right to delete the ref being discussed here on TP. Cheerio, XavierItzm (talk) 14:46, 11 December 2021 (UTC)

@XavierItzm: NODEADLINE is just an essay. It is somewhat disrespectful to ignore responses from your opponents. I expect you either to answer to my full responce, or at least to acknowledge that you have read and understood it. Paul Siebert (talk) 04:25, 13 December 2021 (UTC)

• I reviewed the RSN dicussion regarding the source and it seems the main reason for saying it is unreliable is due to the source's bias, but per WP:BIASEDSOURCES, that isn't a valid reason. Clearly they are a reliable source for their own opinion, regardless of how biased some may think it is, and given that their opinion is clearly attributed to them I see not issue here. --Nug (talk) 20:54, 10 December 2021 (UTC)

I hadn't even noticed the RSN discussion, which is near-unanimously unreliable. Most of the people there are questioning its reliability (it's a blog, after all), not its bias; but more generally it's a clear consensus of unreliable. If you disagree, feel free to re-open the discussion there, but I'm skeptical you'll convince anyone you can cite a blog. --Aquillion (talk) 21:05, 10 December 2021 (UTC)

WRT "they are a reliable source for their own opinion" - that is unbeatable argument, which is applicable to almost every source. The question is what makes WoC blog notable enough to include it in this section? Are they a scientific or scholarly institution? Do they have some broadly recognised expertise that make their opinion notable? Paul Siebert (talk) 21:09, 10 December 2021 (UTC)

victimsofcommunism.org is a bit more than a blog, it is a website of a educational and research foundation with an academic council with a program of research. Just because people think it is anti-communist isn't a valid reason to exclude it per WP:BIASEDSOURCES. --Nug (talk) 21:41, 10 December 2021 (UTC)

This deserves a serious analysis. I will think about that, and I expect that meanwhile you will respond to my analysis of W&T. The ball is on your side. Paul Siebert (talk) 22:01, 10 December 2021 (UTC)

The "Victims of Communism Memorial Foundation" should be avoided at all costs. Firstly they were founded and funded by the US government so they're far from neutral. Secondly their staff includes a slew of war criminals including George W. Bush, and a bunch of politicians who supported the Contras. Thirdly they are so desperate to boost the number of people killed by communism that they claim that every single global death from Covid-19 as victims of communism. So even people who live under capitalism and died because they could not afford healthcare under a privatised medical system, are counted by the Foundation as killed by communists. It's very simply a government funded propaganda outlet. The History Wizard of Cambridge (talk) 22:19, 10 December 2021 (UTC)

Yeah, no. Just because you think someone is a "war criminal" doesn't mean they are. Whether they were founded by US Gov or not is irrelevant. It's not at all surprising that it includes some people who supported opponents of a communist government. I have no idea what Covid has to do with any of this. Volunteer Marek 21:59, 11 December 2021 (UTC)

@Nug: From one hand, VoC really looks like a serious organisation, and your argument sounds seriously. However, there is another aspect. If read Rummel, one aspect of his estimates (which is methodologically correct) is that each source must be independent: if two articles or books take their data from the same source, only one should be used. Currently, we have data from (i) VoC (founded by the US Congress and supervised by Brzezinski), (ii) Lenczowski (a US official), (iii) Brzezinski again, and all those sources are affiliated with the federal US authorities. Don't you find that the point of view of one singe state is overestimated in the article? And is Wikipedia supposed to represent an opinion of any state at all? In addition, I have a strong reason to suspect these sources are not independent, because it is not clear where they took their data. I think, we should combine these three references together, and to say that US officials and some institutions affiliated with the US federal authorities maintain that the number of victims of Communism amounts to 100 million. This short statement should replace the existing text, and it should be supplemented by these three references (VoC should be re-added to the article). If there will be no reasonable criticism, I am going to do that in next few days. Paul Siebert (talk) 04:05, 11 December 2021 (UTC)

In addition, Rudolph Rummel was one of the members of the VoC advisory council. Therefore, it is quite likely that the number of 100 million may reflect Rummel's views. How many sources that cite Rummel's figures are we going to cite as ostensibly "independent" sources? Paul Siebert (talk) 04:54, 11 December 2021 (UTC)

First you say nobody uses Rummel's data, now you say everyone is using Rummel's data. Nice try, but no, the VoC only cites Rummel for Vietnam and North Korea (and the lower estimate for North Korea at that), they cite other scholars for the other communist regimes. --Nug (talk) 11:08, 11 December 2021 (UTC)

Don't twist my words. The claim that Rummel data are not used by other scholars belongs not to me, but to W&T, and you are perfectly aware of that. And, I don't say everyone uses Rummel's data: I says that many sources affiliated with US federal administration use Rummel. And, you haven't checked the data: thus, Panin's figures (57,000,000 to 69,500,000) are cited by Rummel's "Lethal politics", and Rummel's own estimate (61,911,000) is close to the average of Panin's high and low estimate. Clearly, Rummel's opinion is overrepresented in the VoC data. Anyway, VoC cites the Blacl Book, Rummel, and a couple of similar questionable sources, which makes it non-independent: we de facto cite the same figures twice, thereby giving an undue weight to highly questionable, politicised and outdated sources, and to the position of the US government.

I am going to rewrite this section, leaving these sources in, but combining them together (for they are by no means independent) and explaining their affiliation, as well as political implications of those statements and those figures. Paul Siebert (talk) 21:52, 11 December 2021 (UTC)

No, your conclusions are pure synthesis, you have no evidence that these other authors have used Rummel's numbers and they are not independent and combining them together would be considered disruptive. --Nug (talk) 01:57, 12 December 2021 (UTC)

@Nug: My conclusion is a pure analysis (not a synthesis). I can give a more detailed responce, but I would prefer to do that after you review and answer to my previous posts (to which you promised to give a detailed responce). Paul Siebert (talk) 03:36, 12 December 2021 (UTC)

• I noticed VoC became a subject of an edit war. That is fruitless, and we need to stop it.

Let me explain what, imho, we should do:

• VoC, Rummel, Brzezinsky etc should stay (I mean the reference stay).
• They should be combined together, and its affiliation with the US government should be explained. Something like that Starting from 1980s US state officials and institutions affiliated with the US administration maintain that Communism killed XX to YY people(ref to VoC and other sources).
• With regard to other sources, some of them should be removed (e.g., an obscure 1970s source, teh first line) and a collection of figures made by an amateur scientist);
• A proper context should be provided for Courtois's estimate (I already presented sources, other users also provided several excellent sources that are relevant to that topic).
• A discussion of demography (Kotkin) should be moved to a separate section, for which I can find modern sources of high quality.

If someone disagrees, please, provide rational arguments. If someone wants to join this work, please, let me know. Paul Siebert (talk) 22:31, 11 December 2021 (UTC)

I disagree. There is no basis for combining VoC, Rummel, Brzezinsky etc together, you have no evidence they are not independent, only you conjecture which isn't RS but in fact SYNTH. --Nug (talk) 02:02, 12 December 2021 (UTC)

@Nug: You promised to review my analysis of the sources, but you still haven't done so. It seems you even haven't started, because that is an only possible explanation of this your statement (leaving a bad faith action as a less realistic possibility)

• John Lenczowski was a US Department of State employee;
• Zbigniew Brzezinski was Carter's National Security Advisor (US government employee)
• Victims of Communism Memorial Foundation was established by the Act of Congress, and is currently lead by Andrew Bremberg, a former director of the Domestic Policy Council in the Trump's administration.
• Rudolph Rummel was a member of the advisory council of the Victims of Communism Memorial Foundation.

Of course, all those sources are totally independent on each other, and they are not affiliated with any concrete state.

Or, maybe, I am wrong?

Please, don't forget to ping me if you respond, otherwise I may not notice your post. Paul Siebert (talk) 03:10, 12 December 2021 (UTC)

• This seems to be a continuation of a tag team war. I propose @MarioSuperstar77: to self-revert and join a talk page discussion. The section needs a major revision, and this concrete source is not the major problem. The AE report was submitted against one user who removed this content twice in less than 24 hours, but I am going to inform admins about other users who who de factoare participating in a tag-team campaign over this source. This is a bad tactics, that is a non-productive tactics, and we must avoid it. MarioSuperstar77, please, self-revert, and thank you in advance for your cooperation. Although this source is biased, and it is definitely not independent, we may probably keep it, but explain that this, as well as several other sources, reflects a position of the US authorities on that matter. In that sense, it makes sense to keep it, but place into an appropriate context. All of that is a subject of a talk page discussion, not of a tag-team war.--Paul Siebert (talk) 15:46, 12 December 2021 (UTC)

  It is a bad source.--Horace Snow (talk) 10:03, 13 December 2021 (UTC)

Source Analysis Subpage

This statement largely repeats what I have said at DRNMKUCR. On the one hand, I am pessimistic about the possibility that source analysis will result in agreement about how to improve the article on Mass killings under communist regimes. On the other hand, it seems that some editors either are pessimistic about an RFC on restructuring the article without source analysis, or are optimistic that further source analysis will make it easier to go forward. So I have opened another subpage for source analysis. It is at Wikipedia:Dispute resolution noticeboard/MKUCR Source Analysis. It starts with its own set of ground rules, which largely repeat what I have said at DRN and at DRNMKUCR. If I say to be civil and concise repeatedly, it is because there is a lack of conciseness. But you now have a page for source analysis. Robert McClenon (talk) 01:24, 15 December 2021 (UTC)

The position of one party at DRNMKUCR indicated the we need to analyse what majority sources say about a linkage between Communism and mass killings. That means, the whole discussion must move to Source analysis subpage, and, after this issue will be clarified, we can move back to DRNMKUCR to resume the discussion of the RfC. --Paul Siebert (talk) 15:48, 15 December 2021 (UTC)

Culberston, Rummel, Courtous, VoC, Brzezinsky, White, Valentino&demography

To demonstrate the quality of sources that are used in this article, and how they are used, let's take a look at these sources. The article cites Culberston, Rummel, Courtous, VoC, Brzezinsky, White, and Valentino as INDEPENDENT sources of data. Let's take a look on them

• Culberston: It is not clear where he took figures, but this source is clearly desperately outdated.
• Brzezinski: Again, an outdated source whose author was affiliated with the US government
• Rummel: This author provides a comprehensive and transparent analysis of Cold war era source. However,

- Since these sources are old, this Rummel is outdated.

- His procedure was described as seriously flawed by two authors (Dulic and Nathan, whereas Karlsson notes his strong ideological bias). Detailed analysis of Rummel deserves a separate section (and I am going to do it below), but a preliminary conclusion is that it is a piece of trash. Just few examples: Rummel heavily cites old Conquest data that were reconsidered by Conquest himself after "archival revolution". Rummel calculates "democide deaths" by subtracting outdated figures of WWII losses (20 million) from a total number of deaths; Rummel extensively cites Antonov-Ovseenko, who is considered unreliable now, etc. As Nathan correctly noted, reliability of Rummel's estimates depends on reliability of his sources, and his sources for USSR are extremely unreliable and outdated.

• Courtois; Due to the lack of inline references, it is very hard to understand where he took his figures, but these figures are the most criticised part of the Black Book;
• VoC: Two RSN discussiona almost anonumously agreed that this source is unreliable. Furthermore. This source is not independent, it cites Brzezinski (already in the list), Rummel (Already in the list), Courtois (already in this list), and White (already in this list), and a couple of outdated newspaper publications.
• White: this source is authored by some amateur who just collects various figures and calculates a median value. For Communism, he cites Rummel (already in this list; in addition: Rummel is doing the same, but more professionally. Being an amateur, White even doesn't understand the Rummel's method, otherwise he would never add Rummel's average data to the sample), he also cites Brzezinski (already in the list), Courtois (already in the list), and HIMSELF! (the last is a brilliant idea!)
• Valentino cites:

-Rummel

-Courtois

-Brzezinski

-Matthew White

-Culberston

and says they are considered "The highest end of the range" ("Final solutions", p. 275)

Valentino says that he made his own estimates based on "numerous" sources, but he provides no information about those sources.

In summary, the "Estimates" section contains an apparently long list of "independent" sources that cite each other, and that contains biased and desperately outdated estimates, which should be considered "highest published estimates" according to at least one reliable source. And, a significant fraction of those sources are directly or indirectly linked to the US authorities. Frankly, I cannot see how a good faith user can support this terrible text that undermines any credibility of Wikipedia. Although I am not going to speculate about motives of those who wrote that garbage, it looks like an attempt to mislead a reader and to make an impression that large amount of high quality and independent sources exists on the topic. In reality, this section had demonstrated the opposite: No good quality and non-controversial sources exist on this topic.Paul Siebert (talk) 19:54, 16 December 2021 (UTC)

@Paul Siebert: There also exists a slightly different estimates section in the article Criticism of communist party rule that could have some of its contents copied into this article. X-Editor (talk) 21:02, 16 December 2021 (UTC)

@X-Editor: according to post-Soviet demographic studies (this source is considered reliable by Maksudov&Ellman and other experts (e.g. [2], [3]), total demographic losses (which include not only excess deaths, but migration and fertility decline), during 1927-46 was 56 million. Demographic losses in 1927-41 were 13.5 million, but human losses (actual deaths) were 7 million (ibid. p. 431). That give you an impression of a difference between demographic losses and human losses (excess deaths). A consensus figure of human losses (not demographic losses) during WWII is 26 million (ibid. p. 436). The WWII fertility decline partially interfered with WWI/civil war wave (it was a time when woman born in 1914-1922 came to a fertile age). Therefore, the only figures of excess deaths during 1927-46 are 7 million plus 26 million (WWII time, mostly war related), and some number of post-WWII famine deaths, which cannot be derived from the demographic data, probably, due to their relatively small number.

The question is: where Rummel's 68 million come from? Paul Siebert (talk) 21:39, 16 December 2021 (UTC)

@Paul Siebert: Not sure, that section of the other article might have to be checked out too. X-Editor (talk) 21:44, 16 December 2021 (UTC)

Yes, it is in my to-do list. Paul Siebert (talk) 21:45, 16 December 2021 (UTC)

@Paul Siebert: That's good. X-Editor (talk) 22:57, 16 December 2021 (UTC)

@Paul Siebert:, will you publish your research in a journal so we can use it as a WP:RS? --Nug (talk) 00:33, 17 December 2021 (UTC)

@Nug: WP:NOR says: "Articles may not contain any new analysis or synthesis of published material that serves to reach or imply a conclusion not clearly stated by the sources themselves". Since I am not going to add results of my own analysis or synthesis to any Wikipedia article, I don't need to publish anything. Meanwhile, let me remind you that (i) "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, that means not only we should analyse sources, we MUST do that to achieve neutrality. (ii) If you are incapable of performing such an analysis, or even understand the analysis presented by others, then you are incapable of writing a neutral content, and, therefore, this article is probably a wrong place for you per WP:CIR. Paul Siebert (talk) 00:43, 17 December 2021 (UTC)

@Nug: I can answer in less acrimonious way: imagine some book claims that Hitler killed 20 million Jews (more Jews that lived in Europe before WWII). Imagine that no serious RS dispute this fact (maybe, because this source is just ignored). Would it be correct to add this source to "The Holocaust" article? In my opinion, that would have discredited this article and Wikipedia in general.

I am explaining that according to Rummel, Communism murdered more people in the USSR that died prematurely as a result of WWII, collectivisation, repressions and all other non-standard reasons, and Rummel's figure is about two times greater that demographic data say. If we treat this figure seriously, that means (i) Hitler didn't kill anybody and (ii) Hitler's occupation lead to a dramatic decrease of deaths and increase of life expectancy of the occupied population. That is the only plausible explanation. Can a good faith user accept that? I believe, that is possible only if that user is totally incompetent. Paul Siebert (talk) 01:57, 17 December 2021 (UTC)

If I declared some "crusade" against the source that claimed that Hitler killed 20 million Jews, could anybody call me the Holocaust denier? No, for both a claim that Holocaust never was and the claim that it killed more Jews that lived in Europe equally undermine a credibility of the real tragedy. Similarly, Rummel's false figures undermine credibility of real horror of Stalin's regime. Paul Siebert (talk) 02:04, 17 December 2021 (UTC)

What exactly is “outdated” here? And being “affiliated with US government” (Brzezinski) does not actually affect reliability (though it should be attributed). Volunteer Marek 00:40, 17 December 2021 (UTC)

First, you need to self-revert. Then, read this and related discussions. If you will find fresh counter-arguments, please, present them, and we analyze them. Please, do that. Paul Siebert (talk) 00:45, 17 December 2021 (UTC)

Obviously I *have* read this discussion since I’m asking you to clarify a word you keep using in it, what exactly is “outdated” here and what makes it so? Volunteer Marek 00:59, 17 December 2021 (UTC)

I am not sure I have to answer every question: I am not this page's technical support service, and you are not a customer. You made a revert that is against consensus, and that was explained to you, calmly and politely, by another user. Please, re-read what that user wrote, apologize for being rude, self-revert, and read all related discussions, because these discussions already address most of your arguments. If after that you will come to a conclusion that you have some new arguments, we will gladly discuss them with you.

So far, it is clear from your question that you haven't bothered to familiarise yourself with previous arguments, and your question is tantamount to the request to briefly summarise all previous discussions for you. That is not polite. I think any discussion with you is possible after you self-revert. Again, if I were you, I would apologise before Mario, for not every newcomer is a sock of you-know-who. However, that is optional, because it is an indication of one's self-esteem (which may be different in different persons). Paul Siebert (talk) 01:48, 17 December 2021 (UTC)

No, I am simply 1) asking what “outdated” means in this context and 2) pointing out that a source being “government affiliated” does not render it unreliable (although it should be attributed). I can’t really make it clearer than that. Volunteer Marek 08:54, 17 December 2021 (UTC)

You are repeating questions that had been already asked and answered here (and elsewhere) several times. That is a disruptive behaviour. Paul Siebert (talk) 15:23, 17 December 2021 (UTC)

Rummel's metod was sometimes - source one 100,000, source two 1,000,000, so Rummel's estimation was 100,000 plus 1,000,000 divided by 2 made 550,000. But one of the two sources could be precize and the other one inflated. Xx236 (talk) 13:18, 17 December 2021 (UTC)

Actually, you just reproduce Dulic's and Nathan's criticism. However, Dulic's criticism is deeper:

First, Rummel takes all sources without criticism, so he combines obviously unreliable data and reliable data in one pool. Rummel states that when he uses many data points, unreliable low figures and unreliable high figures will cancel each other. Dulic correctly argues that low estimates have a natural limit (zero), whereas high estimates are limited only with the author's imagination. That inevitably leads to inflated data.

Nathan correctly notes that reliability of Rummel's data depends on a quality of his data sets. You can easily see, that the raw data for Cambodia are pretty reliable and have relatively low dispersion, so Rummel's estimates are quite reasonable. In contrast, the quality of his data for the USSR is awful, and they are dramatically outdated, so no serious expert in Russia ever cited Rummel (maybe, just in a historical context).

That should be absolutely clear to any good faith user. Paul Siebert (talk) 15:22, 17 December 2021 (UTC)

Inaccurate paraphrasing, or synthesis/interpolation of source material

The article takes a quotation from a blog post, by way of this research review, summarizing it like this: "Karlsson describes Rummel's estimates as being on the fringe, stating that "they are hardly an example of a serious and empirically-based writing of history", and mainly discusses them "on the basis of the interest in him in the blogosphere."

I think that this is a biased and innacurate citation, since Karlsson never uses the word fringe, and is clearly referring to Rummel's blog post as the target of not being "serious and empirically based" and not his estimates of the death counts.

A similar mistake is repeated a bit lower, about the Black book of Communism: "Courtois' attempt to equate Nazism and communist regimes was controversial, and remains on the fringes, on both scientific and moral grounds."

The source says that the comparison is controversial, but not "on the fringes". The word "fringe" is being targeted in both cases, although it doesn't appear in the source that is being cited. I think that both references should be changed. The authors should either be quoted directly or have their quotations not synthesized together and interpolated in this way. AShalhoub (talk) 20:16, 13 December 2021 (UTC)

Interesting find. Though I do not personally know would this should be rephrased to be correctly paraphrased or accurately portrayed on the article. Removing the word "fringe" would be a start. MarioSuperstar77 (talk) 00:12, 14 December 2021 (UTC)

I am not sure they are referring only to his estimates on the blog. "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." (p. 35) I hope those edits addressed your concerns and improved wording. Davide King (talk) 11:24, 14 December 2021 (UTC)

Mostly okay, I tweaked your wording to be closer to the source. --Nug (talk)

I think the "blogosphere" part is important because we mention that Rummel revised his numbers up for China but he did so in his blog, and Karlsson, like our policies and guidelines also says, comments that his blog posts and non-academic works are not reliable. Davide King (talk) 01:23, 15 December 2021 (UTC)

But Rummel's numbers are published in his book "Death by Government", so reference to blogosphere is largely irrelevant. --Nug (talk) 01:44, 15 December 2021 (UTC)

How frequently Rummel's figures published in his "Lethal policies" are cited by recent works by experts in Soviet history (for example, in 1995-2020)? And what figures are considered consensus figures for USSR? And what is the difference between the former and the latter? And, if the difference is big, why are you still beating this dead horse?

Instead of stonewalling, try to give answer to these relatively simple question, honestly and neutrally, without cherry-picking and manipulations.--Paul Siebert (talk) 03:14, 15 December 2021 (UTC)

This section is about inaccurate paraphrasing, stop trying to hijack every discussion with your anti-Rummel advocacy. --Nug (talk) 04:08, 15 December 2021 (UTC)

"But Rummel's numbers are published in his book" — No, not his additions for China, which came from his blog posts, therefore that part is relevant and was noted even by the OP, that Rummel's blog posts, including the additions for China, "are hardly an example of a serious and empirically-based writing of history", and mainly discusses them "on the basis of the interest in him in the blogosphere." Either we remove his additions for China, or we must re-add this relevant passage from Karlsson. Davide King (talk) 11:08, 15 December 2021 (UTC)

The Karlsson quote I'm speaking of is referring specifically to Rummel's summary of Mao's intentionality to kill people by way of famine and it doesn't mention anything about death counts. Karlsson has other things to say about that elsewhere in the article. In any case, I'm not sure it makes any sense to mention either Rummel's blog or criticism of Rummel's blog in this article.AShalhoub (talk) 20:01, 15 December 2021 (UTC)

Do you mean this quote:

"R. J. Rummel is one of those who claim to have identified an extreme intentionality in Mao, and who insists that even the mass starvation was in some way ‘intentional’. He writes: ‘M ao's policies caused the famine. He knew about it from the beginning. He didn't care! Literally. Indeed, he wanted to take even more food from the mouths of his starving people in order to increase his export of food. It was all he had to export and he was after power... Those in the top circle of the CCP tried to alleviate the famine. They were arrested, some tortured, some executed or allowed to die horribly... So, the famine was intentional’. It could, quite rightly, be claimed that the opinions that Rummel presents here (they are hardly an example of a serious and empirically-based writing of history) do not deserve to be mentioned in a research review, but they are still perhaps worth bringing up on the basis of the interest in him in the blogosphere."

Yes, that is a useful quote: Rummel's opinion is popular in the blogosphere, and that raises a question of we are going to reflect the opinion of the blogosphere or we write the article based on the opinia of experts. Paul Siebert (talk) 20:16, 15 December 2021 (UTC)

I mean, the review is 15 years old; is it really pragmatic to pick apart an outdated blog post when Rummel has heavily cited published material? It's like searching for the weakest possible argument to shoot down. AShalhoub (talk) 21:24, 15 December 2021 (UTC)

As I wrote, if we are going to mention Rummel's addition for China, we must inevitably add Karlsson. In regards to Rummel's published material, we must distinguish between that published in the academic press, which was much better, and that published by the non-academic press. By the way, as was noted here, that review is the true source that makes this article, as currently structured, notable, and that is an argument against Karlsson, which I do not think that is the path you would want to make. Perhaps the fact there has been no other such review, or similar source (that is the kind of sources we need for an article of Communism as a single phenomenon rather than chapters in works about mass killings in general), proves that there have not been much development despite over a decade to do so. Davide King (talk) 21:35, 15 December 2021 (UTC)

@AShalhoub: Yours "is it really pragmatic to pick apart an outdated blog post when Rummel has heavily cited published material?" is a quite reasonable argument. However, two considerations must be taken into account.

• This "outdated blog post" (if I correctly understand you, you mean Karlsson, aren't you?) is a core source for Crimes against humanity under Communist regimes article. The same source is used in this article too. Do you propose to remove it completely, or you assert that it is unreliable only in this aspect? If the latter is correct, what is the reason?
• You implicitly assume that Rummel has is heavily cited. However, is it a proof that he is an expert in China? How frequently Rummel (the source that is even older that Karlsson) has been cited by modern experts in Chinese famine? Is hes opinion on Mao shared by experts? I found no discussion of Rummel's views by experts in Chinese history: no criticism, no support - nothing. That means if we remove Karlsson, we have even more reasons to remove Rummel. Paul Siebert (talk) 22:01, 15 December 2021 (UTC)

I didn't implicitly assume he has citations, I looked, and he has thousands of citations on google scholar. But anyways, the blog post I meant was the literal blog post that Karlsson cites at http://freedomspeace.blogspot.com/2005/11/reevaluating-chinas-democide-to-be.html which doesn't even exist anymore. The article says he caused a stir in the blogosphere, in 2008. People seem to be complaining that this page is bloated and that's exactly the kind of fat that should be trimmed. But if it's left in, you can't confound that specific criticism with the authors' criticisms of Rummel's academic work, so therefore just quote him correctly. All these other questions you're asking here seem to me to be distracting and bloating the topic at hand. AShalhoub (talk) 22:21, 15 December 2021 (UTC)

WRT "I didn't implicitly assume he has citations, I looked, and he has thousands of citations on google scholar." Whom are you talking about? Paul Siebert (talk) 22:50, 15 December 2021 (UTC)

I am pretty sure they were referring to Rummel, who has indeed many citations; of course, the mere numbers of citation is better seen as a sign of notability rather than reliability, unless such citations are analyzed to check what they actually say and what is the context. I am going to guess that majority of those citations come from his work on democratic peace theory, which is Rummel's expertise and is mainstream on that, rather than Communism, which is the subject of this article, and the context will be either outdated sources who still relied on Rummel pre-1989, sources that did not account for the archives revolution, and sources that mentions how his estimates are no longer accepted or relied on by majority of country experts. Davide King (talk) 23:29, 15 December 2021 (UTC)

The question is how frequently Rummel's China's bloody century is cited by experts in Chinese history. I didn't find much. One old review (there is no fresh reviews) says that the main point of the author is that "absolute power in hands of tyrants" is the primary cause of deaths, and he combined late Qing, Nationalists and Communists into a single category.

In addition, the reviewer (Nathan) discusses Rummel's method:

"Rummel's method is to break democide down into chronological and functional parts (land reform, suppression of counter-revolutionaries, three and five-anticampaigns, collectivization, labour camp deaths, the Cultural Revolution, and so on); to seek high and low estimates for each part from among what he regards as responsible sources; to split the difference, after throwing out any highs or lows that he thinks are clearly exaggerated, among those that remain to arrive at a midpoint esdmate; to readjust, in cases of doubt, this midpoint downward to arrive at a 'conservative' number, and then to label this midpoint estimate the 'most likely' figure, with the averages of the high and low estimates retained to provide a probable range of error.

In other words, the spirit of the enterprise is to use relatively mechanical computational procedures to find a rough average among the numbers in available sources, hoping to avoid injecting personal bias into the evaluation and interpretaton of the sources. While Rummel admits that the sources themselves are often not very accurate, he hopes that their errors will more or less cancel one another out, leading to what he calls a 'prudent estimate'.

How convincing one finds this procedure depends on how much confidence one has in the sources from which he draws his raw numbers."

That is literally the same that I was trying to explain to @Nug: Paul Siebert (talk) 23:57, 15 December 2021 (UTC)

It seems this is the link. Blogosphere ii full of that. Paul Siebert (talk) 23:47, 15 December 2021 (UTC)

Paul Siebert, this is really getting the flavor of "you're removing X, so I'm removing Y". I think the conversation about Rummel must have been had dozens on times on this wiki, so if you want to take down all of his references, I would follow up on where that conversation in a better place than this thread, which is very specifically about misquotations in the Karlsson paper. AShalhoub (talk) 08:12, 16 December 2021 (UTC)

Absolutely not. I am just saying that we must be consistent. If Karlsson is outdated, then Rummel is even more outdated. And, if we remove Karlsson's opinion on Rummel, this source is not acceptable in other context too (Karlsson is used not only as a source about Rummel, but for other statements)?

But if we decide that Rummel is acceptable, why Karlsson is not?

And I am not going to get rid of Rummel. His contributions into science are important and widely acknowledged. However, his figures are highly unreliable, and, importantly, outdated, and his "Democide" concept is challenged by many authors, including Valentino and such a renown sociologist as Michael Mann. Paul Siebert (talk) 16:07, 16 December 2021 (UTC)

I describe Rummel's 'statistics' below. 'his figures are highly unreliable' - exactly. He tried to describe the whole world using mostly English language sources. Xx236 (talk) 13:24, 17 December 2021 (UTC)

Not exactly. Sometimes, English sources are quite reliable, for example, for Cambodia. But that is not applicable for Cold war era data on the USSR. Paul Siebert (talk) 15:39, 17 December 2021 (UTC)