Talk:Mass killing

Proposed merge to Mass murder

"Mass murder" and "Mass killing" are different terms, the "Mass killing" article describes a scientific term proposed by some genocide scholars (Valentino et al) in an attempt to explain genocidal events and predict their onset in future. In that sence, "mass killing" is a totally different from "mass murder". Many authors, for example, Steven Wheatcroft, insist that "mass murder" should be discriminated from "mass execution", "manslaughter", etc. Moreover, genocide scholars do not use "murder" at all. In connection to that, it is totally incorrect to merge these articles.--Paul Siebert (talk) 23:43, 30 August 2018 (UTC)

Merger with mass murder

Paul Siebert: first, thanks for your addition of lots of informative and well-referenced material.

Though some scholars may use "mass killing" to refer to events involving over 50,000 deaths, and though "murder" has a particular legal definition (which varies by jurisdiction) readers commonly understand "mass killing" and "mass murder" to be synonyms. News media commonly describe a shooting of 5-10 people as a "mass killing".

Even if the phrases were used in distinct ways, differences in terminology do not inherently justify separate articles if there would be a lot of overlapping content, since those differences could be explained in a short "terminology" section. The current content of mass killing overlaps completely with the current content of mass murder, which includes killings of people by both governments and individuals, from small mass shootings up to the largest genocides, repression campaigns, and indiscriminate wartime killings. If there was a useful distinction to be made in our coverage, we could perhaps come up with a more specific topic to separate out under a different name, but for now I think these should just be merged. -- Beland (talk) 23:49, 30 August 2018 (UTC)

I respectfully disagree. "Mass killing", as well as "democide" or "politicide" is a scientific term, it was introduced as a tool for creating statistical models (which imply some inclusion tresshold), and the difference between "mass killing" (a scholarly term) and colloquial "mass killing" is not less than between force and force.

"Mass killing" (a scholarly term) is not just "killing of many people", as well as "genocide" is not just a brutal murder (this term has a very concrete meaning). "Mass killing" is the term which is used by scholars to understand the roots of mass violence, there are special databases where this type events are selected for subsequent analysis, and the conclusions different authors draw from these databases collected using different inclusion criteria may be different (for example, conclusions Rummel draws from his database are different from conclusions made based on some other database. --

--Paul Siebert (talk) 00:10, 31 August 2018 (UTC)

What's the distinction between "mass killing" and "mass murder" that you find important? The number of deaths? -- Beland (talk) 00:12, 31 August 2018 (UTC)

Not only. For example, Wheatroft separates "mass murder" (extermination of Jews by Nazi) from "mass executions" (killing of political opponents by Stalin, when some minimal legal procedure was observed) and from "mass manslaughter" (deaths as a result of a criminal neglect). In addition, "mass murder" always implies clear intention. There are many differences, and only journalists and superficial writers ignore them.--Paul Siebert (talk) 00:21, 31 August 2018 (UTC)

Would you want separate articles for "mass killing", "mass murder", "mass execution", and "mass manslaughter"? -- Beland (talk) 00:29, 31 August 2018 (UTC)

@Paul Siebert: And what would we call the article that provides an overview of all these things? -- Beland (talk) 00:50, 31 August 2018 (UTC)

I am not a native English speaker, but it is obvious for me that "murder" is much more emotionally loaded (and more narrow) term than "killing" or "manslaughter", which means, "mass murder" is hardly a good overview article. I also think that our policy does not allow us doing synthesis. In special literature, the authors who study mass killings are colloquially called "genocide scholars", although "genocide" is a very strict (and very narrow) term. In general, no common terminology exists, "democide" is not a generally accepted term, Valentino's "mass killing" is too easy to misinterpret (for example, the redirect to "mass murder" existed until today), "classicide" and other "-cides" are also not universally accepted terms. I think, we can live (at least, for a while) without a general article. In my opinion, the "Mass killing" article should tell about various theories that explain onset of various geno/politicidal events based on an analysis of existing data and draw casual linkages between regime's type and genocide.--Paul Siebert (talk) 01:06, 31 August 2018 (UTC)

Well, if you don't see a good name for the general topic of killing many people at the same time, the next most general overview article is homocide, which actually come to think of it really should have a summary of that subtopic there. It does at least to a few mass* articles via Template:Homicide, but if we are creating more top-level articles within this subtopic, we should link them from there.

If we are not using the everyday meanings of "mass murder" and "mass killings", we'll need to add hatnotes to explain the scopes of those articles to readers. Are these the sort of distinctions you'd want to make:

• Small-scale unlawful killings by individuals and non-state actors (less than 4 per session) - murder is the main article, that explains legal categories and exceptions like manslaughter, and euthanasia. It could simply link to the "mass" articles via Template:Homicide.
• Medium and large-scale unlawful killings by individuals and non-state actors (4 or more per session) - mass murder.
• Large-scale killings by governments (~50,000+) - mass killing. I'm assuming all events of this size have had government support?
• Small and medium-scale killings by governments - extrajudicial killing?

This would involve eliminating and splitting the contents of the "By states" section of mass murder and moving some examples to mass killing and some to extrajudicial killing or a list of war crimes or something.

I'm assuming you don't want to create "mass manslaughter"? Mass execution redirects to massacre, which seems a bit vague. Though because it is a broad term, "massacre" could be a spot to park an overview of all types of mass homicide. Ooo, "mass homocide", there's a nice sum-of-parts phrase, though maybe too much of a neologism. -- Beland (talk) 08:21, 31 August 2018 (UTC)

Yes, I see no reason for creation of "mass manslaughter" because I am not aware of any scholarly source that defines this term or discuss it as a separate category.

I see the scope of this concrete article pretty narrow. When some scholars decided to study the causes of the Holocaust and similar event, they realized they need a good statistical model to find correlations between various factors that might lead to similar tragedies in future. To do that, they created databases that combine all events where many people were killed in some particular society, and some parameters that describe that society (regime type, ethnic composition, etc). Then they draw conclusions about casual linkages and try to make predictions. This article, for example, should describe Rummel's theory (his "democide" is equivalent to "mass killing" except it has no lower limit; he conclude that totalitarian regimes are more likely to commit mass killing), Valentino's theory (different types of mass killing exist, for example, "strategic mass killing" is used as a tool by auto/totalitarian leaders to achieve their strategic goals; ideology plays almost no role in that), Semelin's theory, etc.

What I would like to avoid is Wikipedians's own creativity when new instances of mass killing are added to the table (or tables), and various journalism dilutes more or less rigorous theories. It should be explained in the lead that this article summarises existing databases (published in reliable sources) and existing theories of geno- and other "-cides", and it is not a place for unlimited creativity.--Paul Siebert (talk) 15:39, 31 August 2018 (UTC)

OK, I have removed the proposed merge tags and put a hatnote on mass killing scoping it to the scholarly definition of non-combat killing of 50,000 people or more. It would be nice to add more information to the table and article generally as to whether all of these incidents were perpetrated with government support, and which type of governments were involved. I have also re-scoped mass murder to focus on criminal mass murder (applied to individuals and groups only) and made Homicide#By governments the overview for government killing in general. I guess we'll see how things develop from here. -- Beland (talk) 20:51, 5 September 2018 (UTC)

Beland, I've just copied the table from the article I cite. No additional information was there, but I see no problem with adding it, for example, by providing links to specialised articles (many of those events have their own articles, or they are described in other Wikipedia articles). The only thing I will object to is addition of new items to this table, because this table is a work of a very concrete author, and it reflects her vision of summary of all genocidal events that happened globally during that period of time.

I generally like what you are doing with the "mass murder" article. With regard to this article, I think we need to explain the reason for development of this terminology (and why there is still no universally accepted terminology), to describe Valentino's theory, to talk about the difference between Rummel's "democide" and genocide/mass killing (Harff discusses it in her recent work), about "second generation genocide scholars", etc. Unfortunately, right now I am somewhat busy, but I'll return to that in close future.--Paul Siebert (talk) 22:47, 5 September 2018 (UTC)

Cool, thanks for your efforts! Busy here too, but I'll try to add some links and whatnot when I have a moment or two. -- Beland (talk) 23:31, 5 September 2018 (UTC)

Recent edit

I think all recent edits are definitely an improvement, except this one. Not only that text is relevant, this is an opinion of a renown genocide scholar about themselves. It should be expanded, probably in a separate section (and to add opine of other authors). This point is very important, because genocide scholars are not experts in figures, and, importantly, most of them do not pretend to be experts in figures for each concrete country. They just summarize existing data (including obsolete or marginally reliable) in attempts to derive global trends. That is why it is important to use country expert's works for accurate data for each separate country. --Paul Siebert (talk) 14:26, 6 September 2018 (UTC)

Related discussion

As a courtesy notice, there are discussions active on Talk:Mass killings under communist regimes that concern this article, particularly in the summary style relationship between the two. (not watching, please {{ping}}) czar 08:36, 13 December 2020 (UTC)

Scope

What would this article lose/exclude if its scope was limited to "by governments/states"? The current content appears limited almost exclusively to mass killings by governments/states. czar 19:41, 21 December 2020 (UTC)

I think it is implicitly assumed that we the scope is "by governments/states", because all other categories are described in Mass murder. However, if that is unclear, we can specify that in the lead.--Paul Siebert (talk) 19:55, 21 December 2020 (UTC)

Wouldn't it be more effective to specify it in the title? I think that someone visiting this title would think it's about mass shootings/murder, terrorist attacks, etc. as phrased, so while concision is generally preferred in titles, it's also judged on balance with precision and clarity. czar 20:20, 21 December 2020 (UTC)

Actually, the same problem is with Mass killings under Communist regimes article: when a reader sees that 100+ million were killed, they also think they were victims of mass shooting/murder (similar to what happened during the Holocaust), but majority of deaths were a result of famine and disease, and that happened in the countries, such as China, where devastating and deadly famine was a routine event.

I don't completely understand what title do you propose, but if "Mass killing under Communist regimes" is an acceptable title, then some more general article is needed that will cover "mass killing" in general. Moreover, the term "mass killing" was applied by Valentino to the events in Communist states, but his book was not focused on Communism at all: he pays nearly zero attention to ideology, and his concept is equally applicable to Rwanda, Yugoslavia and China. Therefore, if this title seems misleading to you, the "Mass killing under Communist regimes" is even more misleading.--Paul Siebert (talk) 03:06, 22 December 2020 (UTC)

The idea was to workshop title suggestions at Talk:Democide § Neologism but if I had to start somewhere right now it'd be "Mass killings by states". I think it's okay that famines might not be the first type that come to mind, but that's also why the discussions of whether/how famines "count" (and the role of intent) should get a mention in the lede and its own section for discussion. If something like "by states" is acceptable as a title, then a whole lot of content can start being merged into this article. czar 03:56, 22 December 2020 (UTC)

That sounds reasonable, but, for consistency, that requires renaming of "Mass killings under Communist states" to "Mass killings by Communist states". Both renamings should be done concurrently.--Paul Siebert (talk) 04:09, 22 December 2020 (UTC)

No objections from me but since that likely requires a talk page discussion, given everything else that's going on there, I doubt it's worth the energy right now. Besides, once its content is properly distributed and balanced, it might not even exist under that title. czar 04:53, 22 December 2020 (UTC)

• Going to store some alternative titles here for safekeeping as I come across them (no action needed): mass political killings, mass state murder, state-sponsored murder (also note that "state" by some readings might exclude mass killings by rebel groups) czar 09:11, 22 December 2020 (UTC)

czar before we move further, let's decide what this article is about. I doubt some mainstream political science school studies mass killing as a separate phenomenon, they are more interested in some concrete cases, each of which has a unique nature and rarely shares significant common feature with other mass killings. The discipline that studies general dependencies is "genocide studies", but, as the source provided by me says, it is a minority school of thought. The general books about mass killing are, by and large, intended for general public, and I suspect they tend to be redundantly sensationalists (I recall even Guinness book had a topic "the worst mass killer"). Therefore, it seems we should either present views of genocide scholars on some general model of mass killing onset, or just make a list. In reality, most mass killing events already have their own articles in Wikipedia, and focusing too much at general description would be a violation of NPOV principles, because the experts in each country's history prefer to describe each separate mass killing separately, and they provide facts and explanations that may be significantly different from those provided by "genocide scholars". If we will try to embed the opinia of country experts into the narrative written from "genocide scholars" perspective, we artificially promote the latter to the rank of majority view. If we pay due attention to the opinia of country experts, there article may fall apart. --Paul Siebert (talk) 01:27, 24 December 2020 (UTC)

The alternative names were taken from scholarly papers either before or after the advent of "genocide studies". They're alternative frames for the scope and, as I wrote, I don't think they need consideration. I don't think there is disagreement that the ideal sources in this article's case are scholars of genocide (in the macro, not "country experts") writing about genocide-broadly-construed (i.e., "mass killing") in the macro. If you'd like to talk potential article sections, let's split that out to its own discussion? czar 01:54, 24 December 2020 (UTC)

There is one fundamental problem with genocide scholars: they are too focused on finding some general laws, and as a result they do not pay much attention to details. Sometimes, they prefer to ignore some facts that do not fit into their concepts, use questionable figures, etc. And, as I demonstrated previously (see the reference), "genocide studies" is not a mainstream topic, so the lack of criticism is not an indication of acceptance of their views.

A couple of examples. One of the leading experts in Cambodian genocide is Kiernan. He identified three major roots of the genocide (desperate poverty of rural Cambodians, who hated urban population and intellectuals, a national tradition of revenge, and ultra-Maoist ideology), and pointed at massive bombing of Eastern Cambodia by the US as one of important factors that helps KR to come to power. In contrast, such genocide scholar as Rummel selected just one factor, totalitarianism, which lead to the genocide. There was no dispute between Rummel and Kiernan, probably because the latter did not consider Rummel's writing as deserving significant attention.

One of Rummel's critics, Dulic, pointed at serious methodological flaws in his statistical analysis method, however, he did that only in a context olf Yugoslavia, probably, because he himself is from that country. Rummel failed to adequately address the criticism, which means his conclusion about Yugoslavia are very questionable. The authors from other countries didn't criticize him, despite the fact that his "estimates" are in a blatant contradiction with modern data. That means they do not think Rummel's analysis deserves serious attention.

"Genocide studies" is an emerging discipline, it is not universally recognised, its conclusions may be somewhat valid in general, but they may be very questionable in each particular case, so to presents them as "superexperts in genocide" would be a mistake. Accordingly, the article should be not about the events themselves (each of which had a very complex nature, were unique, and it is hard to fit in some simple general scheme), but about some attempts to find general dependencies. And we should explain that this work is in progress, and all conclusions are just preliminary, and they have not become the majority viewpoint yet.--Paul Siebert (talk) 17:26, 24 December 2020 (UTC)

I don't think Wikipedia intends to present any source as incontrovertible truth. The point is to take the best vetted sources on this mass killing macro concept and show what they say about each other. Where it gets into claims about specific events (i.e., not about the macro concept), yes, the individual articles are better equipped to handle those discussions. czar 18:49, 24 December 2020 (UTC)

I don't know if you read the source that I provided, but it clearly says that "genocide studies" is a minority school of thought:

"This new generation of scholarship has crystallized into the interdisciplinary field of "genocide studies," a community of scholars and practitioners dedicated to researching and preventing genocide. However, genocide studies has emerged as its own research field, developing in parallel rather than in conversation with work on other areas of political violence. Aside from a few important exceptions, mainstream political scientists rarely engage with the most recent work on comparative genocide. Some of the newest genocide research appears in topic-specific conferences and journals like "Genocide Studies and Prevention" and the "Journal of Genocide Research", but not in political science venues. The reasons for this separation are complex, but partly stem from the field's roots in the humanities (especially history) and reliance on methodological approaches that have had little resonance in mainstream political science, as well as the field's explicit commitment to humanitarian activism and praxis. Earlier generations of political scientists and sociologists who studied genocide often found little interest for their work among dominant political science journals and book publishers; they instead opted to establish their own journals and professional organizations. Although the field has grown enormously over the past decade and a half, genocide scholarship still rarely appears in mainstream disciplinary journals."

Therefore, we can use these sources, but a reservation must be made that makes it clear.--Paul Siebert (talk) 23:06, 24 December 2020 (UTC)

No primary sources

I went through a lot of tertiary source material today and found that there's a whole lot of material—this article is not hurting for sources. Given this, I'd like to recommend as a ground rule for this article: "no primary sources". There's no need to quote Rummel on Rummel if there is a secondary source who has analyzed Rummel's analysis for us, giving us a better sense of proportionality for what to include in the article. There's no need to quote directly from studies doing original analysis on patterns and process of genocide—instead we should look to paraphrase what other scholars have written about those original analyses. This saves us from some mental gymnastics of determining which (and how much) of the many opinions to include in a topic's discussion—it's easier to look to what scholars with editorial distance/independence have already determined to be the main perspectives on the topic and what they found noteworthy enough to describe. czar 09:09, 22 December 2020 (UTC)

I am not sure I understand you. When Rummel writes about democide, he is a secondary source, not primary. Rummel was not a historian, he worked with primary and secondary sources and summarised them.--Paul Siebert (talk) 01:14, 24 December 2020 (UTC)

Rummel talking about Rummel's definition is first-order analysis, even if that analysis is based on "primary sources". Same for Staub on Staub. Charny discussing the merits of Rummel's definition/analysis is second-order analysis. When citing sources, we prefer the analysis to be one order removed. czar 01:48, 24 December 2020 (UTC)

If Rummel's definition of "democide" is a subject of discussion, then yes, you are right. However, if the subject are various manifestations of democide (i.e. the concept itself is not a subject of discussion), Rummel is a secondary source. In other words, your point of view is valid mostly for Democide.--Paul Siebert (talk) 17:01, 24 December 2020 (UTC)

It's any scholarly analysis. If I have a new thesis or claim a pattern/trend/etc., that is my original analysis. We look to the scholarly peer community to vet that analysis and tell us (as Wikipedia, a tertiary source) what is noteworthy about that original analysis. We prefer sources that are secondary for that reason. czar 18:53, 24 December 2020 (UTC)

But if I make analysis of somebody else's analysis, it is also a new thesis or claim. Does it mean I am a primary source?

I think you apply PSTS too literally.--Paul Siebert (talk) 23:08, 24 December 2020 (UTC)

It doesn't need to be that meta. The idea is to prefer sources with ideological distance (independence/non-affiliation) from what they're describing. czar 09:16, 1 January 2021 (UTC)

Group citations

@Davide King, might I suggest, in lieu of adding multiple, large, footnoted quotes instead formatting a single group citation similar to this/these, only quoting the immediately pertinent parts? The idea would be to provide the best (not all) evidence supporting the statement. czar 17:51, 22 December 2020 (UTC)

And similarly, some repeated citations, like Wayman/Tago or Harff, need to be cited with page numbers (de-grouped). Otherwise they become difficult for a reader to verify. czar 18:06, 22 December 2020 (UTC)

Czar, I agree. If you could do that for me, also turning all refs into sfns, which should be the default for any good article, I would be very grateful and thankful. :) Davide King (talk) 07:14, 25 December 2020 (UTC)

 Done czar 07:43, 1 January 2021 (UTC)

Exact page references

Articles do not need exact pages, only books need them. At least, that is how the reference system works in most peer-reviewed journals.--Paul Siebert (talk) 01:16, 24 December 2020 (UTC)

Not sure where you're getting that impression on Wikipedia. The convention from the last decade has moved away from "general references" and to exact page citations (usually in short footnotes, leaving the general citation independent) for reader verifiability. (See articles listed in Wikipedia:Featured articles.) Otherwise readers have to take the citation on faith, which we want to discourage. The only exception I've seen to this is maybe medical articles (e.g., Malaria#References), but in those cases the general citations generally have their answers within their abstracts. czar 01:42, 24 December 2020 (UTC)

I don't know what convention are you talking about, I prefer to stick with what most scientific journals do. The reason is simple: in contrast to books, articles are usually focused on one narrow topic, so they should be read in full to avoid a situation when some words are taken out of context. Generally, they are not too long, so that is not too hard. In contrast, books are long, they may discuss several subtopics, and it is desirable to specify the chapter of subchapter the statement was taken from.--Paul Siebert (talk) 17:06, 24 December 2020 (UTC)

If the article/chapter is short, then it's not a huge issue to cite the entire thing, but I think it is generally more verifiable to give specific page number if possible. There was a recent discussion on WT:FAC; most agreed that for <10 pages, the page number wasn't a big deal, but for longer articles giving the page number is more important. (t · c) buidhe 17:15, 24 December 2020 (UTC)

Actually, WP:CITEHOW, says "volume number, issue number, and page numbers (article numbers in some electronic journals)". Note that the article number is equivalent to pages, from which I conclude that the first and last pages are meant, not the exact page number. Similarly, for books, "chapter number or page numbers for the chapter" are considered as two equivalent options.

All of that is not too important, but if we provide an exact page, we create an impression that we just picked up some statement without checking the context (which is not an uncommon practice, especially when users use google.books snippet view).--Paul Siebert (talk) 18:27, 24 December 2020 (UTC)

For general citations, what you suggest is fine. For short articles and general claims, the general citation is fine. We're talking about using {{sfn}} to give more precise page numbers or ranges within long articles: the full citation. Ours is a controversial topic being read and verified by a general audience. They need more guidance than being told to refer to a 30-page chapter to review a claim. Further down below where you quoted: When citing lengthy sources, you should identify which part of a source is being cited. The idea that precise page references somehow imply cherry-picked evidence is an argument I haven't heard in a decade of editing Wikipedia. Especially for this topic, I'm not seeing a convincing reason to avoid as much precision as we can afford. czar 19:05, 24 December 2020 (UTC)

Buidhe, Czar, and Paul Siebert, I was finally able to turn all refs into sfns. This really needs to be summarized in the body; I have added and expanded it a bit but I think a new section (Comparative research, Historiography, or Overview) needs to be created because many others scholars are mentioned and discussed that could be summarized. In general, there needs to be such a section that expand on comparative research, debates, and historiography, especially in regards to causes and attempts to predict or prevent genocide and mass killing. Davide King (talk) 11:20, 23 August 2021 (UTC)

Related discussion

Discussion on the first paragraph of ethnic cleansing article at Talk:Ethnic cleansing‎#RfC. (t · c) buidhe 18:28, 24 December 2020 (UTC)

Disputed information presented as fact

Nug is trying to add the following sentence: "Tago and Wayman have shown that Rummel's methodology is sound and that the higher number is due to Rummel's use of a broader data set of victims compared to what Harff uses; Harff's geno-politicide data is basically a subset of Rummel's democide data." It's clear from the rest of the paragraph that is is far from generally accepted that "Rummel's methodology is sound", so it should not be presented as a fact. How exactly did they "show" this? (t · c) buidhe 11:17, 16 November 2021 (UTC)

References

The authors literally say:

"Rummel has a broader concept than geno-politicide, but one that seems to include geno-politicide as a proper subset"

and

"We have further shown that (although the onset years vary from Harff to Rummel), when one looks at which sovereign states were involved (and the approximate onset year), the geno politicide data is basically a proper subset of the democide data (as one would expect by the addition of the need to show specific intent in geno-politicide).

"In practice, Rummel is careful to measure the onset of a regime through independent means; for instance, in the case of Japan, the regime did become distinctly more dictator ial and independent of domestic opposition in 1937, and this autonomy logically then made it easier to start up a campaign of mass killing. However, despite this appearing to be a sound method Rummel is employing, it still makes it very difficult to compare his results in a neutral meeting ground against other analyses."

The words "sound" and "subset" are found only in these three sentences. From these quotes, it is absolutely clear that:

• Rummel's numbers are different because he uses different category of data (no lower limit and no intentionality). By no means these data are "broader": they both use a worldwide statistics, but Rummel's inclusion threshold is different.
• The word "sound" refers to Rummel's approach to defining the "the regime" variable. That does not refer to the approach as a whole.

Definitely, the added sentence contains severe misinterpretations of the source. Maybe, it makes sense to read it in full before using? Paul Siebert (talk) 16:00, 16 November 2021 (UTC)

Not at all. This reliably sourced sentence is clearly attributed to Tago and Wayman, not presented in WP:WIKIVOICE. I'm amazed this article uses Tago and Wayman's paper five times for some peripheral text, but ignores their central conclusion and in fact we even have an editor actively blocking its inclusions. I would imagine any casual observer would perceive this as an egregious case of WP:CHERRYPICK to present a misleading idea that Rummel's data is flawed, and as such this current section must be considered POV. A large portion of Tago and Wayman's paper is devoted to a comparative statistical analysis of Rummel's and Harff's data sets. I note Seibert missed the first part of the conclusion:

"We have shown the differences in the emphasis in explanation of mass killing in Harff and Rummel are not due to (1) differences in unit of analysis, or (2) temporal domain, or (3) control variables in the original studies but rather persist when unit of analysis, temporal domain and control variables are made exactly the same. Within that constant research design, we showed that the differences were not due to threshold either. The only remaining difference is the measure of mass killings itself - democide vs. geno-politicide. We have further shown that when one looks at which sovereign states were involved, the geno-politicide data is basically a proper subset of the democide data."

It makes sense, Rummel's higher data due to him using a bigger dataset than Harff does, nothing more. --Nug (talk) 16:27, 16 November 2021 (UTC)

Let me be frank. If you want to push some specific POV, you may continue in the same vein. If you want to figure out a real situation, let's do the following: we both read the article in full and post it summary here. After that, we can discuss how the information should be presented.

In addition, as a person who bothered to take "Death by Government" from a library and read it in full, I am surprised how desperately lousy and outdated Rummel's figures are. Virtually all experts in Soviet history (Rosefielde, Conquest, Wheatcroft and others) re-examined their old figures after new materials become available due to the "archival revolution" of 1990s. Rummel is arguably the only author who refused to do so. Actually, from the point of view of figures, his works have only historical interest, and we need to think how to represent them in a correct way, without giving an undue weight to desperately outdated numbers collected by Rummel from old sources and inflated due to an intrinsic flaw in his statistical procedure (as explained by Dulic). Paul Siebert (talk) 17:00, 16 November 2021 (UTC)

From Crimes against humanity under communist regimes – Research review, which is a tertiary source, Rummel is considered to be fringe ("they are hardly an example of a serious and empirically-based writing of history"), and is only mentioned "on the basis of the interest in him in the blogosphere." Davide King (talk) 17:26, 16 November 2021 (UTC)

Great! Whereas I am familiar with this source (which seems to be considered as very reliable by Nug and other users who share his POV), I managed to overlook this paragraph. For Nug's convenience, I reproduce it below:

"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. On the other hand, the considerably lower figures in terms of numbers of Gulag prisoners presented by Russian researchers during the glasnost period have been relatively widely accepted."

It seems Karlsson's statement about ideologically motivated inflation of figures by Rummel in combination with Dulic's detailed analysis of methodological flaws (which was done for Yugoslavia only, although criticism is general) is sufficient to describe Rummel's estimates as methodologically flawed and ideologically motivated. Paul Siebert (talk) 19:51, 16 November 2021 (UTC)

It is necessary to note that Karlsson is a core source for the Crimes against humanity under Communist regimes, so if someone questions his reliability, that may be an argument for deletion of that article. Paul Siebert (talk) 19:53, 16 November 2021 (UTC)

Paul Siebert, it appears that is exactly what Nug did here, where Rummel's status is brought up and is dismissed as "the Karlsson's one line statement about ideologically motivated inflation of figures by Rummel", not just Karlsson but also Dulić. They have again stated that we are actually misreading Tago and Waytman, so I think we should further discuss this. If I may, their full statement still did not answer why we must have a MKuCR article; even if what they said is true, that same database does not make any clear categorization. There is no MKuCR database, and they have failed again and again to debunk Siebert's rebuttal and explanation of how genocide scholars, who are a minority and is often forgotten, do not make such categorization, which is why MKuCR is OR/SYNTH. Davide King (talk) 21:13, 18 November 2021 (UTC)

To make it even more clear, just because the global database lists some events which happened under a Communist regime, it does not mean we must have a MKuCR article or that sources support there is a MKuCR category. It would be like taking all events listed there that were capitalist, Muslim, or happened in Africa or Asia — it would be OR/SYNTH because such source, or any other scholarly source, does not make any of such categorization, and to have such an article we would have to either have a scholarly literature about it or a clear link that connect those events, yet we do this only for Communism. Davide King (talk) 21:52, 18 November 2021 (UTC)

In light of this, do you mind if we discuss any issue here, while continuing the general discussion there? I have no problem doing both. Nug, I ping you again because I would like to hear your comments in light of the aforementioned link from your statements, which shows you were not satisfied by Paul Siebert's latest comments here. Buidhe is also free to address Nug's complaints in their linked comments. Can you clarify what you understand it to mean "reliability of Rummel's dataset"?

My main issue is the use of Rummel to support the current version of MKuCR, which is a different thing (Tago and Wayman do not discuss of MKuCR, even Rummel's categorization is described as "authoritarian and totalitarian government" at p. 5 vis-à-vis Valentino's disagreement), and I do not see what "reliability of Rummel's dataset" changes, other than we may rewrite MKuCR to be about authoritarian and totalitarian governments in general rather than only Communism — so whatever way you look at it, I do not see how that supports MKuCR or how they are misread. Finally, it must be kept in mind that genocide scholars are not as big and mainstream as MKuCR may look like them to be (they are a minority and are not relied by historians when discussing the events) — do Nug also dispute Genocide studies just like Mass killing? Davide King (talk) 03:02, 22 November 2021 (UTC)

In light of their comments here, Fifelfoo could provide further input. If Nug refuse to engage with us, they must concede that their allegations are wrong, and stop using this as an argument against MKuCR's problems. Davide King (talk) 14:35, 22 November 2021 (UTC) [Edited to add] Davide King (talk) 14:38, 22 November 2021 (UTC)

Tago & Wayman 2010 shows that there is a disagreement among scholars, and the solution is certainly not to give too much weight to Rummel by following his categorization, which are criticized by other scholars by Valentino, who is not the only one. When scholars disagree, the solution is not following categorization by a relevant but undue (in light of disagreement and criticism) scholar like Rummel. That we must give WEIGHT and priority to Rummel by having a MKuCR (full Communist-devoted article despite scholars either disagreeing or rejecting ideology and regime type links) is absurd, false balance, does not follow, and is quite frankly beyond me. I cannot possibly be the only to think this — I am well open to the idea of being proven wrong but I just do not see any sufficient rationale that would justify this. Davide King (talk) 16:28, 22 November 2021 (UTC)

Buidhe, Paul Siebert, were those edits by Nug acceptable and/or an improvement from their previous wording attempt? I think they certainly were an improvement, as they did remove "Rummel's methodology is sound" wording, and their chosen wording seems to reflect more what Siebert said, namely that "Rummel's numbers are different because he uses different category of data (no lower limit and no intentionality). By no means these data are 'broader': they both use a worldwide statistics, but Rummel's inclusion threshold is different", hence "Harff's dataset of politicide-geoncide being essentially a subset of Rummel's dataset, where he includes other types of killings in addition to politicide-geoncide." I would remove essentially though. Davide King (talk) 10:49, 3 December 2021 (UTC)

Datasets

Not sure why there is such an emphasis on the datasets in this article, but here is a list from Anderton, Charles H. (2016). "Datasets and Trends of Genocides, Mass Killings, and Other Civilian Atrocities". Economic Aspects of Genocides, Other Mass Atrocities, and Their Preventions. ISBN 9780199378296. fiveby(zero) 16:09, 23 January 2022 (UTC)

Datasets

Political Instability Task Force Geno-Politicide Dataset	genocide/politicide	Marshall, M. G.; T. R. Gurr; B. Harff (2014). PITF—State Failure Problem Set: Internal Wars and Failures of Governance, 1955–2013. Societal-Systems Research.
Ulfelder and Valentino Mass Killings	mass killing	Ulfelder, Jay; Valentino, Benjamin (February 1, 2008). "Assessing Risks of State-Sponsored Mass Killing".
Easterly, Gatti, and Kurlat Data Mass Killings	mass killing	Easterly, W.; Gatti, R.; Kurlat S. (2006). "Development, Democracy, and Mass Killing". Journal of Economic Growth. 11 (2): 129–56. JSTOR 40216091.
Rummel	democide	Rummel, R. J (1998). Statistics of Genocide: Genocide and Mass Murder since 1900. Transactions.
Uppsala Conflict Data Program One-Sided Violence Dataset	mass atrocity and one-sided lower-level atrocity	UCDP Publications
Political Terror Scale	high-level terror	Wood, R. M.; Gibney M. (2010). "The Political Terror Scale (PTS): A Re-Introduction and a Comparison to CIRI". Human Rights Quarterly. 32 (2): 367–400.
Cingranelli-Richards Human Rights Dataset	one-sided lower level atrocity	unpublished?
Minorities at Risk	forced resettlement, systematic killings, ethnic cleansing	unpublished?

Fiveby, that's a good find. --Nug (talk) 20:44, 27 January 2022 (UTC)