The following discussion is an archived debate of the proposed deletion of the article below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.

The result was Keep - Simply no support for nom's position on this one Mike Cline (talk) 12:20, 11 August 2010 (UTC)[reply]

List of battles by casualties[edit]

List of battles by casualties (edit | talk | history | protect | delete | links | watch | logs | views) – (View log • AfD statistics)
(Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL)

I suggest deleting this article as, even though sourced in respect of separate data, the entire tables, which are not sourced per se, consisiting of a mixture of separate numbers, constiute violation of the following Wikipedia rule: WP:SYN. Rubikonchik (talk) 11:04, 4 August 2010 (UTC) Just to make sure everybody undrestands correctly, the violation of WP:SYN consists in the fact that no sources are available for the made up tables.Rubikonchik (talk) 13:25, 4 August 2010 (UTC) The further debate goes, I realize in fact that not only rule: WP:SYN is violated, but also rule: WP:COATRACK and rule: WP:FRINGE. Please, do consider application of all the three rules in your answers. Rubikonchik (talk) 22:58, 4 August 2010 (UTC)[reply]

  • I added that sentence, and the problem is that most sources simply list all casualties. If I were to remove all the entries for which I did not have a source that listed only the number killed, or separate numbers for killed and total casualties, the list would be eviscerated and have fewer entries than it did before I started building it up. While I agree that what you're saying would be better, if the sources for that are out there, I don't have them. – Joe N 17:55, 4 August 2010 (UTC)[reply]
1WP:SYN
The list per se is not referenced. What is referenced are different numbers in each line. By the way, different sources provide different numbers and different values for those numbers. There is no value-number unity even in the separate numbers, and even less in entire tables. The synthesis of these numbers results in the freely made up by editors tables. This perfectly responds to the WP:SYN definition and requirements: Do not combine material from multiple sources to reach or imply a conclusion not explicitly stated by any of the sources. If one reliable source says A, and another reliable source says B, do not join A and B together to imply a conclusion C that is not mentioned by either of the sources. This would be a synthesis of published material to advance a new position, which is original research. "A and B, therefore C" is acceptable only if a reliable source has published the same argument in relation to the topic of the article.
2 WP:COATRACK
As far as WP:COATRACK is concerned, there is clearly an error of mixture of different values represented as one in these tables. It includes sampling bias, in which some numbers and values pertaining to this or that type of casualties are more likely to be included than others. This is parallel to a typical spectrum bias consisting of evaluating the ability of a test in a biased group of values, which leads to an overestimate of the sensitivity and specificity of the test. Alternatively, the bias of an estimator is the difference between an estimator's expectation and the true value of the parameter being estimated. Omitted-variable bias is the bias that appears in estimates of parameters in a regression analysis when the assumed specification is incorrect, in that it omits an independent variable that should be in the model. therefore, in present article, what is the independent variable in these tables? Can anybody answer? We are also dealing here with (i) systematic bias which is characterised by external influences that may affect the accuracy of statistical measurements and (ii) data-snooping bias which comes from the misuse of data techniques. As a result, due to the preferred sources by one editor, Berlin battle may appear as the one with most casualties and eventually imply that all of them were German and all dead, which is not necessarily true. The same applies to all other entries.
Therefore, taking into account all of the aforesaid, here again, it fully correpsonds to the definition of coatrack articles which can be created purposefully to promote a particular bias, and they can accidentally evolve through excessive focus on one aspect of the subject.
3 WP:FRINGE
Along with the above explained bias appears also WP:FRINGE - description of ideas that depart significantly from the prevailing or mainstream view in its particular field of study. Examples include conspiracy theories, ideas which purport to be scientific theories but have little or no scientific support, esoteric claims about medicine, novel re-interpretations of history and so forth. Some of the theories addressed may be hypotheses, conjectures, or speculations.. It goes without saying that this article with its biased lists includes numerous speculations about interpretations of history, and namely numbers of casualties and their relevant values, nations who suffered the most or the least, etc...Rubikonchik (talk) 10:19, 5 August 2010 (UTC)[reply]
Again, your argument for SYN doesn't apply because the list doesn't imply a conclusion. It merely says what the sources said, A and B, without introducing C. And merely having conflicting sources doesn't imply that the list should be deleted, or we'd be AfDing 90% of the articles on anything notable; for the most part, ranges are given when significant discrepancies are found.
I think you missed the point of COATRACK, unless you are implying that the author deliberately sought to subtly manipulate the data for some sort of nefarious scheme... what, I couldn't tell you. A bias in the data collection does not imply coatrack (which is a deceptive subject), that's merely a cleanup issue and not an AfD one. I see on the talk page that no attempts have been made to address this issue, which is simply shameful to suggest burning down a dirty house before cleaning it.
I think the FRINGE argument is just silly. If you think the statistics are wrong, then provide some reliable sources of your own and have them corrected. To say that the Battle of Berlin was a bloody affair is certainly not a deviation from the prevalent view (BTW, I'm not sure why you argue against an implication of German dead when the article doesn't differentiate the nationality of the casualties; such an implication simply doesn't exist, and the article gives a number not too far off).
Sounds like effort would be better spent repairing the supposed shortcomeings, rather than fighting to kill it with fire. bahamut0013wordsdeeds 11:18, 5 August 2010 (UTC)[reply]
My argument does apply because the list items are sorted from maximum to minumum casualties, including other criteria. There is always a conclusion looking at any table - the most casualties took place in X or Y. When I'm speaking of conflicting sources, I'm referring to the sources in respect of separate numbers. The tables as they are present in the article, by way of the aforementioned WP:SYN should be also examined under a larger scope of WP:OR. What we need for these tables and lists - are the sources and references for already existing tables and lists with a clear compatible and applicable to this article criteria (nominal unit). Please, revisit the Wikipedia:No original research.
In respect of the WP:COATRACK, here again, please revisit the relevant Wikipedia page with explanations and instructions. Pursuant to the existing wording of the aforesaid rule as of today, a bias does imply a coatrack. You may certainly propose to change the rules in accordance with your interpretation.
Regarding WP:FRINGE, personally, I don't think calling my argument "silly" gives you more credit, but anyway, looking at these lists and tables, one clearly sees speculation on multiple historical issues - what fully corresponds to the current definition as per WP:FRINGE. Here again, please revisit the current wording of the Wikipedia rule in this regard. Unless it has chaged since yesterday, it didn't become silly since it was adopted by a good number of Wikipedians. As for the numbers, for example, just type in Leningrad casualties in Google and see how many different numbers appear. Why namely the one retained in the list was considered? The same applies to all the other entries. Do you want to check all existing sources with all different numbers and count an average for all entries?
I'm not really fighting, just applying Wikipedia prescriptions. The scope of the article seems simply impossible to attain, although your proposal to improve is very interesting, but HOW and WHAT exactly?.Rubikonchik (talk) 13:52, 5 August 2010 (UTC)[reply]
The sorting defaults to chronological order, not by casualty. If a reader draws a conclusion on thier own based on raw data, how can that be considered biased? If you claim that the data is being manipulated to lead to a certain conclusion, then go ahead un-manipulate it. I also have to note that most of the numbers are taken from a small handful of sources, making SYNTH harder to prove when they are uniformly drawn.
I think it is rather you who needs to re-read COATRACK. The opening line defines it as a misleading article purporting to be about A, but really is about B. There's no interpretation there. This article is exactly what it purports to be: a list of battles by casualties.
Again, you seem to be twisting the guidelines. FRINGE refers to actual fringe theories, not disputing references. If source A says that 200,000 people died in a battle and source B says that 250,000, does that make one depart from the standard and commonly excepted norm? There is no real norm for such wide ranging estimates, and 50,000 is well within the margin of error, depending on how recent the conflict was and how well it was recorded historically. Now, if one of them was claiming that two million Japanese were killed at Iwo Jima, that would be a fringe theory (I doubt a million men could even fit on that island). I also note again that most of the sources appear to be reliable and respectable sources.
Ultimately, I think that would belong on the talk page, but you'd have to enumerate where you think the list is inaccurate or misleading, and then make discussion and consensus on what to display. I might suggest posting upper and lower limits, as well as a typical average. bahamut0013wordsdeeds 16:07, 5 August 2010 (UTC)[reply]
Your argumentation seems irrelevant. Please, reread my lines just above, there is no source for these tables and lists, there are sources (and unfortunately all giving different number indications) for each separate entry, but again, what is proposed to be deleted is this article containing these invented and synthesized on WP lists and not the separate numbers in each article regarding this or that battle. And if the purpose of these lists and tables is not to draw a conclusion, then what it is? Speaking of a concrete example how people draw conclusions, please have a look at how much is written regarding only one person's conclusions in relation to this article containing invented lists and tables. Moreover, basing on these invented lists as an ultimate source for the "truth" and proper reference may create a very dangerous precedent of invented lists.
Here again, you haven't shown any exact wording you were referring to. I will do it for you. WP says: A coatrack article is a Wikipedia article that ostensibly discusses the nominal subject, but in reality is a cover for a tangentially related biased subject. Here, the nominal subject would be an exact number of casualties (themselves undefined), the biased content is well explained just above, and since you haven't argued against it, I understand that you fully adhere to all the bias present in this article. Moreover, WP rules on COATRACK provide for the following: It is inappropriate to "even out the percentage of bias". These are considered scarves, hats, and gloves, and along with the coats, obscure the coatrack, and are also good candidates for removal.
Unfortunately you haven't revisited the relevant pages. Speaking of fringe theories, Wikipedia says The governing policies regarding fringe theories are the three core content policies, Neutral point of view, Wikipedia:No original research, and Verifiability. Jointly these say that articles should not contain any novel analysis or synthesis, that material likely to be challenged needs a reliable source, and that all significant views published in reliable sources should be represented fairly and proportionately. Should any inconsistency arise between this guideline and the content policies, the policies take precedence. Further, Wikipedia says, while identifying fringe theories, examples include conspiracy theories, ideas which purport to be scientific theories but have little or no scientific support, esoteric claims about medicine, novel re-interpretations of history and so forth. Some of the theories addressed here may in a stricter sense be hypotheses, conjectures, or speculations. Presenting a list, one speaks of exact numbers or defines a nominal common value for all entries, otherwise it's enough to mention in each separate article about each battle the appropriate number guesses and explain where do they come from, however one can't invent a common table out of it. The question whether there is or not a norm regarding the error margin is not an issue, as per WP rules a mere speculation is enough, and we definitely do have much more than mere speculation regarding these numbers and the way they are represented in the table (no differentiation between nationalities, dead, wounded, etc...). Again, you are missing the point, each entry per se is not contested here, although one can easily contest each of the indicated numbers as well. What is proposed to be deleted is the article inventing new tables, organising and sorting questionable numbers in even more, newly invented, questionable lists and tables with obscure or rather no sorting criteria.
Taking into account the presently applicable and herein explained Wikipedia policies, this article should be deleted, since, in accordance with the very same Wikipedia policies, not only it should be deleted, but it shouldn't even be modified, taking into account the present bias, absence of a commonly applicable clear nominal value, hence the very object of the article.Rubikonchik (talk) 09:04, 6 August 2010 (UTC)[reply]
What? Are you serious? You put this up for deletion because you seemed to be against consensus at one of the individual articles? That's pretty cheap.
Your argument seems to be that a list that isn't sourced by a single reference is synth? That is absurd; as the goal for multiple independant reliable sources is one of the conerstones of Wikipedia's referencing policy. Multiple independant sources doesn't make SYTH or OR; and incidentally, several of the sources are indeed referenced multiple times, which would seem to go toward your argument. The distinction here is on the presence of a conclusion, and possibly in the case of lists, comparative judgment. But when it comes to raw data, you can't argue bias except in the case of the individual statistics being flawed. Tying them together is not inherantly biased unless the data is arranged to lead the reader to a specific conclusion. Simply put, this list doesn't violate SYNTH because it doesn't advance a position.
You've made my point for me: the list covers exactly what it purports to. There is no deception or tangental manuevering. You can claim the data is flawed or biased, but that doesn't make the article a coatrack. The lead paragraph specifically states that the numbers given are estimates, not exact numbers (incidentally, any reader with more than two brain cells to rub together wouldn't have to be told this explicitly), meaning that there is no attempt to appear A but discuss B.
Where is the lack of scientific support? Where is the reinvention of history? Where is the conspiracy theory? Where does referenced data because a fringe thoery (besides where you disagree with it)? These estimates are not conjecture or supposition, these are respected historians who arrive at them through study and research. If you dispute the individual estimates, then take it to the talk page and find a better one.
Lastly, please don't refactor the discussion. It's harder to follow that way. bahamut0013wordsdeeds 19:25, 8 August 2010 (UTC)[reply]
What? Are you serious? You put this up for deletion because you seemed to be against consensus at one of the individual articles? That's pretty cheap.

- again, wrong understanding for unknwon to me reasons. This article was put up exactly for the reasons exlained above, and namely violation of three WP rules.Rubikonchik (talk) 14:19, 9 August 2010 (UTC)[reply]

Your argument seems to be that a list that isn't sourced by a single reference is synth?

- yes, exactlyRubikonchik (talk) 14:19, 9 August 2010 (UTC)[reply]

That is absurd; as the goal for multiple independant reliable sources is one of the conerstones of Wikipedia's referencing policy.

- not relavant to the argumentation for deletion nomination of this article, nor is this contested at allRubikonchik (talk) 14:19, 9 August 2010 (UTC)[reply]

Multiple independant sources doesn't make SYTH or OR; and incidentally, several of the sources are indeed referenced multiple times, which would seem to go toward your argument.

- again you are missing the point. Not only each number entry may be contested, as already have done commentators on this very page, but what is being contested is this article containing invented lists by WP users.Rubikonchik (talk) 14:19, 9 August 2010 (UTC)[reply]

The distinction here is on the presence of a conclusion, and possibly in the case of lists, comparative judgment. But when it comes to raw data, you can't argue bias except in the case of the individual statistics being flawed.

- Both presence of conclusion and comparative judgment are clearly present and have been already contested and discussed by edtiors (see above for references). That's exactly how the bias comes in - through obscure, flawed and criteria absent statistics.Rubikonchik (talk) 14:19, 9 August 2010 (UTC)[reply]

Tying them together is not inherantly biased unless the data is arranged to lead the reader to a specific conclusion. Simply put, this list doesn't violate SYNTH because it doesn't advance a position

- Exactly, and the reader does reach a very specific clear conclusion, have a look at the provided above references in respect of the Siege of Leningrad article. As a a matter of fact, a reader does adhere to the advanced by this invented table position.Rubikonchik (talk) 14:19, 9 August 2010 (UTC)[reply]

You've made my point for me: the list covers exactly what it purports to.

- I'am afraid you haven't read this talk page. I'm not sure where I have made your point???Rubikonchik (talk) 14:38, 9 August 2010 (UTC)[reply]

There is no deception or tangental manuevering. You can claim the data is flawed or biased, but that doesn't make the article a coatrack.

- Ok, sounds like that kind of argumentation: "whatever you say, I don't like it and won't agree just because!". That doesn't give any credit to your argumentation. Please refer, like I did, to WP rules.Rubikonchik (talk) 14:38, 9 August 2010 (UTC)[reply]

The lead paragraph specifically states that the numbers given are estimates, not exact numbers (incidentally, any reader with more than two brain cells to rub together wouldn't have to be told this explicitly)

- Again, the articles, lacks the very core, the subject, the very common value, which is even more important in statistics and comparative tables and lists.Rubikonchik (talk) 14:38, 9 August 2010 (UTC)[reply]

meaning that there is no attempt to appear A but discuss B.

-A & B argumentation was not argued here. It's not clear to me what's the purpose of advancing this undiscussed issue? It's about historical speculation here, please see above as explained.Rubikonchik (talk) 14:38, 9 August 2010 (UTC)[reply]

Where is the lack of scientific support? Where is the reinvention of history? Where is the conspiracy theory? Where does referenced data because a fringe thoery (besides where you disagree with it)? These estimates are not conjecture or supposition,

- The lack of scientific support is right there: there are no other sources of such table and lists, unless you consider WP editors established scientists. When an editor refers to this table as an ultimate source to consider a battle "the most" or "one of the most", depending on the number retained in the table, costly in casualties - the reinvention fo history happens right there. Conspiracy theory was just part of the multiple criteria definition, please reread the above given argumentation in this regard. I'm not the only one who disagrees with numbers, there are plenty of editors who disagree with numbers already on this very deletion nomination page. An estimate is a supposition per se, you contradict yourself...Rubikonchik (talk) 14:38, 9 August 2010 (UTC)[reply]

these are respected historians who arrive at them through study and research. If you dispute the individual estimates, then take it to the talk page and find a better one.

- I'm not contesting individual entries. Again, your argumentation in this regard is irrelevant. This is not being discussed here. I'm contesting the made up tables and lists with no clear criteria. However, numerous users have raised the question of individual entries here: indeed, the individual numbers themselves are all subject to contestation.Rubikonchik (talk) 14:38, 9 August 2010 (UTC)[reply]

Lastly, please don't refactor the discussion. It's harder to follow that way.

- I hope it's better to follow this way.Rubikonchik (talk) 14:38, 9 August 2010 (UTC)[reply]

I think at this point, we've come to an impasse, and more kilobytes of conversation probably won't be productive. I disagree with your rationale, find your reasoning to be a flawed interpretation of policy and guideline, and it seems that neither of us shall convince the other, nor probably sway other editors in thier votes. As a last ditch attempt, I encourage you to take a look at a few featured lists, and see if you find your interpretation of SYNTH being applied there. In closing, I will express my gratitude in that you have remained quite civil and approachable/open throughout the process, a behavior that is often sadly lacking on Wikipedia. bahamut0013wordsdeeds 20:22, 10 August 2010 (UTC)[reply]
  • When adding things to the list, I did not add battles with fewer than 10,000 casualties. Seeing as I added probably around 90% of the entries here, that kind of became the de facto minimum. Sorry if I should have stated that in the article, I didn't think of it. – Joe N 15:00, 6 August 2010 (UTC)[reply]

Keep Quite an encyclopedic topic, obviously. In World War 2 armies on both sides did keep track of how many people they had that died in each battle, so I doubt those stats are in question. For the rest, you just reference the most trustworthy source. If necessary you can add in two numbers, saying 20000-25000 if one says 20,000, and another says 25,000. This is what the Wikipedia exist for. Good job to those who created and worked on it. Dream Focus 17:55, 6 August 2010 (UTC)[reply]

Could you please bother to explain how clearly was the WP:SYN misinterpreted? And what about WP:OR, WP:COATRACK and WP:FRINGE?Rubikonchik (talk) 08:59, 10 August 2010 (UTC)[reply]
the assertion that although the data is references the article is not.
the assertion that arranging a list of things in numerical order is SYN; it's just this type of arrangement which is the normal counterexample to SYN.
COATRACK, but what is it a coatrack for? The challenger was not able to specify.
the assertion,confusingly listed as part of Coatrack, that the data was not compiled on a single consistent basis. (I gather this is the meaning of the statistical nonsense about hidden variables--that something else may have affected the casualty count; but this would be true only of an argument that did assign some cause to this, while this is merely a list of numbers based on sources, not an attempt at determining causation or correlation.
FRINGE, but what's fringy about battles having casualties? All the sources listed count as RS--(I cannot judge the Chinese ones, of course) The challenger says " looking at these lists and tables, one clearly sees speculation on multiple historical issues - ". However, nobody else has been able to see any. I can't. Looking at the references, the only thing that gives me a hint is the possible comparison of deaths caused by Stalin and Hitler, and the controversy of the role of the USSR in the eventual victory--and some of the links given in the argument suggest that's what is involved. But nothing in the article goes into this.
In summary, I am not sure whether to consider this a good faith nomination, or whether the nomination is the coatrack. The only other editor supporting deletion is a respected editor, though which some tendency to go off on tangents of his own--I do not question his good faith in the slightest. DGG ( talk ) 04:17, 11 August 2010 (UTC)[reply]
The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.
More