Talk:List of anthropogenic disasters by death toll/Archive 7

Article splitting
In reading back through this talk page, it appears that more people than I previously noted have expressed a desire to break up this article in some way. Also I think some of Jdcooper's arguments for doing so are probably better than I originally thought.

Not that I've ever been implacably opposed to splitting anyhow, but in the absence of any concrete proposals for doing so, I've been happy to keep things as they are.

But if a majority is in favour, I'm prepared for the sake of maintaining goodwill to formally drop my opposition to splitting provided that certain categories remain together on the one page (regardless of what that page might end up being named).

The categories I believe should stay on the same page (listed by their current section names, and including subheadings) are: wars, democide, genocide, manmade famines, and also probably human sacrifice and mass suicide. I also think concentration camps should probably be broken out of its current rather awkward section (of individual massacres, air raids and concentration camps) and listed as a subsection of the genocide heading, since they are obviously dealing with the same topic.

As for the rest, I really have no strong opinion, but would suggest that all the sections relating to accidents should simply go into an article called something like "list of accidents and disasters by death toll", while individual battles, massacres, air raids, terrorist attacks and riots could probably go into their own article called something like "list of individual battles and other mass killings by death toll". I would think murders should go to that page as well, but I believe jd has expressed a desire to keep that particular topic separate and I'm not sure I care enough about the issue to oppose such a split (although quite frankly I think it would only be a matter of time before someone else added that category to the proposed page anyhow).

Anyhow, I think that would be a rational way to split this article, would address the concerns of some (which I don't share) that the article as it stands is "too long", and would also, I think I must concede, be a more rational way of dealing with the diverse subject matter that currently exists here. Comments? Gatoclass 19:23, 24 August 2007 (UTC)


 * If nobody objects to this proposal in the next day or two, I will go ahead with the split as outlined above. Gatoclass 01:29, 30 August 2007 (UTC)


 * I think I've mentioned to you before my work-in-progress murderers article, have a look at it at User:Jdcooper/Most prolific murderers by number of victims (working title) and if you reckon its viable as its own article I'll finish it up in the next couple of days and split out the duplicate section here and put it live. I am aware that the rest of the categories are highly sensitive, and I'm sure throughout our discussions you have picked up on the fact that I don't have the specialist knowledge of historiography that you and Phillip Baird Shearer seem to, so I'm quite willing to cede to your better judgement in the grouping of the other categories. Murderers are the only thing on this page I really know about! (bit of a sad reflection on my intellectual status!) Jdcooper 13:48, 30 August 2007 (UTC)


 * You mean you didn't add your article to the mainspace yet? I thought you did that ages ago.


 * I suggest you add your article to the mainspace whenever you're ready, and then just delete the murderers section in whatever article it ends up in after I split this article and leave a link to your page. Gatoclass 09:02, 31 August 2007 (UTC)


 * Ok I will do that. I'm just finishing the last of the references then I will upload it. Jdcooper 12:12, 31 August 2007 (UTC)

Okay, I have split the article as proposed and removed the relevant content to the new pages. They can now be found at List of battles and other violent events by death toll and List of accidents and disasters by death toll (I misspelled the latter's name in the edit summary unfortunately).

There's still a bit of cleaning up and additional links to be added though. Gatoclass 09:25, 1 September 2007 (UTC)

Proposed reorganization of content
The other main problem we are having at this page ATM is under what headings to list the entries currently listed under "genocide" and "democide", which I think most of us have recognized to be inadequate for various reasons.

So here's my proposal: the main heading needs to be something nice and generic, so I propose simply

Mass killings of noncombatants.

That heading would also have the advantage of dovetailing nicely with the existing "Wars" section.

The subcategories could be

Democides and other systematic violence and Genocides

Democides and other systematic violence would be broad enough to hopefully finally put an end to edit wars over which events do and don't belong in the section. I believe genocides must also be included here, for the simple reason that genocide is a subcategory of democide. To put it another way, you can't list democides exclusive of genocides, because the former includes the latter.

But that doesn't stop us also listing genocides as a separate subcategory, and by way of compromising with PBS I am prepared to drop my opposition to it as a separate listing. It probably makes sense to list them separately anyhow because figures for democide and genocide are obviously different in at least some cases (one obvious example being Nazi Germany for which Rummel lists 21 million democides but only 16 million genocides).

As I said in the previous section, I think we should also probably list individual concentration camps as a subcategory of genocides since they are obviously closely related in content.

Anyhow, I think that is a reasonable compromise proposal which hopefully addresses the concerns of all parties. Of course, we can always change headings later if someone comes up with a better proposal, but for the time being I think this arrangement would work a lot better than what we currently have. Comments? Gatoclass 10:04, 24 August 2007 (UTC)

What is the scholastic definition for "other systematic violence" --Philip Baird Shearer 19:27, 24 August 2007 (UTC)


 * One doesn't need a "scholastic definition" of the term. The term is self explanatory - examples of violence which have an element of planning, or systemization, in them. All the examples in the list clearly fit this description. As an alternative, we could just use "organized" - my original proposal - but I thought "systematic" might be a little more appropriate. Gatoclass 06:04, 25 August 2007 (UTC)

So you want to create a list that is of indeterminate size and is not supported by scholastic research. On top of WP:V, WP:NEU, and WP:NOR, please read Lists, and these two essays Lists in Wikipedia, and Listcruft and see if such a list is withing Wikipedia guidelines and consider if this suggestion lies within the self imposed remit of Wikipedia list. --Philip Baird Shearer 10:02, 25 August 2007 (UTC)


 * So you want to create a list that is of indeterminate size and is not supported by scholastic research.


 * I want to do no such thing. I simply want to create a list of events compared by death toll which is the very purpose of the page. What I want to avoid is to create a host of multiple lists whereby the primary purpose of the page becomes increasingly obfuscated.


 * I have suggested that there should be a heading called Mass killings of noncombatants. If necessary I'd be content to leave it at that. But you are the one who wants a separate list for genocide (although I note that some of the events you have listed under genocide are in fact only alleged genocides). So I have been trying to come up with an inclusive heading for all those "mass killings of noncombatants" which are not genocides, that's all. There's nothing inherently POV about doing that. Gatoclass 10:41, 25 August 2007 (UTC)

OR template
Genocides are not necessarily the killing of just noncombatants -PBS

The membership criteria does not actually say that, it says "where the principal targets" were noncombatants. "Principal" does not mean "just" (ie "only"). It means the main target of the violence was noncombatants.

That is surely self-evident. If genocide is the targeting of a group for its ethnicity, that means that they are targeted whether they are armed or unarmed, young or old, male or female, child or adult. And since the overwhelming majority of members of any ethnic group are obviously not young men capable of bearing arms, how can it be called "original research" to state that genocide principally targets noncombatants? Gatoclass 13:20, 26 August 2007 (UTC)


 * This section [Noncombatant killings] lists violent events resulting in significant mortality of noncombatants, excluding victims of collateral damage from war.


 * Genocide is a subsection of that. " how can it be called "original research" to state that genocide principally targets noncombatants" See the Srebrenica genocide the majority of victims were men (and boys) of combat age --Philip Baird Shearer 14:12, 26 August 2007 (UTC)


 * Doesn't matter, I've altered the preamble again anyhow. The new preamble avoids the problem you mention, and says pretty much all that needs to be said IMO, but I may try to tweak it a little further over the next day or two. Gatoclass 15:35, 26 August 2007 (UTC)

Three Kingdoms War
The estimate for number of dead in the Three Kingdoms War is 40 million. But according to our own article (Three_Kingdoms_War) The total population of all three kingdoms was about 7.6 million people total. (See "Population" in the aforementioned article). —Preceding unsigned comment added by 24.60.106.28 (talk) 06:37, August 27, 2007 (UTC)


 * I did try to confirm this one myself last night but couldn't find a figure, but I completely missed the population figures which appear to make this large estimate untenable, and since this is such a big figure I think the only recourse is to delete the entry for now. I haven't been able to find any estimates of casualties from the net either, so I think this one will probably have to be resolved (if possible) by a trip to the library. Gatoclass 08:56, 27 August 2007 (UTC)

In the Three Kingdoms article. It has stated in the introduction section: ''The Three Kingdoms period is one of the bloodiest in Chinese history. A population census in late Eastern Han Dynasty reported a population of approximately 56 million, while a population census in early Western Jin Dynasty (after Jin re‐unified China) reported a population of approximately 16 million. Even taking into account the inaccuracies of these census reports, it is safe to assume that a large percentage of the population was wiped out during the constant wars waged during this period.'' So before the Three Kingdoms period, China had a population of 56 million. After the Three Kingdoms period, China had a population of 16 million. I know that if we do 56,000,000 minus 16,000,000, it would be original research. So like Gatoclass said, the library is the best solution so far. Oidia (talk) 10:18, 27 August 2007 (UTC)


 * Dammit, I thought I'd seen something like that the last time I checked that article, but I must have missed it this time around. That means in fact there is a major contradiction in that article, because according to the Han Dynasty article, the Three Kingdoms ruled over the whole of China - so how could their combined populations total only 7 million? I think I'll ask for an explanation on the Three Kingdoms page, thanks for pointing that out Oidia. Gatoclass 11:46, 27 August 2007 (UTC)

Great Leap Forward
I think we should remove the Great Leap Forward famine figures from China's political repression in the "Non Combatant Killings" table. Because Mao did not deliberately created a famine. According to the cited source, GLF famine was created by poor agricultural practices combined with natural disasters of drought and flood. So Mao did not have the intention or aim to create that famine. Oidia (talk) 09:08, 1 September 2007 (UTC)


 * The problem is that we have sources who specifically claim that the famine was deliberate - such as RJ Rummel, who has specifically labelled the famine as a democide (deliberate mass murder by government). Though I personally also feel his categorization is bullshit, because he's a reliable source there is really no way to remove his estimate. Gatoclass 10:48, 1 September 2007 (UTC)


 * And sometimes a reliable is not very neutral. I feel his catagorisation is crap too. I really think that having Rummel classifying the famine as "a delibrate act of the government" is violating the policy of NPOV. Or maybe I can go and find another reliable and "more neutral" source. Oidia (talk) 11:37, 1 September 2007 (UTC)


 * NPOV doesn't apply to reliable sources, it only applies to Wiki editors. NPOV means that we can only report what reliable sources have said, not base article content on our opinions about what they have said - even when we're sure we're right.


 * Rummel is obviously biased but as a reliable source we have no way of excluding his estimates. But the good news is that because we are reporting a range, we get to report the other side of the coin too. Gatoclass 13:30, 1 September 2007 (UTC)

OK. Oidia (talk) 23:39, 1 September 2007 (UTC)

Yeah I just had a thorough read through WP:NPOV and WP:V and you're right. Rummel and the PRC government are both creditable sources so we insert both high and low estimates. Oidia (talk) 00:25, 2 September 2007 (UTC)

Noncombatant killings
Non combat killings seems to be a random list of unsubstanciated facts. For example why is there a list of "Executions 	England 1509 1547 Henry VIII" why not one for every king in every country that has ever lived? Why not one under every presidential term of every country that has had a president and one or more prisoners are killed? Why does it include some slave trades which are not to do with noncombatant killings? What do the "Reign of Terror, France, 1793,1794" and Witch hunts, Europe, 1450-1700" have in common? What is significant about the period from "1450-1700" where no witches in Europe killed before or after that date? Why only witches in Europe were no withes else where killed? Similar criticisms can be leveled at all the items in this list.

The list is unencyclopaedic and should be deleted. --Philip Baird Shearer 07:51, 3 September 2007 (UTC)


 * Henry the VIII executions are there basically because IIRC they were related to religious/political persecution (rather than executions for criminal conduct). But perhaps that should be made more clear in the entry. The slave trades are there because they resulted in "significant numbers of noncombatants" being killed, ie on the middle passage, the "seasoning camps" etc. Witch hunts = persecution of minorities.


 * But if you think the membership criteria are still too vague, I guess we could refine it further. I'd just prefer to try and keep it as open-ended as possible though, as we may find estimates for other mass deaths that are not easily pigeonholed, or else decide to expand the list in other ways. Gatoclass 08:48, 3 September 2007 (UTC)


 * Okay, I've changed the membership criteria to better reflect the contents of the list. Gatoclass 10:39, 3 September 2007 (UTC)

You can try to rationalise the list here on the talk page, but it does not get away from the fact that it is a random collection of events with no clear definition as to why these events should be chosen over many many others. It fails almost every test that Wikipedia has for such lists. See for example Lists in Wikipedia "When creating new lists, think of the reader: Does the list add value? Is the list's criteria so open-ended as to welcome infinite results or abuse? ... Is there a reason for creating the list other than "it would be cool" or "just for the hell of it"? Lists should enhance the encyclopedic value of content rather than diminish it." --Philip Baird Shearer 18:15, 3 September 2007 (UTC)

The new wording "This section lists campaigns either aimed at or resulting in significant mortality of noncombatants, excluding victims of collateral damage from war." does not help as most of the entries were not campaigns and as the list is still a a random collection of events with no clear definition as to why these events should be chosen over many many others. --Philip Baird Shearer 08:29, 4 September 2007 (UTC)


 * So far, the listed events "appear" to be randomly selected. And that's because the list is incomplete. If you can find more sourced info and figures to add to the list, then please do. For now the list should stay. Oidia (talk) 11:53, 4 September 2007 (UTC)


 * The reason I used the word "campaigns" is to try and alert editors to the fact that the events listed here are not one-off isolated instances like individual massacres and so on, but established patterns or conditions in which multiple acts of violence occurred. "Campaigns" is not entirely satisfactory I agree, and I'm still trying to think of a better alternative, but I'm certainly open to suggestions. "Regimes" would be more correct, but then some people would probably interpret that to mean ruling regimes rather than "class of conditions" or "established pattern", and I'd like to find a word or phrase that is less ambiguous.


 * As to the list being a "random collection of events" I have to disagree. It's basically just a list of death tolls which some scholars have characterized as substantially the result of deliberate killing or murder of noncombatants. Gatoclass 18:33, 4 September 2007 (UTC)

Oidia, I think it it is a random open list -- as the definition is an open -- ( "When creating new lists, think of the reader: Does the list add value? Is the list's criteria so open-ended as to welcome infinite results or abuse?") If it is not a random list then explain the connection between these events: Russian programs, Witch hunts in many European countries, executions by Henry VIII and the Political repression by Fidel Castro; that would not include hundreds of thousands of other events.

Also Gatoclass you write "Henry the VIII executions are there basically because IIRC they were related to religious/political persecution (rather than executions for criminal conduct). " who was executed for religious reasons? AFAICT they were executed for political/criminal reasons if you include that one you will have to include all regime executions.

Further what are we using as the criteria for including events like "Harrying of the North", given that William did not break any international laws during his invasion of England, why does this event get into this list and not every pacifying campaign in all of history when one of more non-combatants die? Again if we include all of them then such a list it will be far to large for any one article and of little use as an encyclopaedia article. --Philip Baird Shearer 08:13, 7 September 2007 (UTC)


 * So are you suggesting that we should eliminate the entire non-combatant killings section all together? And I'm still unhappy that I spent 4 hours putting the War death tolls into a sortable table and someone deleted a huge part of the table. Oidia (talk) 09:55, 7 September 2007 (UTC)


 * Philip did that, and I feel your pain, LOL. I thought about restoring it, but to avoid acrimony decided to leave things as they are right now until I can organize myself to track down some bulletproof references for some of these entries. It's just that I've been too busy with other things to take a trip to my local library recently. I intend to restore the entries once I have the references though. Gatoclass 20:24, 7 September 2007 (UTC)

if you include that one you will have to include all regime executions - PBS

Yes, I realize there are *potential* problems with such a list but until we have an *actual* problem then I see no reason to start deleting entries. The Henry executions are there because they are described as a fearful series of political executions in the Henry article. The point is that 72,000 executions at this stage of the game appears to be an exceptional number. If it turns out, for example, that many English kings executed similar numbers, then there is probably no point in listing such events, at least not in this list.

what are we using as the criteria for including events like "Harrying of the North", given that William did not break any international laws during his invasion of England - PBS

Okay, first of all I think part of the problem is that you continue to see this list through the prism of international law. I assume you have come to this page with that mindset from your oversight of the "genocides in history" page, where international law is obviously an important factor in deciding what does and does not merit listing there.
 * No, I am very aware that there is a dichotomy between an act of genocide and the crime of genocide. But that is OK because genocide is well defined (It was I who put together the article on genocide definitions), and scholars have discussed what in the past was a genocidal act and what is not. The problem you have here is that the distinction you are making between combatant and non-combatant and what combatants could do to non-combatants under the rules of war as they were at the time is clearly confused, and putting together such a list is full of non NPOVs. For example when William first landed near Hastings on the south coast he sent his men to slaughter the local inhabitants because he knew that as Harold was their Lord he was duty bound to respond quickly to such attacks. As there were no international laws against this strategy to argue that such a strategy was not within the laws of war is a little difficult yet that is what we seem to be doing if we have a lead that says "This section lists campaigns either aimed at or resulting in significant mortality of noncombatants" --Philip Baird Shearer 23:39, 7 September 2007 (UTC)

But this is a very different list that does not rely on such a stringent yardstick. Basically this is a list of killings that are widely regarded as unjustifiable or reprehensible in some way, either because it is self-evidently the case or because we have scholars who have essentially characterized the events in such a way. By way of example, we hardly need a scholar to label the killings during the India-Pakistan partition as mass murder because it is self evidently the case that that is what occurred.
 * "widely regarded as unjustifiable or reprehensible in some way" very difficult to prove without a WP:SYN. Yes you do need a scholar to label the killings between Indians and Pakistanis (Not India and Pakistan) as mass murder otherwise it is WP:SYN: Lots of people died therefore it is mass murder so we should include it in this list is a "Synthesis of published material serving to advance a position". --Philip Baird Shearer 23:27, 7 September 2007 (UTC)

As for the inclusion of "events like Harrying of the North", I confess I am in still two minds about whether or not such "pacifications" should be listed here, listed separately, or not listed at all. It gets back to the earlier conversation about famines. If one lists every famine that occurred "through human agency", doesn't that mean you have to end up listing practically every war since famine is a common byproduct of armed conflict? So yes, I am a little concerned that pacification operations could potentially overwhelm the list.
 * Exactly but that will continue to happen unless you define what the list is see below. --Philip Baird Shearer 23:27, 7 September 2007 (UTC)

However, my point as with the Henry VIII listing above is that such events are certainly not overwhelming the list right now. Nor do I think, realistically, that they are ever likely to do so. However, if it turns out I'm wrong about that, I think we could always try listing them under a separate heading, although again, coming up with clear criteria for membership might prove a bit tricky. Gatoclass 20:24, 7 September 2007 (UTC)
 * If this is listed then every regime that has ever executed one or more persons should be listed if the list is to be comprehensive and not non NPOV. --Philip Baird Shearer 23:27, 7 September 2007 (UTC)


 * If we go back a month to before I started to edit this section 20 August 2007 it was called Genocides and Democides. Now that was name that could be justified because, there is an article on genocides in history and it is fairly easy to look up democides. Once the list was split into two then then all that should be left in this section were demodices. But splitting the list into two (by taking out the genocides) did not leave a list of democides it left a rat-bag of items and it clearly shows that the previous list was not, as the list titled implied a list of "Genocide and Democide". If the rat-bag it is meant to be a list of democides then make it such a list, but instead Gatoclass you have tried to retrofit a description to a rat-bag list. Gatoclass IMHO your new definition has not changed the fact that the list that does not fit Wikipedia criteria because "When creating new lists, think of the reader: Does the list add value? Is the list's criteria so open-ended as to welcome infinite results or abuse?" --Philip Baird Shearer 23:27, 7 September 2007 (UTC)

Okay, the problem with "democide" as a heading, as I've noted before, is that only one scholar uses the term. This means that if we are going to be legalistic about it, the only scholar whose estimates we could list under such a heading would be Rummel himself, which would clearly in my opinion violate WP:UNDUE.

If we are not going to be legalistic about it, then we could list others' estimates under that heading, but then if we are not going to be legalistic, we are essentially facing the same problem which you have highlighted, which is that we would no longer have a clear criteria for including those other estimates.

That's why I've looked for a broader alternative title that can include both Rummel's estimates and estimates from other scholars. Now as I was writing the previous post, it occurred to me that probably the most straightforward alternative heading would be simply "mass murder", because that is effectively the kind of events we are listing. That is to say, this is a list of events in which large numbers or a siginficant proportion of the dead were either innocent or else unarmed and defenceless noncombatants, deliberately (as opposed to inadvertently) killed by another group.

I think that all the events listed in this category can very reasonably be characterized as mass murder, or at least unjustifiable homicide. I don't think we need a scholar to spell out for us that when an innocent person is deliberately killed by someone else, that is an act of murder. Nor do we need someone to spell out that killing someone for their political or religious beliefs is an unjustified killing. So I think there *is* a quite obvious criteria for membership in this list, everyone except you appears to see that, and no-one but you is complaining that the list is too open ended.

The fact that we are yet to agree on or come up with a precise wording for the membership criteria outlined above, is another issue altogether. But perhaps if you were prepared to engage a little more constructively, we could do this fairly readily.

If this is listed then every regime that has ever executed one or more persons should be listed if the list is to be comprehensive and not non NPOV - PBS

No, you don't need "every regime" to be listed in order to satisfy "NPOV". That is sheer nonsense. Wiki is rife with lists that are incomplete, and this repeated claim of yours that lists are somehow invalid just by virtue of being incomplete is in my view quite idiosyncratic to you personally and not at all supported by Wiki policies or practice. So I wish you would stop regurgitating this point, it's just a distraction. Gatoclass 02:01, 8 September 2007 (UTC)


 * I agree with Gatoclass. If you want total NPOV, then go and restore deatht toll from wars back to the original very long list. Oidia (talk) 07:22, 8 September 2007 (UTC)

It is not that all lists are complete -- true-- but list should be defined in such a way as that they can be completed. That is the difference between a listing in "Wars and armed conflicts" more than a million, (half a million or what ever the limit is set to) but not all wars which would be a very very long list and difficult if not impossible to complete + lots of points of view problems (eg was the (Dirty War A war?). For example see this version of military occupations in December 2004 Military occupation and List of military occupations. It would be impossible to create a list of all Historical occupations and because the definition at the top said at that time "The Hague Convention of 1907 and the customary laws of belligerent occupation govern belligerent occupation in international law." any historical occupations before 1907 under the Hague Conventions run into definition problems. Compare that with the articles today which use much tighter definitions (Military occupation List of military occupations) and I think you will see what I mean and I hope agree that the list is much more useful and not so open to POV issues. There can also be selective list based on who can be bothered to add to an open ended list (in the case of military occupations because of the intensity of feelings on the issue the Israeli Palestinian conflict tended to be over represented in the list of occupations. There can also be a bias by choosing a definition that skews a list for example choosing occupations since GCIV (1949) also tends to skew the list in a similar manner.

Another example is list of war crimes. It used to cover the whole of history but is now only from 1907, because before Hague Conventions (1899), what constituted a war crime is much more difficult to define. LWC is nowhere near complete but it is a definition that stands a chance of being completed one day, but a list of all war crimes over all of time never could be.

I put it to you that the current list is not one that can ever be near completion and the definition is not succinct enough to be encyclopaedic. --Philip Baird Shearer 10:21, 8 September 2007 (UTC)


 * I can see there would be obvious problems trying to list military occupations or war crimes before a certain date, because such things had little meaning before then. They are, effectively, modern concepts. But this isn't the case for a death toll, which is simply a statistic. So the only question is in what way to best sort or order these statistics for the benefit of the reader. I mean, we could just have a big category called "Miscellaneous manmade death tolls", which would be perfectly legitimate, it just wouldn't be very helpful to the reader. So the idea is to sort the tolls in some sort of logical way that will make the list a bit more accessible (not to mention more interesting and informative). There are actually probably innumerable ways we could go about this that would not violate any Wiki policies or guidelines. Really, all we need to do is exercise a little common sense.


 * As for your charge that "the current list is not one that can ever be near completion" - I have to disagree. There are only a limited number of events with notable death tolls which scholars have chosen to study, let alone make estimates for, so this is not an open-ended set at all. In fact, it's already proving to be quite a challenge to find more entries, there have been very few new ones added to this list in recent months. Gatoclass 11:55, 8 September 2007 (UTC)

An Shi Rebellion killed 30 million? Give me a break.
This claim is absurd. To think that the An Shi Rebellion killed 1.5 times as many people as the Taiping Rebellion - by far the bloodiest civil war in history - is preposterous. I know where the 30 million number comes from - it comes from the difference between the Tang imperial census counts before and after the war. However, this difference could easily be accounted for by:

A) the fact that the area (and thus the population) of the Tang Empire contracted greatly over the course of the rebellion,

B) people moving out of Tang-controlled areas because of the rebellion (note the high ration of refugees to casualties in all modern conflicts), and

C) deaths due to diseases that followed the war (like the Spanish Flu that followed WW1 and killed 80 million).

I'm highly skeptical of these ancient body counts in general, but this one to me seems just egregious. To think that this rebellion killed as many people in 7 years as the Mongols did in all of Asia in 70 is just beyond my suspension of disbelief. —Preceding unsigned comment added by 68.40.186.184 (talk) 06:19, 8 October 2007 (UTC)


 * Nice original research buddy! Very logical and makes a lot of sense. All you need to do now is to find us a source to verify your claims, thanks. Oidia (talk) 07:32, 8 October 2007 (UTC)


 * Here's my source. —Preceding unsigned comment added by 68.40.186.184 (talk) 04:25, 9 October 2007 (UTC)

Nice work!! The source clearly says that the "An Lushan Revolt" has a death toll of 36 million. So, yeah... Oidia (talk) 23:53, 10 October 2007 (UTC)


 * Yes we have three reliable sources there that give a total population decline of 33 to 36 million. So that pretty much settles it. Gatoclass 01:16, 11 October 2007 (UTC)


 * Reliable sources is stretching it a bit. They're all based on the same original ancient figures that are far from reliable. Who seriously believes Herodotus' claim that Xerxes commanded an army of over two million? Ancient sources when it comes to statistics should be taken not just with a grain of salt, but with a whole shaker of the stuff. The Taiping Rebellion occurred at a time when the population as a whole was many times larger than during the Tang dynasty and yet most sources put the death-toll for it at around 20 million. To accept the figures for the An Shi Rebellion we would have to assume that the killing in the war was twenty-four times as intensive (when the relative populations and durations are taken into account) as the Taiping Rebellion, despite the death toll for the latter itself being extremely high. Also the 36 million figure would imply the An Shi Rebellion was drastically more deadly than either the Rwandan or Cambodian genocides, the two most intensive killings for which we have a reasonably reliable range of figures on the totals killed. This is despite a much more dispersed population during the Tang dynasty as well as a lack of modern communication and transport. In short I feel not merely confident, but certain, that the figure of 36 million is an utterly ludicrous exaggeration.  —Preceding unsigned comment added by 203.212.153.155 (talk) 04:29, 14 March 2008 (UTC)


 * If you're talking ratio's the 'depopulation' of the Australian Aboriginals took their population from a possible 750,000 to 90,000 in 100 years. That outweighs your 'most intensive killings' by a fair margin. —Preceding unsigned comment added by Senor Freebie (talk • contribs) 05:58, 1 January 2009 (UTC)


 * Fine, I'm not saying we shouldn't have it on the page. History is history. I'm just saying I find it much more likely that the "population decline" does not reflect the actual number of people killed in this war (or even the amount by which the population actually shrunk). If you think it's accurate, more power to you - no one will ever be able to prove you wrong. —Preceding unsigned comment added by 68.40.186.184 (talk) 05:39, 13 October 2007 (UTC)

NPOV and sources
From the history of the artice:
 * rv: please justify on the talk page why a neutral point of view should not be included and why you removed sources.

--Philip Baird Shearer 17:03, 8 October 2007 (UTC)


 * I'll be happy to explain that for you. The subject of this article is "List of wars and disasters by death toll". It's supposed to be just a simple, straightforward list of death toll estimates for various events, with links provided for people who want to find out more about any particular event. Lists are not designed to provide detailed explanations of things, that's what articles are for. If you are going to start including extended notes for table entries, it opens the door for others to start including extended notes for their favourite entries, and pretty soon we no longer have a list at all, but a dog's breakfast of a page that is halfway between a list and a text article.


 * Also, you are the one who was complaining before that the page was "too long", extended notes in the tables are obviously going to make the table listings unmanageably long and the page unnavigable.


 * I have no objection to the addition of extended notes per se if you want to include some, but they should go in the footnotes not in the tables themselves. Gatoclass 18:15, 8 October 2007 (UTC)

If there is to be tables that contain controversial subjects such as alleged genocides then there should be a WP:NPOV in the text of the article not in the footnotes. Otherwise we should strip out all the alleged genocides leaving only those for which there is near universal acceptance (where the other view can be dismissed as undue weight) -- Which comes down to those where people have been found guilty in a court of law. But in cases like the Australian genocide debate or the Holodomor famine then clearly NPOV demands that both sides of the argument be equally represented, or such entries should be removed and the proponents are not in the majority and to exclude the other POV is a breach of the WP:NPOV policy. --Philip Baird Shearer 18:35, 8 October 2007 (UTC)


 * I agree that the status of some of these events needs clarification, so I've kept your notes and refs by moving them from the tables to footnotes, while substituting a word or two of explanation in the Notes column describing the event status as indicated. Hopefully that will address both our concerns. Regards, Gatoclass 19:06, 8 October 2007 (UTC)

It addresses neither. The number of words will be the same if the explanation is given in a footnote and by placing them in a footnote you are not giving the two POVs the same weight that WP:NPOV requires. The thing is that for tables about wars although there may be a dispute about the number of dead that there was a war is not usually under dispute. But with things like alleged genocides and genocides, there is a need to present a balanced point of view as a substantial minority or even a majority will not agree that it was a genocide. Hiding one point of in the footnotes while presenting the other in a table is not fulfilling the requirements of the Wikipedia NPOV policy. --Philip Baird Shearer 19:15, 8 October 2007 (UTC)


 * The number of words is the same, sure, but they are not cluttering up the tables and that's the difference.


 * As for your second point, I agree that the alleged genocides weren't being sufficiently differentiated from the generally well established ones, and it's been bothering me for quite a while. I thought my edit was a good compromise but since you disagree, I've followed your suggested alternative and removed the offending entries until further notice. Gatoclass 19:45, 8 October 2007 (UTC)

If you are going to do that then you need to remove far more. The only ones that do not hold a significant divergent POVs are the Holocaust, and two more tried in international courts since 1990 -- See the Genocide section. There are also a few in the Genocide under municipal laws article like those in Brazil and a few more that could be included. But I think you are making a mistake, because editors will come here and add in incidents like the Armenian genocide, even when there substantial disagreement over whether it was a genocide or not. I think it is much better to include them and put in a balanced POV--Philip Baird Shearer —Preceding signed but undated comment was added at 19:58, 8 October 2007 (UTC)


 * Okay, I accept that you have a legitimate concern but lists are not the place for detailed discussions, that's what the links are there for. There's a middle ground between my original minimalist position and your maximalist one, so I've edited the entries accordingly. Gatoclass 04:46, 9 October 2007 (UTC)

There is still not enough detail in the Notes section on many of them. As they are listed it implies there was a genocide, there has to be enough information in the Notes so that the alternative POV is as clearly put across otherwise it is not a balanced POV. --Philip Baird Shearer 08:48, 9 October 2007 (UTC)


 * As I've said, I'm not prepared to accept long rambling notes in the tables. I am however prepared to make other changes in order to accomodate your views. To that end, I have pasted part of the intro from the "genocides in history" page which makes it clear these entries should not be taken at face value. Gatoclass 09:34, 9 October 2007 (UTC)

"I'm not prepared to accept long rambling notes in the tables." please read Ownership of articles. --Philip Baird Shearer 09:47, 9 October 2007 (UTC)


 * Thanks for the gratuitous advice, I'm well aware of WP:OWN. I am simply stating my position that there is a limit to which I am prepared to compromise on this issue. If you want to insist on your POV, then I will expect you to demonstrate a consensus in support of your view. Gatoclass 09:53, 9 October 2007 (UTC)

See WP:NPOV there can be no compromise on this issue:
 * All Wikipedia articles and other encyclopedic content must be written from a neutral point of view (NPOV), representing fairly and without bias all significant views (that have been published by reliable sources). This is non-negotiable and expected on all articles, and of all article editors.

The view of the Government of Pakistan over the issue of the killings in Bangladesh is one of the "a significant views" and removing their view is a breach of WP:NOR.

The removal of information that is not sourced is covered by WP:PROVEIT, no consensus needs to be found for the removal of unsourced information, but you can not stop the development of an article by claiming "I will expect you to demonstrate a consensus in support of your view" when the information is sourced and it is added to fulfil the WP:NPOV policy. --Philip Baird Shearer 11:17, 9 October 2007 (UTC)


 * In regards to the Pakistan government report, here's what WP:RS has to say: A reliable source is a published work regarded as trustworthy or authoritative in relation to the subject at hand. Evaluation of reliability will depend on the credibility of the author and the publication, along with consideration of the context.


 * Now if you can explain to me how a Pakistani government report ostensibly examining the consequences of its own political repressions can be regarded as in any way "trustworthy" or "credible" - particularly given that its findings are massively at odds with those of every independent source which has investigated the matter - by one or two orders of magnitude - then I might concede your point.


 * you can not stop the development of an article by claiming "I will expect you to demonstrate a consensus in support of your view" when the information is sourced and it is added to fulfil the WP:NPOV policy


 * Our current dispute is not about NPOV per se, it's a disagreement about how much information is necessary to establish NPOV. I feel there is more than enough information already extant in the section to fulfill the requirements of NPOV, and that any additional information is superfluous. I also think your edits were adding little if any additional information in any case. What they were doing though was cluttering up the tables in such a way as to detract from the object of presenting information in a clear, accessible and attractive manner. Gatoclass 13:48, 9 October 2007 (UTC)


 * Since you have reverted once again, I am going to start an RFC to try to resolve the matter. Gatoclass 11:27, 11 October 2007 (UTC)

Pakistan government report
''Two users cannot agree as to whether a report commissioned by a government under a given set of circumstances constitutes a reliable source under Wikipedia guidelines. The circumstances are as follows: In 1971, Pakistani troops were alleged to have massacred large numbers of Bangladeshis during the partition of Pakistan and Bangladesh. The Pakistani military junta which ruled Pakistan at the time commissioned a judicial report which found 26,000 Bangladeshis had been massacred by Pakistani troops. The findings of the report were at odds with every independent investigation which found the number of killings to be between one and two orders of magnitude higher.''

''Should the government commissioned report in these circumstances be regarded as a reliable source for the purpose of setting the low estimate for the death toll in a list? Or should it be ignored and the lowest estimate from an independant source be used instead? Here is the discussion so far by the two users who disagree. Please add your own comment. 11:27, 11 October 2007 (UTC)''

See WP:NPOV there can be no compromise on this issue:
 * All Wikipedia articles and other encyclopedic content must be written from a neutral point of view (NPOV), representing fairly and without bias all significant views (that have been published by reliable sources). This is non-negotiable and expected on all articles, and of all article editors.

The view of the Government of Pakistan over the issue of the killings in Bangladesh is one of the "a significant views" and removing their view is a breach of WP:NOR.

The removal of information that is not sourced is covered by WP:PROVEIT, no consensus needs to be found for the removal of unsourced information, but you can not stop the development of an article by claiming "I will expect you to demonstrate a consensus in support of your view" when the information is sourced and it is added to fulfil the WP:NPOV policy. --Philip Baird Shearer 11:17, 9 October 2007 (UTC)


 * In regards to the Pakistan government report, here's what WP:RS has to say: A reliable source is a published work regarded as trustworthy or authoritative in relation to the subject at hand. Evaluation of reliability will depend on the credibility of the author and the publication, along with consideration of the context.


 * Now if you can explain to me how a Pakistani government report ostensibly examining the consequences of its own political repressions can be regarded as in any way "trustworthy" or "credible" - particularly given that its findings are massively at odds with those of every independent source which has investigated the matter (by somewhere between one and two orders of magnitude) then I might be prepared to concede your point. Gatoclass 11:27, 11 October 2007 (UTC)


 * For reference, the article on the event in question: 1971 Bangladesh atrocities. Gatoclass 11:39, 11 October 2007 (UTC)

The problem that you are running into with this debate is that the the number of 3 million is equally political as White, Matthew (Death Tolls for the Major Wars and Atrocities of the Twentieth Century: Bangladesh]) makes clear the 3 million is the "The official estimate in Bangladesh" should that number be removed from the top of the range? After all we have a paper presented that the numbers may have been exagerated for political reasons (Bose, Sarmila, "Anatomy of violence: An Analysis of Civil War in East Pakistan in 1971", published October 8, 2005 in the Indian Journal, Economic and Political Weekly). There is also the Office of the Historian of the United States State Department held a two-day conference in late June 2005 on U.S. policy in South Asia between 1961 and 1972. In which is was reported "Almost all scholars agreed that the real figure was somewhere between 26,000, as reported by the Hamoodur Rahman Commission, and not three million, the official figure put forward by Bangladesh and India."(U.S Department of State South Asia in Crisis: United States Policy, 1961-1972 June 28-29, 2005, Loy Henderson Auditorium, Tentative Program. Anwar Iqbal Sheikh Mujib wanted a confederation: US papers, The Dawn, July 7, 2005 -- this same new article by Anwar Iqbal but headlined US State Department's declassified documents appeared in Financial Express, 16 December 2005.) It is better to include both the upper an lower numbers from both governments until there is general agreement on the issue. --Philip Baird Shearer 15:50, 11 October 2007 (UTC)

The number of 3 million probably is exaggerated. But it isn't only the Bangladesh government which has used the figure. White lists the sources as follows:

WHPSI: 307,013 deaths by pol.viol. in Pakistan, 1971. D.Smith says 500,000 S&S: 500,000 (Civil War, Mar.-Dec. 1971) 1984 World Almanac: up to 1,000,000 civilians were killed. Hartman: 1,000,000 Bengalis B&J: 1,000,000 Bengalis Kuper cites a study by Chaudhuri which counted 1,247,000 dead, and mentions the possibility that it may be as many as 3,000,000. MEDIAN: 1,000,000-1,250,000 Porter: 1M-2M Rummel: 1,500,000. Eckhardt: 1,000,000 civ. + 500,000 mil. = 1,500,000 (Bangladesh) Harff & Gurr: 1,250,000 to 3,000,000 The official estimate in Bangladesh is 3 million dead. [AP 30 Dec. 2000; Agence France Presse 3 Oct. 2000; Rounaq Johan: 3,000,000 (in Century of Genocide: Eyewitness Accounts and Critical Views, Samuel Totten, ed., (1997)) Compton's Encyclopedia, "Genocide": 3,000,000 Encyclopedia Americana (2003), "Bangladesh": 3,000,000

So the lowest figure we have from White's collection of sources is 307,000. The highest is three million, cited by a number of sources, and then we have a slew of estimates at around the 1 million mark, with White quoting 1 mil as the "median" figure.

As for this quote of yours from the US Dept of State, it is not supported by the link you provided, which is simply an itinerary of events held at a conference. Where exactly does the quote that "Almost all scholars agreed that the real figure was somewhere between 26,000, as reported by the Hamoodur Rahman Commission, and not three million, the official figure put forward by Bangladesh and India" come from? The quote also doesn't make sense - the "real figure is "somewhere between 26,000...and not 3 million"? What the heck is that supposed to mean? And who actually made this nonsensical statement? Gatoclass 01:12, 12 October 2007 (UTC)


 * Okay, I overlooked the second link you provided. The statement looks like a blog post, from someone whose credentials are not supplied, on an internet site that appears to be a subsidiary of a Pakistani newspaper. As a reliable source, it's of marginal value - while the statement itself is so vague as to be utterly meaningless. Gatoclass 01:30, 12 October 2007 (UTC)
 * It is not a blog as you put it (and why do you put it that way, would you do the same for a BBC news item?), it was a new item on the date given. Notice that I have also provided a citation that the same article appeared in the Bangladeshi Financial Express under the byline US State Department's declassified documents on 16 December 2005.


 * If it's a "news item", it's remarkably amateurish. It looks more like the casual blog of someone on the paper's staff to me.


 * However, that is beside the point. The statement itself is not only illogical, but devoid of useful information. Assuming the writer meant "almost all scholars agreed that the real figure was somewhere between 26,000...and substantially less than three million" that statement would be true if "almost every scholar" believed that 2.9 million were killed. It reads to me like the disingenuous attempt of a pro-Pakistan writer to say something about the event that appears to mitigate the scale of the atrocities, while actually saying nothing whatever. Gatoclass 09:28, 12 October 2007 (UTC)
 * How do you know he is a pro-Pakistan writer as the article also appeared in a news paper in Ban Financial Express newspaper published in Bangladesh? --Philip Baird Shearer 09:58, 12 October 2007 (UTC)
 * The same article also contains
 * ''They said that the original figure was close to 300,000, which was wrongly translated from Bengali into English as three million.
 * Shamsher M. Chowdhury, the Bangladesh ambassador in Washington who was commissioned in the Pakistan Army in 1969 but had joined his country's war of liberation in 1971, acknowledged that Bangladesh alone cannot correct this mistake. Instead, he suggested that Pakistan and Bangladesh form a joint commission to investigate the 1971 disaster and prepare a report.
 * --Philip Baird Shearer 09:22, 12 October 2007 (UTC)


 * That is certainly a more worthwhile comment, because it comes from a reliable source. However, note that Chowdury is only saying the figure "was close to 300,000", which is still far more than the 26,000 claimed by the Pakistan report. Also according to other articles on this conflict, Chowdury was severely criticized in his home country for the remarks he made at the conference. So again, this does not support your case. Gatoclass 09:28, 12 October 2007 (UTC)


 * Another person mentioned in that article is "Prof Bose, a Bengali herself and belonging to the family of Netaji Subhash Chandra Bose, emphasized the need for conducting independent studies of the 1971 conflict to bring out the facts." The paper by Sarmila Bose (a Harvard-educated Indian academic related to the Indian Netaji Subhash Chandra Bose) presented was published as "Anatomy of violence: An Analysis of Civil War in East Pakistan in 1971", published October 8, 2005 in the Indian Journal, Economic and Political Weekly. As the article makes clear this work has been criticized in Bangladesh and her research methods have been attacked as shoddy and biased (Drishtipat website In this website, we tried to collate information concerning this paper, (it did included Sarmila Bose’s original paper but says that was removed for legal reasons), relevant Bangla articles and rebuttals of Bose’s paper. Drishtipat is a non-profit, non-political expatriate Bangladeshi organization.).
 * It does not matter for the purposes of the range in the Wikipeda table that the minimum and maximum figures are disputed. All we have to do is include them, as we are not able to judge which is the more correct source given the academic cat fight that is going on over the figures. To include one without the other shows a bias and a judgement on the behalf of Wikipedia editors on the correctness of the figures and as WP:V says "The threshold for inclusion in Wikipedia is verifiability, not truth." --Philip Baird Shearer 09:51, 12 October 2007 (UTC)


 * To include one without the other shows a bias and a judgement on the behalf of Wikipedia editors on the correctness of the figures


 * If it was only the Bangladeshi government which cited the 3 million figure, I would agree. But we have numerous independent sources which cite a figure of anywhere from 300,000 to 3,000,000. By contrast, we have just one report that cites a figure below 300,000, which is the official Pakistani government report which cites a figure ten times lower than the lowest figure given by any other source.


 * Just because a report is issued by a government is no reason to assume it is reliable - on the contrary, since governments can always be expected to act in a self-interested manner, we should always be wary of such reports. As the policy states, a source must be "trustworthy" and "credible" in relation to the context, and this is one context in my opinion where a government report should be regarded as not at all trustworthy. Gatoclass 10:18, 12 October 2007 (UTC)
 * Just as you have your opinion, providing the numbers given cite a source, readers can make their own judgement on whether the extremities of the range presented are from reliable sources. It seem to me that from the evidence given both figures are incorrect and that the number lies somewhere in between, but I am not going to insist that we (Wikipedia editors) make such a judgement by suppressing one or more of the sources. --Philip Baird Shearer 11:40, 12 October 2007 (UTC)


 * After reading the above comments, I personally think that the government report is a reliable source. (PBS, your comments are rather long and took me a while to read) Oidia (talk) 13:05, 12 October 2007 (UTC)

In principle I think it's a mistake to treat the testimony of the perpetrator of a crime as reliable. It could lead to all sorts of ludicrous results. We may still get some additional opinions from the RFC, but if not, I'm putting it on notice that if this does look like becoming a larger problem I may revisit the issue. In the meantime, I've added an explanation of the discrepancy in figures to the footnote and that will have to do for now. Gatoclass 13:49, 12 October 2007 (UTC)


 * In this case as far as I can tell whether a genocide was committed or not is debatable. You seem to be stating that Pakistan committed a crime. If so what was the crime? What do you mean by "I'm putting it on notice that if this does look like becoming a larger problem I may revisit the issue" --Philip Baird Shearer 16:12, 12 October 2007 (UTC)


 * Good question. I might well ask what you mean by removing the footnote I just added explaining the large discrepancy between the two figures - after I explained my reasons for doing so in only the previous post. I might also ask why it is that you feel entitled to continually revert the work of other editors even after they have expressed objections to your edits, and before a working consensus has been established.


 * But in regards to your question above, I mean exactly what I say - I will raise this matter again if I think it's becoming a problem. Why are you implying that I have no business doing so? Gatoclass 17:40, 12 October 2007 (UTC)


 * You have not answered the first question I asked what is the crime that you think Pakistan committed? --Philip Baird Shearer 18:11, 12 October 2007 (UTC)


 * Evidently, they committed mass murder. And I have no idea what you mean when you say that I "seem to be stating that Pakistan committed a crime". They did commit a crime - even their own judiciary admitted it - who disputes it? But I've made no such statement in the article. Gatoclass 18:31, 12 October 2007 (UTC)

As I wrote in the history it is best to keep sources for this separate as they are not both reporting the same thing. And in my opinion the "but" that was added to the footnote is a non neutral addition, as it was only applied to question the lower Pakistani number although as we have seen in this talk page section,recent research that throws up questions about the 3m official estimate in Bangladesh and that was not being 'butted' in the footnote. Best to leave that sort of thing out of the footnotes and let the reader decide for themselves. --Philip Baird Shearer 18:08, 12 October 2007 (UTC)


 * The reader may not understand that he is looking at the official Pakistani government report. The reader may also not check both footnotes, and therefore not discover that there are multiple independent sources estimating 300,000 to 3 million and only one source (the Paki report) estimating 25,000. Or he may be confused into thinking that the better presented Hamoodur report is somehow more valid. That is why some sort of explanation in the footnote is necessary.Gatoclass 18:31, 12 October 2007 (UTC)

People seem to have addressed the factual issues here quite well, so as a trained historian I would like to address not the actual facts/estimates/etc., but rather how to present these.

I think that when it comes to reporting this kind of death toll in this kind of conflict there can be no really accurate figure. Rather, all we can do is to cite the total range of deaths, note that the extremes of this range are both very unlikely to be accurate, and then report a narrower "more probable" range for the actual number based on independent estimates, providing a number of these actual estimates within that range. Sources should be cited for all of these figures. This way, individual readers can assess for themselves the probable accuracy of the estimates as well as where the more important figures of "supposedly" neutral parties fall within it, particularly if a number of these figures fall into clear clusters.

While I don't think this is a perfect solution, it seems to me to be the closest we can come to accuracy as well as to neutrality.

I hope this is helpful. —Preceding unsigned comment added by Lsisson (talk • contribs) 22:30, 17 October 2007 (UTC)

Mongol conquests: A single war?
I'm not sure if the Mongol conquests should be counted as a single war. It is true that the Mongols never paused in their pursuit of conquest. However, that period involved separate conflicts between the Mongols and a variety of independently acting opponents: the Tanguts (Xia), the Khwarazmians, the Abbasid Caliphate, the Southern Sung, various Russian principalities, Poland, Hungary, the Fatimids, and others. If we lump these together as a single "war", we might as well lump all the conquests of the British empire together, or the Spanish empire. I realize this is largely a matter of semantics, but it's hard to compare an extended period of empire-building with a "war" as we know it today. —Preceding unsigned comment added by 68.40.186.184 (talk) 05:47, 13 October 2007 (UTC)


 * Interesting point. But I suppose it's ok to "lump" the wars together as one event. Just like how Japan's conquest of East Asia, South East Asia, and the Pacific, as well as Germany's conquest of Europe and northern Africa are lumped together as WW2. Oops!...I did it again (talk) 07:06, 13 October 2007 (UTC)


 * The overall toll is lumped together in that way basically because it's what many historians have done. The further away in time the event, the more this sort of thing tends to be done, because the individual events are seen as less significant to the overall shaping of history than the sum of the events as a whole.


 * So a single entry for this table is fine by me. That doesn't mean though, that someone can't create another table in a subsection giving a breakdown of the various campaigns. I did that earlier with the Vietnam War and the Arab-Israeli conflict, but user Philip Baird Shearer deleted it all. I intend to restore it if and when I get time to find some bulletproof references (actually I think the tables are well ref'd enough already, but I have a distaste for edit wars). But if you or somebody else wants to create such a table, I will certainly support the move. Gatoclass 10:10, 13 October 2007 (UTC)


 * Ideally, this whole table should have sub-tables, for example listing the deaths from the European vs. Pacific theaters of WW2, or, possibly, by type of death (combatant, civilian, or famine/disease). Of course, neither type of breakdown is available for ancient wars and conflicts, so this table is kind of like comparing apples and oranges...


 * I want to eventually have a full set of wars in history listed at this page, or if not a full set (depending on practicality), then at least a set of the major wars in history. The current limitation of "wars of over a million dead" is so restricting as to be next to useless. No-one who is interested in a list of wars or manmade mass mortalities is going to come to a page that lists only a tiny subset of all the wars in history.


 * And yes, we should have breakdown tables where appropriate, I think that is important. I'm not sure that listing deaths by type will be at all practical though, this kind of information is often sketchy and hard to come by. I don't think we should try to include too much information in the one page. For more recent wars for which there is a lot of information, it would probably make more sense to have separate main articles for those - like for example the casualty list pages for WWI and WII, which are highly detailed and extensive. Gatoclass 01:30, 14 October 2007 (UTC)

I'm having a think about the direction this page should go in. There are over 1000 wars listed at the List of wars page and its subpages. Of course, we could put them all in here - or as many as we have death toll estimates for - there are plenty of military lists with 1000 or so entries on a single page - but it is going to overwhelm the other existing sections like "noncombatant killings", "genocide" and so on. Maybe it would make more sense after all to have just a modest subset of wars listed here, as PBS has suggested, and then to have three or four pages dedicated to "wars by death toll", separated by era and/or size of death toll or some other criteria. Gatoclass 01:56, 14 October 2007 (UTC)


 * I'd also think about separating them by time. Is it really instructive to compare ancient-war death tolls to modern ones? It ends up looking like modern wars are nothing compared to medieval ones, which may or not be true. Just like ancient historians took great liberties when relating the sizes of armies (Achaemenid Persian army numbered in the millions? I highly doubt it), they may have just "guestimated" or wrote down a high number just for shock value when reporting casualties. Also, reliable surveys and statistics were just not available, and famine and disease deaths were often lumped in with war deaths (which, as the table notes, would give WW1 a death toll of 180 million!). If I were making this page, I'd limit it to wars after, say, 1700 or 1750. That's also around when "genocides" started to be counted as special events. Then you could put the ancient wars in a separate table or even in a separate page. —Preceding unsigned comment added by 68.40.186.184 (talk) 03:16, 14 October 2007 (UTC)