Talk:Siege of Suiyang/Archive 1

Not for Civilians' Sake
"Initially, many soldiers and civilians refused to eat human flesh, but Zhang Xun begged of them to do so. At the end, around 20,000 people were fed. Many of the soldiers were greatly moved by this, and swore to fight till the very end."

This is not correct. The Old Book of Tang mentioned that the soldiers were invited to the cannibalistic feast, and the New Book of Tang did not say that the civilians were also invited. There's evidence that Zhang did so for feeding soldiers and no evidence that he did so for civilians' sake. No need to glorify it. The meat was for keeping the war machine working unless evidence suggests otherwise. Qrfqr (talk) 08:23, 2 October 2008 (UTC)

I was rather shocked by the twisted and glorified cannibalism part in the old version of this article. So I just translated what is recorded in The Old Book of Tang and The New Book of Tang.

It should be noted that, although 食 can refer to "to eat" and "to feed", the word here means "to eat", or it would be incoherent in the context, and it would be hard to explain why only 400 people were left after 20,000 to 30,000 people were "fed". Qrfqr (talk) 10:39, 2 October 2008 (UTC)

NPOV removed
The article has been substantially changed since 2007, I've removed the NPOV template, please use or better yet  for sentences, then detail issues here. This will help address them in a timely manner. - RoyBoy 00:08, 3 January 2012 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 2 one external links on Battle of Suiyang. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20050923123431/http://www.starblvd.com:80/mem/l/a/lastwit/war-state/oldtonshu/old194.htm to http://www.starblvd.com/mem/l/a/lastwit/war-state/oldtonshu/old194.htm
 * Added archive https://web.archive.org/web/20070927120023/http://www.1-123.com/0suitang/zhangxun1.asp to http://www.1-123.com/0suitang/zhangxun1.asp

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers.— InternetArchiveBot  (Report bug) 21:19, 28 October 2016 (UTC)

Date Terminology
Since this is a non-Western, non-Christian battle, should we change the date terminology from AD to CE? Tpsreport84 (talk) 20:56, 28 January 2020 (UTC)
 * So why have you inserted CE to articles like The Battle of Ravenna and the Battle of Toulouse which do involve only Western, Christian societies then? Why not rather leave your prejudices and stop editing to push your bias?--154.59.156.93 (talk) 13:48, 11 February 2020 (UTC)

Using ancient sources is primary research
This article is in very poor shape since it leans heavily on ancient chronicles. This is especially bad considering that Chinese sources often appear to be especially prone to inflated figures. In any article about European history, this would likely never be tolerated. When figures in the tens or hundreds of thousands are given, everyone know this is a matter of estimates, or simply a way of conveying "a lot of people".

For this reason, I'm labeling this as original research.

Peter Isotalo 12:28, 22 May 2023 (UTC)


 * I've completely rewritten the Cannibalism section, relying on newer secondary sources instead of the primary history books. I therefore think that the maintenance template on primary research (meanwhile changed to "unreliable sources") can be removed now, and will do so in a few days if there are no objections. Gawaon (talk) 18:17, 9 July 2023 (UTC)
 * I have now removed the maintenance template. Gawaon (talk) 12:07, 14 July 2023 (UTC)
 * 资治通鉴 was already THE newer secondary source with much more moderate estimates. 185.148.3.234 (talk) 07:42, 30 July 2023 (UTC)
 * The problem is that the books that gave the outlabdish "30000 were eaten" estimates were not in the chronicle format but biographies (of the commander Zhang Xun). In this case it was actually the chronicles that remained doubtful of the inflated figures. 185.148.3.234 (talk) 07:47, 30 July 2023 (UTC)
 * The actual problem is that you're again doing original researching (OR) which Wikipedia is not meant for. You can't just read and translate ancient sources and try to make sense of what they meant. Well you can, but that's a topic for a historical paper that you can then hopefully publish somewhere, it's not a topic for Wikipedia. In the earlier (revised) version we simply gave the estimations of victim counts which are also cited in newer secondary sources (such as Chong and Pettersson) – that's neutral enough. Your claim that the Zizhi Tongjian deliberately revised these estimations downward (rather than, say, simply not mentioning them because they were not considered relevant by its authors) it problematic because it seems to be entirely based on your own reading of that work. Hence it's OR, which does not belong into Wikipedia. Do you have newer secondary sources (scientific/historical works, preferably peer-reviewed) that confirm your reading? If so, please add them. Gawaon (talk) 15:53, 1 August 2023 (UTC)
 * I'm sorry, but I had to revert your edits since they are not only OR, they are also misquotations. Pettersson attributes the opinion on Zhang Xun to the New Book of Tang, and I see no reason to doubt him. You also found it in the Zizhi Tongjian, which was published later. OK, presumably it was copied from the New Book of Tang, as happened frequently enough at that time. Moreover, the chapter from the Zizhi Tongjian you referenced itself marks this text as a quotation, attributing it to a "biography" written by "his friend Li Han". I suppose that's the bio already included in the New Book of Tang. Now, the bio, as cited by Pettersson, discusses a hypothetical situation: "If Xun in the beginning of his defence had already planned to ... lose a few hundred for the purpose of unification" – he doesn't say that's what actually happened, he merely talks about what Xun's plan might have been. So interpreting this short hypothetical snippet as a "much lower estimation" of the victim count is a distortion – it's not talking about the actual victim count at all!
 * Moreover, when reading over the rest of the page with the help of Google Translate and Deepl, I noticed that it too accepts the estimation of 10,000s of victims. It says that the siege went on a long time, that it led to cannibalism, that "there were 10,000 soldiers and tens of thousands of people living in the city" when the siege started, while "there were only 400 people left" when it ended. That's in fact very similar to what the Old and New Books of Tang say.
 * But this is all original research, hence none of it belongs in the page. Still it's rather a confirmation of the secondary sources cited in the article instead of raising grave doubts about them. From now one, if you make any further edits, please cite reliable sources which were published (say) less than hundreds of years ago, and please refrain from distorting what they actually say. Thank you! Gawaon (talk) 17:30, 1 August 2023 (UTC)
 * Fine. I'll follow the OR thing. But, several things:
 * a) Patterson attributes the quote to the new book of Tang, or Zhang Xun's biography more specifically: https://zh.m.wikisource.org/wiki/新唐書/卷192, where I didn't find the quote. Maybe he misattibuted, or maybe it's somewhere else in the New Book, maybe not a significant one but there's this small reason to doubt his credibility.
 * b)Li Han was talking in a subjunctive tone, but tell me, had he and his peers in the 760s been told the 30,000 figure, why would he not use several 萬 as the hypothetical snippet instead of 百? This ain't distortion but a reasonable inference from the source.
 * c)mentioning having 10,000 when the siege began and 400 when it ended does NOT say that all 9,600 were cannibalized. You are the one distorting the source by skipping months of details which I cited (or at the very least you probably shouldn't call seen native speaker distorting the source when you yourself rely on DeepL and google translate). In fact Zizhi Tongjian does contain a very similar passage that had several verbatim phrases from the New Book of Tang (hint: it's near the 400 figure, try find that with DeepL), but guess what? It was exactly the 30,000 figure that Zizhi Tongjian deleted, at which place more detailed processes of how they were reduced to cannibalism were added.
 * But again, despite how little sense I think OR makes, Ili follow the rule this time. However, given that I've observed quite often on other articles verbatim quotes that cited primary sources (for instance here https://en.wikipedia.org/wiki/Sack_of_Rome_(410)#cite_ref-Jerome127_8-0 where they cited Zosimus and St. Jerome), I'm suggesting that we could perhaps reach a compromise where I nonetheless leave the timeline reconstructed in Zizhi Tongjian here, but deleting the part where I interpreted it as a lower estimate. 185.148.3.234 (talk) 13:26, 2 August 2023 (UTC)
 * Why to you say you "follow the OR thing" but then do the opposite, re-adding the OR? That's not acceptable! We can discuss here whether there is anything from the Zizhi Tongjian that should be added, but we'll have to reach an agreement here on the Talk page first. If you don't respect that and continue adding your OR without prior agreement, I'll have to call for the page to be protected/semi-protected. Gawaon (talk) 15:21, 2 August 2023 (UTC)
 * YOU, sir, were the one who first deleted the entire thing before reaching an agreement here. I deleted all the original analysis and left only paraphrased timelines. Do you think verbatim translation of the whole passages like the 2 quotes could have been better though? 185.148.3.234 (talk) 15:58, 2 August 2023 (UTC)
 * We can sort this out, but I'll first kindly point out to you that you can't remove maintenance templates such as the OR warning before the underlying issue has been resolved (see WP:MTR). Gawaon (talk) 17:35, 2 August 2023 (UTC)
 * I don't think it would necessarily be OR to add a literal quote from the Zizhi Tongjian. What do you, however, is an interpretation of it, and a highly problematic at that.
 * Here are the paragraphs you give as reference for your claim that they "eventually resorted to cannibalism in October, [only] one day before Suiyang fell":
 * 尹子奇久圍睢陽，城中食盡，議棄城東走，張巡、許遠謀，以為：「睢陽，江、淮之保障，若棄之去，賊必乘勝長驅，是無江、淮也. 且我眾饑羸，走必不達. 古者戰國諸侯，尚相救恤，況密邇群帥乎！不如堅守以待之. 」茶紙既盡，遂食馬；馬盡，羅雀掘鼠；雀鼠又盡，巡出愛妾，殺以食士，遠亦殺其奴；然後括城中婦人食之；既盡，繼以男子老弱. 人知必死，莫有叛者，所餘才四百人.
 * 癸丑，賊登城，將士病，不能戰. 巡西向再拜曰：「臣力竭矣，不能全城，生既無以報陛下，死當為厲鬼以殺賊！」城遂陷，巡、遠俱被執.
 * Which DeepL translates as: 'Yin Ziqi had been besieging Suoyang for a long time, and when the city ran out of food, he suggested abandoning the city and leaving for the east. Zhang Pan and Xu Yuan planned to do so, thinking, "Suoyang is the safeguard of Jiang and Huai, if we abandon it, the bandits will take advantage of the victory and drive away, and there is no Jiang and Huai. If we abandon Suyang, the bandits will take advantage of the victory and drive away, and there will be no river or Huaihu. In the ancient times, the warring lords still saved each other, not to mention the generals of the Secret City! We might as well hold on and wait for them." When the tea and paper were exhausted, they ate the horses; when the horses were exhausted, they dug up rats and finches; when the rats and finches were exhausted, they sent out their concubines and killed them to feed on the soldiers, and Yuan also killed his slaves; then they included all the women in the city to eat them; and when they were exhausted, they continued to feed on the men, old and weak. Knowing that they would surely die, no one rebelled, and there were only 400 people left.
 * 'At the end of the day, when the robbers boarded the city, the generals were sick and could not fight. Patrol bowed again to the west and said: "I am exhausted, I can't take the whole city, I can't repay Your Majesty in life, but I shall die as a ghost to kill the robbers! The city fell, and both Patrol and Yuan were executed.'
 * A few sentences earlier "October" is mentioned – I suppose that's when the city fell. But then the text goes back to what happened before. There is no reason to suppose that everything described in that first paragraph – paper eaten, horses eaten, finches eaten, finally concubines, slaves, and women eaten – happened in October. But if these events started earlier, as seems plausible enough, then your claim that cannibalism started only in October is entirely unsourced!
 * Even more doubtful is your claim that cannibalism started "one day before Suiyang fell". I suppose it's based on the wording "At the end of the day". But I don't have to understand Middle Chinese to know that that wording is to be read metaphorically. It means something like "finally, in the end" – it does NOT imply that all the events described in the preceding paragraph happened in a single day! Also, even I can read (thanks go DeepL) that they ate first their concubines, then their slaves, and finally all the women as well as the old and weak men. How should they have accomplished all that in a single day? Such a suggestion of a single day of cannibalistic mass murder followed by immediate surrender is a much more ludicrous interpretation than the more plausible assumption that these events took place over a longer, unspecified period of time while the siege lasted. Which however means that your whole "timeline of the siege" claim becomes irrelevant since there is simply no start time for the cannibalistic events (nor the horse eating etc.) given.
 * I'll leave it to others to hopefully chime in, but I have the strong impression that the whole paragraph on the Zizhi Tongjian is not only OR, but also a gross misinterpretation of what the source actually says – which is why it should be deleted. Gawaon (talk) 18:21, 2 August 2023 (UTC)
 * First of all, that's terrible even for machine translation. It seems that DeepL took everything as modern Chinese.
 * Second of all, you don't seem to understand the format at all. Zizhi Tongjian lists events month by month if not day by day (when possible). Two paragraphs before your quote it says "十月" or October, several more back it says September. There's no word like "初" to indicate that they were recounting the events before the time scale. For the one day before the fall part though, i don't blame you. I'm not an expert on how to converting Sexagenary calendars to modern months and days either, but the very basics i know that 壬子 precedes 癸丑 by 24 hours. It's like i don't have to know what the 5th of Ramadan translates to in Gregorian calendar to know that it's one day before the 6th of Ramadan. That probably wouldn't even need to be a mass murder given how many corpses were there already (another point of my OR that won't go into the edit) And even if we freezed this part of the debate, volume 219 clearly wrote that there were already only 600 left back in August eating tea paper. Maybe you could tell me whether 30,000 human bodies were many times more than the required calorie intake of that 600, plus an unspecified number of enemies Zhang Xun managed to convert back to the Tang camp with his eloquence.
 * Or perhaps, yes, we should wait for someone else who's not an Anglophone monolingual and have basic understandings of non English or even non 19th~20th century historiography. Just in case you didn't realize, Zizhi Tongjian was, together with the 24 Histories which the New and Old Books of Tangs, one of THE 2 most important pieces of Chinese historiography (yes I'm confident enough to not even say "perhaps" or "arguably". If you really needed that basic some knowledge i suggest perhaps starting with the Oxford History of Historical Writing), not to mention how one of the motives of its author was to revise what he deemed haphazard of the official histories like the 2 Books of Tang. If the books of Tangs go in here, Zizhi Tongjian would definitely have a place too just to crossreference, especially when I had ditched the speculative (albeit informed) part. 185.148.3.234 (talk) 19:16, 2 August 2023 (UTC)
 * Of course DeepL is not specifically trained for Middle Chinese, what did you expect? It's not a language one encounters frequently in the web today. Still DeepL is good enough to get a good impression of what the source says. And I'm not monolingual, for the record.
 * I agree that the Zizhi Tongjian is a fine source to mention if it can shed new light on the events. But none of us here is a professional historian, and it's not our job to try to make sense of ancient sources. So either we summarize what actual modern historians say – that would indeed be best – or at most we limit ourselves to a faithful summary of the relevant things actually and explicitly said by the Zizhi Tongjian. No reading between lines, that's not our job, nor our right while we're here. So, as a faithful and concise summary of what the book has to say on this topic, I'd suggest:
 * "The Zizhi Tongjian, a work in chronicle format published a few decades after the New Book of Tang, does not repeat the claim of 30,000 eaten, which its authors may have considered exaggerated. Otherwise it is largely in agreement with the Books of Tang, noting that of 10,000 soldiers and several 10,000 civilians in the city at the start of the siege, only 400 were alive when it ended. It confirms that the cannibalism started with the generals' concubines and slaves being butchered for the hungry soldiers, followed by all the women still alive in the city at that time, and finally by the old and weak males. "
 * Do you agree that that's a good summary? We can't include your claim that cannibalism lasted just a day or at most a few weeks since there is nothing in the source that explicitly says so, but your observation that the authors don't mention how many were eaten, possibly due to doubts about the earlier statements, is there.
 * Besides this new summary of the Zizhi Tongjian's pertinent contents, I'll also try to look up a few modern secondary sources (academic publications), to see how they interpret the work's statements about the siege. I already have found one or two relevant publications and will summarize their findings soon. Gawaon (talk) 16:23, 3 August 2023 (UTC)
 * I appreciate your summary but isn't this "…which its authors [may] have considered exaggerated" also OR though?
 * Also corroborating with what Zizhi Tongjian has in agreement with the Books of Tang isn't its best value (in fact it's totally possible that Tongjian itself was just paraphrasing the Books of Tang), which as I mentioned was how it supplemented with a timeline which particularly highlights how many soldiers were still standing at which point of the siege.
 * I gotta admit that the starting one day before the fall is a bit suspicious, but NOT because how "single day of cannibalistic mass murder followed by immediate surrender" was ludicrous (it wasn't. It's totally possible that it's an act of desperation before their last attempt to break the siege. Also if you read closely they didn't surrender, just exhausted), but how while the Zizhi Tongjian mentioned the events in Suiyang in July, in August and in October, but not anything in September, during which the last call for help was sent out and the city was basically an information black hole. Still I'd argue that how the 600 soldiers left in August eating 30,000 within 60 days is perhaps as hard to believe as how they did everything within a day. This corroborating the cited contemporary discussion where "few hundreds" was talked about in a subjunctive tone I would say is rather plausible alternative explanation.
 * But that's OR, so I think the best solution is to add the first part of your summary, and then the timeline. If I got time I would like to replace the paraphrase with the translations of the relevant passages in full. I happen to be a historian (in training) and didn't need translators to read Middle Chinese like most people with junior high or above education in China. 185.148.3.234 (talk) 05:07, 4 August 2023 (UTC)
 * I have now replaced your paragraph with the summary suggested above. I have also added a reference to a newer academic work (PhD thesis by Xian Wang) that mentions what the Zizhi tongjian says on the cannibalism. Unsurprisingly, Wang doesn't give a timeline, nor all the details, but she says: "Similar accounts of the killing and eating of Zhang Xun’s concubine also appear in Xin Tang shu and Zizhi tongjian. Unlike Yao’s play, the historical books all mention that wartime cannibalism on a monstrous scale followed the death of the concubine. In addition, the horrifying detail that only four hundred people were left alive in the city by the time Suiyang fell is added to the Xin Tang shu and the Zizhi tongjian versions." (pages 44–45)
 * As for the timeline, I suggest you check and adapt the Beginning section. It is very much a timeline already and several of the dates you give are already there. Indeed I think the whole section could well be renamed to "Timeline". However, it is currently in quite bad shape – there are no references whatsoever and I suppose it's the main reason why the article has the "lacks sufficient inline citations" maintenance template on top. Still, I suspect the section mainly follows the Zizhi Tongjian as that seems to be the only work with a detailed timeline of events? So maybe, if you have the time, you could go through that section, correct or adapt it as needed and add some references to the relevant sections in the Zizhi Tongjian. Having a newer academic work (whether in English or Chinese) to reference here would be even better, but I don't know if any such can be found. In any case, references to the Zizhi Tongjian are definitively an improvement over having none at all, and I suspect that any newer work would tend to closely follow the Zizhi Tongjian anyway.
 * Regarding the OR question: I think an authoritative "which its authors considered exaggerated" or your "the authors ... discarded the 30,000 estimate" would be OR, since they attribute to the authors motives which we simply cannot know for sure. So such statements, if they can be justified at all, would require an explicit justification which would be OR. However, merely pointing out a fairly obvious possibility ("which its authors may have considered exaggerated") is not OR as I would understand it – the NOR rule is not a prohibition to think and to use the results of one's thinking to make the text clearer. At least that's my interpretation of the rule, others might disagree. Gawaon (talk) 16:44, 4 August 2023 (UTC)
 * I didn't write "which the authors considered exaggerated". How's "may have considered exaggerated" less authoritative than "presumably unconvinced" when both reduced the certainty of the tones in different ways?
 * It does sound better phrased than mine so I'll leave it here. However regarding the timeline, I believe it's still more constructive to leave the parts of the timeline that's most pertinent to cannibalism at the cannibalism section. I might find a time to revise the beginning section later as you suggested. 185.148.3.234 (talk) 19:01, 4 August 2023 (UTC)
 * Sigh, I thought we were past this unconstructive edit warring and rollbacking. I think I have been constructive enough, why can't you be too? And now you're starting to totally falsify what the sources say – Pettersson translated from the New Book of Tang, not the Zizhi Tongjian, so you can't change that and claim the opposite, so you now do. Also, the timeline you attribute to Wang is of course not from her thesis, which contains no timeline at all! Nor is at clearly in the Zizhi Tongjian as you now again suggest in the article, as you yourself admitted in your earlier reply. Gawaon (talk) 19:33, 4 August 2023 (UTC)
 * I'll get past this when you get past accusing a native speaker of falsifying and distorting primary sources, while you yourself have to rely on secondary sources and machine translation. Do you see I sounding calmer and having less grammatically screwed stuffs like your last sentence the sign of you were being constructive enough and I'm the trolling one?
 * How is it even falsifying when the passage wasn't at all concluding the events? It wasn't even in the same volume (https://zh.m.wikisource.org/wiki/新唐書/卷192). Either Petterson got the citation wrong after translating it, or your phrasing describing the quote needs a reworking.
 * Sorry about the Wang part. It should be after the first sentence that you wrote. 185.148.3.234 (talk) 20:42, 4 August 2023 (UTC)
 * Also, your rollback totally destroyed the bib entry for Wang plus the correctly formatted references to the Zizhi Tongjian (Sima et al. 1084.), which you'll see in the References section of my latest version. I'll have to roll you back too and must insist that if you make to make further changes, start with what's already there, don't remove references, don't misattribute statements, and do use the pretty citation templates such as sfn|Sima et al.|1084|loc=... . Gawaon (talk) 19:43, 4 August 2023 (UTC)
 * Can we agree that you won't revert me again and we resolve this peacefully? Please??? Gawaon (talk) 20:06, 4 August 2023 (UTC)
 * I thought we were to become peaceful when we settle at your summarizing Wang's summary, as well as the timeline. 185.148.3.234 (talk) 20:44, 4 August 2023 (UTC)
 * Don't fret, I'll re-add the timeline, except for the unsourced parts. Just give me a minute. Gawaon (talk) 20:53, 4 August 2023 (UTC)
 * I sourced everything in the timeline. Also could you send me the screen shot or photo of the Petterson paper? 185.148.3.234 (talk) 20:57, 4 August 2023 (UTC)
 * I don't have access to the Petterson any more, I had borrowed it. Misattributions can of course happen, but in this case I'm 100% sure that it was NOT the Zizhi Tongjian, since Petterson's whole work only deals with the 24 Official Histories, and the Zizhi Tongjian is not one of them. Gawaon (talk) 21:00, 4 August 2023 (UTC)
 * That's fishy but fine. Could you at the very least rephrase the "The New Book of Tang contains a similar account of the events and concludes by discussing their ethical implications" part? While both the events and the discussion exist in the new book of Tang they were in two very distant volumes. Your language made sound as if the author concludes the narration of the events with the quotation right after. 185.148.3.234 (talk) 21:06, 4 August 2023 (UTC)
 * I have changed the wording. Gawaon (talk) 21:16, 4 August 2023 (UTC)
 * Looking good now finally. Though why did you remove the footnotes that included the original quotes? Does the OR thingy also say you can't make citations easier to verify? 185.148.3.234 (talk) 21:26, 4 August 2023 (UTC)
 * I don't think they are all that helpful, considering that most readers here likely don't speak Middle English. And especially I wanted nicer looking citations, using the author–year style used throughout the article. But if you want, you can re-add them in parentheses after the "chapter XX" reference. Or if you prefer it, I can do it tomorrow. Gawaon (talk) 21:43, 4 August 2023 (UTC)
 * Also, I think in this case the solution is fairly obvious, isn't it? Both sources contain the same material, possible from a third original source. If you consider that fact relevant, just add it together with a reference to the original source you found, but don't destroy that which is already there, and don't misquote Petterson. Gawaon (talk) 21:02, 4 August 2023 (UTC)
 * If you can't show how Petterson actually wrote and cite, I'd say it's much more likely that you misquoted Petterson. 185.148.3.234 (talk) 21:08, 4 August 2023 (UTC)
 * Why? Didn't you just write yourself that the New Book of Tang contains the passage as well, if in a different volume? And like I said, Petterson only covers the Twenty-Four Histories. How should I have misquoted him? The only possible confusion could be between the Old and the New Book of Tang – but I don't think so and, in any case, what difference would it make? Gawaon (talk) 21:19, 4 August 2023 (UTC)
 * I don't know. Maybe you remember it wrong, or maybe you distorted what he meant by conclusion and made it sound as if it was in the same volume. After all he was writing about some subjects that you needed google translate. 185.148.3.234 (talk) 21:34, 4 August 2023 (UTC)
 * I have now re-added your reference to the original location and the other places where the statement can be found. It's in any case interesting that it spread so widely. But could you help to make the naked URL https://zh.wikisource.org/wiki/全唐文/卷0430 nicer formatted? Author, translated title, and publication year of that source would be good to have. Gawaon (talk) 21:40, 4 August 2023 (UTC)
 * Author: 李翰 or Li Han, b&d years unknown but he lived at least to 801 AD to write a preface of another book published that year
 * Title: 進張巡中丞傳表 "memorial to the emperor (accompanying the) biography of Zhang Xun the Zhongcheng (a bureaucratic title)"
 * Publication year: December 757, only two months after the fall of Suiyang 185.148.3.234 (talk) 22:09, 4 August 2023 (UTC)
 * Still, when did you borrow the book and did you have your device to edit this article? I'm asking because I just noticed how bad the last part of Petterson's translation was: "...and also ask whether it is not a question of eternal ideals", which is not at all what "況非其素志乎" mean. It should be "not to mention that (cannibalizing few hundreds for strategic goals) wasn't even """his""" (Zhang Xun's) long-cherished/long-standing will/ambition/desire". Are you sure this was Petterson's translation? Or is it someone he cited? Or your machine translation? 185.148.3.234 (talk) 22:30, 4 August 2023 (UTC)
 * Not only that, "I look upon [Zhang Xun] humbly and in painful compassion" also doesn't seem to be a phrase anywhere in the memorial. 185.148.3.234 (talk) 23:09, 4 August 2023 (UTC)
 * I can 100% guarantee that this is exactly the translation made by Petterson, including the first sentence (there is something omitted between it and the rest, as marked by the "..."). I'm very careful with these things. I guess it indicates that the version in the New Book of Tang and the ones you found are not exactly the same after all – at least they seem to have added a new introduction, including that first sentence. If you want me to, I can re-borrow the Petterson to find the exact reference he gives, but it would probably take me a few weeks before I can get it.
 * In any case I'll change the article to indicate that these different versions are not exactly the same. Gawaon (talk) 06:06, 5 August 2023 (UTC)
 * Nah all three were almost identical except for the book of tang version that omitted the possessive pronoun "其", which in that context still can mean no other's long-standing will. Not only wikisource, ctext also have the same versions as in https://ctext.org/wiki.pl?if=gb&res=182378&searchu=素志乎&remap=gb. 185.148.3.234 (talk) 12:38, 5 August 2023 (UTC)
 * I have re-added the timeline, removing just the irrelevant parts and those really not supporting by the source (the claim that cannibalism started in October, when you yourself already admitted that it might at lest have been September). Gawaon (talk) 20:57, 4 August 2023 (UTC)
 * The thing is that Zizhi Tongjian follows a rather strict chronological format that doesn't do flashbacks unless marked by the particle "初，"(which you can find in the same volume too). I admit that a blank September sounds suspicious, but the fact that the cannibalism part is written at the section of the day before the fall still is indicative. Your re phrasing now is fine though. I'm probably adding back the melee combat part back since by refraining the most exhaustive activities, the garrison could have required much less energy intake in order to survive hence a later, probably October start to cannibalism. 185.148.3.234 (talk) 21:20, 4 August 2023 (UTC)
 * I have noticed and added the hardly irrelevant detail that they received outside support in August, which was already mentioned in the Beginnings/Timeline section. (至寧陵，與城使廉坦同將步騎三千人，閏月，戊申夜，冒圍，且戰且行，至城下，大戰，壞賊營，死傷之外，僅得千人入城. ) If you want to play the numbers game, that matters a great deal of course. So please remember that you must be honest and can't just selectively quote the numbers that fit your agenda! Gawaon (talk) 06:43, 5 August 2023 (UTC)
 * Yep I noticed that later too. Impressive that you could find it with translators. Although I also mentioned earlier that Zhang Xun managed to convert many enemy soldiers back to their camp right after when I found it. 185.148.3.234 (talk) 12:28, 5 August 2023 (UTC)
 * I didn't inform you right away because first you were already offline, and second I didn't know what to make the corroborating narration in the New Book of Tang which also mentioned that Nan Jiyun's thousand men also brought with them several hundred ox's into the city (得城使廉坦兵三千，夜冒围入. 贼觉，拒之，且战且引，兵多死，所至才千人. 方大雾，巡闻战声，曰：「此霁云等声也. 」乃启门，***驱贼牛数百入***) which wasn't in Zizhi Tongjian. If I was really being selectively dishonest I'd have selected these several hundred ox's but not the 30,000 figure in the same book. 185.148.3.234 (talk) 13:09, 5 August 2023 (UTC)
 * All right, fair enough. I think we're good here. Gawaon (talk) 13:11, 5 August 2023 (UTC)
 * Next time learn to question your machine-aided reading comprehension of a foreign language text first before some native speaker's "dishonesty" or "distortion". 185.148.3.234 (talk) 14:11, 5 August 2023 (UTC)
 * Speaking of ctext, it's actually a much nicer primary source not only because it provides physical versions each text was based on every time (instead occasionally like in wikisource) and colorlabels sidenotes, but also how it converts traditional calendar to the modern one. Thanks to it I've also noticed that there's a leap month in between (traditional) August and (traditional) September. So the timeline would be converted to modern calendar as:
 * slightly after 757-02-18: Zhang Xun and Xu Yuan joined forces in Suiyang, totaling 6800 soldiers
 * 757-07-25: Yin Ziqi again mustered several tens of thousands to siege Suiyang, which by the end of the week had 1600 troops left and started eating rice alongside tea paper and tree barks
 * 757-08-20~09-17: at some unknown (due to the siege) point (probably right before Nan Jiyun's departure to Linhuai) during this month,
 * ·Suiyang had only 600 soldiers left
 * ·Zhang Xun was eating tea paper with them
 * ·Down the wall melee combat was given up (relevant since in that way they needed a lot less calorie intake to survive and more horse to spared for food and other uses)
 * ·Zhang persuaded 200 enemy soldiers to rejoin
 * ·Nan Jiyun led 30 horsemen to break through the siege and called for reinforcement from Linhuai (307 kilometers southeast to Suiyang), and were rejected
 * ·Nan's men went to Ningling (30kilometers Northwest to Suiyang, relevant since Nan must have taken a much larger detour to avoid being spotted by the siege) and received 3000 reinforcement of both calvary and infantry
 * 757-09-20: Nan and the 3000 returned to Suiyang at midnight, 1000 managed to survived into the city (with several hundred bovines stolen from the enemy, if we were to still selectively trust New Book of Tang), so 1000 plus most likely much less than the 800 when they left
 * 757-11-23: the entire process where the troops were reduced from tea paper to horses (relevant because it indicates how it happened after Nan's return in 9-20 with some horses), to birds and mice and finally to cannibalism, was narrated here
 * 757-11-24: Yan troops climbed up the walls only to find soldiers too exhausted to fight
 * Yes with this charted out I retract my insistence on mentioning the one day before the fall thing. Please feel free to add this part to in between 9-20 and 11-24, but:
 * 757-11-27: Tang reinforcement regained Suiyang (relevant since they'd be able to interview the survivors on spot)
 * 758-01-23: emperor Suzong issued amnesty and awarded living and posthumous titles to those whose loyalty merited. It was this point when some at the court criticize Zhang Xun's nomination for inducing cannibalism and how it was worse than abandoning the city, which Zhang's friend Li Han responded with a biography of Zhang and a memorial (probably spoken? I'm not entirely sure) to the emperor that "If Xun in the beginning of his defence had already planned to eat people and lose """a few hundred""" for the purpose of unification, I would still say that merits and faults cancel each other out, (fixed translation) not to mention that it wasn't even his long standing will". (I side with the critics btw, but I don't see why Li would make up a random number even when it's in subjunctive tone.)
 * I don't expect you to add what's inside the brackets, which are the OR, but please feel free to clarify the timeline and cannibalism sections with what's outside the brackets. 185.148.3.234 (talk) 16:30, 5 August 2023 (UTC)
 * This is sure interesting, but I don't think we can use any it, since it's totally OR especially because of the calender shift. The sources we have say the city fell in October and so that's the date we must, otherwise total confusion would result. Also, if you publish a paper on these detailed in an academic journal or in similar suitable place, we could cite and use it, but OR made on talk pages obviously can't be used, as I'm sure you know by now. Gawaon (talk) 17:07, 5 August 2023 (UTC)
 * Again we seem to have very different definition for OR. My understanding is that while my extended analysis of what subtexts each part of the primary source imply would no doubt be OR, simple paraphrases from primary sources would not be OR even with converted dates since it isn't even a controversial thing. But you seem to define everything that's not 20th century secondary sources as OR which looks a lot like stretch of its definition. The article at this point looks decent and I hope we can settle on this. I'll probably add the converted dates to the timeline section some time in the future.
 * I'm probably not going to make the paper on so morbid a topic one of the first few publishes on my CV but thanks for your (as well as several of my friends') suggestion. 185.148.3.234 (talk) 18:25, 5 August 2023 (UTC)
 * And I found another highly suspicious remark in chapter 219: 城中食盡，一鼠直錢數百，餓死者相枕藉. (...food is exhausted, a mouse costs hundreds of money...). That seems to have been in May or June, no exact date is given. Clearly, at that time the food situation must already have been extremely dire. Now I get your point that one might expect that they would have mentioned if at that time cannibalism had already happened in addition to mouse eating, but can we be sure of that? Hence I'm still, and now even more than before, very suspicious when it comes to the timeline thing in the cannibalism section. In any case, the current wording "according to which food supplies started to run out in July" might not just be OR, it is also wrong – the source clearly states that provisions were already running out at least one or two months before that, when the number of soldiers was still higher. Gawaon (talk) 07:08, 5 August 2023 (UTC)
 * "山南東道節度使魯炅守***南陽***，賊將武令珣、田承嗣相繼攻之. 城中食盡，一鼠直錢數百"
 * It's a different city. Guess the translator isn't always that helpful. 185.148.3.234 (talk) 12:23, 5 August 2023 (UTC)
 * But now you've gotten destructive again, sigh. To avoid edit warring, I urgently ask you to undo your last edit, where you deleted the info on what the accounts of the Books of Tang and the Zizhi Tongjian have in common. As you yourself said, the latter set out to correct what it considered as misleading in the former works, and together these three are surely the most relevant historic book on the Tang era, right? If it's not an minor details but highly relevant if they agree on something, since it gives considerably more confidence that the given descriptions are reliable. So, if we show where they differ, we must also point out where they confirm each other. Gawaon (talk) 17:18, 5 August 2023 (UTC)
 * If the latter sets out to correct the misleading details, all it needs to do is what's different, in this case all the events in chronological order. What both sources have in common can be sufficiently summarized in that one phrase "is largely in agreement with the Books of Tang" without repeating everyhing single point already mentioned above. I'm not being destructive but you are being overly territorial if not distorting the original text using an obviously wrong translation from a secondary source that you can't provide and distracting the important differences with redundant repetitions. 185.148.3.234 (talk) 18:37, 5 August 2023 (UTC)
 * For what it's worth, you're in the wrong. @Gawaon is in the right. They're not being territorial, you're just not following Wikipedia standards and they're pushing back against your aggressive edits. toobigtokale (talk) 22:51, 23 September 2023 (UTC)
 * Also, didn't you yourself agree that my summary was a good one, expressing most of what the source actually and explicitly says about cannibalism? So, regardless of what to do with the timeline (we can resolve that separately), why do you even want to revert/delete it? Gawaon (talk) 20:09, 4 August 2023 (UTC)

Several hundred eaten?
, : About the statement "several hundreds ~30,000 civilians eaten" in the infobox. We had discussed this before and the problem with this figure of "several hundred" is that there is no source that actually talks about the siege to back it up.

The claimed source is the one also cited in the "Cannibalism" section, translated by Pettersson: "If Xun in the beginning of his defence had already planned to eat people and lose a few hundred for the purpose of unification, I would still say that merits and faults cancel each other out...." But this, as the wording makes clear, is a hypothetical statement. It asks a "what if" question and then answers it. It doesn't say anything about how many people were actually eaten during the siege and we cannot claim that it does. Here at Wikipedia we're obliged to stick to Verifiability – we report and summarize what RELIABLE sources say, but we have no right to make our own facts, or impose our own interpretations on sources that don't already clearly say so.

So without any reliable source that explicitly states that several hundred were eaten during the siege, we cannot draw that number out of thin air, nor out of a hypothetical statement that doesn't even claim to be a fact. So I'll revert the page to the old version, which simply stated that we don't know a reliable lower bound: "Up to 30,000 civilians eaten". And I will continue to do so, if need by, unless you find a reliable source that clearly and explicitly gives a lower bound, or an estimation that we can use as such. Gawaon (talk) 11:58, 11 November 2023 (UTC)


 * I'd charted out a timeline for you months ago to show how implausible the 30,000 figure that came up 200 years later was, and you've yet to answer why a same year source would randomly come up with a different number even for a hypothetical scenario. 31.205.18.96 (talk) 15:11, 11 November 2023 (UTC)
 * But now you're again trying to do WP:OR, and you know that we're not allowed to do that. Plus much of what you say is already in the article. It says that "the Zizhi Tongjian ... may have considered the earlier claims of 30,000 eaten as implausible." But since they didn't provide any estimates of their own, we can't cite specific numbers from that book, since it doesn't have any. And that non-knowledge is very clearly specified in my preferred version of the infobox entry, which would start with "Up to". You realize that "Up to 30,000" also means that theoretically just one or two people or even nobody was eaten? Not likely, maybe, but that wording allows it. So where exactly is your problem with it? Gawaon (talk) 18:54, 11 November 2023 (UTC)
 * 1.The purpose is not only from as you said " in the "Cannibalism" section, translated by Pettersson", but also in New Book of Tang, Zizhi Tongjian, 全唐文 and so on. I have given the cites even the exact sentences by quote;
 * 2.WP:OR:On Wikipedia, original research means material—such as facts, allegations, and ideas—for which no reliable, published source exists. This includes any analysis or synthesis of published material that reaches or implies a conclusion not stated by the sources;
 * 3.WP:NPOV:Editors, while naturally having their own points of view, should strive in good faith to provide complete information and not to promote one particular point of view over another. As such, the neutral point of view does not mean the exclusion of certain points of view; rather, it means including all verifiable points of view which have sufficient due weight;
 * 4.WP:V:If reliable sources disagree with each other, then maintain a neutral point of view and present what the various sources say, giving each side its due weight.
 * 4.1 WP:RSUE:Citations to non-English reliable sources are allowed on the English Wikipedia.
 * 5.According to 2ed point to 4th point, you should prove the purpose is not being showed in New Book of Tang, Zizhi Tongjian, 全唐文, even you must prove these books are not reliable or wrong by reliable sources.
 * 6.Please give the sources, which said that 30,000 civilians eaten; We have given the sources for several hundred eaten, but now I cannot find the cite, which can prove what you belived;
 * 7.A historian has written an article "史料解析：安史之乱张巡守睢阳到底吃了多少人" to ifeng.com's historical column. If you have interest you can read it;
 * 8.What you have done is not only feeling possessive about material you has contributed to Wikipedia), but also original research. MINQI (talk) 17:09, 11 November 2023 (UTC)
 * Re 1. Sure, and these references are already given in the article text itself. But they all reproduce the same text, a short bibliography written by Li Han, right? That it's reproduced in several sources doesn't make this text more authoritative, nor does it change the fact that Li Han explicitly asks (and answers) a hypothetical question about what the army leader might have intended ("If Xun in the beginning of his defence had already planned to eat people and lose a few hundred for the purpose of unification, I would still say that merits and faults cancel each other out") – he doesn't even claim to make a statement about what actually happened. Interpreting that hypothetical statement as a fact about the actual number of the eaten is OR, which is not allowed here. Deriving from this hypothetical that "several hundred were eaten" is even more curious, since the text only speaks about "eat[ing] people" without giving a number of the eaten. The "a few hundred" that would be "los[t]" might refer to the number of the eaten, or it might refer to the total number of people who would die during the siege – and we all know that the latter number was much higher. So deriving from this short hypothetical remark that hundreds were actually eaten, while a much higher number did die (but was not eaten) presses something out of it which it really does not contain.
 * Re 6. The source for the 30,000 eaten (about half of the original population of 60,000 people) is given in the article text, you can look it up there. The quoted reference is to Chong, who in turn attributes it to the Old and New Books of Tang. The number was in the article (and in the infobox) long before I started making my first edit to it, by the way.
 * Re 7. That article is interesting. If it gave its own estimate of a few hundred eaten, then we could use it to support your currently unreferenced claim, but as far as I can see, it does nothing of that sort. Rather it says that the number of eaten may well have been high, but that the cannibalism didn't start with Zhang Xun sacrificing his concubine, but may well have started long before that time. Which is all well and good and I won't dispute it, but it's not in disagreement with anything this article says, nor can be be used to support your claim.
 * Re 8. I don't think I feel particularly possessive about the whole article (most of which was anyway written by others), nor about the two words "Up to" I added to the infobox entry in question. And as for OR, can you point out a single phrase or sentence I added to the article that could be considered OR? As far as I know, everything I added or substantially rewrote is backed by reliable sources. And I want it to remain that way, which is why I don't like a claim unsupported by what the source actually says to slip into the article. Gawaon (talk) 10:36, 12 November 2023 (UTC)
 * TO RE1.Please remember that "original research on Wikipedia means the information or explaination from an Wikipedia's editor". Which also means the purpose from Li Han is allowed here;
 * TO RE6.The purpose of 30,000 eaten are also from an old dokument, just like the purpose of several hundred. The original is "士感其誠，皆一當百. 待人封鎖所疑，賞罰信，與眾共甘苦塞暑，雖廝養，必整衣見之，下爭致死力，故能以少擊眾，未嘗敗. 被圍久，初殺馬食，既盡，而及婦人老弱凡食三萬口. "(New Book of Tang——列傳第一百一十七　忠義中);
 * TO RE7.It seems you have problem with understanding chinese. Excuese me, but may I ask are you reading the article with google translater? The original :"闰八月后，一千多守军，有贼牛战马垫肚子，若还不够，哪里需要吃那么多人. 就算上七、八两月，善藏腌制，也决计吃不到三万人！"
 * TO RR8.The explaination of "a short bibliography written by Li Han" is from you, right? "Interpreting that hypothetical statement as a fact about the actual number of the eaten is OR" is your opinion, right? I and MartjnMap take "several hundred" as a purpose just like the purpose of 30,000 eaten, please note that no one, even the ancients has done the actually statistics. MINQI (talk) 14:55, 14 November 2023 (UTC)
 * The figure of (up to) 30,000 eaten doesn't just come from an old book, it's also reproduced by Chong, who apparently considers it credible or at least possible – so it's vindicated by a modern academic publication (secondary source), which is better than just relying on an ancient primary source alone. But even more important than that is that the estimate of c. 30,000 is stated as a fact, while Li Han's figure isn't. He merely mentions this number when discussing what Zhang Xun's initial plans might have been, without saying that it was the actual number of eaten (or whether it indeed refers to victims of cannibalism at all, as opposed to those who died during the siege). Interpreting it nevertheless as a statement of an actual number is, in my understanding, very clearly OR. Our OR guidelines demand that "the information is present explicitly in the source", but interpreting this number as actual number of the eaten is very clearly reading between the lines, since the text doesn't explicitly say so.
 * As somebody (MartjnMap?) wrote when reverting my last edit: "why would Li Han come up with a random number even in a hypothetical scenario?" That's very clearly OR: speculating on the state of Li Han's mind instead of just sticking to what he explicitly wrote.
 * Well, obviously that's my interpretation of what OR is, it's possible that if more editors get involved here, they would mostly see things differently – but I would be surprised if that were the case. I think the OR guidelines are clear enough that a misunderstanding is unlikely. Which is why I think the range "Up to 30,000 civilians eaten" is best – it gives the upper limit which (while quite possibly exaggerated) is clearly given in primary and secondary sources, while avoiding a lower limit for which we have no explicit source. But well, I guess I can live with the "several hundred" if it's so important to you. It's OR, and it's pure guesswork, but ultimately it's not that important that we need to continue fighting over it. Gawaon (talk) 19:16, 14 November 2023 (UTC)
 * Well, obviously that's my interpretation of what OR is, it's possible that if more editors get involved here, they would mostly see things differently – but I would be surprised if that were the case. I think the OR guidelines are clear enough that a misunderstanding is unlikely. Which is why I think the range "Up to 30,000 civilians eaten" is best – it gives the upper limit which (while quite possibly exaggerated) is clearly given in primary and secondary sources, while avoiding a lower limit for which we have no explicit source. But well, I guess I can live with the "several hundred" if it's so important to you. It's OR, and it's pure guesswork, but ultimately it's not that important that we need to continue fighting over it. Gawaon (talk) 19:16, 14 November 2023 (UTC)