This is the talk page for discussing improvements to the Tang dynasty article. This is not a forum for general discussion of the subject of the article.
This article is written in American English, which has its own spelling conventions (center, color, defense, realize, traveled) and some terms may be different or absent from other varieties of English. According to the relevant style guide, this should not be changed without broad consensus.
This article is within the scope of WikiProject Former countries, a collaborative effort to improve Wikipedia's coverage of defunct states and territories (and their subdivisions). If you would like to participate, please join the project.Former countriesWikipedia:WikiProject Former countriesTemplate:WikiProject Former countriesformer country
This article is within the scope of WikiProject East Asia, a project which is currently considered to be defunct.East AsiaWikipedia:WikiProject East AsiaTemplate:WikiProject East AsiaEast Asia
This article is within the scope of WikiProject China, a collaborative effort to improve the coverage of China related articles on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.ChinaWikipedia:WikiProject ChinaTemplate:WikiProject ChinaChina-related
This article is part of WikiProject Vietnam, an attempt to create a comprehensive, neutral, and accurate representation of Vietnam on Wikipedia. If you would like to participate, please visit the project page.VietnamWikipedia:WikiProject VietnamTemplate:WikiProject VietnamVietnam
This article is within the scope of WikiProject Military history. If you would like to participate, please visit the project page, where you can join the project and see a list of open tasks. To use this banner, please see the full instructions.Military historyWikipedia:WikiProject Military historyTemplate:WikiProject Military historymilitary history
Latest comment: 5 months ago2 comments2 people in discussion
In these papers, like "The Dabuyid Spahbeds in Tang Texts: The Position of Tabarestan in the Political-Military Strategy of the Tang Empire in the Eighth Century AD," by Hamidreza Pashazanous, an Iran professor deeply involved in the exploration of ancient Iranian history, argues that the Tang Dynasty nominally and strategically incorporated Iran (Tabarista, south of the Caspian Sea) into its jurisdiction to counter the Arabs. This involved diplomatic recognition, inclusion in administrative maps, and the establishment of a strategy to defend against the Arabs. [1][2][3][4][5][6][7][8][9][10]
The Dabuyid rulers were not merely allies but were recognised as officials within the Tang military-political hierarchy, signalling that even the heart of the Iranian plateau was envisioned as a frontier of the Tang world.
If it is real, then this is even larger than the Tang Dynasty maps drawn by Japan, which did not include Iran.
First and foremost, the map discussion has already happened already (probably quite a few times now, ive lost track) and consensus was to just keep the status quo maps. Also I tried checking your sources but a lot of them appear to be in Farsi + its difficult to access some of your sources (some of them didn't load in their websites). Id strongly caution against overreliance on one source to make a point. Sunnyediting99 (talk) 21:19, 22 April 2026 (UTC)Reply
Could you quote the authors you listed and what exactly did they say against the protectorates beyond Anxi? -Esiymbro
For Denis Sinor, he spends only about 1 page talking about the period from 630 AD to the beginning of the 2nd Turkic Khanate and does not detail any such protectorates beyond Anxi and the tribes he does mention appear to be in the Tien Shan mountains area very far from Kabul, Herat, Bukahara, etc.
p.309-310, The Cambridge History of Early Inner Asia
"In 630 T'ung yabghu was a powerful man. *snip for brevity* However, in his pursuit of glory and conquests - in the words of the Old T'ang Annals - T'ung yabghu was no longer "good to his people, and the tribes hated him." Led by the Turkic Karluks, they revolted and murdered him. Moses Dasxuranci, who also speaks of the plight of the "destructive lion of the north," puts these words of belated regrets in the mouth of T'ung yabghu, whom he calls Jebu kaghan: "Brigands have fallen upon me, and you shall never see my face again, for I did not consolidate my position but imprudently dissipated myself over kingdoms unsuited to me. My pride has thus caused me to fall from my exalted position.
The death of T'ung yabghu created a political vacuum no one was able to fill; self-appointed rulers fought against each other with means inadequate for the fulfillment of their ambitions. The Western Turk confederation was disintegrating, though the five Nu-shih-pi tribes in the west and the five Tu-lu tribes in the east kept their respective cohesion. Jointly they were called in Turk On Ok "ten arrows," while the Chinese referred to them as the tribes of Ten Clans (Shih Hsing). Using for their benefit the incessant conflicts between tribes, the murderous quarrels that set their respective leaders against each other, T'ai-tsung and his successor, the equally gifted Kao-tsung. The establishment and dissolution of the Turk empire longer menaced by the Eastern Turks, gradually extended Chinese rule over the territories only lately under Turk domination. The agony of the Western Turk state was more prolonged than had been that of its eastern counterpart which - as we have seen - collapsed quite suddenly. The process of disintegration and absorption by the Chinese dragged on for a quarter of a century. In 657 Ho-lu - the last de facto ruler of the Western Turks — was made a prisoner and taken to the Chinese capital where, it would seem at his own request, he was symbolically sacrificed on the tomb of the emperor T'ai-tsung. His life was in fact spared, and he died two years later, in 659, to be buried beside the tomb of the kaghan Hsieh-li who, it will be remembered, also died a prisoner of the Chinese. The five Tu-lu tribes were the power base of Ho-lu. The yabghu Chen-chu, his erstwhile enemy and the last chief of the Nu-shih-pi faction of the Western Turks, fell in battle against the Chinese in 659. With his death, which followed by a quarter of a century that of Hsieh-li, last ruler of the Eastern Turks, and with its territory taken over by the Chinese, the Western Turk realm ceased to exist."
For Peter Golden, he spends about 2 sentances on the topic and uses the vague term "protection" in quotations. No mention of any garrison, or any form of adminstration center, oversight, or presence by the Tang.
p. 136, An Introduction to the History of the Turkic People
"In 659, A-shih-na Ho-lu (653-659) of the Tu-lu, which bad dominated On Oq politics, was captured by Tang forces which were once again extending the Middle Kingdom's direct authority to Central Asia. The Western Qaganate now formally came under Chinese "protection" and was divided into 10 prefectures."
For Nicola Di Cosmo, he also only mentions it in the vaguest passing and only described consolidated Tang control over 'modern Xinjiang'.
pp. 182-185 (184 in particular), Military Culture in Imperial China
"Helu retreated far to the west with several hundred horsemen to Shâsh (Shiguo, modern Tashkent in Kirghizia). The combined forces of Su Dingfang, Ashina Mishe, and Ashina Buzhen hunted down Helu, inflicted a final defeat on his forces, and captured him. The victory consolidated Tang rule over much of modern Xinjiang and allowed the dynasty to establish temporary suzerainty over the region further to the west."
I've already given quotes and source for Hans van de Ven and Christopher I. Beckwith previous, van de Ven explicity disagree and Beckwith uses language such as 'allegedly' to describe these claims.
The only author/source I'm not currently able to find an online copy for to quote of the names I previously gave is Michal Biran's The Empire of the Qara Khitai in Eurasian History: Between China and the Islamic World but I'll contiune to look for that one as time allows and maybe make a Request at Resource Exchange.
Meanwhile the current page for the Protectorate General to Pacify the West says, "In 659, Sogdia, Ferghana, Tashkent, Bukhara, Samarkand, Balkh, Herat, Kashmir, the Pamirs, Tokharistan, and Kabul all submitted to the protectorate under Emperor Gaozong of Tang.", in the Intro. A level of detail and naming cities that generally is not reflected in any of the sources I have.
Meanwhile the current page for the Protectorate General to Pacify the West says, "In 659, Sogdia, Ferghana, Tashkent, Bukhara, Samarkand, Balkh, Herat, Kashmir, the Pamirs, Tokharistan, and Kabul all submitted to the protectorate under Emperor Gaozong of Tang.", in the Intro. A level of detail and naming cities that generally is not reflected in any of the sources I have.
I will say, the Tang definitely had some sort of real influence over regions directly adjacent to the Protectorate General to Pacify the West such as Gilgit and Ferghana as listed above. The Tang general Zhao Xiaosong is listed in the protectorate page as having re-installed the local ruler of Ferghana in 715 after defeating the invading forces, who themselves installed another ruler. This is supported by more recent sources such as Zhidong Zhang (2025) in Springer Nature and Victor Cunrui Xiong (2017) in his Historical Dictionary of Medieval China. Gilgit and Ferghana were areas of real military contest between the Tang, Tibetan Empire, and the Arabs and Turks. The Tang played kingmaker in these areas for a period, however brief. Qiushufang (talk) 23:36, 30 June 2026 (UTC)Reply
I agree, but I believe that influence and control in Gilgit and Ferghana is better captured by this map, the long time incumbent, than the current map from 2023 with the maximalist claims of protectorates and their size. That isn't to say that map is perfect and cannot have some imnprovements as you noted previously; but it does seem to better confrom to the majority of sources and what can be verified beyond court claims.
Rather than the map that encompasses all of "Sogdia, Ferghana, Tashkent, Bukhara, Samarkand, Balkh, Herat, Kashmir, the Pamirs, Tokharistan, and Kabul", namely Bukhara, Sogdia, Samarkand, Balk, Herat, Kashmir, and Tokharistan don't seem to have any evidence really and most are hundreds of kms from the nearest evidence of Tang Imperial presence. Guile's Theme (talk) 23:47, 30 June 2026 (UTC)Reply
The area in the north along the Yellow River in Inner Mongolia should be added to reflect where most of the Protectorate General to Pacify the North's seats were located. Parts of Korea and Gilgit could perhaps be added in a different color to reflect short occupations. I generally prefer that map to show areas of real administrative control, but it is by no means perfect, as you pointed out, and should not be considered as a preferred version prior to changes or a new map is made. Qiushufang (talk) 00:08, 1 July 2026 (UTC)Reply
The area in the north along the Yellow River in Inner Mongolia should be added to reflect where most of the Protectorate General to Pacify the North's seats were located.
The seat of the Andong Protectorate was situated in Pyongyang from 670 to 676 when it was moved to Fushun in modern Liaoning due to conflict with Silla, although the area was heavily contested during those years as well with much back and forth. I'm less sure about Ungjin Commandery since I didn't write that page. Qiushufang (talk) 00:33, 1 July 2026 (UTC)Reply
Although the area north of the Taedong River was not part of Silla even after 676 and likely came under the control of Parhae more than 20 years later. The year 670 is generally considered the year when Silla started attacking the Tang, so it would make sense that was when Tang control of Ungjin ended since it was cut off from the Tang by sea. Qiushufang (talk) 00:41, 1 July 2026 (UTC)Reply
My initial draft of the map we discussed;
Tang Dynasty 660CE Initial_Draft
My initial thought is that maybe the Chanyu, Dongyi, and Anxi protectorates should all also be in the lighter shade as these were still formally protectorates and not direct annexations in 660 CE which might be a difference worth reflecting in the map? Interested in any thoughts or feedback from you and others concerned. Guile's Theme (talk) 06:07, 1 July 2026 (UTC)Reply
The reworked map seems fine to me.
The rationale for having the Chanyu, Dongyi, and Anxi protectorates in a different shade based on legal distinction is wrong because the the understanding of "protectorate" is wrong. It did not imply protection of local polities as independent entities or a separate part of the Tang empire under "less administration".
Protectorates with stationed military garrisons were military commands meant to solidify and stabilize Tang control over a militarily precarious frontier region, not to denote their protection of local polities, although that could also be used as justification. For example, the Andong Protectorate was created as an occupying military administration of the annexed Goguryeo kingdom and the Anxi was the occupying administration of the conquest the oasis states. The Tang never intended to protect either Goguryeo or the oasis states, much less let them go, but to control them as their de facto ruler. Certain elements of local rule were retained such as members of the local royalty, but they were removed by the Tang when they displayed signs of autonomy. Throughout the duration of these protectorates, especially Anxi, all commander titles were held by appointed generals of the Tang. Effectively, it was a military occupation of conquered territory.
Arguably, this state of administration could be applied to all parts of the empire for which numerous rebellions arose and required putting down by the military. The difference therefore seems arbitrary to me. Is there a discernible difference between Annan, controlled by the military and suffered from numerous invasions and rebellions, versus Anxi, Andong, and Chanyu? They all held the same title at one point or another. The Tang certainly did not consider Annan to be a separate part of the empire legally distinct from other parts.
Protectorate titles were not exclusive to other administrative titles. When Karakhoja/Gaochang was conquered in 640, it was turned into a prefecture and direct administrative control was established there (per Hansen (2012) in The Silk Road: A New History p. 91). The following is a description of Turpan/Xizhou's administration post-conquest:
Since Turfan was a prefecture exactly like the three hundred other prefectures in the Tang Empire, officials implemented redistribution of land as stipulated by the regulations of the equal-field system, which was in effect throughout the empire. They were required to update household registers every three years. The household registers listed the head of each household, all family members and anyone else living in the same house, and their tax obligations... Drawing up registers for each household in 640, officials counted 37,700 people living in 8,000 households in Turfan. (One hundred years later the number of households had increased to 11,647.)
—Valerie Hansen (2012) p. 91
Protectorate garrisoned cities such as the above were treated as normally administered cities of the Tang empire. No doubt there were also cities under protectorate rule that did not have the same level of direct rule, but that would be due to other factors such as logistics, duration of rule, and military conditions, rather than simply because they were part of a protectorate. The distinction between protectorate and non-protectorate is therefore not a legal distinction predicated on core vs. non-core territory, or more vs. less administration, but rather whether it was considered part of a territory administered by a military command. Qiushufang (talk) 07:36, 1 July 2026 (UTC)Reply
I just realized you mentioned Dongyi Protectorate rather than Andong Protectorate as one of the three protectorates that you think should be colored in a different shade. However I checked the sources for that page and couldn't find anything to support its existence. The Jiaole Area Command given to a Xi chieftain in 648 is too vague to be of any use. It could be a merely symbolic title to recognize the power of a local ruler. The Songmo Area Command created in 648 is directly stated to be imaginary. So neither should be part of Tang territory. However I don't see it as solidly defined in any of the maps. Most of them put it in the upper northeast corner of Tang territory, covering southern Inner Mongolia.
Got it, thank you. I'm content with leaving the map draft as-is as well then if you are. Now hopefully I can reach a similar agreement or compromise with Patali and Esyimbro as well. Guile's Theme (talk) 08:51, 1 July 2026 (UTC)Reply
There is a case to be made, given the way the map of Tang China is presented in most published sources, that the Map should be of 750 AD. What is your opinion on this idea?
As to the date 675 AD, while this might expand Tang control in Korea it would mean a shrinking of the Tang to an equal or greater degree in the West as by 670 the Tibetan Empire had begun or already conquered the Tarim and the Tang would not reassert their previous strength there till 692.
If nothing else the 750 date seems a good way to avoid this confusion about which date exactly between 660 and 675 should be the map date. And has the added bonus of having the famous Battle of Talas as an anchor point. Guile's Theme (talk) 02:29, 9 July 2026 (UTC)Reply
In 673 the Tang consolidated control over the Wuduolu Turks living in the area that came to be known as Dzungaria. The Tang also captured Qiuci and re-established control over the Western Regions in the same year. - Protectorate General to Pacify the West
These developments encouraged the king of Khotan, who had remained loyal to China, to expel the Tibetan forces from his country at the end of 674. In recognizing his contributions, the Tang court in 675 appointed him head of the Pisha government-general at Khotan, with jurisdiction over ten prefectures. In the same year, the ruler of Kucha also pledged loyalty to the Tang, thus allowing the Tang court to move the seat of Anxi protectorate back to Kucha. To reinforce Tang authority in the Western Regions, the Tang court also established governments-general at Kashgar and karashahr.
per Wang (2013) p. 147-148 in the source you listed below
I do not favor most of the maps you have used as examples. To be frank, most of them are awful and have obvious problems that make them inferior to what can be found according to the sources on Wikipedia. I recognize for your argument that Tang authority did not extend past Anxi that they are useful, but for any other purpose they are woefully inadequate and the date 750 is almost certainly chosen for the sole reason that that was right before the Battle of Talas as a convenient cut off date even though the "full extent" of Tang borders had been reached decades prior if not almost a century. Qiushufang (talk) 02:40, 9 July 2026 (UTC)Reply
From the Tibetan Empire article, "In the spring of 670, Tibet attacked the remaining Chinese territories in the western Tarim Basin after winning the Battle of Dafeichuan against the Tang dynasty. With troops from Khotan they conquered Aksu, upon which the Chinese abandoned the region, ending two decades of Chinese control. They thus gained control over all of the Chinese Four Garrisons of Anxi in the Tarim Basin in 670 and held them until 692, when the Chinese finally managed to regain these territories."
Also from Beckwith on the part regarding Kucha in 674, pages 36, 40-42, and 146 from the same source from his above;
"At the beginning of 674, according to Chinese sources, the kings of *Kóngül and Kashgar came in person to the Tang court, then in Chang-an, to "surrender."The Chiu Tang shu states that the T'ang subsequently established a Governor Generalship in Kashgar. Soon afterward, in early 675, Fu-she Hsiung, the king of Khotan, came to China with some seventy followers; later in the same year a Governor Generalship was also created for Khotan. The jurisdiction was named P'i-sha after the native dynastic name and had ten prefectures, with Fu-she Hsiung as Governor-General. Coincident with the arrival of Fu-she Hsiung, the pretender to the throne of Persia, Péróz, whom the Chinese had appointed Governor-General of Persia a decade before, also came to the Chinese court. The "king of Khotan" and the "king of Persia" are treated in the sources as exact equals, even though the latter had never ruled Persia, and his father, the last ruler of the Sassanid dynasty, had been killed in the course of the Arab conquest forty years before. Since there is no reference to Chinese military activity in the Tarim during this period, and since foreign kings never traveled by choice to the Tang capital, it seems clear that these kings were refugees rather than victorious rebels who had received Chinese assistance against the Tibetans. If it is true, however, that the Tibetans had let slip their control over Khotan and Kashgar, in 675 and 676 their dominion was restored."
As for the maps; I honestly just tried to find as many maps in published sources as I could. The largest number of these came from textbooks as scholarly works usually don't provide maps and they are more for the beginner, casual, or young learners. But such maps in textbooks are almost always representative of how the mainstream/majority scholarly view of Tang Chinese territory is depicted in maps if not a good indicator and they seem to be in rather broad agreement while the source being used as the infobox map presents something of an outlier in published sources.
But regardless, noted that you do not support the 750 AD as the map date, and I suspect the other 3 would also dislike it. But this doesn't get us much closer to figuring out a new consesnus date for the map unless we just opt to keep 660 with edits to reflect the Source Survey. Guile's Theme (talk) 03:05, 9 July 2026 (UTC)Reply
After having read the page you linked - Tibetan conquest of the Tarim Basin - I can see where you got the impression that 675 would not be a good year to place the map since it does say the Tibetan conquest of the Tarim Basin lasted for the 20 years following . However unfortunately this the result of a simplification by the author (presumably since I haven't checked) that is so prevalent in these generalist histories. According to sources that devote more effort to detailed descriptions or simply anecdotal mentions of battles and the taking and retaking of cities, the Western Regions exchanged hands between the Tibetan Empire and Tang dynasty at least twice in the years between 670-692. As mentioned above, sometimes years are chosen simply for convenience. 692 is one of those years because that was when the Tang re-established control over the region for a more prolonged period of time. In reality based on more detailed sources, the Tang retook the region in 675, the Tibetans in 677, the Tang again in 679, the Tang willingly relinquished the region in 686, the Tibetans again in 687, and then the Tang again in 692. Owing to situations like this, I'm not a big fan of choosing 750, since realistically it's not the date at which the Tang reached its greatest extent, but merely the most convenient as a cut-off date. Qiushufang (talk) 03:03, 9 July 2026 (UTC)Reply
I see, in light of the Beckwith quote I gave it sounds like there is disagreement in the sources about what exactly was going on in the Tarim between 674 to 687 depending on their interpretations of the Tang court documents.
I'm not entirely against a 675 date, it just makes the situation in the West even more difficult to gauge and that is the area of the map I believe needs improvement to reflect the weight of sources on the topic. Whichever date we choose the situation in the East/Korea seems generally less murky. Guile's Theme (talk) 03:41, 9 July 2026 (UTC)Reply
I've read the pages referenced to Beckwith now and in hindsight, it was erroneous to call the source a generalist history. However, there is a lot of conflicting information as well as some missing information between Beckwith and the Wang source. Now that I have read Beckwith in more detail, I am amicable to the idea that the kings who visited the Tang were refugees. That is plausible but there are also issues I find questionable.
On p. 36, Beckwith claims that the Tang protectorate was moved back to Xizhou, but this conflicts with Xiong 2008 p. 45 which states that the seat was actually moved to Suiye (Suyab). Beckwith also talks about other explanations for why the oasis kingdoms' rulers may have gone to the Tang like maybe the Tibetans just left after conquering the area in 670 and "actually began its second wave of invasions in 674" (p. 42). Oddly, Beckwith talks about the Tibetans teaming up with Ashina Duzhi on p.42 as well and how the Chinese sources report that such action led to the loss of the Four Garrisons. If the Four Garrisons had been lost back in 670 how could they lose it again in 677 if they did not have control of it again after 670? Wang 2013 p.148 mentions the re-establishment of the protectorate seat of Anxi at Kucha, which Beckwith makes no mention of but later states that Kucha was outside Tibetan territory. Beckwith denies that the Tang retook the Western Regions in 679 with Pei Xingjian's defeat of Ashina Duzhi's Turks, but attributes the Tang capture of Suiye to that event whereas Xiong puts the relocation to Suiye in 670 directly following the Tibetan conquest.
Beckwith also isn't the only source conflicting with others. Wang 2013p. 148-149 agrees with Beckwith that the 679 action by Pei Xingjian resulted in the capture of Suiye, contradicting Xiong 2008 p. 45, but otherwise conflicts with Beckwith in claiming that the Tang retook the Western Regions in both 675 and 679. Wang 2013 p. 149-150 also mentions the retreat of Tang forces as a deliberate decision in 686 and the Tibetan reconquest of the region in 687, which Beckwith makes no note of. However Beckwith does say that in 687, Tibet invaded Kucha, which he admits lay outside of Tibetan territory (p. 50). I suspect that really they are both talking about the same sources but interpreting it differently. The only parts near the Western Regions that Beckwith believes was controlled by the Tang during this 670-693 period was Suiye and Xizhou.
So this period certainly is contested as to what actually happened. I still think a 675 date at least makes some sense for the eastern side of things since that was the last time the Tang held some control in Korea whereas the 675 date for the west has at least one source that claims it was under Tang control. A later date for 693 is also fine since the lighter parts could be used as a way to show parts that used to be under Tang control briefly. I am however still against using 750 since it is merely a date picked as a result of the Battle of Talas, which isn't even the end point for Tang expansion outside of Anxi or when they retreated from the protectorate. That would be in 755 with the An Lushan rebellion. Qiushufang (talk) 04:42, 9 July 2026 (UTC)Reply
I've been thinking on how to tackle this question regarding the date. I was thinking the method that would deliver the best result would be to create a timeline for each these individual places from 658 to 755 to form a breakdown. At least 3, but maybe 5 or more, preferably specialist sources, for the most contested dates/locations. And the answer should be clear(ish).
And while I would support such a map, I think the criticism that would make it difficult to justify as the infobox map the next time someone comes along and takes issue with it is that instead of using any of the published maps or dates we elected an entirely novel date and map.
So thus in order to avert that issue I think we should probably limit ourselves either to the currently accepted date of 660/661, the date of the previous 15 year map of 700, or the commonly used date of 750 in order use one of the published maps as the 'basis' we can refer to and then justify minor edits to the 'base' using a weight of specialist sources in order to reduce the risk of being accused of Synth/OR. This also avoids the sources disagreeing in regards to the aforementioned dates in Tibet.
Do you agree with this line of thought? And if you do, given you do not support 750, which of the other 2 dates would you elect for? For my part, under the previously given rationale of use a published map as a 'base', I could only support 700 or 750 as there is no published map in the 7th century that could be used as a basis aside from the one currently used for the infobox map. Although maybe my fears of Synth/OR accusations for using a novel date and not using a published map as a basis are overblown?
Alternatively, I have created what I think would be the breakdown for the timeline approach to making a map with a novel date.
In the West this breakdown would include:
The Tarim Basin
Could be broken down into subsections for each city/garrison.
The Western Turk Realm in Semerichyne
The Duolo and Nushibi tribes
Gilgit and the Pamirs
Ferghana Valley up to or just outside modern Tashkent.
Turpan and Beiting Protectorate
In the East:
The Ungjin Commandery
North Korea (Pyongyang)
Liadong Pennisula
Andong Protectorate North of the Yalu river? (Manchuria)
A timeline would be far outside the scope of a front page map, especially one that goes in the infobox. The proposed timeline is also too detailed. 700 is fine for the current map since all the areas are still occupied with the exception of Korea and Gilgit that are represented in a lighter shade. Perhaps add c. to 700. Qiushufang (talk) 11:00, 9 July 2026 (UTC)Reply
When you reject maps from reliable sources and try to construct a replacement from unreferenced maps, your own interpretation of sources, other Wikipedia articles and the opinions of other editors, you will produce something that is completely unmoored and indefensible when the next editor proposes their preferred version. (And the current version is not maximal – for that, see the yellow map at right or the versions in the previous section, which have been proposed and added to the infobox many times over the years.) That is the reason for the Verifiability and Original Research policies. They are even more important in a Featured Article, like this one.
There is no contradiction between the account of Twitchett in Cambridge History Of China and that of Sinor in Cambridge History of Early Inner Asia. They agree that the territory of the Western Turks came briefly under Chinese control. Our sources stress the difference between this area and the longer-lived military administration of the Tarim Basin, and between the latter and the area under civil administration. So should we. It might be useful to add footnotes indicating what the sources say about the first two, though. Kanguole20:09, 1 July 2026 (UTC)Reply
First, hello, and thank you for coming to partcipate in discussion.
When you reject maps from reliable sources and try to construct a replacement from unreferenced maps, your own interpretation of sources, other Wikipedia articles and the opinions of other editors, you will produce something that is completely unmoored and indefensible when the next editor proposes their preferred version.
This characterization I feel is unfounded and unfair; the map I'm trying to bring back was the incumbent for 15 years from 2008 to 2023, the whole time the current map existed in a book published in 1979, but was not made the infobox map until May of 2023.
So while I'm willing to update the older map to satisfy others and reach consensus I feel it is totally unfair to say that I am "constructing my own replacement from unrefrenced map, your own interpretation of sources."
(And the current version is not maximal – for that, see the yellow map at right or the versions in the previous section, which have been proposed and added to the infobox many times over the years.)
And I'm sure there is a map out there that shows the Tang as smaller. The fact a version of a map exists somewhere is neither persuasive nor relevant especially because they can't all be right.
There is no contradiction between the account of Twitchett in Cambridge History Of China and that of Sinor in Cambridge History of Early Inner Asia. They agree that the territory of the Western Turks came briefly under Chinese control.
I feel that is a stretch to say that Denis Sinor agrees with the maximalist map when quite literally the ONLY thing he says on the topic at all is; "With his death, which followed by a quarter of a century that of Hsieh-li, last ruler of the Eastern Turks, and with its territory taken over by the Chinese, the Western Turk realm ceased to exist." That statement could just as readily support the incumbent map from 2008 to 2023 depending on what territory the Chinese puppets actually controlled during that supposed 1 to 2 year period. Again, no actual evidence of Tang presence, no garissons, no admin centers, no officials, no actual realistic logic basis of control has been brought up or discussed neither by any source or any editor that would consitute what a Chinese protectorate adminstration was supposed to be anywhere beyond Anxi. Guile's Theme (talk) 00:30, 2 July 2026 (UTC)Reply
I was wrong to say that the 2008 map was unreferenced. It was cited to Albert Herrmann (1935), History and Commercial Atlas of China, though it seems to be a composite of the 700 and 750 maps from that atlas, with some changes. When you change the date to 660 and add other pieces to satisfy other editors, it is certainly no longer supported by sources.
I said that Sinor agrees that the territory of the Western Turks came briefly under Chinese control (for that is what he says). You have cited him as opposition to Twitchett's map, but this is an argument from what he does not say, which is so weak as to be insubstantial. Kanguole09:47, 2 July 2026 (UTC)Reply
Descriptions of the Chinese protectorates in the west
Map of the Tang Empire and Central Asian Protectorates circa 660 CE. It was through the conquest of the Western Turks that the Tang dynasty was able to reach its maximum extent, although for just a few years.
Again providing some of the main sources for the "Map of the Tang Empire and Central Asian Protectorates circa 660 CE", since the discussion moved here. Still, Tang control in the western areas was quite weak and short-lived, essentially stemming from the Tang control of the Western Turks. Per James A. Millward, it is fairly standard academic practice to show Tang's greatest expansion in this way ("Cartographers tend to choose this moment of the Tang's greatest extent to depict historical maps of the Tang" hereafter):
"Further Tang expansion followed. The Western Türk Khaghan He-lu (r. 651-7) briefly succeeded in reuniting the Western Türks and retaking the Tarim, as well as lands as far as Persia. When Tang forces defeated the Western Türks at Issyk Kul (in modern Kyrgyzstan) in 657, the Tang emperor installed two rival khans to rule the vast Western Türk empire, and scattered Tang protectorates-general and garrisons throughout it. Tang suzerainty thus extended thinly from Talas and Tashkent in the north, over Samarkand, Bukhara, Kabul and Herat, and as far south-west as Zarang in modern Iran. Tang thus abutted the frontiers of the expanding Arab empire. Cartographers tend to choose this moment of the Tang's greatest extent to depict historical maps of the Tang, but five years later, by 662, the western Central Asian regions nominally under Tang protectorates and prefectures were in rebellion, and after only a few years the Western Türks succeeded in overthrowing their puppet khans and restoring their independence throughout their former territories. Although the Tang would regain its sway over western Türk tribes on occasion thereafter and Tang maintained a Zungharian foothold in Beiting (Beshbaliq, north of Jimsar in the vicinity of modern Urumchi), direct Tang influence beyond the Pamirs was momentary." in Millward, James A. (2007). Eurasian Crossroads: A History of Xinjiang. Columbia University Press. p.33. ISBN978-0-231-13924-3.
"After some inconclusive fighting lasting several years, in 657 the T'ang general, Su Ting-fang, finally broke the Western Turkish military power in a battle fought near the Issyk-Kul. Sha-po-lo fled to Tashkent, whose people turned him over to the Chinese. Kao-tsung then divided the Western Turks once again into two federations, the Tu-lu and the Nu-shih-pi, and placed rival qaghans, both vassals of the T'ang court, at their heads. The home territory of the Western Turks, in the region of the Ili valley and the Issyk-Kul, was put under the control of two Chinese protectorates-general (tu-hu fu) whose territories coincided roughly with the previous eastern and western divisions of the Western Turkish empire. Other foreign people previously under the suzerainty of the Western Turks in western Sinkiang, Russian Turkestan and the valley of the Oxus, were also placed under formal Chinese control in 659 and 661. The T'ang empire then stretched from the China Sea all the way to the borders of Persia, and part of the new territory under Chinese control was named the 'Persian' government-general (Po-ssu tu-tu fu). These new territories were, however, so vast, and T'ang forces spread so thinly, that it is not surprising that this further extension of T'ang authority lasted only briefly. At the beginning of 665 both the Tu-lu and the Nu-shih-pi tribal federation rebelled against their pro-T'ang qaghans, and regained their independence from the Chinese." The Cambridge History Of China, Volume 3 p.280
Blunden, Caroline and Elvin, Mark (1983), Cultural Atlas of Chinapp 92-93, Oxford: Phaidon Press, ISBN0-7148-2309-0, with text:
"Overextension of the empire. The Chinese empire reached its maximum western extension under the Tang emperor Gaozong in the later 7th century. Its protectorates stretched across the Pamirs into Soghdiana and Tukharistan and as far as the frontiers of modern Iran. This far-flung hegemony was impossible to maintain for long. The rise of Islamic power in the west, symbolized by the Arab victory over the Chinese at the Talas River in 751, and the increasing strength of the Tibetans and of the kingdom of Nanzhao in the southwest created strains on resources that were impossible to resist after the internal weakening caused by An Lushan's rebellion in 755..." 92
"Of the protectorates shown, only the Anxi protectorate was a real administration governed from Lucha, and with garrison armies. The others were "spheres of influence" with no permanent officials, no fixed administrative centres, and no military forces. The boundaries shown are very approximate and not to be taken as indicating that these areas were in any way controlled from Chan'an. They give a notion more of Gaozong's imperial pretensions."
On Millward, the issue I see with him is that he is less influential/authoritative than just about every other author/source we've spoken of. Namely because just about every other name is considered an expert in Turkic, Sogdian, Mongolic, or early Inner/Central Asian history; Denis Sinor is a foundational source in Inner Asian history, Nicola Di Cosmo is a foremost contemporary expert on early Inner Asian history, Beckwith is a foremost expert on History of the Tibetan empire, Peter Golden an expert on early/medieval Turkic history, Michal Biran a Mongolic expert, de La Vassiere an expert on Sogdian and Pre-Islamic central asian history.
While Millward's speciality is in Qing and Modern Xingjiang, IE over 1000 years after the period being discussed and thus why he is citing other authors for his claims, one of which inlcudes Beckwith who is not very supporting. (Although not directly relevant in my recent research I've also found Millward to be subject to a lot of modern/mainland Chinese criticism because of his expertise in more recent Xinjiang history)
Of all the names mentioned, if anything, Nicola Di Cosmo should be the most authoritative since he bridges the gap between Chinese histories and the early Turkish the best and he did not even mention the protectorates further west but only mentions 'consolidation' of Xinjang and "temporary suzerainty over the region further to the west" with nothing to go as to what that 'region' constitutes.
"Other foreign people previously under the suzerainty of the Western Turks in western Sinkiang, Russian Turkestan and the valley of the Oxus, were also placed under formal Chinese control in 659 and 661."
No mention of who those 'foriegn people' were in context of the aftermath of things like the Muslim conquest of Khorasan
"Merv was the capital of Khurasan and here Yazdegerd III held his court. On hearing of the Muslim advance, Yazdegerd III left for Balkh. No resistance was offered at Merv, and the Muslims occupied the capital of Khurasan without a fight." (651)
The only description we have the lands occupied of by the tribes the Tang claim to have installed puppet Khan's in are one group 'East' of the Chu River and the other 'West' with their collective territory centered on modern day Semirechye which still gives us no basis to draw their scope on a map.
Blunden, Caroline and Elvin, Mark (1983), Cultural Atlas of China pp 92-93, Oxford: Phaidon Press, ISBN 0-7148-2309-0, with text:
This source is an illustrative Atlas, a little old from 1983, and the blurb over the map is talking in vast majority about the 8th century battle of Talas in 751, the An Lushan rebellion in 755, and even goes on to mention 791 and the 9th century with only a cursory mention of the 7th century.
Ven, Hans van de (26 July 2021). Warfare in Chinese History. BRILL
As mentioned previously, even 'with caveats', I do not believe this source can be cited in support of the maximalist map.
In fact I actually believe that between "only the Anxi protectorate was a real administration", "with no permanent officials, no fixed administrative centres, and no military forces", "The boundaries shown are very approximate and not to be taken as indicating that these areas were in any way controlled from Chan'an", and "They give a notion more of Gaozong's imperial pretensions" that Hands van de Ven is actually doing the most to explicity refute the maximalist map out of all the sources/authors mentioned in this discussion.
Overall, I don't doubt there are sources that reflect the Maximalist map, I just currently believe the majority of the most mainstream and authoritative sources don't really reflect such a map nor is a map a substitute for the absence of all other types of supporting/corroborating evidence.
As far as I'm concerned, The Cambridge History Of China, Volume 3 is by far the most authoritative and mainstream source here, and their map of the Tang and their Protectorates (map 8 p.281) perfectly supports what you call the "maximalist map", which is actually a standard academic map of the Tang Empire and their Protectorates in 657-662 at their greatest extent, and is therefore perfectly legitimate on Wikipedia. Now it is just that, and it seems likely that the actual control over several of these "Protectorates" was quite flimsy, largely nominal, and in any case short-lived ("The T’ang empire then stretched from the China Sea all the way to the borders of Persia (...) These new territories were, however, so vast, and T’ang forces spread so thinly, that it is not surprising that this further extension of T’ang authority lasted only briefly."The Cambridge History Of China p.280"). Pataliputra!(talk)10:42, 1 July 2026 (UTC)Reply
I'm not certain the The Cambridge History Of China would be more authoritative than The Cambridge History Of Inner Asia on this topic where they intersect; if anything Inner Asia is technically the location of the western protectorates in question. Arguably the History of Inner Asia would the more focused work of the 2 in both geographic and temporal terms. If anything I suspect the former tends to take the view of the Chinese court records and the latter less so as matter of perspective given each of their focus.
I'd continue to point out that Denis Sinor and his Cambridge of History of Inner Asia aside I've named 6 other notable authoritative/influential authors that don't appear to support it or only mention it with no detail in the vaguest passing reference.
IE Nicola Di Cosmo(1), Peter B. Golden(2), Christopher I. Beckwith(3), Étienne de la Vaissière(4), Hans van de Ven (5), & although I've not found an online copy yet I am familiar with Michal Biran's work (6) I've used her book for research before and don't recall anything different from the others. Even were Cambridge History Of China were the most authoritative it stands relatively alone with Millward being in my opinion weak support outside the period he specializes in.
I am willing to concede that the map can perhaps be used in the body of articles with the appropiate caveats/citation of the Hans van de Ven type although I still do no believe it can be argued to be the majority/mainstream view and thus is not appropriate for the infobox map.
However I recognize I may be no closer to convincing you than you are to convincing of anything beyond that and that belaboring the point may not be productive; so how shall we proceed in regards to the Tang Dynasty infobox map? Do you think an RFC is in order or wait a few days to see if others comment organically? Thank you. Guile's Theme (talk) 11:26, 1 July 2026 (UTC)Reply
Thank you for providing the quotations, but I still fail to see how they contradict the Cambridge History of China map. Specifically, the Cambridge History of Inner Asia description
...with its territory taken over by the Chinese, the Western Turk realm ceased to exist.
seems consistent with the territorial extent given in the current Infobox map. The fact that the Tang briefly took over Western Turk territories was also mentioned by the other authors. It is true that they did not describe the protectorates. But since it is well known that Tang control over these regions only lasted a few years, the authors likely felt that it was not needed to write about details of administrative structures that did not leave a lasting impact on the Tang or Central Asia. That should not be taken as evidence that the authors made the claim that the protectorates did not exist. In fact, the Cambridge History of China did not discuss the far western protectorates in detail either.
I would also like to point out that there have been many previous discussions over the Infobox map on this talk page, and most of them were in fact complaints about territories presented in the map being too limited rather than too extensive. Regarding how actually maximalist maps would draw the borders, I believe the one from the highly influential Historical Atlas of China by Tan Qixiang et al. (largely the same as the second file in this talk section) is a good starting point as it was also frequently cited in the previous discussions. There are publications that claim even more extensive territories as well but I would not go into those here. Esiymbro (talk) 13:07, 1 July 2026 (UTC)Reply
I still fail to see how they contradict the Cambridge History of China map (and) seems consistent with the territorial extent given in the current Infobox map.
The issue is none of the 7 sources I’ve listed, including the Cambridge History of Inner Asia, give any scope or detail of the protectorates that would support such a map and that the majority of sources do not support such maximalist claims. Thus it cannot be said to be the majority/mainstream view and would be inappropriate for the infobox. 1 of the 7, Hans van de Ven, explicitly does contradict it but probably only does so because he is the only one of those 7 sources to show the same map so I'd argue it makes sense that he is more explicit on his opinion regarding it than the others.
Generally historians are not going to list all the things they do not support as that isn't practical. An omission of information from a single source or 2 can be discounted, but that the information that would justify the scope of the map is missing for all 7 and so more much difficult to discount in my view.
It is true that they did not describe the protectorates.
They not only do not describe them, they do not even name the supposed protectores, nor name almost all the cities and territories lised in the intro of the General to Pacify the west, or generally give any real info as to their scope, they also give no evidence of any Tang imperial presence, garrison, administrative centers, officials, or any physical archaeological evidence beyond Anxi either.
the authors likely felt that it was not needed to write (and) details of administrative structures that did not leave a lasting impact on the Tang or Central Asia.
I’m not sure trying to infer ‘what the authors felt’ is usually a good way to interpret sources. And it is more than that just they didn’t have an lasting impact; there seems to be little to no actual evidence of the protectorates beyond Anxi existing in any form but on paper or a map.
“...with its territory taken over by the Chinese, the Western Turk realm ceased to exist.”
Rather than being consistent with the maximalist map (just so I conveniently have something to refer to as) I’d argue that quote is problematic. The Tang had no presence or administration in the claimed protectorates beyond Anxi thus their control and authority would rely entirely on the loyalty of the local Khan; but if the Western Turkic realm ‘ceased’ to exist it can’t very well have been a body through which the Tang could have claimed the maximum former borders of said realm. Instead they would be in fact only control whatever fragments of that realm their puppets controlled. Thus in my opinion the maximalist Tang claims in the West rely on the Western Turkic Khanate not disingtegrating precisecely because of the lack of Tang presence and real control through adminstration centers, official, garrisons etc.
I'd also quote a concised version what Qiushufang said earlier when he described Chinese protectorates; from what he said it doesn't sounds like the protectorates claimed beyond Anxi could have ever been protectorates in fact.
"It did not imply protection of local polities as independent entities or a separate part of the Tang empire under "less administration. Protectorates with stationed military garrisons were military commands meant to solidify and stabilize Tang control over a militarily precarious frontier region, not to denote their protection of local polities, although that could also be used as justification. The Tang never intended to protect either Goguryeo or the oasis states, much less let them go, but to control them as their de facto ruler.
Protectorate garrisoned cities such as the above (Turfan) were treated as normally administered cities of the Tang empire. No doubt there were also cities under protectorate rule that did not have the same level of direct rule, but that would be due to other factors such as logistics, duration of rule, and military conditions, rather than simply because they were part of a protectorate. The distinction between protectorate and non-protectorate is therefore not a legal distinction predicated on core vs. non-core territory, or more vs. less administration, but rather whether it was considered part of a territory administered by a military command.” - Qiushufang Guile's Theme (talk) 15:14, 1 July 2026 (UTC)Reply
@Guile's Theme: It is really impossible to dismiss a monumental source such as The Cambridge History Of China, Volume 3 as "not mainstream", simply because you manage to find 7 or so sources that don't go into the same amount of details about the Chinese Protectorates in the West. It is not just a matter of arithmetics: this most authoritative source, also being the largest (15 volumes for The Cambridge History Of China), naturally goes into deeper research and presents details which the others won't bother with. A few other sources still repeat this level of detail though, such as Blunden, Caroline and Elvin, Mark (1983), Cultural Atlas of Chinapp 92-93, Oxford: Phaidon Press, ISBN0-7148-2309-0 (Mark Elvin was one of the leading Western historians of China). As such, what you call the "maximalist map", as corroborated by The Cambridge History Of China (map 8 p.281) is probably simply the most precise and best documented map we have of the maximum extent of Tang influence in Central Asia. I suggest you self-revert your recent deletions in the many articles where that map was present . Pataliputra!(talk)18:30, 1 July 2026 (UTC)Reply
It is really impossible to dismiss a monumental source such as The Cambridge History Of China, Volume 3 as "not mainstream"
I am not dismissing it, I'm noting that the majority of sources do not offer anything to support it and it is thus such a map represents a minority view and that the infobox map should represent something a larger collection of high reputable sources/authors could agree on.
Instead, I'd argue you are dismissing how little there is in terms of any evidence or writings to support the scope of such a map in the grander picture and putting that one map from one source of equal standing to the names of 6 experts on the time and place I've brought up does not feel like proper balance and therefore not proper NPOV.
this most authoritative source
Actually, published in 1979, the one you are championing is the single oldest source out of all the ones talked about and listed by either of us. It is impossible in my view to somehow argue this single older work is more authoritative than those of 6 bigger contemporary names in the field, including Denis Sinor's The Cambridge History of Inner Asia (1990), or that any single source could outweigh all 6 respected names.
I also find the history of the infobox to disprove the notion the current map is a mainstream view, the older map was incumbent from 2008 to 2023; How could a non-mainstream map on the most high profile history article of China been wrong for so long when the map/source you are pushing was published 1979? For over 15 years the map you are talking about existed in that book but was never chosen over the incumbent infobox map which I'm trying to bring back an updated version of. And when I look at the discussion that lead to it being replaced in May 2023 I saw a concerning lack of actual/serious scrutiny at the time. IE here.
Who in that discussion was putting the map change at the time under any kind of scrutiny? No one that I can see. Is there another discussion before that I'm not seeing? I don't think there is a single credible editor who put it under any rigor at the time and if there is another relevant discussion please link it to me or point me towards it.
The Cambridge History Of China), naturally goes into deeper research and presents details
What is the source of this claim? The Cambrudge history of China is longer because the temporal span and geographic extent of China's history is much bigger/longer. If anything I'd say length is not an asset here as the shorter work, The Cambridge History of Inner Asia is the more focused on the specific time and place being discussed.
A few other sources still repeat this level of detail though, such as Blunden, Caroline and Elvin, Mark (1983), Cultural Atlas of China
This is the first time I've ever seen a illustrative Atlas used as a scholarly resource, also from 1983 and older than all 6 of the names/source I've cited. Also, Blunden is an illustrator, Mark Elvin was a professor and historian but per his wiki article, "he specialised in the late imperial period."
The Cambridge History Of China (map 8 p.281) is probably simply the most precise and best documented map
On what exactly do you base this arguement on other than it is now a requirement for the maximalist map to stand since it is not supported by the 6 other sources I listed? And in the case of Hans van de Ven, the only one of the 6 to show the map, explicity deprecates the claimed map.
So to sum up, I've cited 6 of the biggest and more authoriative names/authors on the time and place possible, all these sources are newer than the single one you are championing published 1979, which amounts to just a map with little to no explanation or realistic implementation or evidence, and you claim that single oldest source brought up out weighs ALL others because you claim it is the most aurhoritative for... what reason exactly?? Guile's Theme (talk) 23:39, 1 July 2026 (UTC)Reply
I suggest you self-revert your recent deletions in the many articles where that map was present [1][2][3][4][5][6][7][8][9][10][11][12]
I am willing to revert all those edits and then review each implementation of your map again on a case by case basis and willing to agree to the map appearing in the body of articles with appropairate edits as to the caveats of said map if you are willing to endorse a less maximalist map as the infobox map for the Tang Dynasty (or any article that would use the same map in the Intro/Infobox) and a return to something more in line/based off the 2008 to 2023 incumbent map. I feel that would be a fair and equitable compromise in the context of the many sources we've discussed and what they say, or more often don't say, on the topic. Guile's Theme (talk) 01:30, 2 July 2026 (UTC)Reply
Again, a source doesn't disprove a fact by not mentioning that fact, and an author doesn't become fringe simply because they provide more details than others. Based on the content you provided, there is no inconsistency between the other sources and the Cambridge History of China. The "caveats" mentioned only appeared in van de Ven's map legend, but its claim of no Tang presence beyond Anxi clearly contradicts most other sources that suggest a Tang takeover of Western Turk territory. I don't understand why the unsourced claim that the CHoC is maximalist is still here either, despite evidence to the contrary.
I support Pataliputra and Kanguole's last comments. The deleted map needs to be restored in all articles where it was present. Esiymbro (talk) 13:19, 2 July 2026 (UTC)Reply
(Tagging because I believe this can serve as a response or follow-up for all 3)
The only author/source I'm not currently able to find an online copy for to quote of the names I previously gave is Michal Biran's The Empire of the Qara Khitai in Eurasian History: Between China and the Islamic World but I'll contiune to look for that one
To update on this, I did eventually find a copy of Michal Biran’s book here. Unfortunately her book did not mention anything on the period as it only picks up after the 9th century and the later part of the Tang period at the earliest.
However as recompense I attained copies of:
1) Jonathan Karam Skaff, 2012 , Sui-Tang China and Its Turko-Mongol Neighbors: Culture, Power, and Connections, 580-800 (bought as I could not find an online copy but this is a highly respected work of interest to me anyway)
Both authors are very highly reputed, authoritative on the time and period, would be at the top tier or near the top of all the names mentioned in the discussion thus far, and widely cited by other authors. Skaff’s work is also the most focused on the exact time and place out of all of the sources brought up so far as it deals specifically from 580 to 800 CE. While Sören Stark is a leading expert on archeology and material culture in Inner Asia during this period.
What they say/quotes;
1) Despite Skaff’s work being entirely focused on the 220 years most relevant to our topic, just like all of the 6 previously quoted authors, he does not mention by name or city/territory any of the supposed protectorates further west than Anxi and the only 3 quotes relevant to our discussion in his entire work that I could find were:
p. 177 “After the Tang conquest of the Western Türks in 657, Gaozong successfully invested two client qaghan’s to rule over their ten subordinate On Oq tribes. Ashina Mishe headed the five eastern tribes and his cousin, Ashina Buzhen, controlled the five western tribes”
Nothing New, same lack of any detail that could support either maps just as well.
p. 248 “Other efforts to establish communications systems are evident for the Western Türks around 660 [11^ “After the conquest of the Western Türks, Gaozong established an official position at Tingzhou (Beiting) called canjiang, to handle correspondence with the submitted tribes]"
That is modern https://en.wikipedia.org/wiki/Jimsar_County, and could not have realistically been used to administer communication effectively to the claimed expansive map.
p. 279“The illusion of a China-centered world that infuses the traditional Chinese sources has obfuscated the multipolarity of medieval Eastern Eurasian politics. Sui-Tang empires faced competition for the allegiance of subordinate Turko-Mongols from great powers based in Mongolia and Tibet. Despite the customary ideal that patron-client bonds were exclusive, borderlands and other politically unstable environments provided clients with opportunities to form dual or multiple loyalties either concurrently or serially”
IE arguably Tang patronship of Turkic leaders/warlords did not result in exclusive loyalties or even create a true vassal relationship and perspective of the relationship might have been very different from the Tang court to the frontier reality.
2) Sören Stark didn’t have any archaeological or material evidence to offer in support of the claimed Tang protectorates beyond Anxi either nor any words to that effect. What he did talk about that I found interesting was that the ‘disintegration’ of the Western Khanate began in the ‘early 630s’, somewhere between 20 to 30 years before the Tang defeat of Ashina Helu, which makes the maximalist map claims less believable since they’d rely on the Western Khagante being intact in lieu of any Tang garrisons, officials, administration centers, etc.
p. 257 “a look at the political and social situation in the steppes of the Seven Rivers region: from the early 630s onward, the Western Qaghanate entered a crisis and progressive disintegration, until in the year 658 it fell to the armies of the Tang and their allies. Until the end of the century—until the rise of the Türgesh— the state/realm of the Ten Tribes existed in a largely shadowy existence under Chinese Ashina puppet rulers.
A crucial key to answering this question lies in examining the political and social situation in the steppes of Semirechye: beginning in the early 630s, the Western Qaghanate experienced a crisis and progressive disintegration, ultimately succumbing to Tang armies and their allies in 658. Until the end of the century—and the rise of the Türgesh—the *On Oq Bodun* (10 arrows) led a marginal existence under Chinese-backed Ashina puppets. In this context, the supra-tribal cohesion of the Qaghanate was likely significantly undermined in both political and social terms. At the same time, a large portion of the Qaghanate’s former military-administrative elites—who operated outside traditional tribal structures—lost their established political ties; this effectively "released" a substantial pool of military retinues, which gathered around the Qaghanate’s former military commanders. We can only surmise that, in the ensuing decades, some of these commanders found employment in the military service of the small and micro-principalities flourishing in Transoxiana. Given the fragmented political landscape of the region, there was almost certainly a significant demand for mercenary forces of all kinds. The political situation in Ferghana around 630 serves as a telling example:
"For roughly a decade, the region had been without a supreme ruler. Local nobles ruled by virtue of their own power and were independent of one another."
Conflicts among these numerous regional powers were likely commonplace, even though the sources offer little in the way of detail.
The beginning of Muslim raids from 654 onward certainly further increased the need of the Transoxanian states for mercenary troops. In return, the widespread availability of Turko-nomadic warrior groups and aristocratic retinues also ensured the political survival of even smaller regional powers during the 7th and early 8th centuries. This made it possible even for minor dihqāns to surround themselves with a powerful military following and to pursue a policy relatively independent of the traditional centers. Steppe mercenaries thus appear to have further stimulated political fragmentation in pre-Islamic Mā warāʾ al-nahr and in a sense acted as a catalyst.''
To summarize I now have a list of 8 authors in total, all of which more recent than the 1979 map source being championed, at least 7 of whom are experts on the topic, time, and place in one form or another, none of whom name any of the supposed protectorates beyond Anxi by name or the territories/cities claimed by the map, and none of whom describe anything that could be said to support maximalist map anymore than it supports the 2008 to 2023 incumbent map. The 8th author, Hans van de Ven, the only one to also show the more expansive 1979 map in his 2022 work, explicitly deprecates it. I’d note usually when historians don’t mention something that means they don’t support it. Histories are told by listing all the things that did happen and not by listing everything that did not. I'd also note at this point I'm also struggling to find names of new authors who are experts on the time and place in question who are as well regarded as the names already presented.
I’ve also noted that there are other Wikipedia articles the ‘maximalist’ map is somewhat incongruent with as it overlaps in claimed territory to some extent in regards to Merv, Nishapur, Herat, Balk, and more.
I'd point out most of the source for those articles are Arab, Iranic, or Muslim and they don't mention Tang suzerainty over the areas beyond Anxi. Also multiple articles for the claimed cities/territories in the maximalist map never mention any Tang control of any sort:
Arguably, implicitly, in a disagreement between a conservative and a more expansive map the more conservative map will have more support in my opinion. This is because the expansive map supporter does not actually contest any part of the conservative map, rather it is the conservative map supporter that is contesting what the maximalist wishes to add; thus they implicitly agree on the conservative map while disagreeing on the more maximalist version. IE, if we both agree we want a pepperoni Pizza but one wants to add green pepper to half and the other disagrees we still implicity agree on the pepperoni half.
All that said, I've collected 5 Maps of Tang China from the most popular textbooks I could find, one of which was the example James Millward gave when he said, p.33, “Bentley and Ziegler's popular history textbooks represents the Tang with a map of the same four year period”. Notably, not a single one of these maps supports the more maximalist version of the map with the protectores claimed beyond Anxi, and all the maps but 1 opt to show the Tang at 750 AD and not 660. All 5 maps and all 8 sources mentioned thus far are also more recent than the source of the 1979 map. I believe this goes some way towards proving what the 'mainstream/majority' view map of the Tang Dynasty in sources tends to look like.
There has been no response for 6 days but I’ve not been idle. I’ll still plan to make a change to the infobox map in a few days if there are no responses. If you revert, I would appreciate it if you could explain why in response to the consolidated argument & survey of sources below. My intention is to focus on the sourcing and most relevant Wikipedia policies from here on.
To begin, it would be helpful to review the Wiki policies most relevant here; foremost the WP:WEIGHT subsection of WP:NPOV. To quote; “Neutrality requires that mainspace articles and pages fairly represent all significant viewpoints that have been published by reliable sources, in proportion to the prominence of each viewpoint in those sources. Giving due weight and avoiding giving undue weight means articles should not give minority views or aspects as much of or as detailed a description as more widely held views or widely supported aspects. Generally, the views of tiny minorities should not be included at all, except perhaps in a "See also" to an article about those specific views."
And, “Undue weight can be given in several ways, including but not limited to the depth of detail, quantity of text, prominence of placement, juxtaposition of statements, and use of imagery”.
Thus, my concern is not whether 1, or a clear minority, of reliable sources support the existing infobox map. Rather, the question is whether the current infobox represents the balance of the modern specialist literature and published maps, or whether it gives disproportionate prominence to one expansive interpretation. To answer this question I’ve compiled a rather exhaustive survey of maps in addition to written sources ie texts.
I’ve found 9 additional maps to add to the previous 5 bringing the total to 14 maps; most from the most popular textbooks I could possibly find that contained maps of Tang China, and a few from books or Atlases. Including a map from The Cambridge World History Volume V. The source of most maps can be seen within the images. (Note: citations for all 14 maps and additional ones near the beginning of 'Response' section. Also, the same map appears twice because it was in 2 different books)
Taken together, the maps present a remarkably consistent picture. I'd note that none of the 14 support the current infobox map. Most instead depict the Tang at 750 AD, at least one at 700 AD, and none explicitly use a date in the seventh century. Even the map that most closely resembles the current infobox is dated to 750 AD. In three examples the accompanying text explicitly describes 750 as the Tang's "peak" or "greatest extent". Because textbooks are intended to synthesize mainstream historical literature rather than advance new interpretations, they provide one useful indicator of how the prevailing scholarly view is represented. In my view, this already raises a substantial WP:WEIGHT concern, even before considering the textual evidence below.
This is much more complicated to present than the maps. In total there are 19 sources I’ll divide into 3 groups. I’ll only provide the citations, page numbers, short quotes where practicable, or a summary of the cited pages instead of exhaustive quotes for the sake of brevity. Most have already been cited/quoted with links to the source in the previous discussion.
1) Denis Crispin Twitchett, John King Fairbank (1979), The Cambridge History Of China, Volume 3, pp. 280-281
A well regarded work that is the source of the Infobox map and thus the principal support for the current infobox map although it is the 2nd oldest source in the survey. The work is not particularly focused on the exact time and place being a broader history of China. I would also note that the map's inclusion of Zaranj appears difficult to reconcile with the accompanying statement that Tang territory extended "all the way to the borders of Persia." Zaranj lies only 5–10 km from the modern Iranian border and, as shown in the accompanying map, was well within the borders of every major Persian empire.
2) Caroline Blunden, Mark Elvin (1983), Cultural Atlas of China, pp. 92-93
An illustrative Atlas whose contributing historian, Mark Elvin, per his Wiki article is, “specialised in the late imperial period”. Also somewhat old, and an identical copy of the map from the 1979 source.
3) James A Millward (2010), Eurasian Crossroads: A History of Xinjiang, p.33
Most of Millward's scholarly work focuses on the Qing period and modern Xinjiang. Eurasian Crossroads is a broad survey rather than a specialist study of seventh-century Tang expansion. Nevertheless, it is one of the clearest sources supporting the current infobox map with the following quote; “Tang suzerainty thus extended thinly from Talas, Tashkent, Samarkand, Bukhara, Kabul, and Herat, and as far south west as Zarang in Modern Iran.”
4) Zhenping Wang (2013), Tang China in Multi-Polar Asia, A History of Diplomacy and War, pp. 16, 42-46, 257
Wang appears to be an expert in Tang History & Politics although I could not find a faculty page/wiki article to quote.
His account of the 651–658 Tang campaign places the Western Turk realm north and south of the Tian Shan, with Tang operations largely confined to modern Xinjiang and the furthest advance reaching Tashkent. He nevertheless notes the establishment of the Kunling and Mengchi protectorates, the only source to name them, and Tang vassals in the Pamirs, while observing that local rulers could interpret Tang titles as recognition of their own legitimacy rather than subordination.
5) John Haywood (1998), Historical Atlas of the Medieval World, AD 600-1492, Barnes & Noble, p. 3.2.
Of all 19 sources this is the only one I was not able to find any copy of to see what he actually has to say even after exhaustive searching including making a Resource Exchange request a week ago. However, per his wiki article Haywood is, “considered an authority on Vikings and the Celts,” and appears to make Atlases on a very wide variety of topics.
Note: The source Shin, Michael D., ed. (2014), Korean History in Maps, Cambridge University Press is cited for current infobox map but only talks about the Korean aspects of the map and thus not counted in this survey.
Group 2: Sources cited on Wikipedia that, upon inspection, did not support the cited claim or used qualified/skeptical language
1) Christopher I. Beckwith (1987), The Tibetan Empire in Central Asia: A History of the Struggle for Great Power among Tibetans, Turks, Arabs, and Chinese during the Early Middle Ages, p. 39
Beckwith is notable for his expertise on the history of the Tibetan empire during this period. Beckwith refers to Tang control of Tokharistan and Sistan as "alleged," and does not discuss the other cities or territories shown on the current infobox map.
2) Hans van de Ven (2000), Warfare in Chinese History, p. 119
Described in his wiki article as, “an authority on the history of 19th and 20th century China”.
Van de Ven is notable for being one of only 3 sources to show the map the infobox map is based upon and the only modern example of the 3. But he rather explicitly deprecates the claimed map, more so than any other source, with the following quote; “Of the protectorates shown, only the Anxi protectorate was a real administration governed from Lucha, and with garrison armies. The others were "spheres of influence" with no permanent officials, no fixed administrative centres, and no military forces. The boundaries shown are very approximate and not to be taken as indicating that these areas were in any way controlled from Chan'an (Tang capital). They give a notion more of Gaozong's imperial pretensions.”
3) Harold M. Tanner (2010), China: A History (Volume 1): From Neolithic Cultures through the Great Qing Empire, p. 111
Tanner describes himself; “While I teach courses covering the entire range of China's history, from the Neolithic to the present, my research interests have always been in the history of China in the 20th century.”
Tanner's description appears consistent with the Tang at or near its greatest territorial extent before the Battle of Talas (751) rather than describing the situation in the seventh century ie quote; “At its greatest extent, the might of the Tang Empire extended into modern Afghanistan and Uzbekistan, and to the borders of Iran, even clashing with (and losing to) the Abbasid Caliphate at the Talas River in modern Kyrgyzstan in 751”.
4) Anna Akasoy, Charles Burnett, Ronit Yoeli-Tlalim (2016), Islam and Tibet – Interactions along the Musk Routes, p. 51
This source is cited in support of the following quote from the Tokharistan article, “in the 7th and 8th centuries, it was incorporated into the Tang dynasty”.
Yet in the cited source there is nothing to support this and the cited page describes the travels of a Chinese buddhist monk named Xuánzàn from 629 to 645 to places “outside of the control of the Tang dynasty”.
5) Khodadad Rezakhani (2017), ReOrienting the Sasanians: East Iran in Late Antiquity, pp. 181–182
Khodadad is cited in the Ikhshids of Sogdia article as supporting “The Ikhshids became nominal vassals of the Chinese Tang dynasty.” Yet again the pages cited do not contain any such information.
6) René Grousset (1970), The Empire of the Steppes: A History of Central Asia, p. 119
Cited in support of the list of cities/territories on the Protectorate General to Pacify the West article that supports the infobox map broadly but the cited page, 119, only talked about the 8th century and not the 7th.
1) Peter B. Golden (1992), An Introduction to the History of the Turkic People, p. 136
Golden is a leading expert on early Turkic history and his work is often used as a benchmark on the topic. Despite that, Golden's discussion of the subject is brief, stating only, “The Western Qaganate now formally came under Chinese "protection" and was divided into 10 prefectures.”
2) Denis Sinor (1990), The Cambridge History of Inner Asia, pp. 309-310
Denis Sinor is described in his wiki article as, “one of the world's leading scholars for the history of Central Asia”. He is as sparse as Golden is on this topic; “and with its territory taken over by the Chinese, the Western Turk realm ceased to exist."
3) Étienne de la Vaissière (2005, 2006), Sogdian Traders: A History (Handbook of Oriental Studies); "Les Turcs, rois du monde à Samarcande," in Royal Nawrūz in Samarkand, p. 137, 200 + p. 149
De la Vaissière, a leading specialist on Sogdia and Central Asia, describes the Western Turkic Khaganate as beginning to "disintegrate" after 630 and states that Chinese appointments of Central Asian rulers represented "the fiction of appointment by the Chinese emperor." In his interpretation of the Afrasiab murals, he emphasizes the symbolic "Kings of the World" motif, arguing that the depiction of Chinese embassies bearing silk symbolizes diplomatic equality with the Turkic qaghan rather than evidence of Tang political control over Samarkand.
4) Jonathan Karam Skaff (2012), Sui-Tang China and Its Turko-Mongol Neighbors: Culture, Power, and Connections 580-800, pp. 177, 248, 279
Jonathan Karam Skaff's research focuses on relations between Sui–Tang China and the Turkic states of Inner Asia, and his book is one of the few devoted specifically to the subject. He describes the investiture of two client Kaghans over the Duolu and Nishibi, an office in modern Jimsar County, Xinjiang to correspond with the "submitted tribes," and argues that Chinese sources project "the illusion of a China-centered world" despite the multipolar nature of medieval Eastern Eurasian politics and the prevalence of multiple loyalties among frontier rulers.
5) Nicola Di Cosmo (2009), Military Culture in Imperial China, p. 184
Per his wiki article; “His main field of research is the history of the relations between China and Inner Asia from prehistory to the modern period.”
Di Cosmo describes the Tang 657 victory over Helu as consolidating rule over much of modern Xinjiang while establishing only temporary suzerainty over a “region” further to the west.
6) Sören Stark (2009), Die Altturkenzeit in Mittel- und Zentralasien: Archäologische und historische Studien [The Old Turkic Period in Inner and Central Asia: Archaeological and Historical Studies], p. 257
Stark agrees with other specialists that the Tang client qaghans ruled the Duolu and Nishibi and identifies their collective territory with Semirechye. He describes the Western Turkic Khaganate as disintegrating after 630, releasing Turkic military elites who entered the service of the principalities of Transoxiana, strengthening local rulers and contributing to increasing political fragmentation, a view he emphasizes and notes is shared by Marshak and Raspopova. (2 other authors)
7) S. G. Klyashtorny (1996), History of Civilizations of Central Asia, Volume III: The Crossroads of Civilizations: A.D. 250–750, p. 28
Klyashtorny is a noted Russian specialist in Turco-Mongol history. He notes that, “the (Tang) empire reached its zenith under Hsüan-tsung (713–756),” and afterwards on the same page, “Tang troops advanced to the borders of western Central Asia, but were unable to consolidate their hold,” before then moving on to the 8th century conflicts between the Tang and Tibetan Empire.
8) David A. Graff (2002), Tang Military History, p.194
Per his Kansas State Faculty profile; “His research focuses on Chinese military history, especially that of the Tang dynasty (618-907)”.
His only input on the topic in question is when he described Tang "authority" extending, “into Central Asia as far as the north eastern border of Persia,” as a result of Ashina Helu's defeat.
The Western Turkic Khaganate underwent almost 30 years of continuous civil wars and disintegration until the Chinese defeat of Helu and installation of ‘puppet’ Khans around 657/658.
The Tang-backed client qaghans governed the Duolu and Nishibi, whose territory is broadly identified with Semirechye (the Seven Rivers region).
Among the surveyed works, the furthest west Tang military operation is described as reaching Tashkent in order to capture Ashina Helu. None discusses Tang garrisons, administrative centres, or any permanent Imperial Tang presence beyond Anxi.
The meaning of Tang titles and "submission" is treated cautiously by some specialists while others describe Tang territorial claims using terms such as "alleged", "fiction", "imperial pretension", or "nominal". Stark further emphasizes that the collapse of the Western Turkic Khaganate released military elites who strengthened the increasingly fragmented principalities of Transoxiana with Skaff and Vaissière making similar points regarding the multipolar nature of the region.
Beyond Tashkent and the Anxi Protectorate, discussion or even mention of the remaining claimed cities and protectorates is generally sparse if not entirely absent outside Millward in regards to the most distant places who himself only uses "thinly" to descirbe Tang control. Particularly among the specialist works surveyed, only a minority explicitly describe any of them as belonging to the Tang sphere in the seventh century.
Taken together, I believe this survey demonstrates a significant WP:WEIGHT issue regarding the current infobox map in addition to the miscited sources. It is not that the current infobox map is wholly unsupported as several reliable sources, at least one of which is a specialist, do support an expansive interpretation. Rather, among the broader body of modern specialist scholarship and published maps surveyed here, that interpretation appears to represent a minority position. I therefore believe the infobox should better reflect the balance of modern scholarship. I also note that the previous infobox map depicting 700 AD remained in place from 2008 to May 2023. I have not found evidence in the archived talk-page discussions that the May 2023 change received the kind of source-based scrutiny reflected in the survey above. Guile's Theme (talk) 16:04, 8 July 2026 (UTC)Reply
You have assembled a lengthy collection of authors giving few or no details about the Tang defeat of the Western Turks. However, constructing a majority view from all that silence is unsound. Sometimes people don't write about something because they're more interested in something else.
It is also unsafe to rely on Wikipedia articles, including for authors' biographies. For example, Mark Elvin is best known on Wikipedia for his High-level equilibrium trap theory, but his work ranges across the imperial period, with a particular interest in the constraints of geography and the economy.
As for the maps of the Persian empire, the Sasanian empire was destroyed by Arab armies in the decades before the period in question. Kanguole13:43, 9 July 2026 (UTC)Reply
Before responding to the specific points you raised, I'd note that there are several parts of the survey that I don't think your reply addresses:
The collection of 14 published maps
The various miscited sources
The WP:WEIGHT argument presented in the survey
The undue prominence being given to an interpretation that appears much less frequently in the broader body of published maps and texts.
I particularly don't think your reply addresses the survey of published maps. Regardless of how individual historians summarize the period in prose, fourteen published maps drawn from textbooks, atlases, and general histories converge on a markedly different representation from the current infobox. Because the issue here concerns the lead image rather than only textual description I believe the survey of published maps is directly relevant.
giving few or no details about the Tang defeat of the Western Turks.
On the contrary, collectively they provide considerable detail on the disintegration of the Western Turkic Khaganate from the early 630s onward, the 651–658 campaign against Ashina Helu, the identification of the Duolu and Nishibi territories with Semirechye, and the political situation in Semirechye and Transoxiana. What is largely absent from their discussions is an independent description of Tang authority extending across the same territorial configuration depicted in the current infobox map.
Sometimes people don't write about something because they're more interested in something else.
I agree that silence alone does not establish rejection. My survey therefore does not rely on silence alone, but on the cumulative pattern formed by the published maps, the specialist literature, the explicit caveats and qualified language used by several specialists, and the sourcing issues identified above. Accordingly, when a large body of specialist literature discusses this exact period in considerable detail without independently describing Tang authority extending across the same territorial extent shown in the current infobox map, I believe that is relevant to assessing the due weight that interpretation should receive in the infobox image.
It is also unsafe to rely on Wikipedia articles, including for authors' biographies
Fair enough. Where possible I relied on publishers' descriptions, faculty profiles and other sources. The Wiki articles were conveniently summarized information already available elsewhere and Mark Elvin did not stand out as a specialist on the exact time/place among the other 18 names surveyed.
In closing, I'd also point out that the sources I assembled represent four different scholarly disciplines: Archaeology, Sinology, Turkology, and Iranistics (particularly Sogdian studies). My intention was to avoid relying on any single scholarly tradition and instead survey how this question is treated across the relevant fields. Guile's Theme (talk) 15:21, 9 July 2026 (UTC)Reply
Hans van de Ven & Beckwith were both cited in support of it originally when the former explicity deprecates it and the later uses qualified lanugage such as "allegedly" to mention 2 out of the 4 or 5 claimed protectores beyond Anxi.
There are only two sources attached to the western portion of the infobox map itself; the third citation relates only to Korea and the eastern portion.
The first is the original 1979 source from The Cambridge History of China. The second is the 1983 Cultural Atlas of China, which reproduces essentially the same map and is an illustrative Atlas. I was therefore not referring to the citations attached directly to the infobox map itself when discussing miscited sources.
Rather, I was referring to sources cited elsewhere in support of either near identical maps or wiki article text to the same effect. Six such sources either did not support the cited claims on the referenced pages, qualify them, or deprecate them.
My broader point, however, is not whether the current map has citations, but whether that particular interpretation should receive the prominence of the lead infobox image under WP:WEIGHT. I don't think your replies have yet addressed the survey of published maps or the broader due weight argument.
At least in the context of Wikipedia, reviewing and summarizing literature like this is a clear-cut case of WP:SYNTH. As you have pointed out yourself, you are trying to draw a conclusion based on what the sources do not say. Many of the summarised statements also apparently refer to other periods rather than 658-661. In any case, none of the new specialist texts added here offer maps that are contradictory to the one in the Cambridge History of , or contain descriptions of Tang territorial extent in this period that are detailed enough for us to construct one.
As for the maps in the Imgur link, I cannot verify the sources of them, but most if not all seem to be from very broad works on world or regional history rather than specialist works on medieval China or Central Asia. The quality of Asian history maps from general historical atlases is, in my experience, inconsistent at best: even in the link you posted you can see how much difference there is among the maps, even the 750 AD ones. I would never pick a map from such publications over the current one. Esiymbro (talk) 19:05, 9 July 2026 (UTC)Reply
I don't believe this is WP:SYNTH or at least not in the sense intended by the policy.
I'm not combining unrelated facts to reach a novel historical conclusion, nor arguing that silence amounts to rejection. Rather, I'm assessing due weight, a core Wikipedia policy, by surveying how reliable sources represent the subject. The question is whether the current lead image fairly reflects the balance of published scholarship. I believe the survey raises substantial WP:WEIGHT concerns that it does not.
I also don't think the survey of published maps can simply be set aside. Because the issue concerns an infobox image rather than only text, the way historians and publishers choose to depict Tang territorial extent visually is directly relevant. The survey includes university press textbooks, historical atlases, and The Cambridge World History, not merely popular world histories.
While individual maps naturally differ in detail, the broader pattern is consistent: none reproduce the western territorial extent shown in the current infobox map, and several explicitly identify 750 AD as the dynasty's greatest territorial extent. That seems relevant to whether the present image gives disproportionate prominence to one interpretation under WP:WEIGHT.
Furthermore I do not believe the absence of alternative specialist maps necessarily resolves the issue in favour of a single 1979 Cambridge map. If anything, it makes it more appropriate to consider how subsequent reliable syntheses and published historical maps have represented the same subject in the 47 years since.
Finally, as I state in my previous reply, this survey also presents a balance on several different disciplines on a topic/time/place they intersect; namely Archeology, Turkology, Sinology, Sogdian, and Tibetan studies. One reason for doing so was to avoid relying on the perspective of any single discipline. Taken together these disicplines present a more limited or nuanced picture of Tang authority in Central Asia than is reflected by the current infobox image. Consistency between related articles is also important. Some wiki articles seem to clash with the Infobox map such as the 652 siege of Herat. If overlapping articles rely on different interpretations of the same events without clear sourcing, readers are presented with conflicting accounts.
Ultimately, I don't think the question is whether the existing Cambridge map is reliable. Rather, it is whether that particular 1979 interpretation should receive the prominence of the lead infobox image in light of the broader modern body of published maps and scholarship discussed above. I still don't think that question has been directly addressed by anyone. I yet again note that the current infobox map replaced the previous incumbent map in 2023 of 15 years which depicted 700 AD. From archived discussions I have not found evidence that the current map received any kind of scrutiny at the time.
In closing I also feel left with some very poignant questions of my own I believe the defenders of the infobox map should feel compelled to answer; if the 1979 Cambridge map is so authoritative and accurate as you claim why is it so isolated? Why are there no sources that support the map published it in 43 years? Why are virtually all published modern maps completely different, why are almost all more similar to one another than to that map, and why does the only modern publication of the 1979 map, from van de Vens (2000), explicity deprecate its claims? These are the sorts of questions that should have been asked before the map was changed in 2023. But since they were not I am asking them now. Guile's Theme (talk) 02:09, 10 July 2026 (UTC)Reply
Something else interesting has come up after more research, namely that the author of the only published map I know of that is more expansive than the isolated 1979 map, the 1982 map by Tan Qixiang, has this to say about his map per Wilkenson (2012):
“In his 1987 afterword to volume 8, Tan points out that many non-Chinese local rulers, particularly in the South, were simply not included. He also states that the editorial policy of drawing the boundaries of China to inlcude all Jimi in the Tang and Song, and rusia in the Yuan and Ming, was a weakness. The result is, for example, maps of Yunnan in volumes 7 and 8 give the false impression that tusi regiemes were under Ming control and that the "national boundary" included these regiemes, despite the fact that there were almost no demarcated borders between Yunnan and Southeast Asian kingdoms until the 1890s” as cited in The Historical Atlas of China wikipedia page.
Even though that quote is in reference to a different map it seems relevant because Tan's western depiction beyond Anxi is among the closest independent published maps to the 1979 Cambridge map.
And what he says also seems to be the best explanation for the discrepancy between all these maps. Combined with the similar things other authors have to say about Chinese 'pretensions', 'fiction', and 'alleged' control specifically in reference to some of the western claims of the current infobox map. Taking into account what the weight of sources say about the different interpretations between the court and the frontier of what acceptance of Tang titles meant, and the lack of physical evidence of Tang presence beyond Anxi. As well as the example Beckwith gives in regards to a refugee ruler of Khotan around 674/675 I discussed yesterday with Qiushufang.
Taken together, these sources seem to show that the 1979 map is not wrong if you mean to depict the borders according to Tang courtly declarations and records as opposed to if your intent is to depcit the borders where there existed any realistic degree of Tang control beyond what existed on paper in Chang'an. That also in my opinion best explains why the 1979 map differs so markedly from the majority of sources in the survey. Guile's Theme (talk) 06:41, 10 July 2026 (UTC)Reply
Simply put, there is hardly any disagreement between the text descriptions in the Cambridge History of China and the other sources. The broad version of facts are the same: Western Turk fragmentation, the battle with Helu, installation of client qaghans. The "caveats" of Tang control, its brevity and thin military presence, are mentioned. The Cambridge History stated Tang extent as reaching up to the borders of Persia which is corroborated by multiple others.
As I said before, there is no major contradiction between these sources, except the presence of a map in some publications and the absence of one in the others. The WP:WEIGHT concerns are, therefore, mainly about balancing between one highly authoritative source and your interpretation of the others. Are you really sure that had the other sources made maps, they would look significantly different? If wording such as "alleged" is enough to reduce Tang territorial extent shown in maps, how much should it be reduced? Had the Cambridge History not already made such adjustments for its map based on its own evaluation of Tang control (compare e.g. the Historical Atlas of China)? Without explicit contrary descriptions from sources, most of this discussion has essentially been WP:OR attempting to put the results from such conjecture into an infobox of an FA.
At the same time you are glossing over the major differences in the alternative maps you are proposing and their own differences. You have repeatedly said that some of maps from world histories, atlases, etc. claim that the Tang reached its maximum extent in 750 AD, but ignored the fact that most Tang specialists who discussed the matter clearly agree that it was 660 AD instead (see Pataliputra's earlier comment). In my opinion this is a good argument against using such non-specialist sources. Furthermore some of the maps you linked are clearly wrong even at the first glance, e.g. the one missing Anxi, or the one claiming all of Korea. 700-750 presents its own challenges, e.g. there are major disputes surrounding the Tang's northeastern borders in this period (also reflected in the disaggrement among the linked maps), so the credibility of any general map not citing specialist sources is in my view highly questionable. It is worth emphasizing the importance of the Cambridge History again here. None of the alternatives come from sources that come even close to it in terms of impact.
I am also not convinced that your source survey is complete or balanced, either. I don't have the time to do a major literature search here, but in one of the first comments I quoted the Son of Heaven and Heavenly Qaghan by Yihong Pan which is one of the few specialist works on Tang foreign relations and frontier administration. It broadly agrees with the currently cited sources, and also confirmed the description of Tang borders reaching Persia and the fact that the maximum extent was in 660 not 750. Some of the other sources you listed, e.g. Start, also actually mentioned the protectorates by name despite their brief descriptions. Esiymbro (talk) 07:55, 10 July 2026 (UTC)Reply
I think your response conflates several distinct issues. There is indeed considerable agreement concerning the fragmentation of the Western Turkic Khaganate, the defeat of Helu, and the installation of client qaghans. The disagreement concerns what those developments amounted to in territorial and administrative terms, how they are described and represented in published maps, and consequently how they should be depicted in an infobox.
The “caveats” of Tang control, its brevity and thin military presence, are mentioned.
That is precisely the issue. Those caveats are significant in the specialist literature, but they are not apparent the current infobox map. There is also a notable inconsistency here. If van de Ven is regarded as a 'non-specialist' in this context, the same characterization should seem to apply to Mark Elvin, whose illustrative atlas reproduces essentially the same map. Ironically, van de Ven himself, who is the only modern source that reproduces the map, explicitly cautions that, apart from Anxi, "the boundaries should not be understood as indicating control from Chang'an" and instead convey "more of Gaozong's imperial pretensions." Perhaps more importantly, several specialists employ similarly qualified language, including de la Vaissière’s “fiction of appointment,” Beckwith’s “alleged,” and Skaff’s criticism of the “illusion of a China-centered world.” The question is whether the present infobox map adequately reflects those distinctions and nuances made by specialists.
Are you really sure that had the other sources made maps, they would look significantly different?
Asking editors to speculate about maps that were never published isn't constructive. That is why I have focused on maps that were actually published. Those maps are themselves secondary source representations, not borders reconstructed by me from prose.
Nor does resolving the present WP:WEIGHT concern require editors immediately to construct a wholly new map from textual descriptions. Rather, it raises a substantial WP:WEIGHT question as to whether the current map should have replaced the previous infobox map in 2023.
The WP:WEIGHT concerns are, therefore, mainly about balancing between one highly authoritative source and your interpretation of the others.
I do not think that accurately describes the argument. The concern is based partly on the textual qualifications discussed above, but also on a substantial body of published maps from reliable sources. None of the maps surveyed reproduces the western extent of the current infobox map, and nearly all employ markedly more limited territorial depictions. Those are the judgments of their respective authors and publishers, not my own reconstruction.
Without explicit contrary descriptions from sources, most of this discussion has essentially been WP:OR
Considering the weight of published maps when deciding which map should occupy the infobox is not the same as original research. Nor does the absence of a modern specialist map automatically establish that the most expansive published interpretation should be preferred.
You are glossing over the major differences in the alternative maps
I acknowledge that the maps differ from one another, sometimes substantially. My argument does not depend on treating them as identical or claiming that every map is equally strong. The relevant pattern is that their disagreements largely occur within a more limited range, while the present infobox map’s western extent remains unusual among the published maps surveyed. Weaknesses in individual maps do not by themselves answer that broader WP:WEIGHT concern.
Most Tang specialists who discussed the matter clearly agree that it was 660 AD instead
I do not think that claim has been demonstrated. Pataliputra cited 3 works, one of which was the 1979 Cambridge source itself. The other two, Elvin and Millward, would seem to fall into the same 'non-specialist' category applied to van de Ven. Those three works alone do not demonstrate the specialist consensus you described. Nor do vague text descriptions by other specialists necessarily justify the specific territorial boundaries shown in the current infobox map."
None of the alternatives come from sources that come even close to it in terms of impact.
Reliability and impact do not automatically settle the separate question of due weight for a lead image, particularly where the visual implications of the map are qualified by later scholarship and are not widely reproduced in subsequent published maps.
the Son of Heaven and Heavenly Qaghan by Yihong Pan
Pan provides an important Sinological perspective, but the due weight question also benefits from a balance of scholarship in Turkology, Sogdian studies, Tibetan studies, archaeology, and Central Asian history, all of which contribute relevant perspectives on the nature and practical extent of Tang authority beyond the frontier of Anxi.
I am also not convinced that your source survey is complete or balanced, either
On the contrary, I believe the survey is inherently broader in scope and thus balanced because it incorporates scholarship from multiple disciplines. While the infobox map appears to derive primarily from the 1979 Cambridge map that has not been supported in a new source in 43 years. By contrast, my survey deliberately incorporates specialists from several disciplines in order to assess how the subject has been represented across the broader body of published scholarship. My concern is therefore not with any individual source, but with the collective weight of reliable published representations under WP:WEIGHT.
In closing I'd just ask one question: why should this one interpretation receive the prominence of the lead infobox when the broader body of reliable published maps represents the subject differently? I still do not think that question has been directly addressed. Not now nor when it was made the infobox map in 2023. Guile's Theme (talk) 10:05, 10 July 2026 (UTC)Reply
Because there are no alternative published interpretations in the form of a published map or text descriptions detailed enough to allow making one, except the substantially more expansive one in the Historical Atlas of China. Your other maps are 1. for other periods and therefore cannot be said to "represent the subject differently", and 2. from clearly poorer sources. Esiymbro (talk) 10:32, 10 July 2026 (UTC)Reply
Because there are no alternative published interpretations in the form of a published map or text descriptions detailed enough to allow making one
That is simply untrue, there was the previous 15 year incumbent which seems an outright better presentation of the balance scholarship overall as to what territory the Tang actually controlled vs claimed to control on paper.
None of this answers another very salient question; if this 1979 map is so good why has no other source deigned to use it in 43 years except to deprecate it? That is where Tan's explanation on the "weakness" and resulting "false impression" feels relevant as it is the only explanation offered up by anyone for the glaring discrepancy that is at the center of this dispute. Guile's Theme (talk) 10:54, 10 July 2026 (UTC)Reply
A map for 700 is a better representation of what the Tang controlled in 660? Not to mention the previous Wikipedia map's placement of many labels like Western/Eastern Turks and the Khitans is questionable or outright wrong. Esiymbro (talk) 11:23, 10 July 2026 (UTC)Reply
A map for 700 is a better representation of what the Tang controlled in 660?
Such a comment takes for granted the date should be 660 instead of 700 or 750 for no particular reason I can see as more sources seem to depict or explicity call the 8th century as the peak of the Tang. I feel 700 would be a good compromise between 750 and 660 on top of being the 15 year incumbent map that was changed without any scrutiny.
Not to mention the previous Wikipedia map's placement of many labels like Western/Eastern Turks and the Khitans is questionable or outright wrong.
I'd say that is a lesser issue compared with the Tang borders, the principal purpose of such a map, being "quesitonable or outright" wrong in terms of actual realistic territorial control in light of the weight of sources which is what I believe the infobox map should be depicting.
Also, such intransigence to admit the current infobox has notable issues while knitpicking every other map doesn't feel very balanced. I'm willing to work towards a better map, but your refusal to do so is not a valid defence of the current map that should have arguably never been made the infobox in the first place and never earned it through any scrutiny at the time. Guile's Theme (talk) 11:38, 10 July 2026 (UTC)Reply
That comment on the Historical Atlas of China is what an actual disagreement would look like. Despite the lengthy discussion here, there is nothing resembling that except maybe van de Ven who is, as you noted, not a specialist. But instead you are trying to use comments on the Historical Atlas of China to discredit the Cambridge History of China, whose version is not the same at all. Esiymbro (talk) 08:02, 10 July 2026 (UTC)Reply
there is nothing resembling that except maybe van de Ven who is, as you noted, not a specialist.
By that logic, the 1979 Cambridge History of China source would largely stand alone, since the same criticism would also apply to Millward and Mark Elvin, i.e. the two other sources you referenced from Pataliputra's comment. More importantly, it is not van de Ven's expertise that is most relevant here, but the fact that the only modern publication of essentially the same map felt the need to qualify how it should be interpreted. That is noteworthy in itself. Van de Ven explicitly cautions that, apart from Anxi, the boundaries should not be understood as indicating control from Chang'an, but rather convey Gaozong's "imperial pretensions". That is directly relevant to whether the map is suitable as the lead infobox image.
But instead you are trying to use comments on the Historical Atlas of China to discredit the Cambridge History of China, whose version is not the same at all.
I do not believe that engages with what I actually said. My point was that Tan's 1982 map is the closest published analogue to the 1979 Cambridge History of China map in its depiction of the Tang western frontier among the maps surveyed and they are also the 2 most 'expansive' versions of the map among all the ones published I was able to find. That is why I consider Tan's later comments regarding the "weaknesses" of that mapping methodology and the "false impression" it could create to be relevant as it is the only thus far possible explanation by any source for the discrepancy between the 1979 map and all the others surveyed rather than 'every other map but the 1979 map is just wrong even though the 1979 map has not been used in 43 years by a source that supports it'. Guile's Theme (talk) 10:17, 10 July 2026 (UTC)Reply
Van de Ven's argument of no Tang presence beyond Anxi is an actual minority position, as I have argued above.
Tan 1982 is not close to Cambridge 1979 at all, otherwise the many previous discussions on this talk page would not have happened. They appear similar because your own interpretation of what a 660 CE map should look like diverges too far from real published maps. Esiymbro (talk) 10:42, 10 July 2026 (UTC)Reply
Tan 1982 is not close to Cambridge 1979 at all
This is a bald statement. Which map among the 14 surveyed is closer in its western borders? None.
They appear similar because your own interpretation of what a 660 CE map should look like diverges too far from real published maps.
This statement is a strawman. The maps are not what I think they should look like, rather the issue is the current infobox map diverges too far from the clear majority, if not all, "real published maps" outside the 1979 source. Guile's Theme (talk) 10:50, 10 July 2026 (UTC)Reply
This comment shows exactly why your source survey is SYNTH. You retroactively project general historical atlas maps for 700/750 or unspecified time points back to the situation in 660, which naturally leads to a difference with real specialist maps for 660 because the Tang's territory was not the same one century later. Then, specialist sources that do not draw their own maps for 660 were used as argument that the published 660 maps represent a minority position! Esiymbro (talk) 11:15, 10 July 2026 (UTC)Reply
This comment shows exactly why your source survey is SYNTH.
Trying to dismiss the rather exhaustive survey by trying to call it 'synth' hasn't been to your credit in my opinion nor your repeated attempt to mischaracterize the 14 collected maps as 'my interpretation' instead of acknowldging they come from reliable published sources. That is arguably just intellectually dishonest and I've clearly been arguing from a WP:WEIGHT perspective. I'd also offer to try to find more such maps and round it out to 20, although I was having trouble with finding new maps thus why I stopped at 14, but I doubt that would sway your opinion.
You retroactively project general historical atlas maps for 700/750 or unspecified time points back to the situation in 660
This is a mischaracterization as I've done no such thing at all in the consolidated arguement. In fact, just yesterday I argued against such a thing when Qiushufang tried to suggest the date 675 for a new map.
Again, you are taking for granted 660 should even be the date for the infobox map. I believe Qiushufang and I agreed yesterday 700 was the best date among 660/661, 700, or 750 and probably represents the best balance and compromise among all published texts and maps on top of being the previous 15 year incumbent. And notably the 1979 maps seems to be the only one set in the 600s among published maps. Even Tan's map represents a range of dates across the entire Tang dynasty. Guile's Theme (talk) 11:47, 10 July 2026 (UTC)Reply
Ah, you found the book by Haywood! Thank you. Unfortunately he is probably the single least relevant name among the 19 mentioned in the survey (he is #5 in group 1) as he appears to be the only name that is not any sort of China expert (or any of the other mentioned disciplines) at all, thus I never considered his absence any sort of loss, and he was only named because he was already cited in the Protectorate General to Pacify the West article. So i don't believe it changes the balance in terms of Due Weight.
Furthermore, in light of the balance of the 14 maps and 19 sources presented in the survey, there is also an arguement to be made that we are beyond the point where any single source is going to change the balance under WP:WEIGHT. The survey already includes all of the most high profile specialists on the time/place in question I could find in a full week of searching and the presentation in maps by modern published sources is very disproportionate. The caveats/qualified language such as, "alleged", "fiction of chinese appointment," “illusion of a China-centered world,” and "pretension", 3 of which are from specialists and the last 1 explicity in regards to the infobox map, are also not going to go away. Nor will Tan's explanation which is the only one that properly explains the discrepancy that is at the center of this dispute. Guile's Theme (talk) 08:11, 11 July 2026 (UTC)Reply
John Haywood (British historian) is probably quite a good source for historical maps in general: "Dr John Haywood is a Fellow of the Royal Historical Society of Great Britain. A highly experienced cartographic editor and author, his many publications include The Great Migrations, The Penguin Historical Atlas of Ancient Civilizations and The Penguin Historical Atlas of the Vikings."Thames&Hudson. Pataliputra!(talk)10:31, 11 July 2026 (UTC)Reply
Perhaps you should update his wikipedia article then; because right now it seems to say his expertise lies on the other side of the world from China. And his Atlases cover such a wide range of topics, times, and places he could only be described as a generalist.
Also, much like Haywood's name is out of place among the other 19, the publisher of that work, Barnes and Noble, is similarly conspiciously out of place among a list of university presses and academic publishers like Brill. I maintain he is in all probability the single least relevant name among all 19 in the survey. Guile's Theme (talk) 10:42, 11 July 2026 (UTC)Reply
Wikipedia is not a source. I would also note that you conveniently seem to dismiss Blunden, Caroline and Elvin, Mark (1983), Cultural Atlas of Chinamap pp 92-93, Oxford: Phaidon Press, ISBN0-7148-2309-0 as a source. The co-author actually is Elvin, Mark "A professor emeritus of Chinese history at Australian National University, he specialised in the late imperial period. He was also emeritus fellow of St Antony's College, Oxford." I know you seem to take Wikipedia's "he specialised in the late imperial period" as an argument to dismiss him, but he is a noted China history specialist nonetheless. And the book has been highly praised: "The wealth of information and ideas in the text and in the maps make Caroline Blunden's and Mark Elvin's Cultural Atlas of China one of the best contributions of its kind now available."Pataliputra!(talk)10:59, 11 July 2026 (UTC)Reply
Yes, it is true that Elvin, Millward, Tanner, and van de Ven are all described as specialists in Qing/modern Chinese history, whether it be on wikipedia or elsewhere. And their list of published works also reflects that. I just keep noting van de Ven because he was he at first cited in support of the map and because he is only modern publication by any variety of expert on China, modern or otherwise, to use the same 1979 map, but only to explicity deprecate it. That is all both relevant and true regardless of if it is 'convienient' for me. Is there such a similar source by an expert, even only on qing/modern chinese history, who says the same about any of the other potential infobox maps or the previous 15 year incumbent? I think not.
Also, while on the topic of 'convienence', I believe the single biggest arguement of convienence made in this entire discussion is that the 1979 map is the most correct and authoritative, that for some reason it has not been widley reproduced if at all in over the last 43 years compared to the greater weight of Tang maps, and that somehow all those other published maps are wrong with no explanation for the discrepancy.
It is also true that there are a lot of names in the survey that are clearly specialists on the time/place/topic, chiefly all the names in Group 3 but also Beckwith and Wang, that should have more weight than those who do not have such a specialty.
But Haywood sits below all of them as he cannot even be described as any variety of China expert nor on the time/place/topic. That much is clear. Which is why I never considered the fact his was the only work I could not locate any sort of loss. Guile's Theme (talk) 11:11, 11 July 2026 (UTC)Reply
I am not sure what you are trying to achieve. We have numerous good-quality sources describing the Tang Empire at its greatest extent circa 660 (at least three excellent-to-good cartographical sources, plus van de Ven, who says these may have been largely nominal claims, but nonetheless reproduces the map; plus several excellent sources confirming this extent textually, "to Herat and Kabul"...). On the other hand you have shown that many sources do not necessarily discuss the 660s, and rather illustrate the Tang Empire circa 750, within rather more consolidated boundaries (essentially stopping with the Tarim Basin). Actually it is quite customary on Wikipedia to show an Empire at its maximum extent in the infobox. And the second "750s" map already appears in the very beginning of the article. This seems to me like a fairly standard and satisfactory setup for a Wikipedia article... Pataliputra!(talk)11:35, 11 July 2026 (UTC)Reply
I am not sure what you are trying to achieve.
I've already made that quite clear and even if I didn't it should be obvious, especially to a veteran editor such as yourself. I'm trying to achieve an infobox map that properly adheres to WP:WEIGHT and WP:NPOV which the current map rather clearly fails in light of the survey of sources.
We have numerous good-quality sources describing the Tang Empire at its greatest extent circa 660
Frankly, you have a single 'good' quality source, but that is also quite old from 1979, and the map from it has not been widley used since which is both notable and relevant in terms of WP:Weight. The 1983 example appears to be a direct copy by someone outside their speciality and an illustrative Atlas. And Haywood we've already spoken of.
This arguement by you also ignores the demonstrated WP:WEIGHT of the sources especially among modern published map and what the specialists have to say. The arguement is no longer about having individual sources, but what the weight of sources actually say about the matter and that is where the caveats by specialists becomes quite relevant. It seems interesting the specialists actually seem to have quite a lot to say against the infobox map in terms of actual/real Tang control in a way that isn't said about other such maps.
plus van de Ven, who says these may have been largely nominal claims, but nonetheless reproduces the map
Van de Ven explicity deprecates the claims of the map as not representing anything real. It is absurd to pretend he can still be cited in favor of the map when he is in fact the single author to most clearly argue against it. Infobox maps are not meant to show mere 'pretensions' of a ruler else we would be putting a map of the universe for the Achaemenid Empire... That you continue to insist on misrepresenting what van de Ven actually said about the map in favor of just the fact he used the map at all is concerning to say the least.
Again, is there any source that so similarly deprecates the previous incumbent map or any alternative infobox map as explicity as van de Ven does the 1979 map? Again, I think not.
plus several excellent sources
In reality what the survey shows is that the best collection of specialist sources mention "alleged", "fiction of chinese appointment," and “illusion of a China-centered world,” in regards to these claims without even including van de Ven.
On the other hand you have shown that many sources do not necessarily discuss the 660s
Apologies but that is a mischaracterization. The sources actually talk quite a lot collectively about everything except this supposed Tang control of territory beyond Anxi. Of course they also wouldn't spend much time talking about something that didn't actually exist. Only van de Ven spares the attention only because he uses the 1979 map explicity to do so. All in all seems like a Russell's teapot situation.
Also, I find the fact you wish to completely ignore the lack of any actual evidence of Chinese Tang presence beyond Anxi a bit odd; no garrison, officials, admin centers, nor anything else. Yet we can find Roman camps from their campaigns to the Weser river in Germany and or near modern Hannover they only occupied for a very short period, places we wouldn't even draw as part of the Roman Empire map if it were set in such a date, yet nothing for this supposed Tang presence in the claimed areas beyond Anxi? What have you to say in regards to that lack of any actual evidence beyond claims in Tang court documents or will you remain mute on that point? Also, do these places really even qualify as Tang protectorates under the definition someone like Qiushufang gave over a week ago in the previous discussion of 'Tang military adminstration'?
rather illustrate the Tang Empire circa 750, within rather more consolidated boundaries (essentially stopping with the Tarim Basin). Actually it is quite customary on Wikipedia to show an Empire at its maximum extent in the infobox.
Since you began pointing them out, that is an actual argument of convenience. Because convention broadly is to show empires at their maximum extent in such maps, as you pointed out, implicity all those 14 maps are depicting the Tang at the maximum extent according to their authors/publishers. Moreover, several of those maps do explicity call the 8th century the peak of the Tang as do some specialists by like Klyashtorny while most of the others authors say nothing either way. Or do you believe Wikipedia invented this convention? Guile's Theme (talk) 12:08, 11 July 2026 (UTC)Reply
I'll summarise my points into a conclusion here, because much of the latest discussion appears to be repeating the same few arguments.
Throughout much of the discussion, two separate topics have been conflated into one misleading narrative in order to discredit maps set in 660 from highly reputable sources and justify removing them from several articles. The first is actual criticism of such maps, and the second is the presence of maps showing more limited Tang territories in the 8th century. Although the differences between the 8th century maps and the 660 maps were frequently used in this discussion to paint the latter as an "expansive", "minority viewpoint", they are irrelevant when discussing the credibility of said maps.
On the contrary, genuine criticism of the current infobox map is rare, and the only explicit one was found in van de Ven which was based on the mistaken belief that the Tang never had a military presence beyond Anxi, a belief shared by Guile's Theme but contradicting most other sources. The several published 660 maps found in this discussion broadly agree with each other in the way they depict the Tang's western border, except the Historical Atlas of China which depicts a substantially more expansive one. Beyond sources with maps, we are making WP:OR by comparing vague texts to maps. Most of the alleged critics' texts in fact show very little difference to texts in publications with maps, such as in the Cambridge History of China. These include the supposed "caveats", and there is little reason to believe that the Cambridge authors had not taken them into consideration when making the map.
WP:WEIGHT has been frequently invoked in this discussion to justify removing the 660 map. However, WP:WEIGHT, as part of the WP:NPOV policy, deals with opposing viewpoints. A source with 750 map does not represent a different viewpoint from a 660 map, unless they claimed that the Tang's borders were static for its entire duration. It matters even less now that it has been demonstrated that 660 maps are not an isolated occurence limited to one publication.
On the other hand, the excessive scrutiny placed on 660 maps but a near free pass on those set in the 8th century, poses major NPOV concerns regarding Guile's Theme's edits. The one criticism from van de Ven has been brought up for no less than 20 times, while they apparently don't see any issue in using non-specialist maps with glaring inconsistencies and mistakes as references for a new map. The pre-2023 Wikipedia map, from a 1930s publication and with highly questionable labels, was seriously suggested as an alternative, when as far as I know no one has even conducted a source check. That map was in the infobox because there were no good alternatives except those based on Tan's maps. Now that there is one, I fail to see any reason to return to that version.
Finally it must be said that almost all of the past discussions were about whether the current maps are too restrictive compared to Tan rather than too broad. Similarly it has been convenient to ignore Chinese-language publications (which constitutes the vast majority of scholar output on Tang history) in this discussion, but that is not a Wikipedia policy. So please be aware of the inherent bias of this discussion. I am opposed to counting sources to settle any dispute because it is impossible to be exhaustive or unbiased in source selection, and if it indeed comes down to counting maps, then it is almost certain that there are far more Chinese-language sources that use Tan and its derivatives. Esiymbro (talk) 13:37, 11 July 2026 (UTC)Reply
Esyimbro continues with intellecutally dishonest micharacterizations of both the survery and my own arguments even after repeated corrections in such a manner that I believe he is doing the side he is arguing for far more harm than good at this point.
Throughout much of the discussion, two separate topics have been conflated into one misleading narrative in order to discredit maps set in 660
At the beginning of his reponse Esiymbro wants to claim the way the texts/maps/specialists comments/descriptions stack in the survey is a 'misleading narrative', yet at the end of response says, I am opposed to counting sources to settle any dispute" IE Esyim wishes to both avoid, and in doing so admit, the fact that the weight of text and map sources, especially from specialists, support the previous 15 year infobox map over the incumbent map while he simulatenously wishes to call my pointing out such a clear fact a 'misleading narrative'. Just as he tried to call the 14 published maps 'my interpretations' earlier. And before that the entire survey 'synth'. At this point I'm not sure how much weight anything he has to say has anymore.
genuine criticism of the current infobox map is rare
What is actually rare is any description or detail of what the Tang actually control beyond Anxi with generally no mention of the claimed territories by name nor the protectorate they were supposedly a part of. Even though there is a lot of detail about everything else surronding the topic up to and after Ashina Helu's 658 capture. While criticism of the 1979 map is explict from the only expert of any variety on China to use the map in 43 years, and the criticsm from 3 specialists on the time/place, one of whom includes Skaff whose work is the single most focused on topic/time in question, are all rather clear when it comes to the question of Tang control beyond court document claims. IE Vaissière’s “fiction of appointment,” Beckwith’s “alleged,” and Skaff’s criticism of the “illusion of a China-centered world.” Meanwhile no such qualifcations exists for previous 15 year incumbent map that is a better presentation of the majority of published maps and has no such controversy surrounding it as this one does.
However, WP:WEIGHT, as part of the WP:NPOV policy, deals with opposing viewpoints.
First, that is a complete fabrication. One of many now. Please quote for me where such a thing regarding 'opposing viewpoints' is refrenced in WP:WEIGHT.
Also, when 2 published maps from reliable sources show 2 different things for the same subject that inherently represents "opposing viewpoints". And clearly, given the 1979 map does not appear widely used at all compared to the previous incumbent map, it does not represent the great weight of scholarly opinion. I know that might be difficult for you to grasp since you've insisted on calling every published map that isn't the 1979 map, IE the vast vast majority of them, my 'interpretation'.
then it is almost certain that there are far more Chinese-language sources that use Tan and its derivatives.
Then all those maps are easy to immediately discount since Tan himself crticized his own work as containing "weaknesses" that lead to "false impressions".
while they apparently don't see any issue in using non-specialist maps with glaring inconsistencies
At this point I know not what to say except that I find it comical Esiymbro wishes to criticize the 14 'non-specialist' maps from the survey, ignoring the due weight argument about what the vast majority of published maps by academic presses show, while also ignoring there is no support for the 1979 map in 47 years outside of similar non-specialists like Elvin and Millward...
Also, he contiunes to ignore the inherent problem that the 1979 is simply not widley used and represents a tiny niche among most maps that present the Tang and thus the due weight issue.
I'm not really going to waste my time replying to rest at this point as it never gets any better and it is repititve to previous discussion. I will be changing the map, and if that results in a need for an RFC or trip to ANI then so be it. I'm confident that the current map should have never been made the infobox over the previous incumbent in the first place, it was never put under any scrutiny for that change which seems absurd, and cannot possibly be said to represent, and I quote WP:WEIGHT, "Giving due weight and avoiding giving undue weight means articles should not give minority views or aspects as much of or as detailed a description as more widely held views or widely supported aspects," and "undue weight can be given in several ways, including but not limited to the depth of detail, quantity of text, prominence of placement, juxtaposition of statements, and use of imagery." The infobox map must be given to a map that present the majority viewpoint of scholarly opinion and that much better describes the previous incumbent map than the current one. Guile's Theme (talk) 14:22, 11 July 2026 (UTC)Reply
You are making ad hominem attacks while fabricating my arguments. "Your interpretations" refer to your WP:OR interpretations on the other texts about the Tang presence in Central Asia in 660, i.e. your claims of "criticism from 3 specialists", based on vague descriptions that do not directly contradict published maps, and the conclusions you drew from the absence of such descriptions in other texts. If you can respond to the points above, then do it, otherwise what you said about "arguing far more harm than good" applies well to yourself.
You are holding on to the absurd claim that a 660 map counts as an "opposing viewpoint" against a 750 map, just as you have made equally ridiculous claims that 660 maps can be called "expansive" (search for this keyword on this page to see the actual claims) because 750 maps show smaller territories. For what actual opposing viewpoints look like, compare the 660s map from the Historical Atlas of China and the 660 one from the Cambridge History of China. That is where WP:WEIGHT is actually relevant and what I am refering to in the last section.
I'm not really going to waste my time replying to rest at this point as it never gets any better. I will be changing the map
I encourage you to read WP:Consensus and determine for yourself if you have one at the moment. It is good for you to bring up ANI because this might be precisely what you need. Esiymbro (talk) 14:51, 11 July 2026 (UTC)Reply
Not going to bother responding anymore; there is no point in assuming good faith when you have so continously mispresented both the survey, what I say, and even going so far as to call the 14 published maps 'my interpretation'. Reaching consenus requires both sides argue in good faith and I cannot honestly say I believe you have demonstrated that in your methods of argumentation. Notice I'm not saying this to Patali or Kanguole. Guile's Theme (talk) 14:54, 11 July 2026 (UTC)Reply
you have so continously mispresented both the survey, what I say
I've done no such thing and you have not explained how I have. I maintain your methods of argumentation have not been intellectually honest nor do I believe you have even honestly engaged with/presented the results of the source survey.
Meanwhile, I can quote where I you do the things I accuse you of, IE, They appear similar because your own interpretation of what a 660 CE map should look like diverges too far from real published maps. IE misrepresneitng the 14 published maps that show what their authors/publishers interpret the Tang borders to be at their greatest extent per mapping convention for such images. And that is only one example among several such micharacterizations you've attempted over the course of our debate to the point I no longer see any point in continuing.Guile's Theme (talk) 14:59, 11 July 2026 (UTC)Reply
I can quote where I you do the things I accuse you of
Ok, let us list your mispresentations from only your 2nd response to the Conclusion.
1) Simply put, there is hardly any disagreement between the text descriptions in the Cambridge History of China and the other sources
False, in fact there 3 specialists who say more to deprecate the expansive claims in terms of actual Tang control than any specialist say in support of 661 map.
2) The "caveats" of Tang control, its brevity and thin military presence, are mentioned.
How are they so mentioned in the infobox map? Another misrepresentation.
3) As I said before, there is no major contradiction between these sources
The 14 maps are from reliable sources, so you are inherenlty misrpresenting the results of the source survey on that point alone. You have contiously refused to acknowledge such a basic fact and thus why I said, "nor do I believe you have even honestly engaged with/presented the results of the source survey."
4) If wording such as "alleged" is enough to reduce Tang territorial extent shown in maps, how much should it be reduced?
Another misrepresentation, such language is a mark against the 660 map in favor of the 700 map that has no such language by any source regarding it on top of being the much more commonly used map and thus the one that shold be used per WP:WEIGHT. Not about 'reducing' the 660 map and making a new map which would be synth.
5) without explicit contrary descriptions from sources, most of this discussion has essentially been WP:OR attempting to put the results from such conjecture into an infobox of an FA.
Completely not what I've been doing, thus yet another mischarazterization.
6) but ignored the fact that most Tang specialists who discussed the matter clearly agree that it was 660 AD instead
There is no such agreement for 660 among the specialists in the survey and yet another misrepresentation if not outright invention on your part.
7) I am also not convinced that your source survey is complete or balanced, either.
A suggested misrepresentation, when the breadth of disciplines beyond just Sinology for the specialists in group 3 is inherently more balanced than using any single source.
Those are all only from your 2nd of like 6 responses. I don't think I have the time or energy to go through every single one but they only get worse from there. Not worth my time. Guile's Theme (talk) 15:42, 11 July 2026 (UTC)Reply
Well there is obviously no consensus on the map change and if there was, the reworked map would be used anyways rather than the old one, so please stop reverting. Qiushufang (talk) 15:30, 11 July 2026 (UTC)Reply
I see no consensus for the original change in 2023, therefore it should have never been done. I'm framing this as a reversion to the map it should have never been changed from, I am still willing to continue working towards a better map. But that isn't going to be possible until we change it from this clearly incorrect map first because people like Esiym will defend it to the death instead or stonewall any new/improved map in favor of this one. So please, in an effort to actually get a better map that adheres to WP:WEIGHT as presented by the survey, do not revert. Guile's Theme (talk) 15:52, 11 July 2026 (UTC)Reply
No, that's not how changes work. Consensus is needed when the changes are challenged as it is now. If the change was made without issue at the time, then obviously the default is that nobody cared enough to challenge it. Otherwise, is an editor supposed to request permission from an imaginary group of editors every time they make an edit to Wikipedia? That is unreasonable and not how editing happens. I do not think the old map was better and am against your revert to it. You do not WP:OWN this article. Qiushufang (talk) 15:58, 11 July 2026 (UTC)Reply
And Esyim and the like would stonewall the improved map we agreed upon as Synth while this map exists. So what would you propose? Was it ever actually your intention to improve the map? I don't own the article but it seems an apparent fact that the infobox map simply fails a core Wiki policy in NPOV by failing to pass WP:WEIGHT.Guile's Theme (talk) 16:02, 11 July 2026 (UTC)Reply
If other editors don't think the revert is an improvement, that is their right to oppose it. I think both the reworked map and current map are fine. Tbh the current map already shows the furthest western areas in lighter colors, which shows that they were either briefly occupied or not part of the Tang's core territories. I prefer having the Anxi Protectorate in the same color as other parts of the Tang empire and northern Korea in lighter color as well, but frankly the difference isn't earth shattering. You should not make WP:Aspersions about other editors' intentions. Either is fine to me and both can be included in the same article. Qiushufang (talk) 16:08, 11 July 2026 (UTC)Reply
The practical result of not supporting a revert to the 15 year incumbent is that we are effectively railroaded with no way to improve the map because those like Eysim will never agree to our improvements while that map is in the infobox regardless of how poorly it adheres to WP:WEIGHT.
And the northern part of the Ordos was almost certainly a part of Tang territory at the time, which is not shown on the old 700 map. Skaff 2012 p. 49 has a map of numerous garrisons all over the Ordos as well as three major ones north of the Yellow River loop in Inner Mongolia. This is illustrated at Protectorate General to Pacify the North. Qiushufang (talk) 16:25, 11 July 2026 (UTC)Reply
I support the Ordos loop inclusion, and even some others we talked about, but I think it is so much easier to revert to the original 700 CE map and then improve because then people like Esiym probably would support any expansion to the base 700 CE map once this current map is out of the infobox. That is my logic. Guile's Theme (talk) 16:26, 11 July 2026 (UTC)Reply
I'm not sure why you think reverting to the old map would be easier. Just because it was used for 15 years doesn't make it better or more likely to be maintained. Qiushufang (talk) 16:36, 11 July 2026 (UTC)Reply
I gave my logic; ie "because then people like Esiym probably would support any expansion to the base 700 CE map once this current map is out of the infobox." Guile's Theme (talk) 16:38, 11 July 2026 (UTC)Reply
That doesn't make any sense. If their concern is synth or that the furthest western parts in Central Asia and Persia aren't included, any reworked map would also be included in that category. I do not support using the old map as some sort of imaginary stop gap at all. Most of the general educational maps you've included are awful with weird "cultural regions" or arbitrary lighter shades that don't make any sense. Even some specialist monographs covering the subject have shoddy maps, and most of the ones you've included are worse. If there's going to be any change, I only support using the reworked map or variation of it. Qiushufang (talk) 16:55, 11 July 2026 (UTC)Reply
Would you support an improved 700 CE map like Qiushufang or are you wedded to the 661 map? It is your creation on the wikimedia, so at least I understand for whatever it is worth. If no one else I expect a creator to defend their work and have nothing badly to think of it. You can create the new map as you created this one if you wish. I just cannot believe the current one adheres to WP:WEIGHT properly as I've made abundantly clear. Guile's Theme (talk) 16:25, 11 July 2026 (UTC)Reply
You need to assemble map(s) from reliable sources for an alternative map before proposing a replacement.
On the WEIGHT issue, some authors depict the empire around 750, while others depict it in 660/661. I don't have access to imgur, so I'm not sure which sources you're citing for the former, but I have seen maps of that type. For the latter, we have the sources already mentioned, as well as, for example Ebrey, Cambridge Illustrated History of China, Fairbank and Goldman, China: A New History.
The current map gives reduced weight to the additional areas by depicting them in a light shade and labelling them as briefly controlled. Kanguole17:07, 11 July 2026 (UTC)Reply
Frankly, most of the assembled maps are just generic grade school level maps with arbitrary colorings. I contend however that this is all making things way too complicated. There's really only one source of major contention, the extension beyond Anxi into Persia and Central Asia beyond Xinjiang and Kyrgyzstan. The furthest portions of the current map. Every single decent map more or less agrees on the core parts of the Tang empire for a 50 year period from 700-750 plus or minus Liaodong or the northern part of the Ordos. Qiushufang (talk) 17:20, 11 July 2026 (UTC)Reply
Since it will become relevant, a list of sources for those 14 maps, and I plan to find another 6 to round it out to 20.
And I will probably find 5 such more within a day or 2. Notice these are all essentially from academic or university presses too. I'll just keep finding more because I'm quite certain at this point the ratio of either 700 or 750 maps vs 660 maps like the infobox presented in modern published sources is quite disproportionate.Guile's Theme (talk) 19:07, 11 July 2026 (UTC)Reply
Find 5 such more of what? For what? Adding a list of maps, one of which (Graff 2002) is a height map without any borders at all, is not very helpful without context. What is the number of maps of certain dates supposed to prove? One of them doesn't even have a specific date. One of them has ALL of Korea has part of the Tang dynasty. I've already noted before that many of these are poor quality. This is starting to feel like a WP:RIGHTGREATWRONGS moment. I'm not sure what the argument even is anymore. Qiushufang (talk) 19:34, 11 July 2026 (UTC)Reply
What is the number of maps of certain dates supposed to prove?
Which map represents WP:NPOV & WP:Weight per, "Neutrality requires that mainspace articles and pages fairly represent all significant viewpoints that have been published by reliable sources, in proportion to the prominence of each viewpoint in those sources." & "Undue weight can be given in several ways, including but not limited to the depth of detail, quantity of text, prominence of placement, juxtaposition of statements, and use of imagery."
Also, not just the maps, but quite a few of the specialists give qualifcations that indicate no actual Tang control in the claimed areas while no specialists explicity describes them as under Tang control to the same extent as the map. And the fact that there are no such qualifications, or even an explicit deprecation in the case of van de Ven, in regards to any other such published maps. And Tan's explanation for the weakness and false impressions from such expansive map making methodology, which is the only potential explanation offered by any source for the striking discrepancy between the info box maps and the clear weight of maps from published reliable sources.
I've already noted before that many of these are poor quality.
An opinion not borne out by the names of the academic and univerisity publishers it would seem who chose to publish those maps in addition to their authors. I also find it inexplicable almost all these univeristy and academic press sources would all have 'the wrong map' instead of Tan's explanation being the correct one for what is at play here. I'd also remind you to not cast aspersions. Guile's Theme (talk) 19:42, 11 July 2026 (UTC)Reply
This seems like a WP:RIGHTGREATWRONGS moment to me because you are saying a lot without getting to the point of the argument in a way that is understandable. If your purpose is to improve the map, then using obviously wrong maps as sources is unhelpful no matter who published it. Maps that have obvious issues like Korea being completely owned by the Tang, that contain only info on the height of geography, are not useful. NPOV or WEIGHT doesn't cover content that is factually wrong or irrelevant. Qiushufang (talk) 19:58, 11 July 2026 (UTC)Reply
This seems like a WP:RIGHTGREATWRONGS moment to me because you are saying a lot without getting to the point of the argument
I'm not going to engage with someone baselessly casting aspersions against wiki policy when they've already been asked to stop. I've stayed very much on topic as well and don't know what you are on about anymore so you'd best provide an actual example of what you are trying to accuse me of.
I also don't know what to say to someone who seems to think every published map, whether it is the 661, a 700, or 750, are all wrong and doesn't seem interested in actually offering any workable path forward towards a map that actually meets WP:WEIGHT and thus WP:NPOV.Guile's Theme (talk) 20:06, 11 July 2026 (UTC)Reply
An example of what I am accusing you of (using poor sources): maps of the Tang owning all of Korea, geography height. These are either factually wrong or irrelevant to the topic. These aren't helpful in a discussion about Tang borders because one is obviously incorrect and the other is not relevant.
An improved map would not be based on WEIGHT of number of maps and NPOV does not support factually wrong content. A map represents specific things it is trying to show. Using a map created by averaging out every map ever would not make it more correct, it would just make it show things like the Tang owning all of Korea, or the northern Ordos being in a lighter shade or uncovered. Both are incorrect, so an average map including maps that show these things would still be incorrect. WEIGHT and NPOV don't factor into this at all. Qiushufang (talk) 20:35, 11 July 2026 (UTC)Reply
In example of what I am accusing you of (using poor sources)
I don't know how to tell you this, but your opinion on the quality of the sources, is just that, an opinion. One not borne out by the list of publishers, if not the authors. Also, is The Cambridge World History among these 'poor sources'?
What the sources represent irresepective of their individual quality is a clear trend among all published maps and academic sources in the last 43 years and thus an arguement against the current infobox map under WP:WEIGHT. Something you've refused to engage with even though the argument has been made dozens of times by now.
You've failed to substaintiate your accusation and there is nothing more to discuss. And to be honest, it kind of seems like you got upset when I provided citations for the maps you were trying to deprecate, probably because the list of names and publishers was not what you wished, and immediately jumped to casting aspersions.... Guile's Theme (talk) 20:46, 11 July 2026 (UTC)Reply
It (the map) is a poor source for the argument (whatever it may be) because it is wrong. It is factually incorrect and you are ignoring the other map I pointed out that is irrelevant to the discussion because it doesn't show relevant information. Neither of these have to do with the publisher. The publisher and the content are not 1:1 the same thing. Authors can be wrong. Sources can be wrong. Otherwise, why even bother discussing sources? Oh, it's from a reputable publisher, I guess I'll just take its word 100%, no need to rely on my brain or independent thinking. The publisher and citation has nothing to do with this. How do you not understand that?
A problem with all your replies, and probably why so many disagree with you, is that you are not making yourself clear in your rush to prove something. I think your second paragraph is talking about the western parts of the Tang map, but an average person, including myself, has no idea that might be what you're on about just from reading the second paragraph. Your sentences and logic are longwinded, convoluted, and simply hard to understand. You entrench yourself in these positions and are willing to die on a hill on what no real academic would ever try to argue against, that it is not good to use incorrect sources to prove a point.
I am also of the opinion that the current map and the reworked map are simply not that groundbreakingly different, so the amount of vitriol and amount of time you've spent arguing seems unwarranted. The western portion is already in lighter colors and it is sourced to something. I am fine with the current map or the reworked one like I said before. Qiushufang (talk) 21:05, 11 July 2026 (UTC)Reply
Funny, you seemed totally aware of what the dispute was over earlier today, ie, "There's really only one source of major contention, the extension beyond Anxi into Persia and Central Asia beyond Xinjiang and Kyrgyzstan"
Also, the word 'west' or 'western' has literally been used over 100 times in the discussion since the first one 11 days ago. Not sure how it could have been missed over all that. Guile's Theme (talk) 21:18, 11 July 2026 (UTC)Reply
YES, that is the main source of contention, based on MY belief and belief of what you are arguing. But I had no idea why you need to go on about NPOV and WEIGHT and maps because neither of those have to do with improving the map or the western borders specifically, which greatly confused me. If what you are intending on saying is that "many maps don't show the territories in such and such place, so it shouldn't in the infobox", then just say it instead of trying to justify the integrity of maps whose creators obviously didn't care nearly as much as you do. You didn't mention the word west or western borders once in this entire section until the recent response to me. Qiushufang (talk) 21:25, 11 July 2026 (UTC)Reply
YES, that is the main source of contention, based on MY belief and belief of what you are arguing
How do you clearly understand what the debate was about but then an hour or so later, while you are trying to cast aspersions and deprecate me, then suddenly you are confused? I don't follow. These things don't magically change. Seems a little convenient and/or an attempt to gaslight.
But I had no idea why you need to go on about NPOV and WEIGHT
I've explained that plenty over what amounts to pages now. You can go back and reread and if you don't understand by now nothing I say will fix it.
"instead of trying to justify the integrity of maps
I will just quote myself from a response to Esiymbro 11+ hours ago; "Weaknesses in individual maps do not by themselves answer that broader WP:WEIGHT concern." Clearly I was not very interested in "trying to justify the integrity of (individual) maps".
Because I didn't need to, I just needed to show the current infobox map doesn't represent the balance of scholarship and published maps by reliable sources. And I only believe more strongly the more I look that the balance is not going to improve in favor of the infobox map and that it simply should have never been allowed to become the infobox map in the first place.
you didn't mention the word west or western borders once in this entire section until the recent response to me.
What do you mean this 'entire section'? If you mean Esyim's 'response', then yea, I'm not going to restate what I'm arguing about every 7 hours (about how long this section has existed) 11 days in just because you don't understand something all of the sudden and refuse to ask the question; although you seemed to understand perfectly well until you needed to cast aspersions. Then, all of the sudden, confused! Guile's Theme (talk) 01:35, 12 July 2026 (UTC)Reply
Ya I was confused, because I don't obsessively read every single argument and comment by every editor that I haven't responded to. Qiushufang (talk) 01:57, 12 July 2026 (UTC)Reply
My god... I just realized that one map shows the Tang owning Samarkand, which is in line with the current map. It's hard to see because of the low resolution and angle. Another map also shows the near maximalist Tang western claims but it's hard to see because of the weird angle. They all look like they're based on the same map. Qiushufang (talk) 21:52, 11 July 2026 (UTC)Reply
My god... I just realized that one map shows the Tang owning Samarkand
Oh my god, it is almost that changes nothing since none of them show Bukhara, Balkh, Tokharistan, Kashmir, Kabul, Herat, or freaking Zaranj as part of Tang control. Thus all those 15 maps are still closer to one another than to the infobox map and the infobox map is still an outlier among all modern published maps in the last 43 years whether by specialists or not. Also, 'own' is an interesting word to use on your part.
Another map also shows the near maximalist Tang western claims
It is hard to see on that one, but it does not show what you just claimed. You can in fact see the city of Balkh (Bactra on that map) is outside the colored area of Tang control, Balkh is further East than Kabul, Herat, and Bukhara, and further north than Tokharistan, Kashmir, and other such areas colored by the infobox map. IE, it does NOT match the more expansive westward claims of the infobox map and clearly doesn't reach down to Zaranj either. Guile's Theme (talk) 01:48, 12 July 2026 (UTC)Reply
So what if it's closer to a certain map? Samarkand is still part of the lighter colored area and its inclusion is based on the same claim as the other lighter colored parts. If you're basing your argument on not including that part based on sources that say it's false, then saying one part of the claim is true (Samarkand was part of Tang territory) while the other parts are not is fundamentally a weak argument. That means whoever included that did it based on arbitrary reasons and shouldn't be taken seriously. Qiushufang (talk) 01:55, 12 July 2026 (UTC)Reply
So what if it's closer to a certain map?
In that case, why 'oh my god' about just Bukhara??
Samarkand is still part of the lighter colored area and its inclusion is based on the same claim as the other lighter colored parts.
By same claim you mean the 1979 map that hasn't been reused by any variety of China expert or acadmic/univeristy publishers for 43 years? And you'd wonder why that would be an issue?
Also, one of the authors in Group 3 is a leading specialist on Sogdians, de la Vassiere, he was examining murals in Samarkand he dates to 660 AD and he states, "Chinese appointments of Central Asian rulers represented the fiction of appointment by the Chinese emperor."
I know you said you didn't read the dicussions between myself and others editors, but did you read the Source Survey?? Because that was in there. As were a lot of other things you seem to be confused about.
You know, Wikipedia also has a common sense rule. How does it pass the common sense test that these places were under 'Tang control' with no evidence of military presence, garrisons, admin centers, officials, you know, all the stuff YOU yourself said came with actual Tang protectorate status, and there is NO physical or acheological evidence of anykind, and that nearby Tibetan or Arab sources didn't record any such Tang control of the places being contested? It doesn't really pass the smell test, and it isn't like the track record of Chinese imperial court records for accuracy and honesty is in anyway unimpeachable as we talked about in regards to Tibet some... idk maybe 24-ish hours ago. Guile's Theme (talk) 02:19, 12 July 2026 (UTC)Reply
If you mean your list of quotations I read some of it but not all because there's just way too much. Even if I did I wouldn't remember half of it and I certainly didn't read your conversations with the other editors. I don't even know what you mean by the 1979 map and had to search for Tibet after you mentioned it, because honestly I forgot. I don't know what Group 3 is or who the specialist on Sogdians is.
Please understand that the more you write, the more formulaic your writing, the more formatted it is, the less likely editors are going to read and care about it. You don't need to green text quote and respond every single reply. Long lists are bad for attention spans. I can't read your mind and I'm not reading every conversation you have with every other editor and memorizing everything. I probably haven't read half the stuff on this page since the topic began.
My position on the maps is laid out below in its own section for clarity. The westernmost portion covering areas beyond Xinjiang are all in a third tone lighter shade, so it already communicates the very loose if any suzerainty that the Tang held over the area. Samarkand is part of that shaded area, so I assume the other areas are part of the same claim, which might be what you're referring to by the 1979 map. The quotation you provide is sourced to a study on Samarkand, so if we are to believe Tang overlordship over Samarkand is false, then the map as discussed in the previous reply would be in the wrong. Doesn't that contradict the validity of the map you're using? Why would Samarkand be part of Tang territory but not the other lighter parts? I'm really not sure at this point what source or argument you're trying to make beyond that "more maps say something therefore it's true". But then you quote a source that contradicts that. It's very confusing and feels like an ad hoc defense and it kinda feels like you're just pulling at things to continue arguing. Of course I haven't read every single reply you've made to every other editor so I might just be missing information.
All of this is just based on your recent replies. I suggest you stop assuming every editor understands what you're talking about automatically as if they've been listening in on a very long phone call and lay things out more clearly in shorter replies. Qiushufang (talk) 02:48, 12 July 2026 (UTC)Reply
So you don't read very thoroughly, or skim. Is it really a suprise that you are a bit confused? I understand this isn't light reading, but the infobox map of a Featured wiki article of high profile isn't supposed to be settled in Twitter post lengths. Much less a wide survey of sources in an attempt to determine what the balance of scholarship actually says under due weight.
You also seem to suggest you can't remember half the things you do read, including entire conversations from just the day before in regards to Tibet, which... idk. I genuinely don't want to be mean, but at the same time that can't be my problem or fault.
I don't even know what you mean by the 1979 map
Which makes evident how little you've read; 1979 has been mentioned 44 times on this talk page. Also, ctrl+f 1979 exists if you need to figure it out. I find it a tad annoying you wish to act confused when the answers are on the same screen you are staring at, and don't even need to entirely read through if you can ctrl+f search by term. Also, why can you be bothered to respond for hours in what amounts to pages but then not read a few pages as well? Maybe the important parts 2 or 3 times so it sticks?
the less likely editors are going to read and care about it.
The funny thing is, I actually went to ENORMOUS lengths to condense everything from the Consoldiated Argument & Sources Survey to the Conclusion. I kept each source to 1 or 2 bullet points and did my utmost to paraphrase or quote only the most relevant parts. It was about 6 pages in docs when I posted it, it started at 11 pages before I edited it down. Please, go ahead and read through and tell me what page worth of info you think I should have left out that was not worth mentioning for my argument. I suppose I could have cut some sources, but I knew the more I added the more representative it could be said to be.
are all in a third tone lighter shade, so it already communicates the very loose
'Loose' is not what most maps show, most maps show no tang control of those areas period. Also, loose is not what a variety of specialist sources like Skaff and Vassiere say between calling Tang control of those places a 'fiction' and 'illusion', doesn't sound like it is supported by Beckwith's 'alleged' either, and certainly not supported by van de Ven's, 'no actual control/imperial pretensions'...
then the map as discussed in the previous reply would be in the wrong. Doesn't that contradict the validity of the map you're using?
The previous map, or any other map, does not have to perfect to be a clear and obvious improvement over the complete outlier that occupies the infobox map. In fact, you said yourself the previous map was better in some sense in Pataliputra!'s talk page.
If the choice is made that only areas with direct control via military garrisons are shown as part of the Tang dynasty, then imo this map shows a more appropriate reflection of that.
Considering Tang military protectorates are according to what you said, "Effectively, it was a military occupation of conquered territory," and the Tang claimed the west beyond Anxi as protectorates, how could such a 'military' occupation ever been a reality with no evidence of any actual garrison? Even the sources that describe the defeat of Helu make clear the majority of the army were Steppe nomads themselves and not Tang forces. I can pull the exact numbers from the sources if you wish.
You yourself also seemed to dismiss other Tang claimed territories in a previous comment in the East in a similar manner to the way the specialists did in regards to the West, IE "The Jiaole Area Command given to a Xi chieftain in 648 is too vague to be of any use. It could be a merely symbolic title to recognize the power of a local ruler. The Songmo Area Command created in 648 is directly stated to be imaginary." - Qiushufang
"more maps say something therefore it's true"
It is more like, "The collected maps show that the infobox map is not representative of the vast majority of published maps. While they vary somewhat, they all fall within a much narrower range in regards to their western borders while the infbox is an outlier among virtually all maps. Having such an image in the infobox therefore gives it undue weight."
I mean, how else do you argue regarding due weight other than seeing what the majority of source, both in terms of quantity, authority, and variety of prespective, have to say? Please explain the alternative if any such exists. Guile's Theme (talk) 03:45, 12 July 2026 (UTC)Reply
Ya, that's why I said I'm fine with either the current map or the reworked map. I recognize that the evidence for "control" of the western part shaded in the lightest color is very light. But that is exactly why it is in a third color, communicating that the "control" is probably nominal in reality, aka only existing by word of mouth. That's why I'm fine if that's not the map used in the infobox. But at the same time I'm not going to get butthurt about it if it s used, because again, it is communicated that's its status.
When I said "previous map", I meant the one showing Samarkand as part of the Anxi Protectorate. Imo such a map would be no different than the current one because it is making an arbitrary distinction between Samarkand and the other lighter shaded parts when they are all part of the same claim for control. If the Tang did not control parts further west of Samarkand, then they did not control Samarkand either. They are a package.
You are perfectly right that I dismiss the Jiaole Area Command and other imaginary territories for inclusion in Tang territory for the reworked map, but that is for that map, the purpose of which is to show all territories that were under effective Tang military occupation. The map in the infobox shows multiple zones of "control" signifying different levels, with the westernmost probably communicating the lowest level, which can easily be interpreted as nominal suzerain or something similar. I want to make it clear that it is perfectly fine to argue that that should not be how the map in the infobox works, and it should only show areas under actual military control but I think you've bundled too many different policies and arguments as support that make it confusing. Whether or not a map is supported by the most maps is ultimately less important than whether or not the map is actually right, which is why I prefer both the current map and the reworked map to the old one. If need be, the western part can be labelled nominal suzerain or something similar as I suggested below.
I'm fine with not using the current map showing the shaded part outside Xinjiang too, but the reworked map or something like it should take its place. I don't know why you are clinging so much to the older map. It is worse than the reworked map, incorrect, and therefore unsuitable for use regardless of it being less of an outlier. Even if it is more supported by more maps, WEIGHT and NPOV do not cover erroneous information (the missing part of Ordos). Anyone who would oppose the old map on grounds of not covering the western portion would oppose the reworked map as well, so it does not matter that the reworked map is used over the older one.
To conclude, I'm fine with the current map because I understand that the lightest shaded part communicates a degree of lesser control (possibly to the point of non-existence or mere acknowledgement of Tang presence). It is perfectly fine to argue that should not be how a map in the infobox works. I am also fine with using the reworked map (which is the same as the old one plus all of Ordos) because it shows all of Tang military controlled areas and that is its purpose. I am not fine with the older map because it is just the reworked map but worse. Qiushufang (talk) 13:15, 12 July 2026 (UTC)Reply
Another frustrating thing about replying to you is that you keep micro-editing your comments, deleting and adding things seconds or minutes apart. While it's not wrong to do that, once somebody has read your reply, especially in high speed conversations, it's less likely that they're going to read the diffs or go back to read your comment, so whatever changes you've made probably won't be noticed. Qiushufang (talk) 02:52, 12 July 2026 (UTC)Reply
This one is from one of those names cited in group 2, and is labeled, "Tang China and Inner Asia 750 AD", and clearly the claimed western territorial extent of the infobox can't fit into that map...
Tang all of Korea
Might be relevant if I wasn't ONLY talking about the western borders presented in the infobox map; which you well know. And again, this isn't about an individual map but what the weight of maps show. Guile's Theme (talk) 21:01, 11 July 2026 (UTC)Reply
The old map's borders don't fit in Graff's map either.
And no, it's NOT obvious that you were talking ONLY about the western borders at all, because you've dragged in so many other maps, many of which are poorly made, based your agument on WEIGHT and NPOV, and didn't mention the western borders for several comments, that it became much less obvious that was what you were talking about. I THOUGHT that was what you were on about, but it still took me more than a couple of minutes to deduce that was your point (or at least I think it is) reading your last reply. Otherwise why wouldn't you just say it in a single short sentence, like "yes I know these maps suck and are shoddy but they all agree the western borders end in Xinjiang"? And to clarify, this doesn't change that it is good NOT to use factually wrong sources to support your argument, even if it's not directly related to that specific thing because it shows that the source did not do their due diligence. If they're wrong on something so basic like Tang territory in Korea, why would they be right elsewhere? Qiushufang (talk) 21:09, 11 July 2026 (UTC)Reply
@Kanguole: After looking at the current map some more, one weakness I find is the lack of topographical features. It's hard to see where exactly the borders end even in Xinjiang. This might be besides the point of the discussion considering that all these borders of early medieval polities are by their very nature vague, but it would at least be useful for better understanding what area is covered. Whatever the decision is, hopefully you can make another map with more base information.
To make my position clear on the discussion around the western borders on the map, I am ok with the current map or the reworked map or something similar to it. Frankly, there are already similar ones available in commons, usually without Liaodong, which is debatable anyways whether or not it was occupied by the Tang by 700. The reason why I don't care about the current map stretching so far west is that it's already pretty clearly communicated based on the triple coloring scheme that the furthest western parts are some sort of vassal or nominal tributary under the loosest terms of control possible. Basically polities that sent the Tang some form of communication acknowledging their power. I really don't think it's that big of a deal and both maps with and without the furthest western highlighted parts can be included. Perhaps the highlighted part can be labelled nominal tributary or something. I do not think the old map is better since it does not include all of the Ordos plateau, which is wrong and a major weakness of many maps of the Tang dynasty. I'd rather have the map showing parts of Central Asia in a light color communicating loose suzerainty than an actual garrisoned part of Tang territory missing. Otherwise all these maps are more or less the same plus or minus Liaodong and the western highlighted part. This really shouldn't have required so much discussion. Qiushufang (talk) 23:23, 11 July 2026 (UTC)Reply
Latest comment: 28 days ago32 comments9 people in discussion
The following discussion is an archived record of a request for comment. Please do not modify it. No further edits should be made to this discussion.A summary of the conclusions reached follows.
After a thorough reading of both this RfC and the discussion that proceeded it, there appears to be no consensus for replacing the current 600/661 map, with either the former 700 map or the newly created 660/700 map; as such, the current map will be maintained for now. While there is substantial disagreement over whether the Western light-shaded areas should be represented in the infobox, editors generally agree that the current map has significantly stronger direct sourcing than the proposed alternatives. Editors also agree that the far western Tang claims were brief and different from firmly administered territory, however disagree on how that distinction should be represented visually. This map may be replaced if consensus emerges to do so in the future. Mit khavershaft, Royz-vi Tsibele(redt mit mir!)21:01, 29 August 2026 (UTC)Reply
Please see the previous discussion here (ie right above) and lend your opinion on the following questions:
Which year should the infobox map depict?
660/661
700
750
Should areas be shaded light to represent either ephemeral or purported/contested control in regards to the West?
Comment from discussion member Qiushufang - The linked discussion section isn't the start of the actual discussion but rather the section where the RFC requesting user created the consolidated version of their argument. There was plenty of back and forth in the previous section where I participated the most, resulting in a reworked version of the map. The number of options seems needlessly complicated to me and some might be hypothetical versions that haven't been created yet. As far as I know, there are currently only three existing options:
The 661 map extending west into Central Asia and Persia created by User:Kanguole in 2023The 700 map created by User:Hko2333 in 2008The 660 map (reworked version of the 700 map with input by User:Qiushufang) by User:Guile's Theme in 2026
Comment by discussion participant Kanguole on sourcing – Map (A), the "661 map", accurately reflects maps in the sources attached to it in the article. Map (B), the "700 map", is based on the 750 map from Albert Herrmann (1935) Historical and Commercial Atlas of China, not the 700 one, except that the northeast part is unsourced. Kanguole17:25, 11 July 2026 (UTC)Reply
notified by SodiumBot Comment - Two points:
(1) This RfC bundles two separate questions - which year to depict, and how to shade nominal/vassal control. Under five combined options, some of which (C, E) don't exist as finished maps yet. Splitting into sequential questions would make this easier to !vote on.
(2) On precedent: other empire infoboxes (Roman Empire at 117 AD, Mongol Empire at 1294) do use apogee maps with a distinct shade for vassal/nominal territory, so that convention isn't itself an outlier. But those apogees lasted decades, the Tang's western extent here (657–665) lasted 8 years before collapse. Whether an 8-year peak merits the same treatment as a decades-long one seems like the actual question, not the year label. Michael Boutboul (talk) 13:28, 13 July 2026 (UTC)Reply
I have split the RFC into 2 different questions per your suggestion which seemed to make it easier to approach by breaking down the issue and may foster more participation. Since no non-participants have yet really stated an opinion I hope this is ok. Guile's Theme (talk) 14:01, 13 July 2026 (UTC)Reply
Significance of the year dates
An explainer on the significance of the different dates related to border changes would probably be helpful.
700 - The Tang was in control of modern Xinjiang as well as Suyab in Kyrgyzstan. The Andong Protectorate had retreated from Korea in 676 and in 677 its seat was relocated to Fushun in modern Liaoning Province. In 699, Tang forces withdrew from Liaodong Peninsula. The Anbei Protectorate's seat was located in the northeastern corner of the Ordos Loop but the Tang suffered from raids by the Second Turkic Khaganate until more garrisons (the three Shouxiang cities) were established along the north of the Ordos Loop in 709.
750 - The Tang still held control in Xinjiang and conquered Little Palola (Gilgit) in 747, recaptured Suyab from the Turgesh in 748, sacked Chach in modern Tashkent in 750, and defeated the Qieshi and Turgesh in Chitral in the same year. The Tang suffered a setback after being defeated in the Battle of Talas in 751 but no major border changes occurred other than losing influence in Transoxiana until the An Lushan rebellion (755-763) when most troops in the west were recalled to deal with the rebellion. The Four Garrisons of Anxi continued to function in isolation until the 790s. The Andong Protectorate's seat had shifted further west to the Old City of Liaoxi/Ying Prefecture in Yi County, Liaoning or Chaoyang, Liaoning in 743 and the protectorate was abandoned in 756. The Anbei Protectorate's seat shifted to Urad Middle Banner in the upper northwest of the Ordos Loop and ceased to function during the An Lushan rebellion (755-763). Qiushufang (talk) 16:25, 13 July 2026 (UTC)Reply
Comment The borders of the Roman Empire were generally more stable during its peak, but Roman control in Mesopotamia around 117 AD was also transient, lasting arguably even shorter than the Tang in its extreme west, so I believe the two current Infoboxes, Rome and Tang, depict very similar situations. (And Mesopotamia is solidly colored, which is not an option in the current RFC.)
Much of the issue with dates comes from the availability of reliable source maps. There are plenty for the core of the empire at various periods – the problem is everything west from the narrow corridor. We have reliably-sourced maps depicting the brief western extension of control around 661. The maps in the tertiary sources listed above are generally rough and marred by errors of detail. The only other source map raised so far is the 750 map in Herrmann (1935). No source map for 700 has been produced here – the green map labelled "700 CE" actually follows Herrmann's 750 map for the western portions. Kanguole10:36, 14 July 2026 (UTC)Reply
There are 2 such 700AD maps, one was the 15th I added later while talking with Quishufang. The maps by that creator appear somewhat widley used in academic works, an ex being this one. The vast majority of maps are set in 750 though. I believe more sources also explicity call the 8th century, or some date therein, the peak, zenith, or maximum extent of the Tang than they do any date in the 7th century. Klyashtorny being one specialist to do so, and 2 of the 750 maps being explicity labeled as such. Also, in my opinion since the 750/700 maps have also been reproduced so much and represent essentially all published maps by academic/university publishers for 43 years you should not say the 1935 source is the 'only' one for an alternative map...
Boutboul was spot on with their analysis of the two-in-one question; I was going to make much the same comment. The natural follow up is whether, per MOS:LEADIMAGE, it should be more natural and appropriate to show the Tang territory at its greatest extent, even if this was largely nominal and for a short period of time, or at a time where its extent was more long-lived, secure, and fixed. From a look at the infoboxes of other empires, it appears that the former is preferred, so on a basis of internal Wikipedia consistency I would have to choose 1a). I would appreciate opinions on whether this is an understandable analysis before I look at the more intricate issue of territorial control. ~~ AirshipJungleman29 (talk) 17:20, 13 July 2026 (UTC)Reply
Uninvolved comment No one is going to read through several thousand words of back-and-forth to cast a simple RfC !vote. May I suggest that the principal discussants summarise their respective positions with short statements of 250 words or less? ~~ AirshipJungleman29 (talk) 13:26, 12 July 2026 (UTC)Reply
Qiushufang's position: Current map or reworked version of the older map for infobox. The furthest western portion of the current map is in a lighter shade showing brief or nominal suzerainty of the Tang empire. The reworked map shows areas under actual military control by the Tang. The old map is missing the northern half of the Ordos Plateau contradicting Skaff 2012 (p. 49-50) showing military garrisons there. Qiushufang (talk) 13:37, 12 July 2026 (UTC)Reply
Thanks Qiushufang. A couple of questions: is it useful to indicate "brief or nominal suzerainty" in an infobox image? Also, is it appropriate to have an infobox image which does not display topographic features, especially when one such feature (the Ordos loop) is a matter of dispute? Finally, when you say "the reworked map", do you mean the third adjacent map in this RfC section or the map in the immediate section above? ~~ AirshipJungleman29 (talk) 14:45, 12 July 2026 (UTC)Reply
The reworked map is the third map in this section with the appropriate caption. I advised User:Kanguole to improve the topographical features of the current map, but mainly to make clearer what parts of the west around Xinjiang is actually covered. The Ordos Loop is a distinctive enough shape that its coverage would be recognizable even without topographic features. I do think displaying suzerainty (vassals, tributaries) or the like is useful for displaying where the Tang's area of influence and since the subject is an empire, I can see an argument made that it is appropriate, but I don't have any strong feelings towards keeping the current map. The reworked one showing actual military control in the form of garrisons is fine as well. Qiushufang (talk) 14:57, 12 July 2026 (UTC)Reply
Guile's Theme position - My position is that the current infobox map is based on a representation that, in the 43 years, has not been reproduced by any variety of China specialist or academic/university publishers, except once to explicitly deprecate its western claims as 'imperial pretensions' and not actual control. Nor does the current infobox map reflect WP:Weight in light of the broader body of reliable academic sources presented in the Source Survey, whether in map or textual form. For those reasons, I do not believe the current map should ever have replaced the previous 15 year infobox map (replaced May 2023). My opinion is that options 1B and 2B be picked; I would be satisfied with either reverting to the previous map and working to improve or reaching consensus on a 700 or 750 map that adheres to WP:WEIGHT to replace the current map with. Guile's Theme (talk) 13:39, 12 July 2026 (UTC)Reply
Thanks Guile's Theme. That's a little vague—can you be more specific? When you say "a representation" or "'imperial pretensions' and not actual control" are you referring specifically to the territories highlighted in lighter shade? I had a quick look at your source survey and while the views seem balanced, I can't see how WP:WEIGHT is seriously violated; can you clarify briefly with reference to the sources? ~~ AirshipJungleman29 (talk) 14:45, 12 July 2026 (UTC)Reply
Yes, the majority of the lighter shade in the west but not necessarily 100% of it. In my opinion WP:WEIGHT is violated because the balance of the sources, both text and maps in survey above, seem to fall in line with something closer to the previous incumbent map, at least in regards to the western borders beyond Anxi. Also, essentially all maps but the one being used as the source for the current infobox map show dates of 700 or 750 AD and not 661.
You can see what the modern maps of the Tang from academic/university publishers look like here, what specialist sources actually say in regards to supposed Tang control in that area here, what van de Ven and other sources miscited in support of the current infobox map actually said here, and the sources in support of the current infobox map I am against here. I emphasize no china specialist of any variety nor any academic/univeristy press has published the map the infobox is based upon in 43 years, except once to deprecate it, nor has its western most light shaded areas been reproduced in other modern maps; making the current infobox map an outlier. Guile's Theme (talk) 14:49, 12 July 2026 (UTC)Reply
Esiymbro's position Several examples against the above claim have been given in sources provided by other editors and some of those Guile's Theme themselves listed. The 660 661 (current Infobox) map has by far the best sourcing out of all options and represent mainstream views on the Tang territory in 660 661. In addition it represents the stable version of the Infobox map after many previous challenges. As far as I can tell no one is opposed to it except Guile's Theme.
My position is a WP:SNOWBALL close to this RFC, and the maps deleted from other articles should be brought back. The time and effort of other Wikipedia editors are better spent on something else. Esiymbro (talk) 14:04, 12 July 2026 (UTC)Reply
That the infobox map has not been reproduced. Multiple other academic sources mentioned in the discussion above support the borders either in the form of a map dated to 660 or text descriptions, e.g. several sources provided by Pataliputra near the beginning of the discussion and here.
I do not believe that NPOV/WEIGHT applies to selecting between 660 maps and 700/750 maps because they are not conflicting viewpoints. We should use the one from the most authoritative source. But as Kanguole has shown several other good sources use 660 maps as well. Esiymbro (talk) 15:22, 12 July 2026 (UTC)Reply
A final issue mentioned in previous discussions but apparently not here: 700 was technically not during the Tang dynasty, but the Wu Zhou interregnum. Whether we can treat the Zhou as part of the Tang is debatable, but the current infobox doesn't, so that should be another discussion. Esiymbro (talk) 15:29, 12 July 2026 (UTC)Reply
Kanguole's position – Any map should reliably sourced in its entirety. Of those proposed, only the current (661) one is.
The point of contention is the lightly shaded area on that map, which represents the former territory of the Western Turks, over which the Tang installed two puppet rulers after their victory in 657. This situation lasted only until the puppets were overthrown in 665. That is brief, and clearly qualitatively different from areas garrisoned with Tang armies. Nevertheless, our sources single out this brief episode (and map it) as "by far the westernmost extension of Chinese power in history".
Some sources show maps of this greatest extension. Others show the state of the empire around 750, roughly corresponding to the two darker-shaded areas of the map. The map attempts to balance the two by flagging the territories that were briefly controlled and showing them in a ligher shade. Kanguole15:38, 12 July 2026 (UTC)Reply
Pataliputra's position: the sourcing for maximum extent of Tang reach in the 660s (although thin, largely nominal, and short-lived) is quite good:
Of course there are many other maps related to the later periods (700-750s, and they already appear in the article) but they fail to show the peak of Tang reach. Pataliputra!(talk)16:39, 12 July 2026 (UTC)Reply
No one is making sense, there is talk of a 660 map above, but that's not an option in the RFC. Are the 660 and 661 maps the same? Is is the 660 one the one put above by Pataliputra and the 661 one is the current one? Also, why is the 700 CE map sourced to a book almost 100 years old? – Aza24 (talk)18:20, 12 July 2026 (UTC)Reply
The 3 current options are the current map which is labeled as 661, the former incumbent map which is labeled 700, and the 660 map Quishufang and I agreed on a week past which is a modified version of the 700 flag. Guile's Theme (talk) 18:26, 12 July 2026 (UTC)Reply
It is based off the 700 map with some modifications, the closest map to it would probably be this one, this one, or maybe the best sourced one is this one although 1 of those is labeled 700AD. Most of the 14 maps linked in the survey show western borders more similar to either the 700 or 660 maps than to the current 661 map. The sourcing for the Korean/Eastern aspect I'm not sure, I left the east up to Qiushifang's discretion mostly and just took his suggestions.
Although in retrospect, after over weeks worth of learning since that map was made, such a map should probably have been set to 700 AD instead of 660 AD to be less synth because there are no maps set in the 600s other than the map the 661 is based off. Guile's Theme (talk) 18:52, 12 July 2026 (UTC)Reply
The 660 map is a variation of the 700 map with the addition of all the Ordos Loop per Skaff (2012) p. 49 showing garrisons in that area in the early 8th century and Twitchett (1979) p. 403 in 742. Xiong (2008) p. 41 also has a chronology of the garrison seats illustrated at Protectorate General to Pacify the North. Parts of Korea are in a lighter shade to represent former and brief occupation in Twitchett (1979) p. 283. I did not choose the 660 year and believe the map better fits 675 or 700, which Guile's Theme seems to be in agreement with. The 700 map according to commons description is based on Albert Herrmann (1935). Historical and Commercial Atlas of China p. 29 and pp. 30–31. I have not checked the 700 map personally so I can't vouch for its accuracy but it seems pretty similar to all the other maps of the Tang dynasty that I've seen. Qiushufang (talk) 19:06, 12 July 2026 (UTC)Reply
The map was originally created in 2008 so I assume the creator did not have as much access to sources as we do now. They could have just been using whatever text they had on hand in the library or elsewhere. Qiushufang (talk) 19:15, 12 July 2026 (UTC)Reply
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
level-3 vital article is rated FA-class on Wikipedia's content assessment scale.
