« previous post | next post »

In media reports, we often encounter titles like this:

"China's Tibet to reward tips on 'terror attacks'"

"China's Xinjiang to ban burqas in public"

"Winter fishing kicks off in NW China's Xinjiang"

"Scenery of Lake Namtso in China's Tibet"

This type of title, with a place name preceded by "China's" is common, whether in domestic or foreign media, though my distinct impression is that it occurs more often in Chinese foreign language publications and in foreign publications that are relying heavily on Chinese news service material.

After decades of exposure to this usage, it still strikes me as somewhat unusual.  For example, "America's Philadelphia" or "the United States' Cleveland" or "USA's Texas" would certainly draw attention if they ever occurred in an article.

Since I first noticed this usage with reference to Xinjiang and Tibet years ago, I thought that it might have something to do with China's anxiety over its ownership of these late acquired pieces of peripheral territory.  So I started to do Google searches on "China's X", where "X" is the name of a province, region, or city.  Although, when I began, I certainly did not intend to cover all of China's provinces and regions, I became so intrigued by the results as I was compiling them that, by the time I had finished, I had probably included all, or nearly all, of them.  I have also included some major cities for comparison, and one tiny island.

Google hits for:

"China's Hainan" 554,000

"China's Sichuan" 359,000

"China's Shandong" 355,000

"China's Xinjiang" 350,000

"China's Guangdong" 290,000

"China's Yunnan" 290,000

"China's Shanghai" 263,000

"China's Zhejiang" 253,000

"China's Hong Kong" 222,000

"China's Jiangsu" 179,000

"China's Henan" 163,000

"China's Fujian" 148,000

"China's Tibet" 143,000

"China's Anhui" 138,000

"China's Hunan" 121,000

"China's Guangxi" 120,000

"China's Hubei" 113,000

"China's Hebei" 110,000

"China's Dalian" 106,000

"China's Inner Mongolia" 105,000

"China's Taiwan" 101,000

"China's Liaoning" 95,300

"China's Shaanxi" 90,900

"China's Gansu" 89,000

"China's Heilongjiang" 84,800

"China's Shanxi" 78,500

"China's Jilin" 74,600

"China's Guizhou" 69,500

"China's Jiangxi" 68,900

"China's Chongqing" 68,500

"China's Nanjing" 65,000

"China's Qinghai" 63,500

"China's Beijing" 59,800

"China's South Tibet" 54,100

"China's Guangzhou" 47,300

"China's Tianjin" 39,900

"China's Hangzhou" 34,500

"China's Ningxia" 31,300

"China's Harbin" 29,800

"China's Xiamen" 24,600

"China's Qingdao" 22,000

"China's Wuhan" 20,400

"China's Diaoyu" 18,700

"China's Xi'an" 18,200

"China's Urumqi" 14,100

"China's Changsha" 11,200

"China's Suzhou" 8,770

"China's Zhengzhou" 8,180

"China's Ningbo" 6,490

"China's Macao" 5,260

"China's Hefei" 5,130

"China's Sanya" 4,750

"China's Jinan" 4,400

"China's Arunachal Pradesh" 3,740

"China's Taiyuan" 3,690

"China's Wenzhou" 3,660

"China's Tsingtao" 2,620

"China's Haikou" 2,420

"China's Lhasa" 1,450

"China's Senkaku" 393

As I was processing these figures and the list grew longer, it appeared that the number of ghits was directly relevant to political, demographic, economic, and cultural factors.  Although the populations of Xinjiang and Tibet are relatively small, they are of great importance politically and economically (resources).  Diaoyu is an example of a place with 0 population, yet tremendous political significance because it is claimed by Japan, China, and Taiwan.  "China's Senkaku" ("Senkaku" is the Japanese name for the islands that China calls "Diaoyu") does occur (the least amount by far of any place name sampled) but almost always in expressions such as "China's Senkaku claim(s) / gas offer / flip-flops / policies," etc.  Hainan tops the list because it is a desirable tourist destination, with the more popular resort town of Sanya besting the capital of Haikou.

In assessing the significance of the above figures, we also have to take into account the fact that some occurrences of "China's X" are not the result of an attempt to identify "X" as belonging to China, but rather as defining a matter pertaining to "X" as an issue for China, e.g., "China's Hong Kong dilemma", "China's Taiwan policy", "China's South Tibet ambition", and so forth.  A finer grained analysis would separate off this latter type of occurrence.

Incidentally, while "China's South Tibet" often is part of an extended phrase such as "China's South Tibet ambition", in some cases it is part of an overt claim or recognition that it belongs to China.  In the eyes of its southern neighbor, the so-called "South Tibet" is Arunachal Pradesh, the northeasternmost of India's 29 states.  Surprisingly, while "China's Arunachal Pradesh" is usually part of extended phrases such as "China's Arunachal Pradesh fixation" and "China's Arunachal Pradesh claim", it sometimes stands as an assertion or recognition of Chinese ownership.  Consequently, it is ironic that a Chinese smartphone maker just a couple of days ago tacitly recognized "South Tibet" as Arunachal Pradesh:

"Xiaomi shows Arunachal is in India at handset launch, triggers anger in China"

Another consideration is the sheer insecurity over whether readers realize that "X" belongs to or is located within China.  Consequently, to be on the safe side, authors and editors, and particularly headline writers, will tend to put "China's" before "X" far more often than we are accustomed to for places in America and other places with which we are more familiar.  In this way, "X" is clearly stamped as being a part of China — but sometimes that is a consciously political statement too.


  1. Jake Wildstrom said,

    February 1, 2015 @ 8:57 pm

    Surely the ratio of "China's X" to "X" is more revealing than the raw number of hits for "China's X"? I rather suspect the tiny number of, say, "China's Ningbo" hits has less to do with an English-language media tendency to refer to Ningbo unprefaced, so much as not to mention it much at all, at least by comparison to the much higher numbers for Tibet, Shanghai, Hong Kong, etc.

  2. Rubrick said,

    February 1, 2015 @ 9:15 pm

    Your analysis rings true. And I'd say that while "America's Philaelphia" certainly sounds odd, "America's Midway Atoll" sounds less so.

  3. Victor Mair said,

    February 1, 2015 @ 9:30 pm

    @Jake Wildstrom

    That's precisely the point. I was trying to get a sense of when a need was felt to specify that a certain place belongs to China. In that respect, Xinjiang and Ningbo are indeed very different.

  4. Victor Mair said,

    February 1, 2015 @ 9:53 pm

    Here are a few results using Chinese rather than English:

    中国的新疆 (China's Xinjiang) 326,000

    中国的西藏 (China's Tibet) 328,000

    中国的台湾 (China's Taiwan) 390,000

    中国的内蒙古 (China's Inner Mongolia) 142,000, 中国的内蒙 (short, and more common, form of the preceding item) 215,000

    中国的上海 (China's Shanghai) 369,000

    中国的云南 (China's Yunnan) 321,000

    中国的海南 (China's Hainan) 312,000

    中国的四川 (China's Sichuan) 316,000

    中国的香港 (China's Hong Kong) 386,000

    中国的宁波 (China's Ningbo) 303,000

    中国的澳门 (China's Macao) 317,000

    It's uncanny that, regardless of the results in English, all of the Chinese items are in the three hundred thousand range, except for Inner Mongolia, which is not far off in the two hundred thousand range. I suspect that this may be due to standardized web pages (for domestic tourism, introduction to government, etc.).

  5. Jeff W said,

    February 2, 2015 @ 12:01 am

    …to be on the safe side, authors and editors, and particularly headline writers, will tend to put "China's" before "X" far more often than we are accustomed to for places in America and other places with which we are more familiar.

    It seems like some of the English language usage as well might be in a travel/tourism context, with the implicit meaning of “yet another thing China has to offer,” combined with an assumption of unfamiliarity—sort of similar to saying something like “Britain’s Lake District” or “Britain’s Cotswolds.” And then it’s baked in through the use of standardized web pages.

  6. Bob Ladd said,

    February 2, 2015 @ 2:48 am

    I think the issue raised in the final paragraph of the post – will readers know that Harbin is in China? – is at the heart of this usage, not any political sensitivities. A few minutes on Google will find you thousands of occurrences of things like "India's Kerala", "Mexico's Oaxaca", "Brazil's Minas Gerais", and "Germany's Lower Saxony". The oddness of "America's Texas" comes from the fact that most people writing news headlines in English are going to assume that readers have heard of Texas and at least know that it's in the US. I'm unconvinced that there's anything specially political about "China's Xinjiang".

  7. Ken said,

    February 2, 2015 @ 10:11 am

    A similar construct is used in sports: "Argentina's Sergio Romero", "France's Olivier Giroud". Some would say that sports is as nationalistic as politics.

  8. Michael Watts said,

    February 2, 2015 @ 10:35 am

    That's precisely the point. I was trying to get a sense of when a need was felt to specify that a certain place belongs to China.

    But you seem to have completely missed Jake Wildstrom's point. Your methodology can't address the thing you claim to be interested in; you'd need the numbers for how often each of these places is referred to without being labeled as "China's".

    For example: if there are 350,000 references to "China's Xinjiang" and 20,000 references to "China's Wuhan", it could be because they both have 500,000 references overall and Xinjiang is 17 times more likely to be explicitly marked as "China's", or it could be that they're both marked as "China's" 100% of the time and Wuhan is much less newsworthy than Xinjiang is. What's really odd is that you call this out in your post

    it appeared that the number of ghits was directly relevant to political, demographic, economic, and cultural factors. […] Hainan tops the list because it is a desirable tourist destination, with the more popular resort town of Sanya besting the capital of Haikou.

    but instead of taking the trivial step of normalizing all your figures to "share of references marked as belonging to China", you just make the post with the meaningless numbers and misinterpret the guy telling you how to get the figures you claim to want.

  9. Michael Watts said,

    February 2, 2015 @ 10:46 am

    For reference, I see 32 million reported ghits for "Ningbo", and 6500 for "China's Ningbo"; all of the front page results for "China's Ningbo" are spurious ("China's Ningbo port"; "China's Ningbo Joyson Electronic Corp."; "China's Ningbo cathedral"; "China's Ningbo Baoxin Stainless Steel"; etc).

    In contrast, I see only 18 million reported ghits for "Xinjiang" and 350,000 for "China's Xinjiang", of which only most of the front page results are spurious. A couple are real! That supports VHM's hypothesis that in fact Xinjiang is much more likely than Ningbo to be explicitly marked as belonging to China, although much of the effect seems to be driven by headlinese (eyeballing those front page results, there are a lot of cases where the headline says "China's Xinjiang" and the article says something more like "China's Xinjiang province").

  10. Victor Mair said,

    February 2, 2015 @ 11:12 am

    Thank you VERY MUCH, Michael Watts! I was just about to perform the operations you did. Now I don't have to do them.

  11. Bob Ladd said,

    February 2, 2015 @ 11:19 am

    I noticed the same thing MIchael Watts mentions when I was looking for non-Chinese examples earlier: the headline would say "India's Kerala" and the text would say "India's Kerala state".

    Curiously, it's *really* hard to imagine any English-language publication saying "America's Wyoming state". Maybe "the American state of Wyoming" would occur, though, in something written outside the US.

  12. Michael Watts said,

    February 2, 2015 @ 1:01 pm

    "America's Wyoming state" is blocked by some sort of convention as to the proper use of 'state', but here are a couple of similar examples:

    "America's Wyoming region" is more or less acceptable to me in a context that clearly refers to areas near Wyoming that may or may not actually be in Wyoming.

    "America's greater Texas area" seems totally standard.

  13. Mr Punch said,

    February 2, 2015 @ 2:22 pm

    "China's Xinjiang province" is what I'd expect to read (and I believe that some Chinese cities are administratively provinces in themselves.) It's true that we don't use "state" that way in the US, but we could certainly write "Vermont's Addison County" — though probably not "Vermont's Burlington." It does seem to have to do with area, primarily, as opposed to precise location or form of government.

  14. Bob Ladd said,

    February 2, 2015 @ 2:46 pm

    @Mr Punch, @Michael Watts

    Nice examples. They make it look like there are two factors involved in the acceptability of this construction – one is whether the reader/listener is likely to be able to identify the proper name (e.g. Addison County) without additional information, and the other is that the proper name should refer to a region within the "possessive" proper name, not a town. Notice that it works nicely with parts of towns: Boston's Little Italy, Edinburgh's Morningside, Tokyo's Shinjuku, London's Mayfair, Berlin's Kreuzberg, etc. etc. These typically have rather fuzzy boundaries. As with Michael Watts's examples, they also work well with added words (Edinburgh's Morningside neighbourhood, Berlin's Kreuzberg district, etc.).

  15. JB said,

    February 2, 2015 @ 4:38 pm

    Tourist brochures and standardised editorial policy in the news are one thing, but the egregious modification of Chinese-published works in English, especially those dealing with religion, history and so on, is particularly irksome. Some of the titles in the Oxford Short Introduction series spring to mind (published in China by the FLTRP), every instance of 'Tibet', for instance, being inevitably preceded by 'China' in the genitive case. That is, when those references are simply not removed entirely.

  16. Jeff W said,

    February 2, 2015 @ 5:26 pm

    @ Bob Ladd
    I was thinking of something similar but I’m not so sure about the fuzzy boundaries or the reader’s familiarity (in some cases): “New York’s Times Square” or “London’s Piccadilly Circus” seem pretty well defined and are very well-known. (There is, of course, a different Times Square in Hong Kong.)

    To me, it seems like there is a sort of normative, rather than empirical, claim going on. It’s not that the person writing is saying the reader is likely or not likely to be familiar with some place being where it is. Rather, the writer is disclaiming the assumption that he or she thinks the reader should be familiar with where a particular place is. If a newscaster says “x happened today in London’s Piccadilly Circus,” it seems that that’s not quite saying “We think you may not know where Piccadilly Circus is”; it’s saying something more like “If we omit ‘London,’ we’re sort of saying that you should know where Piccadilly Circus is—and we don’t want to be seen as making that claim.” (I don’t know if something like that is going on with the various provinces in China but it doesn't seem unlikely to me.)

  17. chips mackinolty said,

    February 2, 2015 @ 7:56 pm

    @ Michael Watts and others in this thread, e.g. "America's Wyoming region" etc etc. What is missed in all this is that the term "America" is automatically assumed to refer to the USA. Have no idea how to track this in Google or any other corpus, but the "America" assumption seems almost universal (unless you live elsewhere on the two continents). The deal about the genitive case applied to China is one thing. Perhaps the Monroe Doctrine has more to answer for, linguistically, than its stance on foreign affairs (or perhaps they are synonymous)

  18. Wentao said,

    February 2, 2015 @ 9:08 pm

    Politics is definitely relevant here – This reminds me of the Mainland translation of "Tintin au Tibet" as 丁丁在中国西藏.

    @Mr Punch
    I suppose "Vermont's" is used here because relative few people know the names of counties in each state and additional information is necessary for understanding. However, Burlington doesn't seem to be that well-known either, but one can easily write "Burlington, VT", which seems to be a convention for US cities/towns. If there is similar system for Chinese place names in English, such as "Ningbo, ZJ", I would imagine the number of hits for "China's XX (city)" to decrease drastically.

    Also I'm happily surprised by your example, since I have rarely come across any mention of "Addison County" outside of Middlebury.

  19. Victor Mair said,

    February 3, 2015 @ 8:53 am

    Wonderfully apt comment about "Tintin au Tibet", Wentao!

    I started out to write a reply, but it became so long and involved that I realized I have to make it into a separate post, though I probably won't be able to complete that for a couple of days because there are too many other things standing in the way right now.

  20. JK said,

    February 3, 2015 @ 6:04 pm

    It's certainly a political statement, and probably comes from copying the Chinese usage. The typical Chinese usage is actually 中国新疆 rather than 中国的新疆. If you look at official PRC media covering Tibet and Xinjiang, when domestic leaders from around China meet local Xinjiang and Tibetan leaders, they just use the short version: Xinjiang or Tibet. But when the local Xinjiang and Tibetan leaders receive foreign visitors, they invariably use the longer version 中国新疆, and the Chinese articles that cover the meetings invariably have the visitors using that exact phrase as well, as if having a foreign visitor come and explicitly state that Xinjiang is part of China is a key part of the exchange. On the other hand, when the leaders of Beijing or Tianjin receive foreign guests, they never make a point of saying 中国天津 or 中国北京 (as far as I know). I think possessives being used for other locations around China can be explained as people simply trying to translate place names without using a comma, even though it doesn't work very well.

  21. Victor Mair said,

    February 3, 2015 @ 9:08 pm


    You are so right!

    From early morning, I was planning to do this, but I didn't get a chance until now:

    中国新疆 548,000

    中国的新疆 326,000

    中国西藏 1,190,000

    中国的西藏 328,000

  22. Chris Sundita said,

    February 5, 2015 @ 8:30 am

    While "America's Wyoming state" sounds off, I think "America's Washington State" is somewhat acceptable, though still unusual. I think it has to do with the fact that we say "Washington State" frequently in order to differentiate our state from the capital (which I do quite frequently here in New York…..State).

  23. JK said,

    February 12, 2015 @ 2:19 pm

    Here's one more example with a double modifier:

    KATHMANDU, Feb. 12 (Xinhua) — The Chinese Tibetan Losar, or new year, has been marked in the Nepalese Capital City Kathmandu on Thursday evening.

    The Chinese embassy in Kathmandu hosted a grand ceremony in celebration of the Chinese Tibetan Losar.

RSS feed for comments on this post