Stephen Owen Translates Du Fu.

Stephen Owen has translated all of Du Fu. Big deal, you say, people are always translating other people? Well, what if I tell you that (to quote Jon, who sent the link to me) “It’s a remarkable piece of work! The translations are beautifully lucid and can be read as poetry, but it’s a critical edition and so has a big, unobstrusive, very satisfying apparatus. Oh, and it’s 3,000 pages long.” Sounds great, you say, I wish I could afford things like that? Well, what if I tell you De Gruyter has put the whole thing online to download for free? Just click the “Table of Contents” link and you’ll get a page from which you can download pdfs of as many sections as you want. Here’s their Aims and Scope statement:

The Complete Poetry of Du Fu presents a complete scholarly translation of Chinese literature alongside the original text in a critical edition. The English translation is more scholarly than vernacular Chinese translations, and it is compelled to address problems that even the best traditional commentaries overlook.

The main body of the text is a facing page translation and critical edition of the earliest Song editions and other sources. For convenience the translations are arranged following the sequence in Qiu Zhao’an’s Du shi xiangzhu (although Qiu’s text is not followed). Basic footnotes are included when the translation needs clarification or supplement. Endnotes provide sources, textual notes, and a limited discussion of problem passages. A supplement references commonly used allusions, their sources, and where they can be found in the translation.

Scholars know that there is scarcely a Du Fu poem whose interpretation is uncontested. The scholar may use this as a baseline to agree or disagree. Other readers can feel confident that this is a credible reading of the text within the tradition. A reader with a basic understanding of the language of Chinese poetry can use this to facilitate reading Du Fu, which can present problems for even the most learned reader.

This is an amazing gift to the world from De Gruyter, and I offer them my heartfelt appreciation. Also, Owen is a good guy and an enjoyable writer; from the “Du Fu Lore and Translation Conventions” section, here’s a sample:

Who’s Hu? Non-Han

By and large people doing Tang studies have fortunately abandoned the blanket term “barbarian” for the non-Han peoples with whom the Tang was engaged. I restrict “barbarian” to the word lu 虜, a contemptuous, pejorative term for non-Han without ethnic distinction. There are archaic terms, there are vague regional terms, and precise designations of peoples and polities. After long brooding I have decided to use the Romanization for Hu 胡. Hu refers primarily to the Indo-European inhabitants of Central Asia, such as the Sogdians, but it was applied more loosely to all non-Han peoples of the north and northwest. In Du Fu it is also used for northeasterners and on rare occasions, for the Tibetans. Du Fu often describes the rebels as Hu, so when he refers to the Uighurs, who were Tang allies, he often does so with ethnic precision, Huihu 回鶻. The northeastern peoples were most commonly referred to as Yi 夷, though sometimes Du Fu uses the more precise ethnic designations “Blond-heads,” huangtou 黃頭, and Xi 奚. Toward the west were the Qiang 羌, between the Tibetans, the Uighurs, and the Chinese. The Tibetans were China’s major adversary in the eighth century. They are often referred to anachronistically as the Rong 戎 or the Dog Rong, Quanrong 犬戎, the ancient adversaries of the Zhou dynasty. The equally anachronistic term for northern peoples was Di 狄. This is similar to the Magyars becoming Hun-garians, or the Germans in World War I being referred to as “Huns.”

(I should note that there is only one “Allusions” section, even though it is listed for each volume, so you only have to download it once.) Thanks, Jon!


  1. The article on Tu Fu in the old Encyclopedia Britannica was hilarious, especially the part where he “became a successful swineherd.” Before I knew that the subject was actually a major cultural figure, I seriously thought the whole entry might have been a fake, a copyright trap.

  2. Sounds like cuneiform sources using terms like Hanaya for “westerners” long after the Hanaeans had vanished from the upper Euphrates and people called HA.NA.HI.A in cuneiform were more likely to call themselves Makedones and come from Yawana. It let them keep using the old omens and save wedges.

  3. David Marjanović says

    Wait. Yellow-headed people northeast of China?

  4. Yeah. I was pretty surprised too when I first learnt that the biggest Sogdian concentration was roughly from modern-day Beijing area upto the Northeast.

  5. ə de vivre says

    Biggest in what sense? The main Sogdian merchant communities in China followed the trade routes through the Gansu Corridor towards Chang’an. Not that the Sodgians were particularly famous in China for being blond.

  6. Sogdians almost took over China during An Lushan rebellion.

    (Edward H. Schafer maintains that An Lushan is probably the Sinicized version of a name derived from Anxi (安息 “Parthia(n)”) and the common Sogdian name ܪܘܚܫܐܢ Roxshan meaning “the Bright”,related to the Sogdian female name Roxana.)

  7. Thanks for this! I had heard about Owen’s edition and that De Gruyter had released it for free (for which, I agree, they deserve a lot of credit and gratitude), but hadn’t been able to figure out how to access the damn thing.

  8. David Marjanović says

    …Oh. I forgot Ān Lùshān was from Liáoníng. Blog post with all the etymology and history anyone could want here.

  9. Jim (another one) says

    “Wait. Yellow-headed people northeast of China?”

    I remember a poem about the sack of Luoyang(?) by the “huáng​tóu​ Xiān​bēi​”, probably around the end of the Norhtern Song. The Xianbei were proto-Mongols and lived in the in and around the area of Yan – northeastward in China. They had been mentioned in records since Han times.

    “Yellow-headed” may be a little extreme. The term covers brown hair too, though that would look nearly as strange I am sure.

  10. Greg Pandatshang says

    Could 黃頭 “yellow head” potentially also refer to some sort of yellow headgear?

  11. There was the Yellow Turban Rebellion (184-205), but I doubt it’s related.

  12. Technically yes – probably not in the case referred to here, but 黄頭 was used to refer to boatmen (黄頭郎) and, by extension, boats and even navies, because (allegedly) boatmen used to wear yellow hats to ward off disaster.


    “Deng Tong was from Nan’an in Shu [Sichuan]. Because he could handle a boat, he was made a boatman (lit. yellow-headed gentleman).”

    A fortuitous appearance in a dream later made Deng Tong a favorite of the Emperor, even though he “had no skills/talents” (無伎能).

  13. In my (limited) experience with De Gruyter, they have generally been very supportive of open access publishing, but according to Owen’s Acknowledgments credit is also due to the Mellon Foundation, who “generously supported these and other initial volumes in this series [the Library of Chinese Humanities], so that they may appear both in standard print format and open­-access on the Web”
    (My version of the page has “Access brought to you by South China University of Technology,” so perhaps they also provided some form of financial support – but I don’t know whether this is location-specific.)

  14. The equally anachronistic term for northern peoples was Di 狄. This is similar to the Magyars becoming Hun-garians, or the Germans in World War I being referred to as “Huns.”

    Historical note: “Huns” for Germans actually originates with the Germans themselves. Specifically, with Kaiser Wilhelm II, who dispatched his troops to the 1900 Peking Expedition (the relief of the Legations during the Boxer Rising) with the words “When you come upon the enemy, smite him. Pardon will not be given. Prisoners will not be taken. Whoever falls into your hands is forfeit. Once, a thousand years ago, the Huns under their King Attila made a name for themselves, one still potent in legend and tradition. May you in this way make the name German remembered in China for a thousand years so that no Chinaman will ever again dare to even look askance at a German!”

  15. I think I’ve quoted this before somewhere, a translation by Rueckert of a Shi Jing song:

    Gekommen sind die Schaaren
    Der nordischen Barbaren,
    Mit langen hellen Haaren,
    Mit Haaren hellen langen,
    Die ihnen wie die Schlangen
    Von beiden Schlaefen hangen!

  16. Schi-King: Chinesisches Liederbuch, p. 44.

  17. ə de vivre says

    While we’re on the subject of Huns and Sogdians, it’s thanks to the 4th century Sogdian letters from Dunhuang that we can establish an identity (at least a nominal one) between Huns (“xwn”) and the Xiongnu. Then by the 5th century, Sogdian graffiti in the passes towards the Indian subcontinent made by Sogdians named “xwn” shows that the Huns have already passed through on their way west (towards Rome?). There’s a great book fittingly called “Sogdian Traders: a History” that details the extent of the Sogdian diaspora. It’s $200+ new, but you can also get it as a pdf on certain sites that rhyme with “ribgen”.

    Generally you (or whoever pays your grants) have to pay publishers like De Gruyter to make your work open to the public. They’re not in the business of giving things away for free. I believe that the exorbitant prices to buy back your copyright was one of the reasons the editorial board left Lingua to form Glossa.

  18. It is by no means the case that the only alternative to the traditional reader-pays model is open access with author payments. As Peter Suber wrote back in 2006: “Some no-fee OA journals have direct or indirect subsidies from institutions like universities, laboratories, research centers, libraries, hospitals, museums, learned societies, foundations, or government agencies. Some have revenue from a separate line of non-OA publications. Some have revenue from advertising, auxiliary services, membership dues, endowments, reprints, or a print or premium edition. Some rely, more than other journals, on volunteerism. Some undoubtedly use a combination of these means.”

    The moral case for OA on the supplier side is that these so-called publishers are really just Raubsritter, the original kind who in defiance of Imperial authority charged arbitrary ship tolls on the Rhine, because they could. They don’t pay their authors, they don’t pay their editors, their readers and advertisers pay them. About the only thing they do that costs money is to run servers and pay printers.

  19. Hear, hear!

  20. David Marjanović says

    Raubritter. And on the Danube, too – there’s the legendary dynasty called Kuenringer.

  21. Eventually the Emperor, merchants on the river, and other stakeholders got really tired of having their ships caught in chains, the Raubritter (thanks, David) were eventually reduced by military force. Monopolists beware.

  22. And of course we know about Caesar and the pirates.

  23. In addition to the multiple PDF links, I note that there is a single EPUB link on the left, which I am guessing is for the entire book — it is 169MB!

    After extracting the files from the epub, I see that most of the space is taken up by PNG image files of the various Chinese characters. I have no idea why they went that route, rather than just using Chinese characters, and including a Chinese font. Surely there was one that would have been appropriate?

  24. A quick web search finds this fun introduction to Chinese typography:–cms-23444

    This part looks appropriate because the PDF files do have an embedded font called AdobeKaitiStd-Regular

    A kaiti font mimics basic brush script lettering–you might loosely translate this as “regular brush”. But a kaiti is not a novelty font, it never gets overly flowery, and it’s still constructed within certain parameters and maintains an upright structure. Here’s Adobe Kaiti Standard which comes bundled with Photoshop and some other Adobe products:

    The image files in the epub look like they were made using the same Kaiti.

  25. @ ə de vivre

    Sogdian Traders: a History wasn’t at anything that rhymed with ribgen but it was findable through a Google search. A fantastic bringing together of material from East, West, and in-between. I only skimmed through it but was struck both by the variety of evidence collated and the large number of statements phrased as surmise or inference.

    The site that I downloaded from was somewhat mysterious. When I typed in the URL of the home page I was redirected to a different URL that announced Website Coming Soon following the successful installation of WordPress. No indication of what kind of site it might be.

  26. ə de vivre says

    Huh, maybe we’re looking at different mirrors.

    It takes some real sticktoitiveness to do that kind of Central Asian history, considering there are sources not only in Sogdian, but also in Chinese, Greek, Pali, Persian, Arabic, and various flavours of Turkish. It’s kind of the area study that makes other area studies shit their pants.

  27. I downloaded from

    I’m curious about your pen name, ə de vivre. It sounds positively Clouseauesque.

  28. ə de vivre says

    By “rhymes with ‘ribgen'”, I was referring to library genesis. Depending on your personal sense of ethics the site falls somewhere between liberating knowledge or theft.

    There’s not much story behind my name. It was just the first stupid pun in the chamber, so to speak, when I first commented here.

Speak Your Mind