More Sinological suffering

« previous post | next post »

[This is a guest post by Brendan O'Kane. See "Sinological suffering", 3/31/17, for background.]

I snapped this picture at the library today:

The best thing about this is that it's annoying on multiple levels:

  1. One title, two volumes, two romanization systems.
  2. The Wade-Giles romanization is wrong, because it doesn't indicate aspiration.
  3. The Pinyin romanization is wrong too, because it uses spaces for syllable boundaries rather than word boundaries.
  4. The Library of Congress rules for Pinyin, which describe the separation of syllables as "generally consistent with Wade-Giles practice," are wrong about Wade-Giles practice.

VHM:  Li Zhuōwú pīpíng Xīyóu jì 李卓吾批评西遊記 (Li Zhuowu's criticism of Journey to the West)

Li Zhuowu is the author's pseudonym.  His real name is Li Zhi (1527-1602). From Wikipedia:

He was born in Jinjiang, Fujian province (in modern-day Quanzhou). His ancestor by seven generations was Li Nu, the son of Li Lü, a maritime merchant. Li Nu visited Hormuz in Persia in 1376, converted to Islam upon marriage to a Semu girl ("娶色目女") (either a Persian or an Arab girl), and brought her back to Quanzhou. This was recorded in the Lin and Li genealogy《林李宗谱》. However, the new faith did not take root in his lineage and the family stopped practising Islam during the time of his grandfather. His father made a living by teaching, and Li Zhi was therefore educated from an early age.


  1. WSM said,

    April 3, 2017 @ 3:35 pm

    In many cases though it isn't obvious how to segment in word boundaries though, right? Segmenting on syllable is much less subjective and quite defensble if you're helping people with searches etc.

  2. JK said,

    April 3, 2017 @ 3:44 pm

    It probably shouldn't bother me, but I am always a little annoyed when books published after 2000 use Wade-Giles.

  3. Guy Plunkett III said,

    April 3, 2017 @ 4:22 pm

    Not my field of study, but these volumes are pretty obviously rebound copies. A little Googling suggests that the originals were softbound, and libraries traditionally rebind those for durability. Binderies are few and far between these days, and if they are like the folks who do thesis binding, you are essentially limited to upper case sans serif stamped lettering on the spine, in white or gold. No excuse for the two volumes using different romanizations, but otherwise it isn't like there were a lot of options.

  4. J.W. Brewer said,

    April 3, 2017 @ 4:29 pm

    I am surmising that the two volumes, even if published in the same year, arrived at the library in question at different times, although I guess it would be more amusing if they had arrived at the same time in the same box but were then passed off for binding/cataloging to different staff members who felt no need to coordinate their approaches to the issue.

  5. Y said,

    April 3, 2017 @ 4:58 pm

    So what did Li Zhuowu have to say about A Journey to the West?

  6. AntC said,

    April 3, 2017 @ 5:36 pm

    Thank you @Guy Plunket III, that puts quite a different slant on it.

    So what are the romanisations used inside, on the title pages? And did the binders just follow that?

    Or is there none? Even so, strange that whoever tackled volume 2 didn't seek out volume 1.

  7. Victor Mair said,

    April 3, 2017 @ 5:44 pm

    China has published official orthographic rules for word segmentation, punctuation, and so forth. An English translation (by John Rohsenow) of the 1996 edition of the orthographic rules is printed at the back of all ABC Chinese-English dictionaries from the University of Hawai'i Press.

    Pinyin orthography is controlled by official Chinese national standards, and — as we know — it was invented by Zhou Youguang and other Chinese experts. The most essential standards are:

    Hànyǔ pīnyīn fāng'àn 汉语拼音方案 (Scheme for the Chinese Phonetic Alphabet) (1958)


    Hànyǔ pīnyīn zhèngcífǎ jīběn guīzé 汉语拼音正词法基本规则 (Basic Rules for Hanyu Pinyin Orthography) (GB/T 16159-1996)

    For those who are interested, I can send a pdf of the latest version of the official orthographic rules from the PRC.

    The Canadian researcher, Clément Arsenault, of the University of Montreal refers to the linking up of syllables as "aggregation". He wrote his dissertation on this subject and has also published several articles about it, paying particular attention to the effect it has for lookup efficiency in library catalogs, etc.

    When the Library of Congress decided about twenty years ago (?) to divide all syllables of Pinyin, except for personal names and the names of countries, I thought they were making a big mistake (it was a huge step backward from what had already been achieved in word segmentation), and I fought hard against it. My arguments against their decision are (or at least were) recorded somewhere on the web. The person in charge of the changeover from Wade-Giles to Pinyin was not a specialist in Chinese, but, as I recall, a specialist in Korean.

    Word segmentation has significant benefits for language pedagogy and computerization of Chinese — helping humans and computers more easily recognize where words begin and end, unlike character texts, which give no clues.

    Sooner or later, word segmentation will be common in Pinyin writing. It is already a significant feature of the emerging digraphia, in large measure because of the role it plays in Pinyin inputting of Chinese texts. Automatic word segmentation is also a feature of many Pinyin typing software programs, and the output is of a high standard that requires very little polishing on the part of the human typist.

  8. J.W. Brewer said,

    April 3, 2017 @ 6:13 pm

    It is hard to fault a US library (if that's where this picture was taken) for following the ALA-LC system rather than try to understand and take sides on criticisms of that system. The domain of the ALA-LC system is necessarily limited to the sort of chunks of text (typically titles of works) that get cataloged and indexed by librarians, and I expect supporters of a rival approach can succeed (and perhaps have thus far succeeded) in pitching their own preferences for use in other contexts. One interesting question is whether if a given book (let's say a facing-page translation) puts its own preferred romanized version of the title on the cover and spine that is enough for that spelling (which could be quite idiosyncratic) to control.

  9. John Swindle said,

    April 3, 2017 @ 6:32 pm

    The picture is funny because it's so typical and so goofy. Regardless of when the volumes arrived, they must have been rebound at different times. In a way it's not as bad as it looks. The minor infelicities in both romanizations, a hyphen missing here, a space added there, aren't random. They're part of a style that will be familiar to users of the library and will pose little additional difficulty to students who are already juggling traditional and simplified characters, Wade-Giles, and Hanyu Pinyin.

    As to what Li Zhouwu had to say, according to Baidu Baike, as translated by Google, "'Li Zhuowu criticized the Journey to the West' (from top to bottom) by the Ming Ye day care Li Zhuowu conducted a comprehensive comment. His commentary theory is not as good as Jin Shengtan comprehensive system, but he inherited Li Zhuowu, under the Jin Sheng sang, the 'Journey to the West' fantasy character, put forward a unique view, than deny the 'Journey to the West' artistic achievements Jin Shengtan, Mao Zonggang, Zhang Zhupeng Clever much more…." That says it better than I could.

  10. Brendan said,

    April 3, 2017 @ 6:39 pm

    @J.W. Brewer – You're right, of course — annoyances (3) and (4) aren't at all the library's fault, and the LoC rules for Pinyin do have a rationale behind them, even if that rationale basically boils down to "It would be too hard to do this the right way." The title of the work Li Zhi is commenting upon is a good example of the difficulties involved in figuring out what the right way of doing things would be: I've seen it romanized as Xiyouji, Xiyou ji, and Xi you ji, any one of which would be defensible. (I don't think I've ever seen Xi youji, but there'd probably be a case to be made for that too.)
    Annoyance (1) might be the result of the books arriving at different times, or possibly being rebound at different times. The library catalogue entry uses Pinyin for both volumes, and searching for the Wade-Giles version of the title turns up no results.

  11. Lupus753 said,

    April 3, 2017 @ 6:57 pm

    For a long time, I preferred Wade-Giles simply because of how non-intuitive some of the consonants in Pinyin are (the English voice actors for the first few Dynasty Warriors game constantly mispronounced Cao Cao as "Cow Cow"), but after learning that Wade-Giles requires superscript numbers to indicate tone, I'm not sure which romanization scheme I would consider less awful.

  12. John Swindle said,

    April 3, 2017 @ 6:57 pm

    And of course I meant Li Zhuowu instead of Li Zhouwu.

  13. Elizabeth Yew said,

    April 3, 2017 @ 7:19 pm

    In Columbia historian Adam Tooze’s recent book, “The Deluge: The Great War, America and the Remaking of the Global Order, 1916-1931” , he has to tackle some Chinese history, a topic with which he is clearly not as at ease as he is in economic history. He mentions “Chang Tso-lin” and “Zhang Zoulin”(sic) , which when reading the text are both clearly identical to 张作霖. It is unclear if he is aware that they are the same person.

  14. J.W. Brewer said,

    April 3, 2017 @ 7:21 pm

    It seems imprudent for the catalog not to have the spelling for vol. 1 that matches what's on the spine of the physical volume. Hopefully there's a scannable barcode affixed to the book that generally makes it unnecessary for whoever checks it back in to even read what's on the spine much less try to correlate it with the library's records, but sometimes those barcode stickers become detached or illegible, and you wouldn't want the volume to pose a mystery to the library staff. (As a one-time low-paid low-level employee of a university library system I should like to stress that one important feature of any system like this is for it to be useful for routine tasks like checking books in and out, reshelving them, etc etc by low-paid low-level library employees who lack knowledge of the language and script of the books and likewise lack knowledge of rival romanization systems and the arguments for and against them. Switching to a new system for newly-acquired volumes when you know you're probably never going to have the time and money to completely redo the many thousands of volumes already in your holdings that were done with a different system is just asking for trouble, and at a minimum you need to figure out how to avoid losing track of works that were cataloged/bound/labeled/etc under the old system as long as you have any material number of them in your possession.)

  15. JB said,

    April 4, 2017 @ 3:15 am

    @Guy Plunkett III: There is an excuse: at least this way you'll be sure to find the book regardless of which romanisation system you use…

  16. flow said,

    April 4, 2017 @ 6:27 am

    Just wanting to point out that while J.W. Brewer's remark that swapping the romanization is just asking for trouble seems obviously right to me, libraries should still do it (i.e. abandon Wade-Giles, which is deeply flawed, and adopt Pinyin, which is somewhat better and has become a de-facto standard, which is good).

    All catalog systems must by necessity always be imperfect, and the records for any sizable collection will always be less than 100% consistent; this is the nature of things. 'Alphabetic languages'—forgive me—are likewise troubled in that spelling used to be less determinate in past centuries, may vary according to place, and is subject to reforms even where regulated by governments. Thus the famous "Brehms Tierleben", first published in the 1860s, was probably spelled "Brehm's Thierleben" in editions up to roughly 1900, and it is not immediately obvious whether a physical catalog should have it show under 'Tier-', under 'Thier-', or both.

    I have seen many library search systems, felt like space-age when using the (advanced for the time) microfiche catalog of my hometown's university library around 1980, and I love card-based catalogs (but haven't had the occasion of working with one for decades); however, as far as ease of use and recall & precision (the two standard gauges for search) go, nothing compares with fuzzy full-text search (the lexico-statistical approach, or call it 'a google-like search engine').

  17. Rodger C said,

    April 4, 2017 @ 6:52 am

    A member of the Comp. Lit. department at Indiana U wrote his MA thesis on the Hsi-Yu Chi and his PhD dissertation on the Xiyou Ji. That was near the beginning and end, respectively, of the Seventies. I used to see both volumes in the departmental library.

    As for these books, I suspect that if you looked at the title pages you'd find penciled-in transliterations by two different staff members, which the bindery then obeyed literally to the letter. I've seen botches worse than this as the result, especially when the transliteration was scribbled and/or in a non-Anglosphere handwriting; surely many of us have.

  18. Rodger C said,

    April 4, 2017 @ 7:35 am

    Codicils to the above:

    1. *staff members or helpful faculty members

    2. Between the title-page scribble and the book spine comes a slip made out for the bindery, which is sometimes still in the book and was obviously typed by a hapless workstudy.

  19. Victor Mair said,

    April 4, 2017 @ 7:51 am

    For a summary of orthographic rules for writing in Pinyin and access to them, see here:

  20. WSM said,

    April 4, 2017 @ 9:23 am

    "Write words as complete words, not as bro ken syl la bles."

    How would you apply this guidance to Bokane's example of 西游记 ?

  21. liuyao said,

    April 4, 2017 @ 9:23 am

    I found it awkward to call it pseudonym, but following links in wikipedia, hao is indeed sometimes rendered pseudonym in English (along with style name, literary name, sobriquet, etc). I thought it gave the wrong impression, when practically everyone literate in the Ming and Qing dynasties had a hao (often many).

    From what I could find online, it is dubious that Li Zhi was the author of this criticism. Some claimed that it was by a certain 葉晝, who had also written other criticisms.

  22. WSM said,

    April 4, 2017 @ 9:36 am

    Also tend to agree about "pseudonym" which would be 笔名 or 化名, not 号 (for which something like "sobriquet"). "Li Zhuowu" is also used for his better known "Book to be Burned" and "Book to be Hidden", both of which are certainly written by him.

  23. J.W. Brewer said,

    April 4, 2017 @ 9:55 am

    I should note that by "just asking for trouble" I didn't mean to imply that no institution should ever switch from an older romanization system to a newer one, only that the decision should not be made without fully weighing the inevitable costs against the anticipated benefits and without fully considering what will be done going forward to deal with a perhaps quite substantial volume of legacy material created/indexed/etc under the old system. It's not just that there will be a confusing transitional period for a few years while people get their bearings in the new system, it's that the transition will never really end, at least not in a serious research collection where you assume that the volumes you acquired 20 or 100 or 200 years ago are going to be kept around essentially forever and occasionally actually be consulted by scholars of future generations, such that both those scholars and some library staff will need to be conversant in all the archeological strata of prior romanization systems.

  24. J.W. Brewer said,

    April 4, 2017 @ 10:02 am

    Presumably, as was noted above, there are parallel complications created by shifts in English orthography. One way to deal with this is by indexing multiple ways. If your Indological collection has old volumes with titles like "A Grammar of the Hindoostanee Language," you index them by title with the "Hindoostanee" spelling and by subject under "Hindustani (or perhaps just Hindi), grammar of." Similarly, a volume that claims on its cover to be an English translation of the Tao Te Ching should be indexed thusly but can also be indexed separately under "Daodejing, translations of." (Or more probably Dao De Jing, if the ALA-LC conventions are followed?)

  25. JK said,

    April 4, 2017 @ 11:09 am

    Sometimes there also may be financial incentives for certain romanizations…

    A lot of times for academic translations the client will want both the pinyin version of a book or article title and English translation, and translators paid by the word may choose to put more spaces in their pinyin.

  26. Brendan said,

    April 4, 2017 @ 3:41 pm

    @JK – Having spent years laboring in the translation mines, I'm a bit dubious about that — in my experience one is paid by the character, rather than by the English word, so Pin Yin isn't any more of a road to riches than Pinyin is.

    @liuyao, WSM – I'd prefer "sobriquet" for 號 too. And yes; from what little I understand of it, the ascription to Li Zhi/Li Zhuowu is grounded more in tradition than in solid evidence. (I guess I might have added a fifth annoyance in the original post: flagrant false advertising.) On the other hand, the ascription is of such long standing that retitling the book "Ye Zhou's/Yeh Chou's Commentary" would probably just add to the confusion.

    @flow – Being able to search the catalogue using Chinese characters really is a life-saver in cases like this.

  27. Jean-Michel said,

    April 4, 2017 @ 4:35 pm

    @Brendan: …annoyances (3) and (4) aren't at all the library's fault, and the LoC rules for Pinyin do have a rationale behind them, even if that rationale basically boils down to "It would be too hard to do this the right way."

    Thing is, the LOC has detailed rules on word-based spacing for Japanese (which employs no such spacing in its native orthography) and Korean (which does used word-based spacing, at least in contemporary writings, but it's often treated as optional in the case of book titles and similar things, and the rules for spacing in romanized Korean differ in some major respects from those used in the native orthography). So I don't really see an argument for treating Chinese differently.

  28. John Swindle said,

    April 4, 2017 @ 7:31 pm

    Are the rules for word spacing in romanized Japanese and romanized Korean followed in Japan and Korea? If so, would that be a difference between the Japanese and Korean cases, on the one hand, and the Chinese case on the other? Not that it would counterbalance the advantages of standard Hanyu Pinyin word spacing.

RSS feed for comments on this post