angling through dictionaries

The most recent rerelease from Sino-Platonic Papers is Tiao-Fish through Chinese Dictionaries (4.3 MB PDF), by Michael Carr.

The tiáo < d’ieu < *d’iôg fish, a classical Chinese happiness metaphor, has been contradictorily identified as a chub, culter, dace, eel, goby, hairtail, hemiculter, loach, mullet, paddlefish, and pike. This paper illustrates the history of Chinese lexicography by comparing tiáo definitions from thirty-five Chinese monolingual dictionaries with tiáo translation equivalents from sixteen Japanese and seventeen Western language bilingual ones.

As Carr explains, “The tiáo fish provides a historical microcosm of Chinese lexicography because every principal dictionary defines it, and because *DZIOG‘s multifarious pronunciations and writings illustrate some unique linguistic problems in Chinese dictionaries.”

This was first published in September 1993 as issue no. 40 of Sino-Platonic Papers.

some tiao fish

Y.R. Chao’s responses to arguments against romanization

Y.R. Chao. Also, FWIW, Wikipedia took this image from Pinyin.Info, not the other way around.Pinyin.Info has a new reading: Responses to objections to romanization, written by the brilliant linguist Y.R. Chao in 1916, when he was a young man of 24.

It’s an unfortunate irony that another writing associated with Chao, the famous “stone lions” (a.k.a. shi, shi, shi) piece, is often mistakenly cited as evidence that the author opposed romanization. In fact, Chao favored using romanization for Mandarin, as his essay reveals.

It’s written in the form of 16 “objections,” each followed by Chao’s reply. For example:

Obj. 8 Alphabetized Chinese loses its etymology.

Rep. 8 This argument is like that often urged against simplified English spelling and is to be met similarly. In actual usage, how much attention do we give to etymology in words like ?, ?, ?, ?, ?, through, draught, etiquette, row, disaster? Of how many of these very common words do you know the original meaning? It is not to be denied, of course, that it is useful to know the etymology of words by looking them up, and our future dictionaries of alphabetized polysyllabic words should no doubt give their derivations.

The etymology of disaster (which is pretty cool) is certainly easy enough for an educated person to guess, if you stop to think about it. But I must admit I never had.

I have added notes following the text.

Writing Taiwanese: 1999 study

This seems as good an announcement as any to end my hiatus from posting. Sino-Platonic Papers has just rereleased a popular issue of likely interest to many readers of Pinyin News: Writing Taiwanese: The Development of Modern Written Taiwanese (2.2 MB PDF), by Alvin Lin.

The table of contents gives a pretty good picture of what’s inside:

The Status Quo: Characters and Taiwanese writing

  1. The Roots of Writing in Taiwanese: Wenyan, baihua and academic Taiwanese
  2. The Missing 15 Percent: Developing a written vernacular
  3. One Attempt at Finding the Missing 15 Percent: Yang Qingchu’s Mandarin-Taiwanese Dictionary

Writing Romanized Taiwanese

  1. The Roots of Romanized Taiwanese: Church Romanization
  2. Church Romanization Today: The Taigu listserver
  3. An Indigenous System: Liim Keahioong and Modern Literal Taiwanese

Linguistic and Social Considerations

  1. Some Linguistic Classifications
  2. Dealing with Homonyms: Morphophonemic spelling
  3. Tones in Taiwanese: Surface vs. Lexical tones
  4. Representing Dialects: Picking a standard written form or representing all dialects
  5. Summary of Linguistic Concerns: Deciding the degree of coding
  6. Writing, Reading, Printing, Computing, Indexing and other Practical Concerns
  7. Social Concerns: Tradition and Political Meaning
  8. Conclusion: Future Orthography Policy on Taiwan


  • Email Survey
  • Pronunciation guide to church romanization

List of Tables and Illustrations:

  • Table 1: Suggested Characters for Taiwanese Morphemes from Three Sources
  • Figure 1: Yang Qingchu’s Taiwanese-Mandarin Dictionary
  • Figure 2: Church romanization
  • Figure 3: Modern Literal Taiwanese
  • Figure 4: Sample e-mail from Taigu listserver

This was first published in 1999 as issue number 89 of Sino-Platonic Papers.

writing four-syllable idioms in Pinyin

cover of Chinese Romanization: Pronunciation and OrthographyThe latest excerpt from Yin Binyong’s book on Pinyin orthography covers how to write four-syllable idioms in Hanyu Pinyin (929 KB PDF). Here’s a key passage:

almost all four-character idioms can be broken in two halves, called y?jié ?? (language segments), on the basis of phonetic structure. The simple expedient of connecting the two y?jié with a hyphen then provides idioms with their own distinctive written form, and assures ease of writing and reading. It is also a simple rule for students of HP to master.

But not all four-syllable idioms follow this rule, as the reading shows.

This is a worthwhile reading for Mandarin learners, even if you’re not particularly interested in Pinyin. There are many examples of idioms here, all given in Hanzi, Pinyin, and English.

Penghu street signs

My wife and I recently spent a weekend in Penghu, a beautiful, stark archipelago between the main island of Taiwan and China.

Since Penghu is under KMT rule, I expected to find street signs in Magong, the capital, in some old system (e.g., MPS2 or perhaps bastardized Wade-Giles) or perhaps even Hanyu Pinyin. (Highway signs, however, are a different matter. They’re put up by the central government, which means that relatively recent ones are in Tongyong Pinyin, regardless of which party might control the area.)

This first street sign, however, is unmistakably in Tongyong Pinyin, giving “Wunsyue” (for what in Hanyu Pinyin would be “Wenxue”).
street sign reading 'Wunsyue Rd.' (Wenxue Road)

But I looked around some more and saw signs in Hanyu Pinyin, such as “Huimin” for what in Tongyong would be “Hueimin” and “Hui[']an” for what in Tongyong would be “Huei[-]an.”
street sign reading 'Huimin Road'

street sign reading 'Huian first Road'

So were there some signs in Hanyu Pinyin after all? Apparently only coincidentally. The previous two hui signs were probably just a mistake, the result of Taiwan’s standard, sloppy chabuduo jiu keyi approach to signage. Here’s a sign on the same street as above; but in this case “?” is romanized huei and not hui. (And “first” is missing, from both the Hanzi and romanization.)
street sign reading 'Hueian Rd.'

Most signs were in Tongyong, such as these. (Note that Penghu, too, has a Hot Milk Road.)
street signs: 'Jhongjheng Road' (Zhongzheng Road) and 'Renai Road' (Ren'ai Road)

So, Tongyong after all. Well, at least they don’t have InTerCaPiTaLiZaTion … or do they?
street signs reading 'JhongShan Rd.' -- note InTerCaPiTaLiZaTion -- (Zhongshan Road) and 'Jhongjheng Rd.' (Zhongzheng Road) -- no intercapping

Fortunately, that sign was a one-off. I didn’t spot InTerCaPiTaLiZaTion elsewhere. Here’s another sign from the same road:
street sign reading 'Jhongshan Rd.' (Zhongshan Road)

So, in short, Penghu’s street signs are in Tongyong Pinyin — but with plenty of mistakes and inconsistencies (e.g., missing apostrophes/hyphens, “first” rather than “1st”, and both “Road” and “Rd.”). It’s especially ridiculous that the KMT-administered Penghu bothered with Tongyong, especially since it was free to adopt Hanyu Pinyin. Now it’s going to have to change its signs over to Hanyu Pinyin. But some of the signs would need to be updated anyway, since many already show signs of age, with letters missing. (My guess is that Penghu put up such low-quality signs that in the annual windy season some of the letters just get blown away.)

Here’s a sign in little danger of having its writing blow away any time soon. This is what a much older Magong street sign looks like. Note that it must be read from right to left: ??? (Fuguo Road — “Recover Atlantis the Lost Country Road”).
old concrete street sign reading, right to left, '???' (Fuguo Road)

Finally, here’s something that isn’t a street sign at all. But it is nonetheless a sign of historic importance, since it’s a stela that commemorates the Ming Chinese official Shen Yourong telling the red-haired barbarians (i.e., Westerners — in this case, the Dutch) to get the hell out of Penghu. (The Dutch were told they could instead go to Taiwan, since back then China didn’t care about it in the least.) The composite photo shows both the 400-year-old stone original and a modern reproduction in wood.

photos of the original stone stela and a modern reproduction in wood

The text reads “Sh?n Y?uróng yù tuì hóngmáo f?n[zi] Wéimálàng d?ng” (??????????????): “Shen Yourong orders the red-haired foreigners under [Dutch commander] Wybrand van Warwijck to withdraw.”

Sino-Tibetan, Indo-European, and the word for ‘wheel’

The latest rerelease from Sino-Platonic Papers is “Sino-Tibetan *kolo ‘Wheel‘” (800 KB PDF), by Robert S. Bauer. Those of you who like historical linguistics should be sure to read this one.


That the horse-drawn chariot appeared suddenly in China in the Shang Dynasty (ca. 1500-1066 BC) has led some Western scholars to believe that it was not independently invented by the Chinese but was introduced there by Western invaders. This paper is based on the premise that there is a connection between the transmission of the horse-drawn chariot from the West into China and the origin of some words meaning “wheel” and “wheeled-vehicle” in Sino-Tibetan languages. In particular, the paper proposes that words for “wheel” in some northern Chinese dialects and Bodic (Tibetan) languages are ultimately derived from an Indo-European source. On the basis of the comparison of words for “wheel” from various Sinitic and Bodic languages, the author has reconstructed the Proto-Sino-Tibetan root *kolo “wheel” which is itself an Indo-European contact loanword.

This was first published in August 1994 as issue no. 47 of Sino-Platonic Papers.

The Art of War: a companion volume

Sonshi, the largest website dedicated to Sun Zi’s (Sun Tzu’s) Art of War, recently selected Victor H. Mair’s new translation as “the #1 Art of War edition.”

In announcing its judgment, the site stated, “how rare a book that courageously stands up to centuries of established thought, proceeds to knock it down with sound logic and proof, and succeeds in convincing even the Old Guard to change their views.”

Professor Mair has just published a free, book-length companion to his translation: Soldierly Methods: Vade Mecum for an Iconoclastic Translation of Sun Zi bingfa, with a complete transcription and word-for-word glosses of the Manchu translation by H. T. Toh (1 MB PDF).

Yes, all that and Manchu too. The appendixes might well supply the longest text in romanized Manchu available online — not to mention the longest one with English translation. (Perhaps someone from Echoes of Manchu can comment.)

And I’d like to note the introduction to the transcription offers a cool word I hadn’t come across before: Mandjurist, which is German for “Manchu philologist.”

Here’s the table of contents:

  • Preface
  • Principles of Translation
  • Guide to Pronunciation
  • Key Terms
  • Abbreviations
  • Discussion
    • The Book and Its Title
    • Authorship
    • Historical Background
    • Dating
    • Stylistics and Statistics
    • Techniques and Technology
    • Taoistic Aspects
    • Eurasian Parallels
    • On the World Stage
    • Notes
  • Appendix I: The Pseudo-Biography of Sun Wu
  • Appendix II: Further Notes on Selected Key Terms
  • Appendix III: Transcription of the Manchu Translation of the Sun Zi with Word-for-Word English Glosses by Hoong Teik Toh
  • Appendix IV: Transcription of the Manchu Translation of the Sun Zi by Hoong Teik Toh
  • Bibliography

This is issue no. 178 of Sino-Platonic Papers.

US post-secondary enrollments in foreign languages and the position of Mandarin

graphs showing the enrollments of Japanese and Mandarin over time, with Italian thrown in by way of comparisonFrom the way the U.S. media talk about the boom in Mandarin classes, it’s easy to get the impression that Mandarin is about to become the most studied language in the United States. So I offer the following overdue reality check.

The data come from the results of a large survey of foreign-language enrollments in U.S. post-secondary schools. The survey was conducted by the Modern Language Association. I started work on this post when the results were released in November 2007; but, well, I got distracted.

This post has lots of tables and figures, so for those who don’t want to scan through everything I offer some basic points up front.

  • Spanish has more enrollments than all other foreign languages put together.
  • By far the biggest enrollment boom since 1990 has not been for Mandarin but for American Sign Language.
  • The boom in enrollments in Arabic also surpasses that for Mandarin.
  • Mandarin is indeed growing in popularity — but in recent years only at the undergraduate level.
  • Japanese continues to be more popular than Mandarin, though by an ever-smaller margin.
  • Mandarin is the seventh most studied foreign language in U.S. post-secondary schools, behind Spanish (which leads Mandarin by a ratio of 16:1), French, German, American Sign Language, Italian, and Japanese.
  • Relatively speaking, enrollments in foreign languages are much lower than they were 30 years ago.

A few summary remarks of my own:

  • I don’t expect the high growth rates for Mandarin to continue for many more years unless the programs are dumbed down (in which case they wouldn’t count for much) or Pinyin gains a much more prominent role in Mandarin pedagogy (and not just at the introductory level). The difficulties of Chinese characters will help keep numbers down, as will the eventual realization that learning Mandarin isn’t an easy ticket to riches (or even a ticket to riches at all).
  • Japanese received a big boost in the 1980s, when the media cranked out story after story about the power of Japan’s rising economy and the need to learn the language. Yet Japanese didn’t become the next big world language. I predict a similar path for Mandarin.
  • A high percentage of those taking Mandarin classes in U.S. high schools are students who are both ethnically Chinese and already familiar with the language. The MLA didn’t provide figures on that for post-secondary students. But I would be surprised if such “heritage” students don’t represent a higher percentage of those in Mandarin language courses than heritage students in most other language classes.

OK, now on to some details.

Look below at the growth for American Sign Language since 1990. If Mandarin had had that sort of growth (4,820 percent!) the pundits would no doubt be telling us that the Chinese had already taken over the planet and were going to rule the entire galaxy within the next decade. (And don’t get me started about the supposed Mandarin in Serenity/Firefly.) But American Sign Language just doesn’t seem to get the same sort of respect, despite the fact that it still has more than 50 percent more enrollments than Mandarin. Arabic, which has also had a much faster growth rate than that of Mandarin, hasn’t received the same level of hype either.

Growth in Enrollments: in declining order of growth from 1990 to 2006

Enrollments 1990 2006 % Growth 2002-06 % Growth 1990-2006
American Sign Language 1,602 78,829 29.7 4820.7
Arabic 3,475 23,974 126.5 589.9
Korean 2,286 7,145 37.1 212.6
Mandarin 19,490 51,582 51.0 164.7
Hebrew 12,995 23,752 4.2 82.8
Portuguese 6,211 10,267 22.4 65.3
Italian 49,699 78,368 22.6 57.7
Spanish 533,944 822,985 10.3 54.1
Japanese 45,717 66,605 27.5 45.7
French 272,472 206,426 2.2 -24.2
German 133,348 94,264 3.5 -29.3
Russian 44,626 24,845 3.9 -44.3
Total 1,125,865 1,489,042 12.7 32.3

Change in enrollments over time: in declining order of total enrollment for 2006

Change between Surveys 1995-98 1998-2002 2002-06
Spanish 8.3% 13.7% 10.3%
French -3.1% 1.5% 2.2%
German -7.5% 2.3% 3.5%
American Sign Language 165.3% 432.2% 29.7%
Italian 12.6% 29.6% 22.6%
Japanese -3.5% 21.1% 27.5%
Mandarin 7.5% 20.0% 51.0%
Russian -3.8% 0.5% 3.9%
Arabic 23.9% 92.3% 126.5%
Hebrew * 20.6% 44.0% 4.2%
Portuguese 6.0% 21.1% 22.4%
Korean 34.0% 16.3% 37.1%
Total 5.0% 16.6% 12.7%

* Modern and Biblical Hebrew combined

Below: Russian may not have the top number of enrollments, but it certainly has some motivated students, given the high numbers of them in advanced courses.

Enrollments in Introductory-Level Courses vs. Enrollments in Advanced-Level Courses

Intro Enr. Advanced Enr. Total Enrollment Ratio of Intro Enr. to Advanced Enr.
Russian 17,527 6,569 24,096 2.67:1
Portuguese 7,387 2,422 9,809 3.05:1
German 72,434 18,758 91,192 3.86:1
French 160,736 40,927 201,663 3.93:1
Korean 5,511 1,397 6,908 3.94:1
Greek, Ancient 13,250 3,176 16,426 4.17:1
Mandarin 41,193 9,262 50,455 4.45:1
Spanish 669,432 142,602 812,034 4.69:1
Japanese 55,161 10,585 65,746 5.21:1
Latin 26,787 4,383 31,170 6.11:1
Hebrew, Modern 7,665 1,250 8,915 6.13:1
Arabic 20,571 2,463 23,034 8.35:1
Italian 69,757 7,593 77,350 9.19:1
Hebrew, Biblical 7,854 705 8,559 11.14:1
American Sign Language 72,694 5,249 77,943 13.85:1
Other languages 27,836 3,478 31,314 8.00:1
Total 1,275,795 260,819 1,536,614 4.89:1

One thing I find particularly troubling is that the number of graduate students studying Mandarin has fallen. (Please click on the link in the previous sentence, since the relevant table is too wide to fit on this page.) The much-ballyhooed but also much-deserved increase in students studying Mandarin has all been at the undergraduate level. Given that the grad enrollment as a percentage of total enrollment for Mandarin is about the same as that for French (2.63 percent and 2.73 percent, respectively) it might appear that Mandarin has simply reached a “normal” ratio in this regard. But native speakers of English generally need much more time to master Mandarin than to master French. Simply put, four years, say, of post-secondary study of French provides students with a much greater level of fluency than four years of post-secondary study of Mandarin.

Also, there is a great deal more work that needs to be done in terms of translations from Mandarin. I do not at all mean to belittle the work being done in French — or in any other language. In fact it pains me that the MLA’s list of languages being studied included neither Old French nor Provençal, both of which I have studied and love dearly. I just mean that Mandarin has historically been underrepresented in U.S. universities given the number of speakers it has and its body of texts that have not yet been translated into English. U.S. universities need to be producing many more qualified grad students who can handle this specialized work. And right now, unfortunately, that’s not happening.

Post-Secondary Enrollments in Select Sino-Tibetan Languages and Classical Japanese: 2002, 2006

Two-Year Colleges Undergrad Programs Grad Programs Total
Language 2002 2006 2002 2006 2002 2006 2002 2006
Cantonese 47 96 128 82 5 0 180 178
Literary Sinitic 0 0 56 101 18 12 74 113
Japanese, Classical 0 0 8 23 11 7 19 30
Taiwanese 0 0 34 21 13 0 47 21
Tibetan 0 0 43 56 35 64 78 120
Tibetan, Classical 0 0 8 11 20 33 28 44

The figures in the table above are probably too low. Literary Sinitic (“classical Chinese”) is probably especially underrepresented because often too little differentiation is given between it and modern standard Mandarin. But at least the numbers can provide minimum figures.

Enrollments in Introductory Classes: 2-Year Schools vs. 4-Year Schools

Language Ratio of Intro Enr. in 2-Year Schools to Intro Enr. in 4-Year Schools
Greek, Ancient 0.00:1
Hebrew, Biblical 0.01:1
Latin 0.04:1
Hebrew, Modern 0.07:1
Portuguese 0.11:1
Russian 0.15:1
German 0.20:1
Italian 0.23:1
French 0.24:1
Arabic 0.26:1
Mandarin 0.26:1
Korean 0.28:1
Japanese 0.39:1
Spanish 0.49:1
American Sign Language 1.47:1
Other languages 0.24:1

American Sign Language sticks out here as the only language that more people take at the introductory level at junior colleges than at universities. Roughly twice as many people take introductory Spanish in universities as at junior colleges. Introductory Japanese classes are surprisingly popular at the two-year college level, well above the level for introductory Mandarin, though Mandarin is not unpopular itself.

Course Enrollments in Some Asian and Pacific Languages

Language 1998 2002 2006 % Change 2002–06
Hindi/Urdu 1314 2009 2683 33.55
Vietnamese 899 2236 2485 11.14
Tagalog/Filipino 794 1142 1569 37.39
Sanskrit 363 487 607 24.64
Hmong 15 283 402 42.05
Thai 272 330 307 -6.97
Indonesian 223 225 301 33.78
Samoan 207 201 280 39.30
Cantonese 39 180 178 -1.11
Tibetan 80 78 120 53.85
Literary Sinitic 32 74 113 52.70
Pashto 14 103 635.71
Punjabi 32 99 103 4.04
Total 4270 7358 9251 25.73

Although more U.S. postsecondary students are studying languages other than English than ever before, that’s unfortunately not because U.S. students as a whole have finally embraced the study of languages. Rather, there are simply more students now. Relatively speaking, enrollments in foreign languages are much lower than they were 30 years ago.

graph showing that present US postsecondary enrollment in foreign languages is relatively much lower than it was in in the 1960s

If “ancient” foreign languages such as Latin and Ancient Greek were included in the graph, the imbalance between the 1960s and the present in foreign-language enrollments would be even greater.

source: Enrollments in Languages Other Than English in United States Institutions of Higher Education, Fall 2006 (PDF), MLA, November 13, 2007