Essays on Writing and Language in Honor - Sino-Platonic Papers

SINO-PLATONIC PAPERS 

Number 27 August 3 1, 1991 $35.00 

Schriftfestschrift: 

<strong>Essays</strong> on Writing and Language 

in Honor of John DeFrancis 

on His Eightieth Birthday 

edited by 

Victor H. Mair 

Order froni: 

Department of Oriental Studies 

University of Pennsylvania 

Phdadelphia, PA 19 104-6305 USA

SINO-PLATONIC PAPERS is an occasional series edited by Victor H. Mair. 

The purpose of the series is to make available to specialists and the interested 

public the results of research that, because of its unconventional or controversial 

nature, might otherwise go unpublished. The editor actively encourages younger, 

not yet well established, scholars and independent authors to submit manuscripts 

for consideration. Contributions in any of the major scholarly languages of the 

world, including Romanized Modern Standard Mandarin (MSM) and Japanese, are 

acceptable. In special circumstances, papers written in one of the Sinitic topolects 

(fangyan) may be considered for publication. 

Although the chief focus of Sino-Platonic Papers is on the intercultural relations of 

China with other peoples, challenging and creative studies on a wide variety of 

philological subjects will be entertained. This series is not the place for safe, 

sober, and stodgy presentations. Sino-Platonic Papers prefers lively work that, 

while taking reasonable risks to advance the field, capitalizes on brilliant new 

insights into the development of civilization. 

The only style-sheet we honor is that of consistency. Where possible, we prefer 

the usages of the Journal of Asian Studies. Sinographs (hanzi, also called 

tetragraphs [fangkuaizi]) and other unusual symbols should be kept to an absolute 

minimum. Sino-Platonic Papers emphasizes substance over form. 

Submissions are regularly sent out to be refereed and extensive editorial 

suggestions for revision may be offered. Manuscripts should be double-spaced 

with wide margins and submitted in duplicate. A set of "Instructions for Authors" 

may be obtained by contacting the editor. 

Ideally, the final draft should be a neat, clear camera-ready copy with high blackand-white 

contrast. Contributors who prepare acceptable camera-ready copy will 

be provided with 25 free copies of the printed work. All others will receive 5 

copies. 

Sino-Platonic Papers is licensed under the Creative Commons Attribution- 

NonCommercial-NoDerivs 2.5 License. To view a copy of this license, visit 

http://creativecommons.org/licenses/by-nc-nd/2.5/ or send a letter to Creative 

Commons, 543 Howard Street, 5th Floor, San Francisco, California, 94105, USA. 

Please note: When the editor goes on an expedition or research trip, all operations 

(including filling orders) may temporarily cease for up to two or three months at a 

time. In such circumstances, those who wish to purchase various issues of SPP 

are requested to wait patiently until he returns. If issues are urgently needed while 

the editor is away, they may be requested through Interlibrary Loan. 

N.B.: Beginning with issue no. 171, Sino-Platonic Papers will be published 

electronically on the Web. Issues from no. 1 to no. 170, however, will continue to 

be sold as paper copies until our stock runs out, after which they too will be made 

available on the Web. 

_______________________________________________

Table of Contents 

Tabula Gratulatoria 

Introduction 


Publications of John DeFrancis 

Hanzi Bu Tebie Biaoyi 

Zhang Liqing 

Typology of Writing Systems 

Zhou Youguang 

Dui Hand de Jiizhong Wujie 

Yin Binyong 

The Information Society and Terminology 

Liu Yongquan 

A Bilingual Mosaic 

Einar Haugen 

The Polysemy of the Term Kokugo 

S. Robert Ramsey 

Memorizing Kanji: Lessons from a Pro 

J. Marshall Unger 

Why Chinese Is So Damn Hard 

David Moser 

Ethnolinguistic Notes on the Dungan 

Lisa E Husmann and William S-Y. Wang 

Korean Views on Writing Reform 

Wm. C. Hannas 

Language Policies and Linguistic Divergence in the Two Koreas 

Ho-min Sohn 

- Okinawan Writing Systems, Past, Present, and Future 

Leon A. Serafim 

Proposal of a Comparative Study of Language Policies and Their 

Implementation in Singapore, Taiwan, and China (PRC) 

Robert L. Cheng 

The Topical Function of Preverbal Locatives and Temporals in Chinese 

Feng-fu Tsao 

Yes-No Questions in Taipei and Peking Mandarin 

Robert M. Sanders 

Patronizing Uses of the Particle "ma": 

Bureaucratic Chinese Bids for Dominance in Personal Interactions 

Beverly Hong Fincher 

Gender and Sexism in Chinese Language and Literature 

Angela Jung-Palandri 

A zhezi Anagram Poem of the Song Dynasty 

John Marney 

Some Remarks on Differing Correspondences in Old Chinese 

Assumed to Represent Different Chinese Dialects 

Nicholas C. Bodman 

Can Taiwanese Recognize Simplified Characters? 

John S. Rohsenow 

Simplified Characters and Their (Un)relatedness 

Chauncey C. Chu 

... 

lll 

v 

vii 

I

The Teaching of Culture and the Culture of Teaching: 

Problems, Challenges, and Opportunities in Language Instruction 

Eugene Eoyang 

The Culture Component of Language Teaching 

K y oko Hijirida 

Thinking About Prof. John DeFrancis 

Apollo Wu 

Wo suo Renshi de De Xiansheng 

Chih-yu Ho 

Two Poems for Professor John DeFrancis 

Richard F. S. Yang . 

Announcement 

Stephen Fleming

Olov B d Anderson 

Ivan Aymat 

Dayle Barnes 

Yung-0 Biq 

David Bradley 

Susan Brennecke 


Madeline Chu 

Alvin P. Cohen 

East Asian Studies, Washington 

University in St. Louis 

Sen Egerod 

Eugene Eoyang 

Mary S. Erbaugh 

Esther ~ iueroa g 

Joshua A. Fishman 

Stephen Fleming 

Feng-fu Tsao 

Martin Girnm 

Christoph Harbsmeier 

C. Y. Ho 

Henry M. Hoenigswald & 

Gabriele L. Hoenig swald 

Vivian Ling Hsu 

Robert L hick 

Betty Jacobsen 

. Frank A. Kierman, Jr. 

Y. C. Li 

J. C. P. Liang 

Louis S. Mangione 

Richard B. Mather 

John Marney 

Cynthia Y. Ning 

Jerry Norman 

Alain Peyraube 

Angela Jung Palandri 

Rulan Chao Pian & 

Theodore H. Pian 

Gilbert W. Roy 

Wayne Schlepp 

C. P. Sobelrnan 

Clara S. Y. Sun 

R. McMillan Thompson & 

Sandra A. Thompson 

H&w-% Wang: 

For the past ten years John's books have 

been an indrspensable part of my language 

program. I have been able to design a 

computer program based on the ciear and 

systematic grammar content, Chinese 

characters, and supplementary matmals 

of John's Beginning Chinese Textbook. I 

am always on the look out for more 

supplementary materials from John. 

John, God bless you for the invaluable 

work you are doing! 

Jing Wang 

TABULA GRATULATORIA

With special thanh to Ivan Aymat, Louis S. Mangione, Feng-fi Tsao, Chauncey Chu, Hsiao-Min Wang, 

Henry & Gubriele Hoenigswald, Rulan Chao & Theodore Pian, Eugene Eqvang, John M aw, Gilben 

Roy, Frank A. Kiennan, Jr., Easr Asian Studies (Warhington University in Saint Louis), C. Y. Ho, C. P. 

Sobelman, Cynthia Y. Ning, Alain Peyraube, Sundra A & R McMillan ~E?ornpson, OZov Bertil 

Andmson, Vivian Hsu, Yung-0 Big, Angela Jung Palandn', Clara S. K Sun, Susan Brennecke, and Stephen 

Fleming. 

And a big debt of gratitude to John S. Rohsenow.

S~hI7frfls~hc$: <strong>Essays</strong> in Honor of Johrl DeFrancis 

Introduction 



The honoree of this volume may be chronologically eighty years old, but he is 

decidedly a young man at heart A number of associates in China and in the United States 

are currently involved in a strenuous effort to compile the first general-purpose, Chinese- 

English dictionary arranged strictly according to a single alphabetical sequence of whole 

words. None of us is more energetic or enthusiastic than John DeFrancis. When our ABC 

dictionary becomes a reality in a couple of years, no one will have been more responsible 

for its conceptualization and execution than John. And if any proof is needed for that 

statement, I have a stack of memos from him in my office to show how amazing are his 

attention for detail and ability to think through problems clearly. 

But it is not just the ABC dictionary that reveals John to be a man of enormous 

vision and productivity. These characteristics have been evident in his life and work for 

over half a century. It is hard to imagine that John published his fmt book over forty years 

ago and just as difficult to conceive of a man his age continuing to write equally eloquent 

and important works up to the present moment. 

1 do not need to name John's numerous books and articles one by one -- the partial 

bibliography of his works that follows this introduction will suffice to give a sense of their 

wide range and great significance. What has always impressed me about all of John's 

writing, since long before I ever had the good fortune to meet him personally, is his keen 

perceptivity. John has an almost incredible talent for cutting through obfuscation and 

seeing what the crux of any given matter may be. This penetrating insight enables John to 

define issues clearly, to analyze them incisively, and to present his solutions lucidly. 

Furthermore, John's well-organized mind permits him to achieve feats of practical 

scholarship that can only be dreamed of by the common mortal. On top of all these other 

stellar qualities, John is possessed of nearly superhuman industriousness and efficiency. 

Let me put it bluntly: if the necessary facts are out there somewhere, John will be able to 

dig them up - even from the middle of the Pacific Ocean! -- and put them to good use. 

So John is a superb scholar with many excellent works to his credit. Yet there is 

another ingredient, or pair of ingnxhents, that sets John DeFrancis apart from all the other 

fine scholars whom I have encountered -- that is his passion and his compassion. John 

cares. Whatever John does is because he wants to help improve things. His classic 

Ndiomlism and Longrcage Reform in chi^ was dedicated to Old Wang. If we turn to p. 

143 of the same book, we can find out who Old Wang was: 

Known as Old Wang. Age thirty-five. Totally illiterate. Occupation: 

peasant. Lives in a tiny village four and a half miles northeast of Peking. 

Married to the daughter of a peasant from a near-by village. Has three 

- children ranging in age from four to nine. Wife and children likewise 

itliterate. 

People like Old Wang really matter to John. It is to all the Old Wangs of the world that 

John devoted his whole life, and that is why his achievements have such profound 

meaning. 

John's entire being is consumed with the noble impulse to make existence easier for 

everyone. Although I do not wish to breach confidentiality, I have witnessed John's 

extraordinary generosity on numerous occasions. Here is a man who seems to find deep 

satisfaction in joyful self-sacrifice for the greater good. It is an inspiration just to how 

him.

Sino-Plutonic Papers, 27 (August 3 1, 199 1) 

Because John &Francis is who he is and so many people have been touched by 

him, editing this volume was no chore at all. Such an outpouring of cooperation is rare in 

academia. I only learned in December of 1990 that John's eightieth birthday would be on 

August 31, 1991. By late January, letters went out to friends and colleagues around the 

world inviting them to participate in this Schrififestschn~. When I sent them out, I was 

dubious that anyone would be able to respond positively on such short notice. I was 

wrong, of course, and the rich collection of essays which follows reveals unmistakably 

what a high regard the sinological and linguistic communities have for John. In fact, so 

many colleagues offered to submit papers that I regretfully had to decline several fine, 

longer pieces for reasons of space limitations. 

The Tabula Grmlarorin is a good indication of the high esteem in which John is 

held in spite of the fact that my circular concerning this volume was both late and 

restricted in size, almost everyone to whom I wrote asked to have their name listed. And 

many contributed gifts of varying size toward the cost of publication. I wish to take 

advantage of this opportunity to thank publicly all those who have helped to bring this 

volume to reality. I hope that no one's m e has been inadvertently omitted from the list 

and regret that I was unable to make the existence of this project known to a wider circle 

beforehand, for I am sure that many more people would have gladly signed the list had they 

known about it. 

Finally, I wish to apologize both to the readers of Schrij?$estschrift and to its 

authors for the less than perfect appearance of the volume. John deserves only the very 

best, but the short amount of time available for compilation and editing would not permit a 

more deliberate approach. At least we can rest secure in the knowledge that the essays in 

this book are of suitable quality as celebratory offerings for someone of John's stature. 

Above all, what matters most is the thought behind this book and the individual essays that 

go to make it up. Consequently, I am delighted on behalf of the authors, on behalf of those 

who signed the Tabula Grrulatoria, and on behalf of everyone else who wishes John well 

on this happy occasion, to dedicate our Schri~stschriifr to him with utmost respect and 

admiration. Congratulations on becoming a veritable octogenarian, John, and may we be 

priviliged and blessed to share many more birthdays with you! 

SCribendi recte sapere.est et principium et fons. 

Horace, Ars Poen'ca

Schriftfestsch~!: <strong>Essays</strong> in Horwr of John DeFrancis 

Publications of John DeFrancis 

(Partial List) 

BOOKS 

Visible Speech: The Diverse Oneness of Writing Systems. Honolulu: University of 

Hawaii Press, 1989. 

The Chinese Language: Fact and Fantasy. Honolulu: University of Hawaii Press, 

1984. 

Colonialism and Language Policy in Vier Nam. The Hague: Mouton, 1977. 

Annotated Quotations from Chairman Mao. New Haven: Yale University Press, 

1975. 

Things Japanese in Hawaii. Hawaii: Unniversity Press of Hawaii, 1973. 

Index Volume. New Haven: Yale University Press, 1970. 

Advanced Chinese Reader. New Haven: Yale University Press, 1968. 

Intermediate Chinese Reader. 2 vols. New Haven: Yale University Press, 1967. 

Beginning Chinese Reader. 2 vols. New Haven: Yale University Press, 1966. 

Advanced Chinese. New Haven: Yale University Press, 1966. 

Intermediate Chinese. New Haven: Yale University Press, 1 964. 

Beginning Chinese. New Haven: Yale University Press, 1963. 

Character Tea for Advanced Chinese. New Haven: Yale University Press, 1966. 

Character Tex for lmermediate Chinese. New Haven: Yale University Press, 1965. 

Character Text for Beginning Chinese. New Haven: Yale University Press, 1964. 

Chinese-English Glossary of the Mathematical Sciences. Providence: American 

Mathematical Society, 1964. 

Chinese Social History by E-tu Zen and John DeFrancis. Washington, D.C.: 

American Council of Learned Societies, 1956. 

Bibliography on Chinese Social History (with E-tu Zen Sun). New Haven: Far 

Eastern Publications, 1952. 

Talks on Chinese History (with Elizabeth Jen Young). 2 vols. New Haven: Far 

Eastern Publications, 1952. 

Report of the Second Round Table Meeting on Linguistics and Language Teaching 

Monograph Series on Languages and Linguistics, No. I. Washington, D.C.: 

Georgetown University Press, 195 1.

' 1950. 

Sino-Plnronic Papers, 27 (August 3 1, 199 1) 

Nationalism and Languuge Reform in China. New Jersey: Princeton University Press, 

Reprinted Octagon. 1972. 

Chinese Agent in Mongolia, translated from the Chinese of Ma Ho-t'ien. Baltimore: 

Johns Hopkins Press, 1949. 

Beginning Chinese (first edition). New Haven: Yale University Press, 1946. 

ARTICLES 

"Chinese Prehistoric Symbols and American Proofreaders' Marks." Journal of 

Chinese Linguistics, 1 9.1 (January, 199 I), 1 16- 12 1. 

"The Why of Pinyin Grapheme Selection." Journal of the Chinese Language Teachers 

Association, 25.3 (October, 1990), 1 - 14. 

"Tone Indication in Chinese (Zhongwend Biaodiaofa)." Xin Tang (New China), 7 

(August, 1986), 80-84. 

"Homographobia (Tongyinci Kongjuzheng)." Xin Tang (New China), 6 (November, 

1985), 2-17. 

"Vietnamese Writing Reform (Yuenand Wenzi Gaige)." Xin Tang (New China), 5 

(April, 1985), 15-25. 

"China's Literary Renaissance: A Reassessment." Bulletin of Concerned Asian 

Scholars, 17.4 (October-December, 1985), 52-63. 

"Zhongguo wenzi gaige zhong de yixie maodun -- yige Meiguorende kanfa (Some 

Contradictions in Chinese Language Reform - - An American Perspective). " Wenzi 

Gaige, 3 (1982), 7-9, 14. 

"Mao Tse-tung and Writing Reform." In Joshua A. Fogel and William T. Rowe, eds., 

fisays in. Honor of Professor C. Mam'n Wilbur on the Occasion of His Retirement. 

Denver: Westview Press, 1979. Pp. 137-54. 

"My Stupidest Student." Cahiers de Linguistigue dtOrientalisme et de Slavique, 10 

(January, 1978), 37-43. 

"Language and Script Reform." Currem Trends in Linguistics, 2. Thomas Sebeok, 

ed. The Hague: Mouton, 1967. Pp. 130-50. Reprinted in Joshua A. Fishman, ed., 

Advances in the Creation and Revision of Writing Systems. The Hague: Mouton, 

1977. Pp. 121-48. 

"Syntactic Permutability in Chinese." Papers in Linguistics in Honor of Leon Dostert. 

The Hague: Mouton, 1967. 4.23-36. 

"Why Johnny Can't Read Chinese." Journal of the Chinese Language Teachers 

Associahon, 1.1 (February, 1 966), 1 -20. 

"The Mathematics Scene in China." The Mathematics Teacher, 55 (1 962), 25 1-55. 

"Mathematical Competitions in China " The American Mathematical Monthly, 67 

(1 %O), 756-62.

Schrififestschrifl: <strong>Essays</strong> in Honor of John DeFrancis 

"National and Minority Problems." Repon on China, The Annals of the American 

Academy of Political and Social Science, 277 (1 95 I), 146-55. 

"A Missionary Contribution to Chinese Nationalism." Jouml of the Nonh .China 

Branch of the Royal Asimic Society, January-February , 1949. 

"The Atlantic Report on the World Today: The Far East." The AtZamic Monthly, 180.5 

(November, 1947), 3-7. 

"Biography of the Marquis of Huaiyin." Hanard Journal of Asiatic Studies, 10 

(1947). 

"Japanese Language Reform, Politics and Phonetics." Far &ern Survey, 16 (1947). 

"The Alphabetization of Chinese." Journal of the American Orienral Society, 63 

(1943). 

"Far Eastern Type for American Scholarly Publications." Notes on Far Eanern Stzdies 

in America, 7 (June, 1940). 

M I S ~ U S 

"Transformational Decomposition of Chinese, with English Parallels." 

Transformations and Discourse Analysis Project, 56 (University of Pennsylvania, 

1 964). 

Chapter entitled "Nationalities and their Problems." In Sinkiang, by Owen Lattimore. 

Boston: Little, Brown, 1950. 

Chinese biographies in The New Cenrur-y Cyclopedia of Names. New York: 

Appleton-Cen tury-Crofts, 1 954. 

"Asia - Languages." EnqcZopedia Amen'carta. 

Numerous book reviews and papers presented at schohry meetings.

Schriifrfesestschrifr: Eksrzys in Honor ofJohn DeFrancis 

Hanzi Bu T&bie Biaoyi 

Zhang LiqTng 

Swarthmore College 

Dadubshu hui Hanzi de ren renwei Hanzi shi biaoyi wenzi. Jia shi 

shu6 Hanzi gBn biede wenil bu yiyang, bubi yikao fayin huozhB biede yGyh 

tiaojian; yi ge ren zhiyao xuehui le hgn dub Hanzi, kanjian Hanzi xi6 de dbngxi 

jiu zhidao shi shenme yisi. 

Zhe dadubshu ren you kandao liang jian shiqing. Di-yi, Hanzi zai 

ZhBngguo lianxu yong le sanqiZin dud nian, bingqig dao xianzai hai zai yong. Di- 

er, Hanzi zai Dbng-Ya ji ge guojia liuchuan le hgn chang yi duan shijian. Yushi, 

tamen you tuixi3ng ch0 liang ge jielbn. Yi ge shu6 Hanzi chaoyue shijian; 

lingwai yi ge shub Hanzi chiioyue kbngjian. Guibing qilai jiu shi Hanzi bigoyi, 

k8yi chaoyue shi-kbng. Zuihou geng jinyibu, bii Hanyii yg IajinUi, shub Hanzi zui 

shihe HanyG. 

Shangmian de kanfg he jielun "g6n shen di gu", danshi buxing d6u hgn 

pianmian, bh f6he zhenzheng qingkuang. Weishenme ne? Hen jigndan. Renhe 

wenzi dbu biaoyi, y6 dbu neng chaoyue shijian he kbngjian. Hanzi bh tebie 

bisoyi, yg bir tebie chaoyue shijian he kbngjian. shi Xianggang 

chOb5n de yi fenr fantizi zazhi. Zhe fenr z&hi di-3 qT (1 991 nian 2yue, di-108 

ye) y5u yT pian cong waiwen fanyiguolai de wenzhang, jiao "Liljli [t,kw de Jiyi 

~bshen he Youtai Ren de Minzu Jiyi -- Lun ZWngguo Ren he Youtai Ren de 

Minzlj Jiyi>>. Wenzhang jfichfilai yT ge gfilHo de Xibolai ccir , zachor. Zachor 

suiran gBn Hanzi wanquan bbtong, que neng bigoshi hgo )7 ge yiyi: 

Xibolaiwen de zachor, ji "jiyi" zhe ge cir, zai Shengjing zhbng yizai 

chuxian. Ta jiBny6u "ji nian biaozhi"; "ji pin"; "ji lb"; "ji nian"; "mingji" dgng 

du6chong hanyi. Zachor tongshi sheji neixin jingli he waijie shiqing, ...

Sino-Platortic Papers, 27 (August 3 1, 199 1) 

Weishenme Hanzi neng zai Zhdnggu6 lianxir yong le sanqian dub nian, 

bingqig xianzai hai zai yong? Zhe ggn Zhbngguo de zhengzhi sixigng, zhengzhi 

jiegbu, shehui, jingji, 1905 nian yiqian ybng k3o Wenyanwen de jTngshn he shici 

Iai xugnba gu8nyuan dgngd8ng y6u guanxi. Hanzi fgichang shihe Wenyanwen, 

shbliang du6, xingti fansu6 firza, xueqilai yongqilai d6u haofei shijian. Zhe xfidub 

tedign rang zhishi hgn nan f8zhgn chuanbb, que hgn rongyi lbngduan. Wdmen 

kgyi shu6, Hanzi bijiao neng bangzhu peiygng mgnfu Jing Shi Zi Ji, qugsh80 

shijian sik2io de chuanteng xuezhg, yT! neng bangzhh peiyang sixiang danchljn, 

qu8shgo zhishi de guomin. Zhe dui weihu chuAnt6ng wenhua he shehui 

wending hgn y6u gongxian. Jiir xiang yizhidao liu-qishi nian yiqian, Zhbngguo 

nzren yTnwei zhengzhi, jingji, he shbhui yuinyin, b8o le zhishao yiqian niin 

xigojig0 yiyang, Hanzi y8 yinwei zhengzhi, jSngji, he shehui yuanyin, zai 

.) 

Zhbngguo lianxh yong le sanqian dub nian. ZhOngguo nuren bao xiZoji80 shi yi 

ge lishi xianxiang. Hanzi zai ZhOngguo yong le sanqian dub nian yg shi yi ge 

lishi xianxiang. W6men bG neng yinwhi Zhbngguo niren b8o le yiqian du6 nian 

xiaojiao, jiG duinding xiaojislo ybuyue, zui shihe Zhbngguo hen. Tongyang, 

wijmen y6 bC neng yinwei Hanzi zai Zh6ngguo yong le sanqian du6 nian, jilj 

shu6 Hanzi tebie bigoyi, zui shiihe HanyE. Hanzi y6xE gBn ZhBngguo de 

chuantbng he lishi xiangyi w6i ming, hirxiang yingxiang, danshi lishi xianxiang 

he shiwir de bgnzhi shi y6u qgbie de. Yao zhTdao shiwu de bgnzhi, d8i cbng 

shiwir bgnshgn guancha, yao zhidao Hanzi shifdu tebie bigoyi, ye bixir cbng 

Hanzi b8nshCn zhuoygn. 

W6 bu shi xuezh8, bb neng yinjing jirdign. Danshi w6 y6u cong ziji shGnghu6 

IT delai de wir ge zhenshi lizi, d6u biHoming Hanzi bing bir tebi6 bigoyi. 

Di-I ge lizi. W6 lih sui kaishi xuexi Hanzi, chozhbng jiir jiechu Wenyanwen, 

1960 dao 1969 nian ye zai dkue he yanjiusub nian le jib nian Zh6ngguo 

wenxue. W6 changchang kan ZhBngwen sh0 bao zazhi. WT, suiran mei y5u 

xuewen, yuedli Hanzi de jingyan k6yi shu6 sh.1 hgn fengfir de. Danshi yudao mei 

XU~~UO hu0zh8 bb shljxT de Hanzi, bTf~j u $% , 4. % 9 f .gngdgng, 

jiu =enme 

kan yg kanbbdeng. Budan gebie shengpi Hanzi ybu zhe zh6ng qingkuang, 

y6ushihou lian yong shljxi Hanzi xi8 de cir, pianyfi, shenzhiyll jirzi y6 y6u wenti. 

Bin3 ccDT-yi Jiating>> (1 989 nian 5yue, di-195 ye) u 4 $,l.g-kg" lide u 9 u?

Schnfrfestschri!: Essay in Honor of John &Francis 

(,&is,j u kg -@ "? BC zhidao ziinme fenci.); ccYu3njian Z&hi>> (1990 nian 

I I yue, di-92 ye) li de Fpji:&%%T 31 "; cctrshiyi Shiji>> Di-3 Qi, "Liuli de Jiyi 

fishe en" (7991 nian 2yue, di-107 ye) kaitou yinyong de yi duan w6nzi: 

Suiran wB hui xi4 shangmian de m8i yT ge Hanzi, yg cha le zidizn huozh8 

cidign, danshi dui zhhxie ciju haishi bu tai ligoran. Rljgui, Hanzi shi tebie bigoyi 

de, jibian wB mei y6u yuedir Hanzi de fgngfu jingyan, wi, ye yinggai cong 

shangmian zhexie Hanzi bgnshen dedao bangzhu, hgn kuai de jigjue wenti. 

Di-2 ge lizi. 1988 nian wi, zai Mgiguo Bryn Mawr Daxue jiao Hanyue. 9yue 

zh6ngxun de yT tian, yi ge D6ngfBng xuesheng Iai zhso w5, yao w6 g8i tCi kai yi 

zhang Zhbngwen chengdu zhengming. W5 tingjian ttia shu6 yao Zh6ngwen 

zhengming, jiu hen ziran de yong Hanyfi dui 18 shuij, "Ni ha5! Ni jiao shenrne 

mingzi?" TB li5ng zhi ygnjing zhgng de hen da, rnSn lian rnihuo jingqi. W8 

dgngdai le yihuir (=yihuir), chongfu shu6, "Ni hgo! Ni jiao shheme mingzi?" TEi 

haishi hen jTngqi de dgngzhe wb. Yllshi w8 jiu yong Yingwen wen t2 shi zai n3r 

xu6 de Hanyil. Ti3 shuci tB xiatian zai ChaoxiEin Hancheng yi sub daxue de shfiqi 

xuexiao xu6 le yi ge xiatitian de Hanzi; shu6zhe jiu b3 na ge shfiqi xuexiao de 

zTliao dig& wB. 

Wb kan le ziliao, you wen le 18 yixig wknti, faxian t% shi yong xiandai 

ChaoxiBnyLi xu6 de Hanzi. W6 jiu gaosu ttia bu neng ggi 19 ktiai zhengming. TEi 

y6udiSn qifen, wen wB weishenme. When y6u zheyang de duihua : 

"WB xu6 le yi ge xiatian de Hanzi, ybu chengjzdan. Weishenme ni bu neng 

ggi w6 k8i zhengming?" 

"Yinwei we ~ C d6ng J Chaoxianyfi. W6 mei fazi ceyan ni de chengdu. Ni 

yinggei qing yT wei ChaoxiBnyG jiaoshou g6i ni k8i zhengming." 

"Bryn Mawr rnei yBu Chaoxianyfi ke, bingqig Hanzi b9 dbu yiyang ma?" 

"Ni shuci de h8n dui. Hanzi dafi shang shi yiyang de, danshi ni yong Hanzi 

xuexi Chaoxianyil, yong Hanzi yuedlj xigzub ChaoxiEinyfi. wenzhtiang; w6

Sino-Platonic Papers, 27 (August 3 1, 199 1) 

yong Qanzi xuexi Hanyir, yong Hanzi yuedu xigzuo Hanyir wenzhang. 

Chaoxianyli de fayin, shBngci, yGf8 dgngdgng d6u gBn HanyG de bu yiyang. 

W6 bu neng yong w6 de Zhijngguo y6wen Iai panduan ni de ChaoxiZtnw6n 

hubzhg Zhangwen, dui budui?" 

"We zhi yao ni panduan wT, de Hanzi chengdir ya! Ni bC y6 chengren Hanzi 

dati shang yiyang ma?" 

W6 shubb~ch~ hua Iai le. Xiang le xigng, you dui 18 shub, "Rang w6 yong 

Riben ren zuo lizi. Ni zhTdao hgn dud Ribgn ren de Hanzi shuiping hgn gab. 

Ni xigng t2rnen neng qing yi wgi hui Hanzi, danshi ba hui Riyii de ChBoxiBn- 

yir ladsh? pandu8n tamen de ChaoxiBnwen huozhg Riwen ma? W6 bb hui 

Riylj. R6gu6 zhexie Riben ren ye bir hui Hanyir, ni xi8ng w6 neng zheng- 

ming zhexie Riben ren de Zhdngwen huozhg Riwen zgnmeyang ma?" 

Zuihou zhe wei tongxue bing bu shifen qingyuan de jigshou le wii de jianyi: 

Dao y6u ChaoxiBnyir kt?, zuoluo zai Feicheng shiq0 de Binzhau Daxue qu, qing 

yi wei jiao ChaoxiSlnyO de jiaosh~u ggi tB kZii zhengming. 

W6 zhidao we meiy6u shuifii na wei tongxue. Wb zhi jir le lizi, meiyou tichD 

hen h8o de liyou. Lieng nian du6 lai, wb changchang zu6rno zhe ge wenti, 

zhbngyli huangran da wir. 

Yi ge Hanzi zhi shi yi ge danger flihao. Xiang biede danger fuhao (bZtokuo 

girding de bishbushi) yiyang, yi ge HBnzi zai m6u zhbng qingkuang xia, zhi 

biaoshi yi ge ybuxian de yuanshi yiyi. R6gub zhe zMng fuhao de shirliang bC 

guofen piing& xueqiai bing b" name k"nn6n. Mgigu6 zhirrning de 

da(h6i)xingxing KgkE! jiu zh8ngwo ie wGb5i dub ge bish6ushi de flihao, yonglai 

xiang tZi de xunlian renyuan bigoda tB de yuanwang he ggnjue. KBkE! xue de bh 

shi Hhnzi, dgnshi Kgkg zhZingw6 de f~jhho, z8i jiaoliii ji8nd8n de yuanshi yisi 

shang, zuoybng g6n xiangdui de wirbii duo ge danger Hanzi que hgn leis]. 

Yinwei mgi yT ge danger fuhao biaoda de yiyi hgn ybuxian, hgn yuanshi, jiir 

bixir jiezhu biede tiaojian cO neng chuanda bijiao firza de neirong. Zai yingybng 

wenzi fgngmian, wbmen bix0 yong wi5men shbxi de yUyan bZi danger wenzi 

fGhao zGzhT qilai, pailiecheng cihui, jirzi, he pianzhang. Jibian shi xiang Hanzi 

t

SchnftfBd~n?: &says iri Honor of John DeFrnncis 

zheyang lishi y6ujifi de fuhao yE! bixU ykao yi zhbng yiryan, cai neng bsitui, 

danger fuhao de joshu, jinrir wenzi de lingyu. Huan ju hua shu6, Hanzi bixo 

yTkao HanyG, Riyfi, Chaoxianyir, Yuenanyir huozhg gGdai Hanyfi (jiu shi 

Wenyanwen) cai neng cong tutengshi de danger kfuhao chfijing li z5uchOlai, 

chengwei jiezhu yGyan Iai bigoda sixigng ggnqing de xigzuo gbngjG. 

Liiing nian dub yiqian, w6 zhi zhidao w6 bG neng gGi na wei xu6 Hanzi de 

ChaoxiBn tongxue kiii renhe yfiwen zhengming, danshi bu zhenzheng mingbai 

daoli. Xianzai w6 mingbai le. WT, bu neng kai, ylnwei w6 neng yong de gbngju 

zhi shi jib% ge yiyi yuanshi de danger Hanzi. Mei ybu gongtong yGyan, na wei.. 

tongxue g8n w6 jiu xiang ligng zhi xuehui le jib5i ge Hanzi de dkingxing. 

W6men jiaolili de shuiping tingdun z&i yingyong danger fuhao de jieceng, bir 

neng jinxing zhenzheng de yirwen hubdbng. Zai zhe zhhg qingkuang xia, w6 

neng wei tB zhengming shenme? 

Zhe y6 shuaming, zuowei danger flihao, Hanzi suiran bi8oyi, que bu tebie 

bi2oyi. Zhe zhbng bisoyj de g6ngju bayiding f8i shi Hanzi bu kg. Rligu6 d8ngchU 

wi, he na wei Chaoxian tongxue yiqi xuehui le KGkg zhangwo de 500 du6 ge 

fljhao, w6men zhaoyang kgyi jinxing yuanshi yiyi de jiaolifi, bingqig hai kgyi gBn 

Kgkg jiaolili ne!. 

ZBngjieqXai shu6, rugu6 mei ybu yiyan zuo jichfi, gebie Hanzi jiu zhi shi yT 

dui bigoda yuanshi yiyi de danger f6hao. SBnqian dub nian yiqian, danger Hanzi 

IT ybu shgoshir shi xiangxingzi, flihao b6nshGn hubxG hO kgyi zhijie tigdng yidign 

mohu de yisi, kgshi zheyang de Hanzi jlntian yijing bu cunzai le. Jiwan ge 

xiandai Hanzi de xingzhuang, fayin, jiben yisi dgngdgng, d6u bixO tangguo 

qianghua xbnlian, yong siji de banf3, yi ge yi ge de beih80. Danger Hanzi 

b6nshGn de bu tebie biaoyi shi ming bgizhe de. 

Di-3 ge lizi yG fasheng zai Bryn Mawr Daxue, yg shi 1988 nian de qiotian. 

Bryn Mawr Daxue Zhdngwen Yi-nianji yong de shir shi . 

Zhe bGn sh9 cong dl-yi ke jia jikshao Hanzi, danshi t6u 10 ke de kewen dbu shi 

yong Pinyin xi8 de. Ban shang yigbng 26 ge xuesheng, qizhang san wei shi 

pangting de jiaoshou, ligng wei nd de, yi wei n h de. Yi wei ni jiaoshou dayu6

Sino-Platonic Papers, 27 (August 3 1,199 1) 

sanshi ji sui, jiao Fgwen. Lingwai yi wei n6 jiaoshou kanqilai si-wirshi, tB he na 

wei liushi du6 de nan jiaoshou shi f9fC; liHng ge ren d6u jiao zhexue. 

Yinwei nianling he chengnian ren y6u ge zh6ng shiwa chAnshGn, s8n wei 

jiaoshou xueqlai bi shib8-jiG sui de xuesheng xinkij. Danshi tamen lijigli gao, y6 

feichang ytjujuexin, feichang niili, subyi qitou dbu xuexi de hgn hgo. Liang wei 

nil jiaoshou tebie chuse, chljle Hanzi xi6 de hGn gbngzhGng, yGf5 he fayin yB bi 

ban shang dadu6shh de xuesheng h8o. Tamen tingxig Pinyin jazi bMan hgn 

kuai, ye hgn zhitnque, jih0 rnei y6u cuowu. 

Shang di-11 ke yiqian, san wei jiaoshbu dbu gBn xuesheng yiyang de lianxi 

chaoxig Hanzi, moxie Hanzi, kan ji3ndBn de Hanzi jhzi. TBmen d6u reai Hanil, 

xu8 Hanzi xue de h8n qijinr. 

Cong di-11 ke qi, kewen quanbir gaicheng Hanzi. Na tian shangke de 

shihou, si-w6shi sui de ni! jiaoshou hnran bi5oxian de hgn fanchang. Ta bu zai 

zhuanxin. Suizhe kecheng de jinxing, tB de lignse yuelai yue nankan, zuihou 

gancui dTzhe t6u bd kengshgngr, IhchO yanwu shangke de yangzi. WI, hen 

nanguo, xinli xigng: W6 shu6cuo shenme hua le ma? Ta dui wi, jintian de 

jiaoxuefg y6u yijian, shiqu xuexi de xingqu? Ta ji8 IT fashgng le shenme shiqing 

ma? KBshi tB de zhuangfu bing mei y8u shenme yiyang a. 

Xia le ke, wb zuihou z5uch0 jiaoshi, kanjiari s8n wei jiaoshou hai meiyou 

IikZii, dbu zhan zai 16uti zhuiinjigo de difang. Na wei si-wirshi de jiaoshou 

rengran dizhe t6u, lingwai ligng wei zhbngzai anwei ta. W6 z5uguoqu, wen 

tamen daodi fasheng le shenme shiqing. Si-wilshi de n6 jiaoshou taiqi t6u 15 

shu5, "Wi, xi5ng jintian shi w6 zuihou de yS jie HanyG ke. WI, yihou bu shang le." 

Ta ygn li jingran ybu Ieihua. W6 hen jingya, ggndao hgn kgxi, yk hen bu shi 

ziwei. W6 wen tB weishenme hOran jueding bir xuexi le. TZi hgn shangxin de 

shu6, "Bu shi wb bfi yao xue, shi wT, tai ben! Wb chabbdub shenme dbu bG hui 

nian! W6 zgnme neng xuexiaqh!" W6 geng jTngya le, yiqie wanquan chayu w6 de 

liaoxigng zhiwai. Tou yi tian ti3 hai xue de hsohgor de, zgnme yixiazi biande 

chabirdu6 shenme dBu bh hui nian, bu neng jixu xiaqu! Zui zaogao de shi, tB 

ji ngran juede ziji "tai ben"!

Schriiftfestschri$: <strong>Essays</strong> in Honor of John DeFrancis 

Zixi yi wen, cai zhidao shi kewen hUran biancheng le Hanzi, til genbushang 

tangr le. Wb xi3ng le yihuir, jid dui tB shu6, "Bu shi ni ben, shi ni muqian dui 

Hanzi bu gou shlixi. Rtigub wb bB di-11 ke yong Pinyin xigchirlai, ni hui nian de 

bi jintian biin shang de tongxue nian de geng hgo. Wenti shi zheyang de: Rugub 

wbmen yBoqi6 de shi ligojig neirong, ni bu bi renhe xuesheng cha, yinwei ni kgyi 

yong Pinyin h8n kuai de kanddng t6ngyang de cailiao; danshi rfigub wemen 

y8oqiG de shi kan Hanzi, ni jiu yinwei mirqian dui zhe tao flihao bu shlixT er lian 

bgnlai hui de juzi y8 nian buchUlai, kanbudbng le. Xuexi Hanzi xiryao h8n dud 

shijian, gBn c6ngming bu cangming mei ybu zhijie de gu3nxi." Zuihou tB jueding 

jixu xuexi. 

Zhe wei ni jiaoshbu jianjue rniandui Hanzi de tiaozhan, yihbu b2 dabhfen 

shijian hu8 zai Hanzi shang. Kgshi xuenian moligo de shihou, tB de yCifa, tingli, 

huihua he yued6 nengli d6u luohou le, Hanzi yS! meiyou xueh8o. Na wei Fawen 

jiaoshou hgn 280 jiir fangqi Hanzi, yizhi yong Pinyin; yirwen fangmian, 18 yizhi 

zai ban shang lingxian. Na wei nan jiaoashou xueguo Hanzi he Wenyanwen; 

buxing dedao de ganriio que bi bangzha dub, zhihou de chengji yg bu tai hao. 

Muqian zai Zhbngguo bu hui Hanzi jiir shi wenmang. Danshi c6ng s8n wei 

jiaoshou xuexi Hanyfi de jiegub kan, Hanzi bing bu tebie bi8oyi. Yong Hanzi 

xuexi Hanyir de ligng wei jiaoshou, zuihou neng zhangwo de Zh6ngwen neirong 

fiin'er sh5o. 

Di-4 ge lizi. 1989 nian, yinwei jiankiing de yuangc, ye zhenghao y6u jnui, 

w5 zhugndao zai wT, jia fujin de Swarthmore Daxue jiao HanyG. 1990 nian 

chOntiZin, wi, zub le yi ge xigo shiyan. W6 qing Swarthmore he fujin ji ge daxue 

Zhdngwen yT-nianji dao si-nianji de tongxue b3 shi ge Zhbngwen jirzi fanyicheng 

Yingwen. Zhexie jirzi y6ude shi w6 xi6 de, ybude shi cong shU shang zhaolai de. 

Y6u wfi ge Hanzi jirzi, wfi ge Pinyin jhzi. Hanzi he yCif5 dubban shi yi-nianji 

xueguo de. Hanzi jljzi de di-si ge ytju Gugngdbnghua yfifii, di-wi ge dabufen shi 

Wenyanwen. Da'anjuan bir jiming, danshi yao zhichir shi Zhdngwen ji-nianji. 

Xiamian shi yT ge yT-nianji he yi ge er-nianji tongxue de fanyi (w5 meiyou jia

enhe gsidbng) : 

1. 5$%&&&($3kk90 


Yi-nianji: Ding Li is a stubborn and big. 

~r-nianji: Ding Li really grew up. 

Yi-nianji: He speaks Chinese---duishi he writes Chinese characters not too 

well. 

~r-nianji: Although .... spoke Chinese extremely well, he did not write Chinese 

characters very well. 

Yi-nianji: What! .... wGyuan (six kilometers?) 

Er-nianji: What! .... 5 dollars! This is really expensive. 

4. -&4~2%3 

Yi-nianji: This sentence 

-A-tg *& - 0 

~r-nianji: This child has understood one day. (I don't understand the 

.... construction 

5. '?* h3iI.J. " 75. &&bt.A&&." 

Yi-nianji: Person said, "don't---------. 

~r-nianji: person said, "....outside people know also? 

1. T2 de didi jiao Aid~hua. ~idehua jinnian shiliu sui; t3 yT tian bi yi tian 

d6ngshi le. 

Yi-nianji: His brother's name is iidehua. This year he is sixteen years old. He 

will understand things more as he grows up." 

~r-nianji: His younger brother is called Aidehua. Aidehua is sixteen this year; 

he understands more each day. 

2. Suiran An Ddngfang zhi xue le jii3 ge yire [de] Dewen, kgshi tB shu6 

Dewen shud de feichhg lilili.

Schriffemchrif: Essay in Honor of John DeFrancis 

Yi-nianji: Although Ann Dongfang studies only 9 month of German, she could 

speak German extremely well. 

~r-nianji: Although An Dongfang only studied German for nine months, he 

speaks German extraordinarily well. 

3. Yinwei xianzai xueqT kuaiyao jieshc le, warnen dgi xi6 hgn du6 baogao, 

kan hen dub sho, su6yi w5men xianzai b6dh ddu f8ichang mang, erqig 

ye dbu feichang lei. 

YT-nianji: Because now we almost finish school, we must write many term 

papers, read many books. That is why we are extremely busy and 

tired. 

~r-nianji: Because now the semester is coming to an end, we must write 

many papers, must read many books, therefore we all now are 

extradinarily busy though also all are extraordinarily tired. 

4. Zhanggub chD le yT ge Qin Shihuang. 

Yi-nianji: There is one person name Qinshihuang in China. 

~r-nianji: From China is a Qin Shihuang. 

5. Ni zhende shenrne dbu bir yao ma? 

Yi-nianji: You really don't want all these things? 

~r-nianji: You really do not want anything? 

Zhe ge shiyan feichang ji3nlou. Birguo .kgyT shaowei rang w5men kanchirlai: - 

suiran Hanzi duoban shi yi-nianji xuesheng xueguo de, danshi nigu5 xuesheng 

dui yfiyan de zhsngwo bh gou, jibian shi er-nianji de xueshgng, yg h8n nan 

cong Hanzi dedao bangzhu. Tebie shi na ju Wenyanwen, Hanzi jThir dbu shi yT- 

nianji hui xi8 de, danshi cong yT-nianji dao si-nianji de xuesheng, d6u fanyi de 

luanqi baiio. Weishenme ne? Yinwei tamen d6u mei y6u Wenyanwen de yiryan 

beijing. Zhe. biaoshi Hanzi gBn biede wenzl yiyang, bizoyi de g6ngneng jianli zai 

yi3yan de jichi shang. Wd geren juede xuexi shgoliang de Hanzi y5u hZiochir, 

danshi Hanzi shizai shi Hanxuejiii de wenzi, bir shihe yiban xuesheng, y6 bu 

shihe yiban Zh6ngguo ren.

Sino-Platonic Papers, 27 (August 3 1, 199 1 ) 

1989 nian xiatian, yi wei Ribgn xiaojig weile xue Yingwen, dao w6men ji8 Iai 

zh" le yi ge yue. TB shi RibEtn Sendai xian ren, zai dangdi yi su5 yinyue 

xueyuan nian s8n-nianji, zhuEing6ng gangqin. TB hui hgn due Hanzi, zi yB xi6 de 

fgichang qingxiu. W6men yi ji8 ren changchang gBn t8 yiqi tan ta jig li de ren; 

Ri bgn, Zh6nggu0, he Mgiguo de xuexiao, shGng huo xiguan, wenxue, lishi, yiryan 

wenzi shenmede. 

Yi ti8n chiguo wirfan, w6men tanqi Zh6ngwen he Riwen de wailaiyir yiji Riben 

ren zgnme yong Hanzi dgngdgng. Na shihou zhudzi shang bgizhe yi b8n zazhi, 

y5u pian wenzhsng de biaoti shi " 

4% ". Wb wen tB khdedbng 

kanbud5ng. TB chenyin le yihuir, wgixiaozhe shub, "Dayi wb kanded6ng." W6 

hgn xigng zhTdao tB daodi neng kandbng dubsh20, jilj qing ta shiyishi bg na ge 

biaoti fanyicheng Yingwen. TB fanyicholai de shi "Powder break (or broken) 

blood (sorry, I don't know this one [ fig I) town pressure". Fanyiwan le, tB pianzhe 

t6u, weixiaozhe shu6, "En--[ don't know, what does this really mean?" 

Zhe wei Ribgn xigojig de fanyi zai yi ci zhengming: Jibian shi Hanzi, ye bixU 

ykao yiryan, cai neng chuanda bijiao zhengque de yisi. 

Hanzi zai Zhbngguo yong le sanqign du6 nian, y6 zai D6ng-Ya ji ge guojia 

lirjchuan le h8n chang yi duan shijian. Cong Hanzi fazhSn chfilai de shidaifu 

whnhua hgn shenhbu. Hanzi neng anfir Zhbnggu6 ren de gSnqing. Hanil shDf5 

geng shi yi zhbng ddte de yishh. Zhe d6u shi buzhgng de shishi. Dan zhe yiqie 

d6u bh neng zhengming Hanzi bgnshgn tebie biiioyi, g&ng bu neng zhengming 

Hanzi zui shihe xiandai Hanyc. 

Hanzi shhliang pangda, bigoyin bbioyi xit6ng bhdan fhza, erqi6 hhnluan. Zhe 

tao wenzi rang Hanxuejia gsndao qcwei wljqiong (yinwei hgn shenmi shen'ao -- 

zhe bing mei ybu shenme bh ~SO). KEtshi dui yiban Zhbngguo ren lai shub, ~he. 

tao wenzi shizai hgn bu ybushan. R~jgu5 w6men zh8nxTn yuanyi yiban Zh6ngguo 

ren d6u neng yingyong Zhdngguo wenzi, congshi xiandai yfiwen shGnghuo, 

wbmen jih bu neng bC kgolu bg Pinyin nalai zuowei Zhbngguo de di-er zhbng 

we nzi .

Schniftfestscschriji: Ersays in Honor of John DeFrancis 

In conmemoration of 80th anniversary of Prof. John DeFrancis 

TYPOLOGY OF WRITING SYSTEMS by Zhou Youguang 

Every writing system has its individuality of external appearance and its 

generality of internal structure. It is easy to see the external appearance 

and difficult to discern the internal structure. Typology of writing systems 

is based not on individuality but on generality. 

Writing systems are like prisms, each having three phases: I, the phase of 

symbol form, 11. the phase of speech segment and 111. the phase of expressing 

mode. 

I, THE PHASE OF SYMBOL FORM 

The phase of symbol form has three layers: I. picture symbols, 2. charac- 

ter symbols and 3. alphabet syabols, 

Picture symbols are either transparent and understandable by seeing, or 

opaque and not understandable by seeing. Picture symbols are not possible of 

being decomposed into strokes. 

Character symbols do not look like any real thing, They can be decomposed 

into a number of strokes, but the total number of strokes is indefinite, 

Alphabet symbols have a small number of units and the number of units is 

generally defini te. Alphabet symbols have three layers: a. sy llabic alphabets, 

b,, consonantal alphabets and c. phonemic alphabets. 

Picture symbols generally belong to the earliest stage of writings. 

Character symbols are generally developed from picture symbols. Alphabet 

symbols are mostly the simplified forms of characters 

11. THE PHASE OF SPEECH SEGMENT 

The phase of speech segment may have Long segments or short segments. Long 

segments may be thesis segments, paragraph segments or sentence segments, with 

one integrated symbol chart representing a whole story, a section of a story 

or a complete sentence. Short segments have three layers: 1. word segments, 

2, syllable segments and 3. phoneme segments, 

I I I. THE PHASE OF EXPRESS I NG HODE 

The phase of expressing mode has three layers: 1. picture drawing mode, 

2. idea narrating mode and 3, sound denoting mode. 

Picture drawing mode requires the reader to understand the writing without 

previous Learning. With this node. functional words are not possible to be 

written down, and human speech cannot be recorded in word sequences. Such 

writings can be read in any human speech with super-speech nature.

Sino-Phonic Papers, 27 (August 3 1, 199 1) 

Idea narrating mode makes the symbol carrying definite idea arbitrarily 

and the reader must learn the connection between idea and symbol first. This 

mode can write functional words and record definite human speech according to 

word sequences, but still retaining a limited degree of super-speech nature. 

Sound notating mode makes arbitrary connection betweem sound and symbol, 

not between idea and symbol. It has no super-speech nature at all. 

expressing mods . . 

The relation of the three phases are not synchronous, but intricatcd. For 

instance, picture symbol can be used in picture drawing, idea narrating or 

sound denoting, to represent a word, a syllable or a phoneme. 

picture drawing picture symbol word segment 

idea narrating character symbol syllable segment 

sound notating alphabet symbol phoneme segment 

From the above explanation, we can get a Table of Typology of Writing 

Systems according to the three phases classification as follows: 

(name in short) I (expressing mode) I (symbol form) 1 (speech segment) 

a, picture writing I picture drawing I picture symbol I word/sentence 

ab. picturwidea I picture drawing/ I picture/ I word/sentence 

writing I idea narrating I charac. symbol 1 

b. idea writing I idea narrating I charac. symbol I word 

bc. idea/sound I idea narrating1 I character/ I word/sy 1 lab Le 

writing I sound notating I alphabet I 

------------------------------------------------------------------------------ 

c.phoneticwriting I soundnotating I alphabet I sy 1 lable~phoneme 

cl. syllable writ. I sound notating 1 syllabic alph. I syllablc 

c2, consonant wri t. l sound notating I consonant alph. I syl lable/phoneme 

c3. phoneme writ. 1 sound notating I phonemic alph. I phoneme 

------------------------------------------------------------------------------

S~h~ftftdtchri!: ,%sap in Honor of John DeFrancis 

Like the Periodic Table of Elements, there are vacant positions in the 

able of Typology of Writing Systems. The common types are: 1. picturelidea 

writing, 2. idea/sound writing, 3, syllable writing, 4. consonant writing and 

5. phoneme writing, The rest types are .rarely seen, Any writing system can 

fin.d its position in this table. 

Now Let us try to find the position in the Table of Typology of a few 

writing systems. 

(1) Ojibwa Love Letter (see Page 39, John DeFrancis: Visible Speech) 

The meaning of this love letter is translated to be a Chinese poem by the 

author of this paper as follows: 

%%ma%, #%AWEJ? 

%%Z%Rp %EeBi&o 

%@aAm, BF*NrPRo 

.'J\@FJ'@@t Bi!W#jlkn 

In respect of symbol form, this letter is an integrated picture symbol. 

In respect of speech segment, it is thesis segment. the longest of long 

segments. In respect of expressing mode, i t is picture-drawing/idea-narrating. 

It can be named in short as a picture/idea writing. Bear and Mud Puppy are 

totems with weak function .of idea narrating to represent clans. The three road 

lines have weak function of idea narrating in indicating directions.. Three 

crosses have full function of idea narrating to represent three christians. 

The hand in t.he west tent has full function of idea narrating to mean weicome. 

This Letter is a picture with idea expressing to be understood between parties 

with tacid agreement. Writings of tacid agreement have no social function of 

communication, 

(2) Yukaghir Love Letter (see page 25, John DeFrancis: Visible Speech) 

The meaning of this Love letter is-translated to be a Chinese poem by the 

author of this paper as follows: 

33kE.5t-ZBAI E@&*%@*o 

%@**%**? %**+rn$)LO 

S%lL*%%%t &i$%+~Gsl;FgO 

In respect of symbol form, it is not picture symbol, but symbols with the 

nature of artificial characters knit together. In respect of speech segment, 

it j s thesis segmeqt, not divided into paragraphs. In respect of expressing 

mode, there is idea narrating without picture drawing, understandable between 

tacit agreed parties. Umbrella forms indicate human beings, roof like lines 

indicate houses, curve lines indicate love thinking, broken lines indicate 

disappointment, cross lines indicate sorrow, and zig-zag net Lines indicate 

quarrels, This letter is super speech. There iq no picture symbol in it. 

neither there is any sound symbol. It can be named in short as pure idea

Sino-Plato~zic Papers, 27 (August 3 1, 199 1) 

writing. Pure, idea writing is very rare, often considered as non-existing. 

('0 Chinese Writing Sys tem. 

Chinese writing system is divided into ancient and modern. In ancient 

Chinese writinq, pictographic, ideographic and even phonetic symbols all 

belong to the category of picture symbols in broader sense, not possible to 

be deromposed into strokes. Roughly speaking, from the time of Qinshihuang, 

shell-and-bone style and hig and small seal styles were changed to clerk and 

regular styles. This is a change from picture symbol form to character symbol 

form. Pre-Qin writings used picture symbols, while post-@in writinqs hegan to 

USP chararier symbols. In respect to speech segment, Chicese characters 

mostly represented word qegments in classic Chiaese for most words were mono- 

syl labjc by that time. It is different in modern Chinese, for most words have 

become polysyllabic, and Chinese characters mostly become syllabic and not 

logographic. A recent strtdy by the author reveals that among 1000 currenlly 

used characters, there are about one third word( free)-characters and two 

thirds syllable(bound)-characters. So modern Chinese writing is a system of 

word,/syllatle scrip! in idea.,#sound writing. There is neither picture nor 

alphabet in formal modern Chinese writing. The Chinese alphabet (Sound- 

Notatins Symbol or Pinyin Alphabet) is not a part of formal writing, 

(4) Japanese Writing System 

Chinese characters were introduced to Japan after they were ripen in 

chnng ing from picture symbol form to character symbol form. Japanese wri t ing 

system is divided into two periods: !he period of character and the period of 

character/alphahet. so far as symbol form is concerned. Japan used Chinese 

classic written speech for about 500 years, and in .the later 1000 years tried 

to write Japanese speech with Chinese characters, This again is divided into 

two pcriods. First, Chinese characters were used in their original forms to 

write Japanese word, root of word and inflexion of word. Chinese characters 

were original ly a1 1 eonosyl labic symbols. For writing Japanese speech, they 

were pronounced as polysyllabic symbols. For inflexion of word. $hey became 

monosvl labic alphabet symbols called Manyo Gana. Later, Manyrr Gana were 

simplified to he modern kana alphabet of pure syllabic nature. But up to 

today, the official Japanese writing is a mixed script of characters and kana 

syllabary. not purely syllabic alphabet. Modern Japanese is word,/syllablc 

writing, not pure syllable writing as often calledmistakenlg by people. In 

complete term, Japanese wri t ing is of charactar./alphabet ic symbol farm, word/ 

syllable speech segment, and idea-narrating..sound-natating expressing mode. 

. In short, it is idea/sound wri t ing.

Schnftfestsnifr: fisays in Honor of John DeFrancis 

(5) Korean Writing System 

Chinese characters were introduced to Korea earlier than Japan, but also 

after the big change from Seal Type (picture symbol form) to Clerk Type 

(character symbol form). After the second world war, the north Koreans use 

only On-mun alphabet, while the south Koreans still use the mixed script of 

Chinese characters and On-nun alphabet, The On-mun alphabet is a peculiar 

alphabet of phoneme symboLs arranged into syllabic squires, The alphabet 

symbols are only forty in number. hut the syllabic squires amount to more than 

20110. For the North, the writing of today is svllabic/phonemic alphabet that 

writes word/sy l lab lel.phonerne speech segments in sound-notating mode. For thc 

South, it is characterlsy 1 labic,~phonemic alphabet that wri tes wnrd/sy l lab L P/ 

phoneme speech segraents in idea-narrating/sound-notat ing mode. In short, the 

North .has sound writing, while the South has idea,.sound wri ting. It is an 

intermidiate stage between idea writi~ig and sound writing. 

(6) Amharic Writing System 

Amharic writing system is the only pure syllabic writing system used as 

the official national writing of a state of the present world, 11 has 247 

syllabic symbols derived from 37 consonant symbols with 7 vowel signs attached 

jointedly. In Japan there is syllabic alphabet but no official syllabic - 

writing, Japanese syllabic symbols are integrated signs impossible of being 

decomposed into phonemic signs. Amharic syllable symbols can be dissected 

into consonant and vowel signs though not very regularly. Amharic writing 

system is, in terms of typology of writing systems: syllabic alphabet symbol 

form, sy l lable speech segment, and sound notating expressing mode. 

(7) Arabic Writing System 

In 1 974 Arabic was made the 6th working 1an.guage-of the United Nations. 

It is the official national writing system of 18 Arabic states and the script 

adapted to other Languages such as Persian, Pashto, Urdu, Sindhi of India and 

Urghur of China, The present Arabic alphabet of 28 Letters consists hasically 

of consonants, the vowel signs being detouchable marks placed above or below 

the letters, They are generally omitted, though used in elementary school 

books and the Koran bible, It is a writing system of consonant alphabet 

symbo 1 form, of sy l lab l elphoneme speech segment, and of sound not at ing 

expressing mode. 

(8) English Writing System 

English writing system is made of phonemic alphabet symbol form, of 

phoneme speech segment, and of sound notating expressing mode, but neither 

pure nor reguiar. 

(9) Finnish Writing System 

The Finnish alphabet contains 21 letters, 13 consonants and 8 vowels.


There is only one sound for every letter, and one letter for every sound. It 

is the typical and pure phonemic sound notating writing system, using phonemic 

alphabet that writes phonemic speech segment. The complexity of Finnish 

speech must be distinquished from the simplicity and regularity of the Finnish 

writing system.. 

Conclusion: In order to save printing space, the examples -given here are 

too few and the explanation too concise. But it shows already that the 

classifiration of writing systems can bc made more clear with the three phases ' 

analyzing method. ! 9 9 I - 03- 1 7

Sino-Plutonic Papers, 27 (Aupt 3 1,199 1)

Schriftfestschnif: fisq in Honor of John DeFrmcis 

. *%@, congress (A*) &+%&&con (#R)Bgress (G*) "%" (-8%) 

%1&#~, UASgPJ-&" "82" "$" @J;$;Xi2-#tEIo super 

(s) jAl+son (9%) , -@&_t- ic (#J) , B%&TB-f;-@ supersonic (#li%?i!#@) 

J 

rnW~Brl#i5I;h-&@+Hf~o fia%EirJEsFr:aF*E*B3m%AP %E%**JC:!!k--B~ 

%S24f-&-7'T&& u#*%f!kn P %*S*,& u#&" 2E?

Sim-Platonic Papers, 27 (August 3 1, 1991)

Schrififestschrifi: <strong>Essays</strong> in Honor of John DeFrancis 

THE INFORMATION SOCIETY AND TERMINOLOGY 

Liu Yongquan 

Institute of Linguistics 

Chinese Academy of Social Sciences 

1. CHARACTERISTICS OF THE INFORMATION SOCIETY 

1.1 Language Information -- the Basis of the Information Society 

There are many kinds of information, of which the most 

important one is language information. Language is ubiquitous and 

all-embracing. Any kind of material or any kind of spirit could 

not be recognized and comprehended without transforming it into 

language information. What is the so-called information society 

like? One of the main features lies in the computer processing of 

language informat ion (storage, retrieval, simulation, 

transformation and transfer, etc.) with the aim of setting up a 

modernized language information system, making optimum use of 

language and making the most of the knowledge involved in it. 

1.2 The Increasing Precision and standardization of Language 

In order to meet with the actual needs caused by the 

scientific and technological revolution, language is getting more 

and more precise and standard. The efforts of language planners 

(including ISO) play an important role in this process. 

1.3 The Extension of Language Integration 

It is virtually a visible trend that dialects are gradually 

being integrated into a common language with the development of 

society, the individual languages into a number of regional 

languages and then into several international language(s). 

2, DEVELOPMENT OF TERMINOLOGY -- AN INDISPENSABLE CONDITION FOR 

THE INFORMATION SOCIETY 

2.1 Terms -- Basic Information Units in the Information Society 

As we mentioned above, a modernized language in£ armat ion 

system will be set up in the information society. Concretely 

speaking, various databases will exist in great numbers, 

unifunctional or multifunctional robots will be seen everywhere. 

Each country or group of countries will establish its own grand 

language system, and through the networks from all directions a 

gigantic language system will be founded in the world. The most 

fundamental elements of these systems naturally are the strictly-

Sirw-Plutonic Papers, 27 (August 3 1, 199 1) 

defined words (terms). It is no exaggeration to say that, without 

such terms there will be no modernized language information 

system, and therefore no information society at all. As everyone 

knows, the prerequisite for language computing is precision, 

formalization and algorithmization 

2.2 Terminological Normalization and Standardization -- the Crux 

of the Information Society 

Natural language is an extremely complicated sign system. It 

is really an arduous task to computerize it. Beyond any doubt, a 

great burden, even an unconquerable difficulty, would be imposed 

on the modernized language system if the basic elements of the 

natural language concerned have many variants (at least the 

storage would be increased, or some other contradictions take 

place in the system itself). Let us take a recent example, the 

book of a famous American industrialist, Autobiography of Iaccoca 

has been translated into Chinese by four publishing houses in 

1986. Four Chinese equivalents have been produced from Iaccoca, 

i.e. (Yakeka), (Aikeka), L $Gfi (Aikeka),&fi+-f 

(Yakek$."Ind this cannot even be compared with the name 

Mendeleev which had 28 Chinese equivalents. The same holds true 

.for technical terms. The little word has many equivalents 

. too,such as rbe, $* SJ#$-~&, s."$4&& , j& . The network of 

information systems eman s a high egree of standardization -- 

this is the crux for ensuring a free flow of information. 

2.3 Terminological Work Should be Modernized 

To meet the requirements of the information society 

terminological work itself should be modernized in the first 

place. The handicraf t-type terminological work and 

lexicographical work cannot be continued any more. A pressing 

matter of the moment is to set up a terminological database and 

relevant termnet and work out electronic dictionaries as well. 

3. CHINESE INFORMATION PROCESSING AND CHINESE TERMINOLOGY 

Chinese information processing is the flight of steps 

leading up to the information society in China, and Chinese 

terminology is the key to the information society. 

3.1 Difficulties in Chinese Information Processing 

3.1.1 Input: The written representation of Chinese language is 

Chinese characters. The defects of Chinese characters lie in the 

huge amount (about 60 thousands), the complex configuration 

(which consist of between one to forty or fifty strokes) and the 

abundance of homonyms and polyphones which are not so easy to 

process as alphabetical writing. In order to input Chinese

Schriffst~~hrifr: fi~rsays in Honor of John DeFrancis 

characters into computer, over 500 encoding schemes have been 

proposed. But only several dozen of them have been adopted for 

input. Generally speaking, there is a translation procedure at 

the time of input with the aid of a small keyboard. Undoubtedly 

it causes a lot of trouble for the user. And what is worse, it 

affects the sharing of data on account of different input 

schemes. 

3.1.2 Segmentation: There is no space between words in a Chinese 

sentence, but the basic unit in information processing is word. 

Therefore the issue of segmentation has necessarily been raised. 

Researchers must segment a Chinese text into separate words at 

present before they do statistical lexical analysis or conduct 

test for natural language understanding. It seems strange to hear 

that the statistical results will vary by as many as the number 

of researchers, for instance there will be three different 

results if the same book is analyzed by three persons 

respectively. The reason for that is the lack of a unified 

principle for segmentation of words. There is no concept of ~ r d 

among ordinary Chinese people. This is the harmful consequence 

caused by the use of Chinese characters over a long period of 

time. It is not easy for machines to do this work either. The 

longest progressive match and the longest regressive match have 

been tested, and still a lot of problems remain. Not only a large 

amount of errors occur, but also a great deal of valuable 

computer time is lost. The issue of word-segmentation may be 

likened to I'a tiger standing in front of the road we cross (in 

other words, a stumbling block)" if it can not be solved the 

proper way. 

3.1.3 Networks: The above state of affairs has already 

constituted a very unfavorable condition for the setting up of 

databases. To make things worse, the databases may be likened to 

a pond of stagnant water if no network can be built up. However, 

the current telecommunication service in China is seriously 

undeveloped. To make significant progress in this field is an 

imperative necessity, otherwise telecommunications will present 

another stumbling block in Chinese information processing. 

3.2 Problems in Chinese Terminology 

3.2.1 Terminological chaos: 1.) The same term is used to express 

different concepts in different fields. For example, yundonq&ijb 

as a physics term refers to mthe act of changing the location of 

an objectw, yet, in philosophy it refers to "the mode of 

existence of matterw, in sports to "the process of physical 

trainingll, and in political life to I1organized purposeful social 

mass movementv1. * 2) Some terms have two meanings in the same 

* corresponding to these terms in Russian are two, dvizhenie and 

SDQK~; in English there are three, tpmotion", flmovementlf and 

llsportl'. 

23


field, such as fujia chenqfen Tif.da&,$ which is used to express 

either the concept of "a££ ix" or the concept of 'tattribute" in 

linguistics. 3) A more serious question is the fact that a large 

number of terms have more than one equivalent in Chinese 

six: 

as with the astronomical term vpolytropeff which was translated as 

uocensaiu 9 A I$. (multilayer ball) at first, and was later 

Lranslatedas duofanu moxina 43 &+ owing to the discovery 

that there is neither wlayerw nor Ifball" in the concept of the 

word. 5) Heterogeneity often comes from the preference for the 

semantic loanwords. A number of loanwords or terms were 

transferred into Chinese by means of a phonetic approach at 

first, but later substituted with semantic loanwords, thus making 

up a kind of parallelism, at least for a period of time. For 

instance, bulai i & 4 (from Russian n4ak.C~) --> Jianviuu&&;& 

ximinana'er ,7 ,f, (from Russian ~run*p ) --> iianaxibanannixilin 

++l from English penicillin) -- ainumeisu +$fL 

(from English laser --> jiauana &kt)~ , etc. T e 

Pv 

last one is typical: msai jkd (leishe $kg Taiwan and 

Hongkong) is a phonetic loanword, while jiuu- 8~ is a 

semantic one, in which the first morpheme fi expresses the 

concept of "stimulatew and the second morpheme quanq "light or 

rayw. 

3.2.2 The Contention of Phonetic and Semantic Approaches in 

Translation: As we noted above, terminological chaos is often 

caused by the contention of phonetic and semantic approaches. It 

is well known that terms may be divided into three subclasses, 

i.e. pure terms, terms and quasi-terms. Pure terms are the .most 

specific ones, while quasi-terms are close to ordinary 

vocabulary. It is apparent that there is no harm in adopting a 

phonetic approach with respect to such pure terms as neutron and 

198; the former might be transformed into niuton (tr --> t in 

this place, as there is no such consonant cluster in Chinese), 

and the latter into yon or ion. There are merits in 

internationalization of terms so that "the agony of a long-term 

hesitation, caused by the establishment of a termw* might be 

avoided and that the burden of students might be reduced and 

academic ' exchange facilitated as well. The internationalization 

of terms is an irresistible trend, but it has not understood by 

the majority yet. As for the mistranslation of the above- 

mentioned term Itpolytropew, it could have been avoided in the 

first place if the phonetic approach had been adopted. 

* & u -- a famous remark made by our great 

translator Yan Fu.

Schriifrestschr!: &sap in Homr of John DeFrarzcis 

3.2.3 Writing System -- A Focus:.The contention of phonetic and 

semantic approaches has a complicated history. Why has this 

problem not been solved for such a long time? Why are there so 

few phonetic loanwords in Chinese and why is it not easy for them 

to exist? There are various arguments.* However, the reason of 

utmost importance lies in the difference between writing systems. 

The phonetic loanwords can not give full play to their 

superiority if the Chinese characters are used. There is no space 

between words in a Chinese text. Phonetic loanwords are obscured 

in their midst. It goes without saying that they are not well 

received. Thus, it can be seen that it is not the Chinese 

language but the Chinese characters which do not tolerate 

phonetic loanwords.** 

3.3 A Common Outlet Applicable to Both Chinese Information 

Processing and Chinese Terminology 

3.3.1 Extension of the Use of Chinese Pinyin: Chinese Pinyin (the 

Chinese Phonetic Alphabet) based on Latin script has an alphabet 

totally similar to the English one in form. Though not an 

official written language at present, it can be used in those 

fields where Chinese characters are not convenient to use or 

cannot be used at all. Not a few scholars hope that it will 

become one of the two-track written languages (in other words, 

that there will develop "digraphia1I, i.e. the parallel use of 

Chinese characters and Pinyin). If this reasonable aspiration can 

be realized, all the difficulties which Chinese information 

processing and Chinese terminology are confronted with can be 

readily solved. However, Pinyin can be used only as a "crutchv' 

(an aid in other words) at the present moment. Naturally this has 

considerably affected its ability to play its role fully. It is 

imperative to extend the use of this powerful tool with great 

effort. 

3.3.2 Input by Means of a Phonetic Scheme: The greatest merit of 

a phonetic (encoding) scheme lies in the input of words instead 

of separate Chinese characters, while configuration schemes fall 

short of this function. Inputting according to words provides 

very favorable conditions for further processing, because it is 

capable of eliminating the additional . process of word- 

segmentation which not only works to no avail, but is also apt to 

create mistakes. Certainly, homographs constitute a problem in 

phonetic input. However, there is nothing to be afraid of in this 

regard. Inputting according to words by itself will reduce the 

amount of homographs. Furthermore, a reasonable orthography is 

...................... 

* Please refer to Liu Yongquan,Modernization of Linguistics and 

the Computer, pp227-241, Wuhan University Publishing House, 1986. 

Also refer to: Liu Yongquan, wTerminological Development and 

Organization in Chinatt, International Journal of the Sociology of 

Language ( 59 ) , Mouton, 1986. 

** See above.

Sino-Platonic Papms, 27 (August 3 1,199 1) 

conducive to differentiate homographs; and the remaining 

homographs may be recognized through analysis of the context. In 

fact, there are a large number of homographs in English too, such 

as: back, badger, bail, bale, bandy, bank, bar, barge, bark, 

base, bat, batter, bag, bear, etc. Yet this does not pose any 

serious difficulty in English information processing. 

In order to meet the needs of those who speak dialects other 

than Putonghua (Chinese common language) and those who use more 

Chinese characters than ordinary users, it is necessary to set up 

another track, i.e. input by configuration. Here is another kind 

of two-track system. * 

3.3.3 Insertion of Pinyin Words in Chinese Text: To eliminate 

both the terminological chaos and the chaos caused by 

transcribing foreign personal names and place-names with the aid 

of Chinese characters, not a few people have proposed the 

insertion of Pinyin words in Chinese text.** But up to now this 

proposal has not been made good except in some academic 

periodicals. Perhaps the overemphasis on language purism is an 

important factor in inhibiting the adoption of Pinyin in Chinese 

text. The purists were against horizonal typesetting*** and 

against simplified characters but failed. They know well that a 

great many "foreign gadgetsv would enter Chinese language, once 

the insertion of Pinyin should be allowed. As a matter of fact, 

their worry is useless. Language is constantly developing with 

the progress of society. There were only two punctuation marks 

(the period and the comma) in our literature several decades ago. 

The new punctuation marks came into being only in 1920 after the 

reformers' hard struggle. The new set of punctuation marks has a 

clear superiority to the old one, but it was still attacked by 

its opponents time and again. Why was it so? It is simply 

ridiculous that the opponents censured these marks as "foreign 

gadgetsw. Almost the same lot befell arabic numerals among the 

pedants. Language and its writing are tools. How to make these 

tools to be of rich expressivity and competent for exerting their 

communicative function in any case (including the man-machine 

dialogue) -- this is the most essential matter. As for what is 

foreign and what is indigenous, it is nonsense to make this 

difference: anything that enhances the communicative function 

should be welcome. Thus it can be seen that purism has become an 

even more serious obstacle in the information era. Without the. 

insertion of Pinyin words there are no phonetic loanwords and no 

more talk about the internationalization of terms. 

Moreover, it is worth pointing out that another merit of the 

-------------------------- 

* Liu Yongquan, "Language Engineering in Chinaw, Proceedings of 

1983 International Conference on Text Processing with a Large 

Character Set, Tokoy, 1983. 

**.Ji Da, '!For the Insertion of Pinyin Words in Chinese Textw, 

Modernization of Language, No 3, 1980 

*** In the past typesetting was carried out vertically from right 

to left, but the new typesetting runs horizontally from left to 

right, just as alphabetic writing does.

Schrijt$estschrifr: Ervap in Honor of John DeFrnncis 

insertion of Pinyin rests with its convenience for computer 

processing. Pinyin words, just like Bat- in Japanese, could 

play the role of word-boundary in the process of word 

segmentation. 

3.3.4 Setting-up of a Wordbank and a Terminological Database: The 

wordbank in essence is no more than a computerized word list, 

used as a norm for word-based processing. The wordbank consists 

of a basic bank and a number of auxiliary banks. The former 

serves users from all trades and professions, and the latter 

provide supplements for users of different specialties. The 

design of this wordbank should be lifted out of the interference 

of the Chinese characters, and based on Chinese Pinyin 

orthography. * 

We have to point out with emphasis that the setting up of a 

wordbank is entirely necessary to our society, because quite a 

few people have no concept of the word in their mind (this is a 

consequence of the long-term use of Chinese characters). The & 

In Its linguistic sense appeared in our country only in the 

beginning of this century.** The aim of this wordbank would be to 

work out a computerized image of Modern Chinese vocabulary, to 

pave the way for establishing various Chinese informati.on 

systems, and to establish a standard for segmenting words by man 

or machine.*** 

As we pointed out before, to achieve the modernization of 

terminological work, the first task is to set up a terminological 

database. Here we would like to dwell on some special questions. 

First, the storage unit should be discussed. Obviously, it must 

be based on the yard rather than the Chinese character, equal to 

a syllable, because modern Chinese is no longer monosyllabic. Now 

in our vocabulary there are only a few monosyllabic words (three 

thousands or so), and disyllabic and trisyllabic ones are the 

overwhelming majority. Secondly, how to sort and look up these 

units is also a consideration. Sort and look-up constitute a 

considerable difficulty in the use of Chinese characters. The 

common indexing methods are as follows: alphabetic, radical, 

stroke and four-corner methods. Only the first of them is easy 

for computing . 

To sum up, without Pinyin we cannot set up the wordbank and 

terminological database in the proper way. 

* "Chinese Pinyin Orthographyqv, Yuwen Jianshe (Language 

Construction'), No 4, 1988. The author of this paper is the 

nucleus member of the Committee of Chinese Pinyin Orthography. 

** Lyu Shuxiang, "A Sketch on the Problem of 'Wordt in ChineseH, 

selected papers on Chinese Grammar(revised and enlarged),1985 

*** Liu Yongquan, 9ome New Advances in Computers and Natural 

. Language Processingn, Proceedings of the 1986 International 

Conference on Chinese Computing, Singapore, 1986. Also Liu 

Yongquan, "Talks about Wordbanksu , Journal of Chinese 

Information Processing, vol. 1, No. 1 Beijing, 1986. .


3.3.5 Three Principles in the rans slat ion of Foreign Words and 

Symbols: In order to get rid of various contradictions caused by 

Chinese characters, to improve our documentation work, and to 

promote international exchange, it is necessary to work out some 

regulations with respect to the translation of foreign words and 

symbols. 

As for personal names and place-names it is reasonable to 

adopt transliteration mode. In fact, it would simply be coping 

word by word on account of the similarity between Chinese Pinyin 

and English alphabets. In dealing with the personal names and 

place names from Cyrillic or other alphabets the ISols romanized 

transliteration schemes should be adopted. 

As to terms derived from personal names or place names, they 

must be treated in different way, for they have been converted 

into Chinese vocabulary as soon as they were introduced. It means 

in more concrete terms that equal attention must be paid to both 

pronunciation and configuration. For instance, the letter 

with pronunciation as [kl should be converted to "kM, and "phn to 

"fq1; and some consonant clusters should be reduced or infixed 

with a vowel, etc.* As the author and his assistant have 

suggested, it is advisable to write "Mendelev zhouqibiao 

(periodic table) in Pinyin instead of "Mendeleev zhouqibiaom and 

to write ltKarington ziwuxian (meridian)" instead of "Carrington 

ziwuxianw, etc. In a word, we have to make the foreign words more 

convenient for pronunciation, and keep the original configuration 

as far as possible for the purpose of being in concert with the 

original names. 

The transplanation of individual terms should be carried out 

in a specific way. As we have said before, the pure terms should 

be transcribed phonetically with the aid of Pinyin. How should 

the phonetic transcription be carried out? As the term is an 

important of our vocabulary and will take root and blossom after 

its transplanation, so it has to be sinicized. Linguistically 

speaking, neither transcription nor transliteration alone will 

do; a compromise* proposal should be adopted. In other words, more 

attention must be paid to the phonetic similarity and less 

attention to the configurative similarity. For example, 

spectrography --> spektogafy; adstringent -->adestinjin or 

adestingen. 

3.3.6 The Necessity of a Transcription-based Transliteration 

Scheme: Many people have noticed the inefficiency of Chinese 

characters in the transcription of foreign personal names and 

place-names and have long advocated the insertion of Pinyin 

words. People have also noticed the superiority of phonetic 

loanwords to sema'ntic ones in the field of pure terms and have 

therefore advocated a phonetic approach. All of these are good 

propositions, but how shall we realize them? These propositions 

---------------------------- 

* Liu Yongquan and Qiao Yi, A Tentative Study of Transcription- 

based Transliteration. (in press)

Schnflestschrifi: <strong>Essays</strong> in Honor of John DeFrancis 

will not work if some concrete questions have not been solved in 

the proper way. For instance; there .is no problem with 

transliteration of names from the visual aspect. But how will 

Chinese speakers pronounce them after all? And what will 

announcers do with them? A 'bridge-like' tool is needed. For this 

reason the author and his assistant have developed a set of 

transcription-based transliteration methods, which has three 

uses: 1) as a phonetic notation for personal names and place- 

names; 2) as an aid to introduce terms derived from names; 3) as 

an important reference for transplanting pure terms (that is to 

say, either adopting its rules directly, or taking the rules as a 

basis). This transcription-based transliteration has 25 ordered 

rules in total at present (actually more than this sum, for there 

. . are a number of subrules), and might be formalized and 

computer ized . 

4. CONCLUDING REMARKS 

As the old Chinese saying goes, "A workman must first 

sharpen his tool if he is to do his work wellR. To establish a 

strong, modern country, we have to make our language more sharp 

and powerful too. 

Chinese information processing and Chinese terminology are 

closely related. They will exert a tremendous influence on the 

course of the development of China's information society. Both of 

them depend upon progress in written language reform. Therefore, 

they should be the subjects of primary importance in China's 

language planning. Though a lot of people have not yet 

comprehended their far-reaching historical significance, I firmly 

believe that the advent of computers will help them get a clear 

understanding sooner or later.

Schnffe~tt~h rip: Ersayr in Honor of John DeFrancis 

A BILINGUAL MOSAIC [for John DeFrancis] 

By Einar Haugen 

Nationalism is a theme of interest to our honoree. Even though his and my fields 

have diverged through the years, it is still a pleasure to recognize a fellow worker whose 

writings have impinged on our common theme of bilinguahsm and bilinguals. 

h this little article I shall compare and contrast two writers from the Ninteenth 

Century who worked in the bilingual area I know best, that of the world that was created in 

the United States by the immigrantion of nearly a million Norwegians between 1840 and 

1900. One of them was an educator, author, and editor named Peter Hendrickson, in short 

a would-be intellectual. The other was S. H. Severson, a small-town merchant, who wrote 

a single book containing acute observations on his fellow Norwegians in the dialect of his 

parents. He was an obsenrer who found their language amusing. Each of these men left 

behind enduring literary monuments, not belles-lettres, but each interesting in its own way. 

We begin with Hendrickson, who was born in Telernark, Norway, in 1842 and 

immigrated with his parents at the age of three. He grew up on a farm in Wisconsin and 

managed to work his way through a local academy named Albion and then entered Beloit 

College, a school founded by New Englanders.' He spent a year at the University of Oslo 

and another at Erlangen. From 1871 to 1884 he was a professor of Modem Languages at 

Beloit College. After that he became echtor-in-chief of the then leading Norwegian- 

American newspaper, Skandinaven from 1885- 1893. It was during this period that he 

wrote his chief publication, a widely read manual of American agriculture for immigrants, 

Fanning med Hoved og H~nder (Farming with Head and Hands). It was published by 

his newspaper, which had its offices in Chicago? 

In his Foreword Hendrickson "makes no claims in a literary or artistic respe~t."~ 

"It is written for a simple and straightforward people, who are seeking advice and not 

entertainment. The language is to be understood, not to be admired. As life is over here, 

- so is the language; it is mixed." One can point to the very title for an example: the word 

"farmingw is a loanword from English, while the rest is Norwegian. 

In the course of his pages Hendrickson pursues all aspects of American fanning, 

from the location of the farm to its cultivation and fencing, its products and animals, ending 

with advice on servants. He actually shows a good deal of incidental humor and history. I 

shall illustrate some instances of his loanwords by topics, offering one or a few examples 

of each topic. He often distinguishes loans by writing them in roman letters, while his 

Norwegian text is in gothic type. Occasionally he used quotation marks. Many words he 

failed to distinguish at all. 

In reference to the word 'fence1 Hendrickson writes: "If by any chance a copy of 

this book should stray into the hands of a 'Norwegian Norseman from Norway1, or even 

turn up across the sea, it might be necessary to explain that fence means 'Gjerde.' In the 

same way many words and expressions that are used in this book have won citizenship in 

the language." (157). 

1. Measures: fire Tons paa Acren af det allerbeste Foder 'four tons per acre of the 

very best fodder' (176). Ton appears with its English plural, acre with a Norwegian 

definite form, foder is a native word. Other measures: bushel, cent, dollar, eighty (of 

land), forty (of land), hands, ounce, peck, pint, rod. 

2. Animals: som kan beskytte Renden mod Harer, Gophers, Muskrats og andre 

Smaadyr 'who can protect the ditch against rabbits, gophers, mushits and other small 

animals'; de maa bzre Hovedet vel op uden Checks 'they [the horses] must bear their 

Lena from Hendrickson in Anderson, Life Story, pp. 119-124. 

212 pages. 

3 Page 3.


heads well up, without being checked [by drivers]'. Other animals: brahmas [chickens], 

chinch bug, cut-worm, frog (Kraake) [part of a horse's leg], tick. 

3. Plants: af saadanne er--Ask--de fortrinligste, 'of such--Ash--is the most 

advantageous' (130). Other terms: American spruce, arbor vitae, michoke, be4 beet, blue 

grass, cedar, corn stalks, cottonwood, cucumber, elm, evergreens, fodder corn, hard 

maple, hickory, Hungarian grass, letuce (Salad), locust, maize, millet, Norway pine, oak 

(Eg), orchard grass, peach, pieplant, pumpkin, red top, rhubarb, Scotch pine, screenings, 

seed corn, soft maple, squash, suckers, sunflowers (Solblom), sweet corn, tamarack, 

tomatoes, walnut, whippletree, white oak 

4. Landscape: Mangt andet har vzret benyttet ti1 Render ... Asp, Tamarak, og 

lignende, endog gamle Rails, eller Brusk 'much else has been used for ditches ... aspen, 

tamarack and the like, even old rails or brush' (81). The word brusk for brush is common 

spoken immigrant terminology. Other terms: bluff, Bottom lands, highway, pasture 

(Havnegang), prairie, river, road, rolling prairie, settlement, slough, spring, subsoil, 

substitute (neuter in Norw.) 

5. Operations: hvis Maisen og Poteterne "cultiveres" ornhyggeligt 'if the maize and 

the potatoes are "cultivated" carefully' (271). The word 'cultivate' is here used in the 

Norw. passive, with attached -s. Other terms: cultivator, drain, draining system, (tobacco) 

harvest, (harvest) arbeide, hill v., job (fmlanger sine $2.00 for "jobbenn asks $2.00 for the 

"jobn 225), shell, shock, soil v., strip v., top v. 

6. Products: farmerne sorterer almindelig i 4 sorter det vil sige fillers, binders -- 

'farmers usually sort in 4 sorts, that is, fillers, binders -- (207). Other products: biscuits, 

bran, buckwheat cakes, cake, cornmeal, fillers, fish oil, gems [muffins?], graham, grain, 

gypsum, lunch, midlings, phosphates, tin (blik), tin-pan, whiskey, wrappers. 

7. Structures: I mit Barn har jeg Basement til Fjes 'in my basement I have a barn 

for cows' (198) (American-Norw. usually en barn). Other structures: barnyard, farm, 

grocery, box-stall, brick, brick house (brickhus), brickyard, brick pillar, cement, chimney, 

cistern, claim shanty, corn crib, creamery, farm, fence, fenceboard, fencestolpe 

[fencepost], fencing, floor, frame house, granary, joist, lattice work, lumber, lumber bill, 

nursery, parlor, pen, pigpen, plates, post, rail fence, rails, roofboard, shed, shed barn, 

shingle, shop, store, strawstack, studding, tobacco shed, wigwam, windbreak, windrow, 

(kitchen) wing, yard. 

8. Tools: Magen fra et Dusin Hms vil fylde omtrent en Barrel paa et Aar The 

manure fTom a dozen hens will fill about a barrel a year' (102). 

Other tools: barbs, barbed wire, board, board fence, box, brace, breaking plow, buggy 

(wheel), buggy spoke, damper, ditching spade, farm hoe, farm team, feedmill, fertilizer, 

drill, girt [girded, gravel, hand shovel, hatchet, hoe, machine fork, Mason sieve, mower, 

muleteam, nitrogen, phosphoric acid, potash, rack, rod pole, scraper, spear, stanchions, 

team (et Team), tile, tile hoe, tile spade, tobacco horse, wagon jack, weather strip, wool 

twine. 

9. General terms: de fleste Farmere har common sense 'most of the farmers have 

common sense' (82). Other terms: broadcast adv., class v., Dakota farmers, failure, foot 

rot, good for nothing, a good investment, grub in the head, gumption, (tree) peddler, 

permanent, plaster, plastered, pole rot, rotation of crops (Vexeldrift), tramp, ventilation. 

In all Hendrickson employed about 222 English loanword terms. He often 

included a Norwegian equivalent (here in parentheses). His pedagogical stance is 

consistantly apparent as well as his familiarity with and enthusiasm for farming. He died in 

Maine in 19 17, after some years (1 894- 1901) as owner and principal of Albion Academy.4 

An entirely different picture appears when we turn to the only book known to be 

written entirely in Norwegian-American ~iialect.~ This was a humorous sketch of life in 

Hofstead, pp. 89-90. 

Cf. Haugen (1953). p. 180.

' 

Schnjifestscschn~: fisay in Honor of John &Francis 

pioneer Wisconsin written in 1892 by S. H. Severson, a merchant of Stoughton, 

Wisconsin. The title is Dei Mates ve Utica (They Met in Utica), which was the name of a 

crossroads not far from Stoughton. It was located in one of the largest Norwegian- 

American districts in Wisconsin, the rural region known as Koshkonong. Severson 

described its language as "pure Telemarking," i.e. the dialect of the Telemark region in 

Norway. He granted that the language was mixed with English, "as the common people 

talk it in the settlement^."^ He deliberately avoided the bookish Dano-Norwegian of the 

immigrant newspapers, because "the topic with which the story dealt seemed especially 

suited to dialect, which should therefore serve to increase the interest of the book." 

Unhappily the book does not seem to have lived up to the author's hopes, for no other 

book by him is known. 

Severson was born in Telemark, like Hendrickson, and was brought over at the age 

of two. He spent most of his life in the Koshkonong region. There is some vacillation in 

the forms he used, but by and large it does correspond with the speech of Telemark, no 

doubt as spoken by Severson's parents in this region of all-Norwegian settlement. It is 

well-known that many of the early immigrants in Koshkonong had come from Telemark 

Severson named two of the towns in Koshkonong and spelled them phonetically: 

Stoughton became Staaten (9) and Fort Atkinson was unsually referred to as Fort (57), also 

more fully as Forteketsen (55), probably a misprint for Fort Etkensen. The double aa in 

Staaten indicates a back-rounded Norwegian vowel used in such words as Baaten 'the 

boat', while the o in Fort makes it rhyme with Norwegian kort 'short' or sort 'sort'. The 

first e in Etkensen reflects the American short open e in 'bet1, while the others are shva- 

sounds. 0 so ska eg reise te Fortetkensen 'and now I'll leave for Fort Atkinson' and 

Staaten for 'Stoughton' thus reflects a Norwegian-American pronunciation as close as one 

can get in the usual alphabet to a Nonvegian rendition of the American names. The words 

in the following section similarly reflect a spoken distortion of American sounds by the 

immigrants. 

The grammar of the English loanwords also reflects a new grammatical system. 

I. Uninflected nouns remain unmarked: tvosita Boggy 'two-seated buggy' (72); 

Bran te Kuin 'bran for the cows' (57); Brand for Kuin 'bran for the cows' (69); ingen te o 

taka care av mine Ting 'no one to take care of my things' (44); &a Pai og Keke 'make pie 

and cake' (10); de va plenty af Drikkevanu o faa 'there were plenty of dnnks to be had' 

(101); der blei License 'there got to be a license' (i.e., for drinks) (101); der fek han seg ein 

heil Kadje mz Beer 'there he got himself a whole keg of beer' (69) (also 01); alle hadde 

hatt Supper 'all had had supper' (107); so ska e snart forminske hans Stock 'then I shall 

soon reduce his stock' (88); ho ville ikje bie te Breakfast 'she wouldn't stay for breakfast1 

(62); & ska Vera so go Pris der sea de blei Timperence 'there is said to be sucb a good 

price there since it became "temperance"' (9). In some cases there is no indefinite article 

where it is needed in English: tvosita Boggy mz Sit for Drivaren frammi 'a two-seated 

buggy with a seat for the driver in front' (72); Spilernnen [sic] stemte up for Kodril 

'fiddlers tuned up for a quadrille' (35?). The same is true when a loan appears as the first 

element in a compound: han ville bes~kje dei norske countrydansane 'he wanted to visit the 

Nonvegian country dances' (35); o saa va de berre Lomber Vognir 'And then there were 

only lumber wagons' (72); ei lite Loghytte 'a little log cabin' (25); eg heve fenje meg ei 

Yankee Kjering 'I have gotten me a Yankee wifet (10). 

11. Inflected nouns are marked by (1) having a preceding indefinite article or 

adjective or (2) a suffixed definite gender marker in the singular: (a) masculine -(e)n, (b) 

feminine -i, (c) neuter -e. 

(la) Indefinite masculine: hadde netop fenje seg istand ein ny barn 'had just gotten 

himself a new barn' (36); so vil& han hava seg ei Bicycle 'he wanted to have a bicycle' (ei 

32/ein 33); so opna Per Bokji si o trekte ut ein 5 Daler Bill 'then Per opened his 

6 Page 7.


(pocket)book and drew out a 5 dollar bill' (63); ska me faa ein Bus? 'shall we take a bus?' 

(158); faa ein Dressmaker te o gjere Kjolen 'get a dressmaker to make the chess' (67); so 

blei me baae tvo tekne for ein Justice of the Peace 'then we were both taken to a Justice of 

the Peace' (22); han ikji maatte gleime o bringe ham ein Plug Tobak 'he must not forget to 

bring him a plug of tobacco' (57); ein fine Pony som han sa at han vilde hava 'a fine pony 

that he said he wanted to havet(32); de va ein Rig 'it was a rig' (72); so reiste han in paa ein 

Temperence-Saloon 'then he went into a temperance saloon' (69); ein Stebel 'a stable' (10); 

la kaan ta kaan ein Vaak yve Farmen let's take us a walk over the fann' (103). Plural: tvo 

Skjilling Yard 'two shillings per yard' (57). English -s as singular: dei tvo reiste afste paa 

Carsen 'these two left on the cars (i.e. the train)' (44); o kan faa noko Kmckers o Cheese 

'and can get some crackers and cheese' (10). 

(lb) Per o Jenny fek kver sin Kopy 'Per and Jenny got each their copy' (150); da er 

du ikke ansvarleg for nokon Expence 'then you're not responsible for any expense' (96) ; 

lnogen Bill 'any bill' (98); ho Jenny ... hadde vore god Huskiper 'Jenny ... had been a good 

housekeeper' (44) ; have ein goe Spri o Moro 'have a good spree and fun' (1 5 1). Plural: ei t 

fint Kjoletai som kosta tvo skjilling Yardi 'a fine dress material that cost two shillings the 

yard' (67); Bra o Biskitur 'bread and biscuits' (1 04). 

(24 Definite masculine: sae han te Bartenderen 'he said to the bartender' (135); te o 

gjere Bisnissen 'to do the business' (154); han selde Kalkoen saa billeg 'he sold the calico 

so cheaply' (1 1); lat ikje den Chancen gaa forbi 'don't let that chance slip by' (147); o ete 

noko taa Krackersen o Cheesen 'and eat some of the crackers and the cheese' (13); for heile 

Krouden 'for the whole crowd' (156); da dei kom forbi Depoen i Staaten 'when they came 

past the depot in Stoughton' (101); sit for Drivaren 'seat for the driver' (72); so kom 

Polisen o sette han paa Jailen 'then the police came and put him in the jail' (15); o vere me 

dei te Judjen 'go with them to the judge' (149); denne gjek han neaat Laken 'with that he 

went down to the lake' (143); so sae Polisman 'then the policeman said' (63); me maa sjaa 

yve Recorden 'we must look over the record' (129); o so reiste dei bedje te Recorder 

Officen'then theyboth wenttotherecorder'soffice'(l29); Saloonkiparen/Kiparentthe 

. saloonkeeper I the keeper' (155); du heve for yve $2000 i Tobak i Sjedn 'you have over 

$2000 in tobacco in the shed' (171) 1 Sheden (65); den Tie at den Steambaaten saak paa ein 

Sjs 'the time when that steamboat sank on a lake' (66); Brousleiven, som ho hadde paa 

Stamen 'the loaf of bread that she had on the stove' (142); berre de ikje hadde vore denne 

Trubelen ma: den fyste Kjeringi 'if only it hadn't been for this trouble with the first wife' 

(90) (2b) Definite feminine; Tobaksplantune begjynte o vise seg fint paa Fili 'the 

tobacco phts began to appear fine on the field' (100); te o seine Noti 'to sign the note' 

(133). (3a) Indefinite neuter: so ska me gaa inpaa eit Drugstore 'then we'll go into a 

drugstore' ( 10); gaa in ei t Groceri-S taar 'go in to a grocery store' (1 0); te ei t Milliner Shop 

'to a milliner's shop' (57); eit halvt Ten rnz Brand for Kuin 'a half ton of bran for the 

cows' (69); alt dette me ho Jenny o Son dm kan vere bere eit Yankeetrick 'all this with 

Jenny and your son may be just a Yankee trick' (93). With the definite article and 

adjective: Jenny reiste paa de fy ste Train te Chicago 'Jenny left on the first train to Chicago' 

(151). 

(3b) Definite neuter: da dei kom ne i Pastre 'when they came down into the 

pasture' (104); han Per hitcha up Time sitt 'Per hitched up his team' (43); garnle Meni korn 

bort i Wire Fense 'the old mare got into the wire fence' (141). 

111. Adjectives and adverbs are unmarked; eg tenkje de ;e allright 'I think it's all 

right' (1 10); de va so forbanna "dull" me Bisnissen naa 'it was so damnably "dull" with the 

business now' (128); dei va alle enige om at de had& gjenge forstrait 'they were all agreed 

that it had gone first rate' (76). 

IV. Verbs are wnj ugated in full: infinitive e or -a; present tense e, preterite -a, 

perfect participle -a

(a) Infinitive: no maa eg close up 'now I have to close up' (158); for o joine eit 

Teater-Kompani 'to join a theater company' (172); o so ska Anne vere me meg og pikke 

seg ut ein fine Kjole o Hat 'and then Anne will be with me and pick out a fine dress and a 

hat' (55); han vilde plise Kiparen 'he wanted to please the (saloon)keeperl (157); 

Kiparen ... va so flinke te o introduse han te alle dei fremmine 'the keeper was so skilled at 

introducing him to all the strangers' (157); te o seine Noti 'to sign the note' ( 133); som kan 

komme te o spile kaans goe Haap 'which can spoil our good hope' (??); ho ze ikje enno 

gamal nok te o vote 'she isn't old enough to vote yet' (14). With inf. in -a: saa kan du 

meka Man din skaffe din Del 'then you can make your husband furnish your partt (3 8) 

(b) Resent: du chaqe meg so mykje for ein halv Dags Ti 'you are charging me so 

much for a half day's timet (130); me ranne ein Resko 'we are running a risk' (148); de a 

best, du setle me ho Jenny 'it is best that you settle with Jenny' (130). 

(c) Preterite: jagu fila eg naa berre, he1 da eg hadde deg 'dam if I didn't feel better 

than when I had you' (115); han Per hitcha up Time sit 'Per hitched up his team' (43); 

Prokuratoren hanordra ein goe Slurk te 'the lawyer ordered one more good swig' (128); 

derme so tok hin honom i Skjegge o pulla de 'with that he took him by the beard; and 

pulled it' ( 1 46). 

(d) Perfect participle: eg trur ikje, at eg blir badra me noke fleire 'I don't think I11 

be bothered with any more' (100); dei fek alt sammen fixa idag 'they got everything fixed 

today' (67); eg kunne ha seiva heme for tjuge Daler 'I could have saved her twenty dollars' 

(76); naa hhre du o ser haas Alting ha tma ut, Pastor 'now you hear and see how 

everything has turned out, Pastor' (86); tvosita Boggy 'two-seated buggy'(72). 

Except for occasional vacillation Severson was remarkably accurate in his rendition 

of the actual speech of his Telemark family. On the basis of my own field work in 

Wisconsin I can endorse every one of his sentences. They could all have been spoken by 

Nonvegian immigrants in America in the old days. 

Our two authors reflect different backgrounds and purposes in their writings. Both 

are thoroughly bilingual. But while Peter Hendrickson is instructive, Severson is 

entertaining. Hendrickson uses all the words that deviate from Norwegian usage in 

reference to farm work, in order to enable Norwegian-American farmers to take an active 

part in the process of scientific farming. He often uses the Nonvegian words beside the 

English, while Severson is only concerned with having an amusing time in rendering what 

he has heard around him in his lifetime on Koshkonong Prairie. These two authors 

represent the opposite extremes of Norwegian-American writing. But neither is a purist, 

unlike many Norwegian authors with literary pretensions. 

BIBLIOGRAPHY 

Anderson, Rasmus B. Life Stov Madison, Wisconsin: 1915. 

Hendrickson, Peter. Fanning med Hoved og Haender Chicago: John Anderson, 

1886. 

Haugen, Einar. me Norwegian Longuage in America Bloornington: Indiana University 

Press, 1953. I 

Hofstead, John A. American Educators of Norwegian Origin Minneapolis: Augsburg 

Publishing House, 193 1. 

Severson, S.H. Dei rnottes ve Urica Stoughton, Wisconsin: Allberg, Swan and 

Company, 1892.

Schnfifeschrifi: Essap in Honor of John DeFrancis 

THE POLYSEMY OF THE TERM KOKUGO 

S. Robert Ramsey 

It has been more than forty years now since the appearance of John DeFrancis's book 

on nationalism and language reform in China. Few works before or since document so vividly 

China's struggles to come to terms with the linguistic realities of the modem world. But, at 

the same time, what is for many of us even more remarkable about DeFrancis's book (and 

many of his later writings) is the unabashedly passionate way in which DeFrancis combines 

linguistics with historical and sociological research. DeFrancis infects us with an enthusiasm 

for the study of language. In his hands scholarship comes to life. 

The East Asian word for "national language" is a concept DeFrancis helped us better 

understand. The term had been used in the Sinitic world since around the sixth century,' but, 

DeFrancis tells us, it took on a new meaning in the China of the 1930s. Before the May 

Fourth Movement, p op was used in reference to the languages of various non-Sinitic ethnic 

groups, or "nations"; after that time, and especially from the 1930s on, it came to mean the 

official language of the Chinese state. DeFrancis demonstrates the importance of this 

difference. He tells us the change came about as Chinese leaders moved towards nationalism 

along the lines of federali~m.~ 

As is well known, the new usage was an import from Japan. In 1902 Wu Rulun, one 

of the most famous Chinese scholars of the day, had just been appointed the Superintendent 

of Education and the new head of the reorganized Peking Imperial University. In this capacity 

Wu went on an inspection tour of Japan and, deeply impressed with the progress of 

modernization in that neighboring East Asian country, came back advocating the development

Sim-Phonic Papers, 27 (August 3 1,199 1) 

of a unified national language along similar lines.3 Wu is said to have coined the phrase 

"Unification of the National Language," and thus the popularization of the new meaning began 

with him.4 

The word Wu was so impressed with was of course kokueo. Kokueo, the Japanese 

word for their national language, is written with the same characters as the Chinese p oy. 

The term is at the most general level an example of what Victor Mair calls a "round-trip" 

word--that is, a word the Japanese first borrowed from classical Chinese sources; subsequently 

altered the meaning, tailoring it to fit some new (usually Western) concept; then finally sent 

in its new meaning back to its country of origin, China. The description in many ways fits 

kokueo. But there is more to the story, as we shall see, in part, in what follows. 

In traditional times the Japanese on occasion wrote the word in its classical Chinese 

sense. Mair cites a Japanese Buddhist work from 1714 in which the word appeared; there it 

referred to colloquial spoken Japanese; but the usage was one any educated Chinese of the 

time would surely have understood. The Japanese were after all, at least from the Chinese 

perspective, a local ethnic group. Their speech was not yet an official state language of any 

country. 

Somewhat later, in the nineteenth century, we begin to encounter kokugo in what are 

clearly non-Chinese uses. By this time the Japanese have been stimulated by ideas of 

nationality from the West. In his 1815 book Rangaku kotohaiime Sugita Genpaku provides 

a fascinating example of what it meant in those days: 

Thus, because of an eccentric nature, I became a disciple of Mr. Aoki and studied 

Dutch horizontal writing and twelve nation's lanmaees (kokueo) [written with] it6

Schnflestschriij: &says in Honor of John &Francis 

Here Sugita, who did much to diffuse Western medical knowledge through his translations of 

Dutch writings, gives us a glimpse of the foreign learning in which he had steeped himself. As 

the example shows, some Japanese scholars in the Edo period were already adapting the 

language to the new frame of reference. However, what is surely most startling about this 

example and most Edo uses of the word kokueo was not that they were different from 

anything in China, but that they were even farther from the present meaning of the Japanese 

word. Sugita's kokueo had nothing to do with the language of Japan, national or otherwise; 

he was talking about Western languages when he used it. 

A half century later the word was still used to refer to the languages of other countries, 

a meaning that from our modem perspective seems peculiarly un-Japanese. In 1866, having 

just returned from a secret trip abroad from Satsuma, Mori Arinori wrote the following: 

I have heard that the national language (kokugo) of Russia is, in Europe, the most 

difficult to learn. 

Of course, by this time kokueo could also refer to Japanese. By the late Edo the more 

familiar, narrower sense had emerged, as we can see from the following quote taken from the 

introduction to an 1856 lexical work by Murakami Hidetoshi: 

This book is compiled for the purpose of examining Western language using [our] 

nation's language (kokueo).

Sino-Plutonic Papers, 27 (August 3 1,199 1) 

Another early mention of kokueo in its modem meaning comes from a petition 

submitted in 1866 by Maeshima Hisoka7 (the man who later founded the Japanese postal 

system) to the Shogun in which he advocated the complete abolition of Chinese characters in 

Japan. Maeshima asserted that "the national language (kokuglo) should be written as simply 

as p~ssible."~ Chinese characters, which were difficult to learn, should be abolished, he 

continued, and the people encouraged to use "phonetic letters" and write in ordinary 

sentences. This was the way to spread education among the common people, Maeshima 

concluded. (It should probably be noted, however, that Maeshima drafted the petition in a 

text heavy with Chinese characters.) 

But the broad meaning of the word kokugo, referring to the language of any country, 

continued well into the Meiji period. Even much later, in mid-Meiji and nearly at the end of 

the 19th century, it could still mean "(any) nation's language." It did not have to refer only to 

the national language of Japan. The broader meaning disappeared gradually and, though 

archaic, is still given in most Japanese dictionaries--it is the first meaning listed in recent 

editions of KEjien, for instance. 

Around the turn of the century certain Japanese intellectuals were drawn more and 

more towards nationalism. For them, like the rest of their countrymen, the most important 

event of the age was the Japanese defeat of China in the Sino-Japanese War of 18941895. 

The Japanese people exulted in this victory and rejoiced in their nation's new-found power 

and prestige. The ease of Japan's victory over China brought the citizens of Meiji Japan to 

a new height of national awareness. It became a symbol of their cultural independence. 

One of these ultranationalist intellectuals was Ueda Kaz~toshi.~ As a young man, Ueda

. 

Schnfrfestschrifi: <strong>Essays</strong> in Honor of John DeFraracis 

had lived for a number of years in Europe, principally in Germany, and from Germany he had 

brought back German philological methods (Germany was the country then most advanced 

in linguistic science), as well as German influence on certain aspects of his thinking. As he 

matured, Ueda turned more and more toward the Japanese past and tradition, stressing the 

importance of Japanese national character in researching the, national language and 

literature. He considered himself a true patriot, whose mission in life was to "restore and raise 

the status of [the] Japanese language to a level above the 'yoke' of foreign (Western as well 

as Chinese) languages."1° 

Ueda's best-known work is the collection of essays entitled Kokueo no tame ("For the 

sake of the National Language"). Whatever its value for linguistic scholarship, this work is 

important because it represents the credo with which Ueda lived his life. Here are the famous 

opening lines of the first volume: 

The National Language is the bulwark of the Imperial Household; 

The National Language is the blood of the Nation.ll 

Consider also this passage taken from later in the work: 

Just as blood shows a common birth in the realm of the flesh, language, for the people 

who speak it, shows a common birth in the realm of the spirit. If we take the Japanese 

national language as an example of this, we should speak of Japanese as the spiritual 

blood of the Japanese people.12

Sino-Platonic Papers, 27 (August 3 1,199 1 ) 

Ueda's infatuation with kokug~ bordered on the obsessive. But, as we have noted, the 

word did not yet mean only the Japanese language. Let us look at some examples taken from 

Ueda's own writings: In his "National Language and the State" (Kokueo to kokka to), which 

was published in 1894, Ueda sometimes used the word to refer to Japanese and sometimes 

notn--as we see in this typical example: "Because of Luther, he [i.e., an early European 

language reformer] at first wanted to make a national language (kokugo) independent of 

Latin ..." 

When it was necessary to stress that it was Japanese in particular that was meant, the 

word kokueo alone was not enough. Even in the famous passage from Kokueo no tame cited 

above, Ueda wrote in a way that most Japanese today would find extremely curious, perhaps 

even ungrammatical. Here is the line in question: 

[I]f we take the Japanese national language (Nippon kokueo) as an example ... 

To make his emphatic point Ueda found it necessary to modlfy kokug~ with the word Nippon 

'Japan'. He meant to say, in other words, that he was talking about no other nation's 

language, just Japan's. No one would write that way today. The phrase would be nonsensical. 

Since kokueo unambiguously means Japanese and only Japanese, it cannot be qualified. To 

speak of "Japan's Kokugo" is like saying "Japan's Japanese." Obviously, the exclusionist 

meaning of the word had still not gelled when Ueda was composing Koku~o no tame. 

Ueda became the first professor of Kokugo at Tokyo Imperial University and served 

simultaneously as an official in the Ministry of Education. In this latter capacity, he succeeded 

in establishing the National Language Research Committee (Kokueo ch6sa iinkai), now called 

42

. 

S~hnfffestschri fi: <strong>Essays</strong> in Honor of John &Francis 

the Council on the National Language (Kokugo shingikai), which, among other 

accomplishments, prepared the basic materials for a national language policy. In this way 

Ueda and his colleagues moved to create a true national language and to mold a discipline-- 

Kokueogaku ('national language science')--that responded to the needs of that national 

language. 

Their linguistic nationalism was of a very Japanese kind. Central to policy was the idea 

of a national legacy shrouded in the origins of the Japanese people. For Ueda and men like 

him, reflections on the language of their nation, the "spiritual blood" of the Japanese people, 

was a mystical experience. This mystic legacy was turned to meet the needs of the nation in 

the present and the future, and it made language policy effective because the common, 

unlfylng language seemed to stem from the core of nation. This shifting of orientation was in 

many ways what modernization in Japan was all about. 

Thus was the linguistic policy produced that so impressed Wu Rulun. He envied the 

results. China needed such singularity of spirit. In the educational reform of 1903 he included 

the wording: "In every country speech is uniform throughout the whole land. Hence it is very ' 

easy to harmonize the feelings of the people in the same country ."I4 Wu was a convert to 

Japanese-style linguistic nationalism. 

Of course poy~ did not turn out to be quite the same thing as koku~o. Japanese 

policy managed to accomplish the seemingly inconsistent task of setting a rigid standard called 

kokueo and at the same time including all varieties of Japanese under the same rubric; in 

China, on the other hand, guoy referred only to the Mandarin standard. Guop stood, and 

still stands, in sharp contrast with other varieties of Chinese. Okinawan may be part of 

kokugo, but Cantonese, Fukienese, Hakka, etc. are definitely not p op. Still, the difference

Sirw-Platonic Papers, 27 (August 3 1, 199 1) 

is a detail. The important political fact remains that Wu was successful in bringing the 

Japanese concept of national language to China. 

The irony is that the term, as well as most other "round-trip words," spread to China 

only after Japan had asserted its cultural independence. It was a sign of a cultural shift in East 

Asia. China's leading role in the area had come to an end 

.

NOTES 

Schriflestschrifi: fisays in Honor of John DeFrancis 

The same characters were also used to write the name of the pre-Qin book attniuted 

to Zuo Qiuming, of course, but the meaning in thatcontext was completely different. 

See Mair pp.13-14 for details. 

DeFrancis, p. 226. 

Cf. DeFrancis, pp.43-44. 

DeFrancis, p.228. 

Mair, £11.32. 

This and the following two citations comes from KyEgoku, p.65. 

There is some confusion about the romanization of Maeshima's name. In 

Enghsh-language writing it is commonly spelled "Maejima"; cf., for example, the entry 

and biographical sketch in the Kodansha Encyclopedia of Ja~an. However, Robert 

Spaulding, the author of the article, informs me that Maeshima himself preferred the 

sh pronunciation, signing his name, in roman letters, 'W. Maesima." (personal 

- 

communication, dated 1 October 1989) 

The citation is the first one listed under the "Nihon no eeneo" definition of the entry 

for 'Xoku~o" in the Nihon kokueo daiiiten. 

The details about Ueda, his life, and his thought are largely taken from the discussion 

in Kamei et al. 

Doi, pp.267-68. 

Cited in Kamei et al., p.31. 

Cf. Kamei et al., p.204.


13. These observations about Ueda's use of the word come from Chapter 3 of Kamei et 

al. 

14. DeFrancis, p.228.

REFERENCES 

SchrIfifists~hri!j?: Essa~ in Honor of John DeFrancis 

DeFrancis, John, Nationalism and Lanrmaee Reform in China, Princeton University Press, 

1950. 

Doi Toshio, The Study of Language in Japan. A Historical Survey, Shinozaki Shorin (Tolyo), 

- - 

Kamei Takashi, et al., Nihongo no rekishi, Volume 6, Atarashii kokugo e no ayumi, Heibonsha 

(Tokyo), 1965. 

Ky6goku Okikazu, "'Nihongo' 'Kokugo' no heiycwa itsu kara?," Gekkan Nihongo, May 1990, 

pp.64-65. 

Mair, Victor, "Buddhism and the Rise of the Written Vernacular in China: The Making of a 

National Language," 1991 ms. 

Nihon Daijiten Kankolrai, Nihon kokugo daiiiten, ShEgakkan (Tokyo), 1972. 

Shinmura Izuru,' Kejien, Iwanami Shoten (Tokyo), 1955. 

t

Schrijtfestschrij?: <strong>Essays</strong> in Horwr of John DeFrancis 

Memorizing Kanji: Lessons from Pro 


Department of East Asian Languages and Lf teratures 

University of Hawaif 

1. An Extraordinary Man 

I am puzzled by the fact that so few students of Japanese know 

about Harry Loraynee= His book How to Develop a Super-power 

Memory (New York: Frederick Fell, 1957) brims over with sound 

advice on memorization that has stood the test of time. It's a 

minor classic among magicians, amateur and professional. To 

quote from the jacket of my first-edition copy: 

Tie a string around your finger? 

Carry a bulky memo pad? 

Shove pieces of paper in your pocket? 

Never again will you have to resort to makeshifts in order 

to remember an important fact. 

Now, at last, with the LORAYNE "LINK-METHOD" OF MEMORY, you 

will be able to: 

* recall faces and names even years later 

* memorize a speech or a scrdpt in minutes 

* remearber the lay or play of cards in bridge, gin, poker or 

pinochle or other cards games 

* memorize the Morse Code in 30 minutes 

* remember the entire contents of a magazine 

* have a photographic memory for a panel of numbers or objects 

* In short, remember prices, details, codes, dates, calories, 

facts, routes, events, school work, lectures--anything of 

need or interest to you! 

Harry Lorayne, who has trained his own memory to the point 

where he is acclaimed as having the most phenomenal memory in the 

world, has written the most practical, lucid and definite 

memory-training book ever written, . . . 

Now, as many students of Japanese seem to know instinctive- 

ly, the only thing better than being able to count cards inde- 

tectably at a Las Vegas blackjack table is being able to memorize 

all the joyskanji (and as many others as possible) before set- 

ting foottin Japan. To say that they are preoccupied with the 

memorization of kanji would hardly do justice to the burgeoning 

For John DeFrancis, whose perennial good humor, wide-ranging interests, and 

common sense should be a model for us all.

Sirlo-Platonic Papers, 27 (August 3 1, 199 1) 

number of study guides, sets of flash cards, learnersr diction- 

aries, workbooks, and, most recently, computer programs that 

cater to their demand for ever more potent means of making Chi- 

nese characters as instantly recognizable to them as the faces of 

their friends and relations. Leaving aside the wisdom of ap- 

proaching the task of learning how to read and write Japanese 

this way, I cannot help wondering why these students pay so 

little attention to the mnemonic techniques explained by Mr. 

Lorayne, who regularly astounded audiences by memorizing the 

faces of dozens of total strangers--together with their names, 

phone numbers, occupations, addresses, and other particulars--and 

recalling them all, in any order desired, with breath-taking 

speed and flawless accuracy. 

Harry Lorayne himself performed his feats as entertainment: 

he never pretended to be anything other than what we might call a 

mental strongman. Performers who specialize in what is called 

"mentaly1 or "head" magic make use of the same techniques but less 

openly, playing the role of a mind-reader or clairvoyant; by 

injecting a theatrical element of challenge and mystery into the 

proceedings, they make them more entertaining. Yet others use 

the same mnemonic techniques to carry out premeditated deception 

for profit.= But only a snob would pass over these methods 

because of their popularity among actors and criminals. Lorayne 

does not describe them with sophisticated academic jargon, but 

they are far from primitive; on the contrary, they represent the 

fruit of literally centuries of experimentation and practical 

experience3--deceptively simple methods good enough to fool even 

the shrewdest spectator when executed with appropriate misdirec- 

tion and sang-froid. They are definitely worth examining in 

detail, for they have much to tell us about how, and how not, to 

tackle kanji. 

2. The Three Techniques 

Lorayne recommends three techniques, which he refers to as the 

"link system, I' "peg system, and the use of "substitute words. " 

Linking refers to the imaginative association of a pair of tangi- 

ble objects or meaningful words that have no apparent connection 

with one another. For example, suppose you need to link "carpet1' 

with llpaperll--why you might want to do this will become clear 

later. You imagine (literally, try to see in your llmlndls eye") 

a single visual scene in which "carpetn and "papern are somehow 

associated. 

The associatjon must be as ridiculous as possible. For example, 

you might picture the carpet in your home made out of paper. See 

yourself walking on it, and actually hearing the paper crinkle 

under foot. You can picture yourself writlng something on a 

carpet instead of paper. Either one of these is a riddculous 

picture or association. A sheet of paper lying on a carpet would 

not make . a good association. It is too logical! Your mental

Schnj@estschn~: &says in Honor of John &Francis 

picture must be ridiculous or illogical. Take my word for the 

fact that if your association is a logical one, you will not 

remember it.. 

To assure that familiar objects will be combined in utterly 

fantastic ways, Lorayne specifically recommends (1) imagining the 

items out of proportion, (2) picturing the items in violent 

action, (3) seeing exaggerated amounts or numbers of items, or 

(4) substituting one object for another? For example, if you 

need to associate "car1' with "hamburger," you might think of your 

own car smashing into a gigantic hamburger (especially if you've 

been embarrassed by bumping you car into something and squirm 

every time you recall doing so), imagine yourself driving down 

the road behind the wheel of a giant hamburger, or picture a busy 

street filled with hundreds of honking hamburgers instead of 

cars. 

Although lllinksll form the foundation of Loraynels system, 

much of its real power comes from what he calls "pegs." It is 

based on a simple enciphering procedure for numbers. The ten 

digits are associated with consonant phonemes of English as shown 

in Table 1. 

Table 1. Lorayne's Phonemic Cipher for the Decimal Digits 

6. sh, tsh, zh, dzh 

7. k, g 

8. f, v 

9- P, b 

0. 8, 

Vowels, glides, and /h/ don't count; -only the pronunciation of 

words are considered, not their spellings.* In accordance with 

these rules, every integer is associated with a peg; Lorayne 

recommends the pegs shown in Table 2 for one- and two-digit 

numbers. The user can choose whatever pegs are most congenial 

but should stick to one peg after settling on it. Pegs for 

three-digit and longer numbers are easily constructed as needed. 

For example, the l2-digit string 633752741631 (which gives the 

first Sunday for each month of 1957, the year Lorayne published 

his book) can be remembered as "chum mug linger dishmat" (63 37 

5274 1631). 

Now suppose you have to remember an ordered list of twenty 

objects. By forging ridiculous links between the objects and the 

pegs for 1 through 20, you can, says Lorayne, easily recall an 

object given its number in the list or vice versa; what's more, 

the order in which you commit each item to memory doesn't matter. 

Linking objects to the peg words rather than directly to numbers 

is more reliable because numbers are morphemically longer than 

the pegs and lack sufficient individuality to permit memorable 

associations with random wordsT: short, semantically heterogene-


ous, ordinary words are easier to remember and link with other 

words than long, semantically homogeneous numbers. 

zoo 

tie 

Noah 

ma 

rye 

1 aw 

show 

COW 

ivy 

bee 

lace 

lot 

lion 

loom 

lure 

lily 

leech 

log 

1 ava 

lip 

Table 2. Lorayne's Recommended "Pegs" 

toes 

tot 

tin 

tomb 

tire 

towel 

dish 

tack 

dove 

tub 

cheese 

sheet 

chain 

chum 

cherry 

jail 

choo choo 

chalk 

chef 

ship 

nose 

net 

nun 

name 

Nero 

nai 1 

notch 

neck 

knife 

knob 

case 

cot 

coin 

comb 

car 

coal 

cage 

coke 

cave 

cob 

mice 

mat 

moon 

mummy 

mower 

mule 

match 

mug 

movie 

mop 

fez 

fit 

phone 

f om 

fur 

file 

fish 

fog 

fife 

fob 

rose 

rod 

rain 

ram 

rower 

roll 

roach 

rock 

roof 

rope 

bus 

bat 

bone 

bum 

bear 

be1 1 

beach 

book 

puff 

pipe 

Notice carefully that the rational phonetic cipher underly- 

ing the pegs facilitates their translation to and from numbers 

but does not participate in the linking process itself. Linking 

not only does not depend on rationality but actually works best 

in its absence; moreover, it works only if you can easily visual- 

ize both items to be linked. This requires seeing them in a 

familiar context, for the irrationality of the linkage results 

precisely from the clash between the imagined relationship be- 

tween the items and the expectations that normally flow from the 

context. If either item is something highly abstract or hard to 

visualize (e.g. equality, anger, sleep), a context may not readi- 

ly suggest itself or even exist. The "pegsvv for the numbers show 

how one copes with such a situation: first, treat an abstract 

item as part of a system (the numbers in order) rather than as an 

isolated entity: second, use ordinary words that refer to easily 

visualized things as tags for the elements of the system: third, 

associate tag words with elements of the system by means of a 

simple, compact set of rules (phonetic cipher). In short, al- 

though Lorayne does not say so explicitly, he is really offering 

two quite different approaches to memorization: a visual strate- 

gy that depends on imagining irrational gestalts: and a linguis- 

tic strategy that depends on constructing ratdonel cipher connec-

Schn$$estschn!jf: &sap in Honor of John &Francis 

tions between the elements of two sets of speech forms. Each 

strategy is better suited to those situations in which the other 

is hard to apply. 

This complementarity Is Implicit in Loraynets third tech- 

nique, which concerns the problem of remembering information that 

has no intrinsic meaning for the performer. Lorayne explains 

what he means by "meaning1' here with the well-known example of 

the lines of the musical staff marked with a treble clef: 

The letters, E, G, B, D, and F don't mean a thing. They are just 

letters, and difficuJt to remember. The sentence. "Every Good Boy 

Does Fine" does have meaning, and is something you know and under- 

stand. The new thing, the thing you had to commit to memory was 

associated with something you already knew." 

Now suppose the performer must associate faces with names. Names 

like Baker or Lincoln readily call up a visual image, but many do 

not. Loraynets description of how to deal with them is clear 

even without the accompanying cartoon illustrations: 

No. 3 is Miss Standish. I would select her "bang" hairdo. 

You could "see" people standing on the bangs and scratching them- 

selves violently because they Itch. Stand itch--Standish. Of 

course, a ddsh standing, would serve the same purpose, but I like 

an association into which I can inject some sort of action. Now 

look at Miss Standish and see the picture you've decided on, in 

your mind's eye. 

No. 4 is Mr. Smolensky. Don't let the name acare you, it's 

easy to find a substitute thought for it. I would see someone 

skiing on Mr. Smolensky's very broad nose, and taking pictures 

(while skiing) with a small camera (lens). Small lens ski--Smo- 

lensky. See how simple it is? I have chosen Mr. Smolensky's 

broad nose; you might think that the receding chin is more obvi- 

ous. Choose whichever you think is most obvious, and see the 

picture of the skier taking pictures with a small lens.= 

Notice how both the rational/phonetic' and irrational/visual 

techniques play a role here. The "meaningless" proper noun 

(typically a single morpheme) is expanded via a phonetic cipher 

(punning) into a string of morphemes associated with visualizable 

things, which form a memorably absurd constellation linked to the 

features of the face in view. True, we increase the number of 

morphemes rather than reduce it, as with the "pegs," and impose a 

ridiculous image on a real sight rather than see a "crazy" ges- 

talt inwardly, as with "links"; still, the same two techniques 

are at work. 

One application of substitute words that Lorayne recommends 

is memorizing foreign-language vocabulary items: you find an 

English phrase that "sounds like" the foreign word and visually 

linking the word's meaning to the (non)sense of the English 

phrase. A good example of this would be "G.1. Japanese" phrases 

like "a ring a toe" for ardgat8 'thankyou' and "don't touch your 

mustachen for d5 itashfmashite 'you're welcome', accompanied by a 

heroic effort to imagine the handing over of ringed toe as a 

gesture of gratitude or granting permission to leave the upper

Sino-Plutonic Papers, 27 (August 3 1. 199 1 ) 

lip untouched as the epitome of modesty. Needless to say, mne- 

monic techniques of this kind aren't good for anything but ad hoc 

memorization of isolated lexical .items, such as "boning up" for a 

test in high school or college, where foreign languages are all 

too often treated as bothersome requirements that students will 

never use once they graduate. True proficiency can obviously 

never be acquired through the purposeful imposition of an alien 

phonology and irrelevant semantics (not to mention unwarranted 

cultural expectations) onto the structures of a language that 

must be used spontaneously and without self-consciousness. We 

will revisit to this point shortly. 

Now, however, let us return to the topic that prompted this long 

digression into Harry Lorayneb bag of tricks: the memorization 

of Chinese characters. We have all seen books that attempt to 

teach kanji by rationalizing their shapes. They start with a 

picture that somehow illustrates the "meaning" of the kanjf, and 

then, by a process of gradual distortion and transformation, move 

from the picture to the visual shape of the kanji itself. Today, 

there are even computer programs that use animation to enliven 

the metamorphosis. The underlying assumption behind all these 

materials is evidently that the llmeaningless,ll unfamiliar, and 

hard to distinguish shapes of the kanji befuddle the student. By 

seeing how an inscrutable shape like -6 can be derived from a 

picture of a horse, for example, the student allegedly acquires a 

trusty mnemonic link. What do Lorayneb professional insights 

tell us about this strategy of pictorial rationalization? 

In order to make a proper evaluation, we need to specify 

goals in terms of which we can measure success or failure. 

Suppose the goal is to recall the of a hundred ken ji 

at sight as quickly as possible. (This is analogous to Harry 

correctly calling out the name of the first person to stand up in 

a large group of people he had not met before greeting each as he 

or she entered ,the room.) For the sake of argument, let's assume 

that the "meanings" are expressed as English tag wordsLo; the 

only qualification is that the tags must reflect the sense of 

words in which the kanjf are actually found in written Japanese. 

The appearance of a kanjJ is to a person's face as its tag is the 

person's name, so the student needs, according to Lorayne, to see 

in the image of the kanjd a bizarre thing or event that so_mehow 

brings to mind the tag word. For example, I could imagine 3 as 

the face of a man in profile, facing left, wincing in anguish, 

his right hand slapped over his brow and eyes (crossed lines at 

top depict fingers), screaming in despair at the top of his lungs 

(mouth wide open, four dots for teeth and tongue) in a hoarse 

voice. Hoarse--horse. No good seeing a horse ftself in the 

kanjd--too logical! If Lorayne is right, pictorial rationalization 

is wrong.

Schriftesrsch~ij: fisays in Honor of John &Francis 

Next consider a Japanese dictation test. This is like Harry 

running over to a person in the audience as soon as his or her 

name or phone number is called out: the Japanese word or 

word-fragment to be represented by the kanji is the name; the 

kanjd itself is the person's face. Now for persons fluently 

literate in Japanese (even non-natives), the kenjd and their 

readings have become so familiar that a character is often said 

to be or to mean a word it is used to write. Pointing to , 

for instance, one might say Kore we ksku to iu jd desu ~hi; is 

the character for l'write" or Kono ji wa kaku desu 'This character 

is lor means] "writelll by way of identification. In reality, 

of course, 5t is the word kaku, not the character g , that has 

the meaning 'write1. Meaning simply "rubs off" on the character 

by vlrtue ofthe role or roles it plays in the orth~graphy.~~ 

Nevertheless, almost everyone slips into the habit of talking 

sloppily about kanji as if they directly symbolized the sense of 

their readings. What connection could have greater logical power 

than that? So we have a paradox: the naive learner, following 

Loraynets method, needs to forge an fllogfcal link between a 

Japanese word and a character so that hearing the former will 

conjure up the Image of the latter; yet improvement in reading 

occurs precisely as this linkage becomes less and less illogical. 

Indeed, the sign of total mastery is that the linkage is so 

lflogical" that any other linkage becomes literally unthinkable! 

One way out of this paradox is to resort to the "G.I. Japanese" 

already mentioned, i.e., "hearing" Japanese words in terms 

. of another, completely unrelated language system. For the reasons 

stated earlier, this is clearly not acceptable unless one is 

interested merely in passing puzzle-like quizzes and tests. 

A second solution would ask the student to dream up ridiculous 

linking thoughts in Japanese. This might work for native 

speakers, but it's inconceivable that a non-native student who 

knew enough Japanese to play Lorayne's game in Japanese would 

need mnemonic devices to memorize kanjf in the first place! 

There is a third way out, but it requires a curious inversion. 

We give each kanji a name that "fits1' it visually. (Harry 

doesn't have the luxury of giving members of the audiences nicknames 

in his act, but it's a feasible option for the student.) 

This name, or '9neaning," is based on the appearance of the kanji; 

any connection between the name and the actual flmeaningfl or 

reading of the kanji is strictly fortuitous. The readings, 

guaranteed to clash vividly with the names, can now serve as 

linking thoughts. 

This clever strategy is actually used in a two-volume study 

guide, James Heisigls Rememberdng the Kanji.%= Heisig develops 

an elaborate theory for giving English names to each kanjf in 

which traditional "radicals" and other "primitive" graphic components 

are given names based on their shape; whole characters are 

analyzed in terms of these lrprimitivesw and given names accordingly. 

Heisig likewise gives an English name to each kana. Each 

Japanese reading is thought of as the sequence of English words

Sino-Platonic Papers, 27 (August 3 1, 1991) 

corresponding to the string of hlragana used to spell it out. To 

associate a kanji with a reading, "alln the student has to do is 

imagine a single image in which the kanji name and the 

reading-generated phrase all come together into one vividly weird - 

picture. (Weirdness is, in fact, almost impossible to avoid!) 

I once had a chance to talk with Heisig in person. I asked 

him whether he had ever been an amateur magician. He said no, 

suggesting that mnemonic techniques such as Loraynels are period- 

ically "redis~overcd.'~ In any case, Heisigls method is uncannily 

similar to Loraynels. Note, however, that it has nothing whatso- 

ever to do with reading Japanese as Japanese;' it is a 

thorough-going technique for memorizing the equivalent of a 

dictionary, much as a magician might memorize the order of the 

cards in a stacked deck--a neat trick,13 but not the secret to 

winning pokerOiJ 

Returning to Lorayne, note especially how his system, and 

those of many other magicians, makes use of interlocking patterns 

of speech. Bizarre visual imagination needs to be used in con- 

junction with systematic phonetic association; the techniques are 

complementary, with little overlap in their domains of useful- 

ness. This suggests that successful readers and writers of 

Japanese must have a thorough grasp of the spoken language in 

order to reach their level of accomplishment, and that, at the 

outset at least, they make use, not of "logicalw links between 

kanji and readings, but of idiosyncratic, illogical, and perhaps 

even embarrassing -associations of kanji forms with particular - 

words or word-fragments. 

One thing is certain: foreign students who attempt to 

bypass the spoken language and "crack the code" of kanji directly - 

through pictorial rationalizations are virtually certain to hit a 

plateau beyond which they will make little or no progress. The 

best they can do is exploit Heisigfs gambit; unfortunately, that 

maneuver has nothing to do with the actual reading of Japanese. 

If only our kanji-obsessed beginnlng students understood how 

Harry and the other pros pull off those seemingly miraculous 

feats of mental magic! 

Endnotes 

1. This paper deals with Japanese, but the argument naturally applies, muta- 

tis mutandis, to Mandarin as well. 

2. See James Randi, The Faith Healers (Buffalo: Prometheus Books, 1987), 

especially pp. 39-44. 

3. Leaving aside legends of prodigious memory that come down to us from - 

classical antiquity, one can trace mnemonic techniques in the West back to at 

least the Middle Ages. See Martin Gardner, Logic Machfnes and Diagrams, 2nd 

ed. (Chicago: University of Chicago Press, 1982), 1-27 on Ramon Lull. The 

techniques of memorization popular among Renaissance churchmen and their flrst 

encounter with the problem of memorizing Chinese characters are described by 

Jonathan Spence in The Memory Palace of Matteo Ricci (New York: Viking.

Schriffestschriiji: &say in Honor of John DeFrancis 

-1984). Significantly, Lull was a key influence on the "proto-sinologist" 

Athanasius Klrcher, discussed in D. E. Mungello, Curious Land: Jesuft Accom- 

modation and the Origins of Sinology (Honolulu: University of Hawaii Press, 

1989), 174-188. 

4. Lorayne, Super-power Memory, 40; emphasis in original. 

5. Lorayne , Super-power Memory, 45-46. 

6. Judging from examples in the book, it appears that the digraph is 

treated as if it were /t/ + /h/; likewise, is taken as /n/ + g , at 

least when pronounced as in finger. 

7. There are of course rare individuals, like the Indian prodigy Srinivasa 

Ramanujan (1887-1920), for whom numbers seem to be have individual personali- 

ties. British mathematician J. E. Littlewood remembered "once going to see 

him when he was lying ill at Putney. I had ridden in taxicab number 1729, and 

remarked that the number seemed to me rather a dull one, and that I hoped it 

was not an unfavorable omen. 'No,' he replied, 'it is a very interesting 

number; it is the smallest number expressible as the sum of two cubes in two 

different ways'" (G. H Hardy, A Mathematician's Apology (London: Cambridge 

University Press, 1940). If you have memorized the first dozen cubes, you 

might notice that 1,7.29 = 1,728 + 1 = 12" + la = 1,000 + 729 = 10m + grn but, 

unless Ramanujan was simply recalling a result he had found at leisure earli- 

er, it is remarkable he could assert immediatelythat no smaller integer is 

likewise decomposable. 

I have been told that Ramanujan's epitaph reads, "Every integer was his 

friendw--si non G vero, 6 ben trovato. 

8. Lorayne, Super-Memory, 16; emphasis in original. 

9. Lorayne, Super-Memory, 144. 

10. In fact, of course, the so-called meanings of kanji are just a by-product 

of how they are used in the orthography of Japanese. It is the morphemes of 

Japanese that convey meaning; the kanji are merely semantic parasites and 

sometimes correspond to something more or leas than an actual morpheme. 

I 

11. Indeed, -:g stands for a noun in Mandarin, shB 'book ' , and only for the 

invariant ka part of Japanese kaku, which has dozens of other forms including 

kad ta, okaki ni narimasu, and kakaserarenakat tara. 

12. James W. Heisig, Remembering the Kanji I: A Complete Course on How Not 

to Forget the Meaning and Writing of Japanese Characters (1977); 11: A Sys- 

tematic Guide to Reading Characters (1987). T5ky5: Japan Publications Trad- 

ing Company. 

13. Actually, a whole act can be developed around this one technique. The 

best known method Is the so-called Nikola System, which makes it possible to 

execute a mystifying group of seemingly unrelated effects. See Jean Hugard & 

John J. Crimmins, Jr., eds., Encyclopedia of Card Tricks (London: Faber & 

Faber, 1961).


14. At a conference in June 1985 at Nagoya, I heard a story about an Egyptian 

student studying in Japan. He decided to comlt Nelson's Japanese-English 

character dictionary to memory the same way he had memorized the Qu'ran as a 

child. He was greatly dismayed to discover that his monumental and largely 

successful effort resulted in virtually no improvement in his Japanese reading 

or speaking ability.

S~hnftft~ch!: &sap in Honor of John DeFranciis 

Why Chinese Is So Damn Hard 

by 

David Moser 

Dept of Asian Languages and Cultures 

University of Michigan 

The first question any thoughtful person might ask when reading the title of this essay is, 

"Hard for whom?" A reasonable question. After all, Chinese people seem to learn it just fine. 

When little Chinese kids go through the "temble twos", it's Chinese they use to drive their 

parents crazy, and in a few years the same kids are actually using those impossibly complicated 

Chinese characters to scribble love notes and shopping lists. So what do I mean by "hard"? 

Since I know at the outset that the whole tone of this document is going to involve a lot of 

whining and complaining, I may as well come right out and say exactly what I mean. I mean 

hard for me, a native English speaker trying to learn Chinese as an adult, going through the 

whole process with the textbooks, the tapes, the conversation partners, etc., - the whole 

torturous rigamarole. I mean hard for me - and, of course, for the many other Westerners who 

have spent years of their lives bashing their heads against the Great Wall of Chinese. 

If this were as far as I went, my statement would be a pretty empty one. Of course Chinese 

is hard for me. After all, any foreign language is hard for a non-native, right? Well, sort of. Not 

all foreign languages are equally difficult for any learner. It depends on which language you're 

coming fiom. A French person can usually learn Italian faster than an American, and an average 

American could probably master German a lot faster than an average Japanese, and so on. So 

part of what I'm contending is that Chinese is hard compared to ... well, compared to almost any 

other language you might care to tackle. What I mean is that Chinese is not only hard for ur 

(English speakers), but it's also hard in absolute terms. Which means that Chinese is also hard 

for them, for Chinese people.' 

If' you don't believe this, just ask a Chinese person. Most Chinese people will cheerfully 

acknowledge that their language is hard, maybe the hardest on earth. (Many are even proud of 

this, in the same way some New Yorkers are actually proud of living in the most unlivable city in 

America.) Maybe all Chinese people deserve a medal just for being born Chinese. At any rate, 

they generally become aware at some point of the Everest-like status of their native language, as 

they, from their privileged vantage point on the summit, observe foolhardy foreigners huffing 

and puffing up the steep slopes. 

Everyone's heard the supposed fact that if you take the English idiom "It's Greek to me" and 

search for equivalent idioms in all the world's languages to arrive at a consensus as to which 

language is the hardest, the results of such a linguistic survey is that Chinese easily wins as the . 

canonical incomprehensible language. (For example, the French have the expression "C'est du 

chinois", "It's Chinese", i.e., "It's incomprehensible". Other languages have similar sayings.) 

So then the question arises: What do the Chinese themselves consider to be an impossibly hard 

lanpage? You then look for the corresponding phrase in Chinese, and you find "Gzn tio'nshz 

y&ang", meaning "It's like heavenly script." 

There is n th in this linguistic yarn; Chinese does deserve its reputation for heartbreaking 

difficulty. Those who undertake to study the language for any other reason than the sheer joy of 

it will always be frustrated by the abysmal ratio of effort to effect. Those who are actually 

attracted to the language precisely because of its daunting complexity and difficulty will never be 

disappointed. Whatever the reason they started, every single person who has undertaken to study 

Chinese sooner or later asks themselves ''Why in the world am I doing this?" Those who can still 

remember their original goals will wisely abandon the attempt then and there, since nothing could 

be worth all that tedious struggle. Those who merely say "I've come this far - I can't stop 

I am spealung of the writkg system here, but the difficulty of the writing system has such a pervasive effect on 

literacy and general language mastery that I think the statement as a whole is still valid. 

59

Sino-Plutonic P ap, 27 (August 3 1,199 1) 

now" will have some chance of succeeding, since they have the kind of mindless doggedness and 

lack 'of sensible overall perspective that it takes. 

Okay, having explained a bit of what I mean by the word, I return to my original question: 

Why is Chinese so damn hard? 

1. Because the writing system is ridiculous. 

Beautiful, complex, mysterious - but ridiculous. I, like many students of Chinese, was 

frst attracted to Chinese because of the writing system, which is surely one of the most 

fascinating scripts in the world The more you learn about Chinese characters the more intriguing 

and addicting they become. The study of Chinese characters can become a lifelong obsession, 

and you soon find yourself engaged in the daily task of accumulating them, drop by drop from 

the vast sea of characters, in a vain attempt to hoard them in the leaky bucket of long-term 

memory. 

The beauty of the characters is indisputable, but as the Chinese people began to realize the 

importance of universal literacy, it became clear that these ideograms were sort of like bound 

feet- some fetishists may have liked the way they looked, but they weren't too practical for daily 

use. 

For one thing, it is simply unreasonably hard to learn enough characters to become 

functionally literate. Again, someone may ask "Hard in comparison to what?" And the answer is 

easy: Hard in comparison to Spanish, Greek, Russian, Hindi, or any other sane, "normal" 

language that requires at most a few dozen symbols to write anything in the language. John 

DeFrancis, in his book The Chinese Language: Fact and Fantasy, reports that his Chinese 

colleagues estimate it takes seven to eight years for a Mandarin speaker to learn to read and write 

three thousand characters, whereas his French and Spanish colleagues estimate that students in 

their respective countries achieve comparable levels in half that time.2 Naturally, this estimate is 

rather crude and impressionistic (it's unclear what "comparable levels" means here), but the 

overall implications are obvious: the Chinese writing system is harder to learn, in absolute terms, 

than an alphabetic writing ~ystern.~ Even Chinese kids, whose minds are at their peak absorptive 

power, have more trouble with Chinese characters than their little counterparts in other countries 

have with their respective scripts. Just imagine the difficulties experienced by relatively sluggish 

post-pubescent foreign learners such as myself. 

Everyone has heard that Chinese is hard because of the huge number of characters one has to 

learn, and this is absolutely true. There are a lot of popular books and articles that downplay this 

difficulty, saying things like "Despite the fact that Chinese has [10,000, 25,000, 50,000, take 

your pick] separate characters you really only need 2,000 or so to read a newspaper". 

Poppycock. I couldn't comfortably read a newspaper when I had 2,000 characters under my 

belt. I often had to look up several characters per line, and even after that I had trouble pulling 

the meaning out of the article. (I take it as a given that what is meant by "read" in this context is 

"read and basically comprehend the text without having to look up dozens of characters"; 

otherwise the claim is rather empty.) 

This fairy tale is promulgated because of the fact that, when you look at the character 

frequencies, over 95% of the characters in any newspaper are easily among the first 2,000 most 

John DeFrancis, The Chinese Lunguage: Fact and Fmtasy. Honolulu: University of Hawaii Press, 1984, p.153. 

Most of the issues in this paper are dealt with at length and with great clarity in both this book and in his Visible 

Speech: The Diverse Oneness of Writing System, Honolulu: University of Hawaii Press, 1989. 

Incidentally, I'm aware that much of what I've said above applies to Japanese as well, but it seems clear that the 

burden placed on a learner of Japanese is much lighter because (a) the number of Chinese characters used in 

Japanese is "only" about 2,000 - fewer by a factor of two or three compared to the number needed by the average 

literate Chinese reader, and (b) the Japanese have phonetic syllabaries (the hiraganu and k atah characters), which 

are nearly 100% phonetically reliable and are in many ways easier to master than chaotic English orthography is. 

.

Schnfrfesischchrifi: Ersa~ in Honor of John DeFrar~cis 

common ones.4 But what such accounts don't tell you is that there will still be plenty of 

unfamiliar words made up of those familiar characters. (To illustrate this problem, note that in 

English, knowing the words "up" and "tight" doesn't mean you know the word "uptight".) Plus, 

as anyone who has studied any language knows, you can often be familiar with every single 

word in a text and still not be able to grasp the meaning. Reading comprehension is not simply a 

matter of knowing a lot of words; one has to get a feeling for how those words combine with 

other words in a multitude of different contexts? In addition, there is the obvious fact that even 

though you may know 95% of the characters in a given text, the remaining 5% are often the very 

characters that are crucial for understanding the main point of the text. A non-native speaker of 

English reading an article with the headline "JACUZZIS FOUND EFFECTIVE IN TREATING 

PHLEBITIS" is not going to get very far if they don't how the words "jacuzzi" or "phlebitis". 

The problem of reading is often a touchy one for those in the China field. How many of us 

would dare stand up in front of a group of colleagues and read a randomly-selected passage out 

loud? Yet inferiority complexes or fear of losing face causes many teachers and students to 

become unwitting cooperators a kind of conspiracy of silence wherein everyone pretends that 

after four years of Chinese the diligent student should be whizzing through anything from 

Confucius to Lu Xun, pausing only occasionally to look up some pesky low-frequency character 

(in their Chinese-Chinese dictionary, of course). Others, of course, are more honest about the 

difficulties. The other day one of my fellow graduate students, someone who has been studying 

Chinese for ten years or more, said to me "My research is really hampered by the fact that I still 

just can't read Chinese. It takes me hours to get through two or three pages, and I can't skim to 

save my life." This would be an astonishing admission for a tenth-year student of, say, French 

literature, yet it is a comment I hear all the time among my peers (at least in those unguarded 

moments when one has had a few too many Tsingtao beers and has begun to lament how slowly 

work on the thesis is coming). 

A teacher of mine once told me of a game he and a colleague would sometimes play: The 

contest involved pulling a book at random from the shelves of the Chinese section of the Asia 

Library and then seeing who could be the first to figure out what the book was about. Anyone 

who has spent time working in an East Asia collection can venfy that this can indeed be a dfficult 

enough task - never mind reading the book in question. This state of affairs is very 

disheartening for the student who is impatient to begin feasting on the vast riches of Chinese 

literature, but must subsist on a bland diet of canned handouts, textbook examples, and carefully 

edited appetizers for the first few years. 

The comparison with learning the usual western languages is striking. After about a year of 

studying French, I was able to read a lot I went through the usual kinds of novels - La nnuske 

by Sartre, V01~'s Candide, L'btranger by Camus - plus countless newspapers, magazines, 

comic books, etc. It was a lot of work but fairly painless; all I really needed was a good 

dictionary and a battered French grammar book I got at a garage sale. 

This kind of "sink or swim" approach just doesn't work in Chinese. At the end of three 

years of learning Chinese, I hadn't yet read a single complete novel. I found it just too hard, 

impossibly slow, and unrewarding. Newspapers, too, were still too daunting. I couldn't read an 

article without looking up about every tenth character, and it was not uncommon for me to scan 

the front page of the People's Daily and not be able to completely decipher a single headline. 

Someone at that time suggested I read The Dream of the Red Chamber and gave me a nice threevolume 

edition. I just have to laugh. It still sits on my shelf like a fat, smug Buddha, only the 

first twenty or so pages frlled with scribbled definitions and question marks, the rest crisp and 

virgin. After six years of studying Chinese, I'm still not at a level where I can actually read it 

without an English translation to consult. (By "read it", I mean, of course, "read it for pleasure". 

See, for ex., Chen Hegin, "Yutiwen yingyong zihui" [Characters used in vernacular literature], Shanghai, 1928. 

John DeFrancis deals with this issue, among other places, in 'Why Johnny Can't Read Chinese", Jourml of the 

Chinese Language Teachers Association, Vol. 1, No. 1, Feb. 1966, pp. 1-20.

Sino-PEatonic Papers, 27 (August 3 1, 199 1) 

I suppose if someone put a gun to my head and a dictionary in my hand, I could get through it.) 

Simply diving into the vast pool of Chinese in the beginning is not only foolhardy, it can even be 

counterproductive. As George Kennedy writes, "The difficulty of memorizing a Chinese 

ideograph as compared with the difficulty of learning a new word in a European language, is 

such that a rigid economy of mental effort is imperative."6 This is, if anything, an 

understatement. With the risk of drowning so great, the student is better advised to spend more 

time in the shallow end treading water before heading toward the deep end. 

As if all this weren't bad enough, another ridiculous aspect of the Chinese writing system is 

that there are two (mercifully overlapping) sets of characters: the traditional characters still used in 

Taiwan and Hong Kong, and the simplified characters adopted by the People's Republic of China 

in the late 1950's and early 60's. Any foreign student of Chinese is more or less forced to 

become familiar with both sets, since they are routinely exposed to textbooks and materials from 

both Chinas. This linguistic camel's-back-breaking straw puts an absurd burden on the already 

absurdly burdened student of Chinese, who at this point would gladly trade places with 

Sisyphus. But since Chinese people themselves are never equally proficient in both simplified 

and complex characters, there is absolutely no shame whatsoever in eventually concentrating on 

one set to the partial exclusion the other. In fact, there is absolutely no shame in giving up 

Chinese altogether, when you come right down to it. 

2. Because the language doesn't have the common sense to use an alphabet. 

To further explain-why the Chinese writing system is so hard in this respect,-it might be a 

good idea to spell out (no pun intended) why that of English is so easy. Imagine the kind of task 

faced by the average Chinese adult who decides to study English. What skills are needed to 

master the writing system? That's easy: 26 letters. (In upper and lower case, of course, plus 

script and a few variant forms. And throw in some quote marks, apostrophes, dashes, 

parentheses, etc. - all things the Chinese use in their own writing system.) And how are these 

letters written? From left to right, horizontally, across the page, with spaces to indicate word 

boundaries. Forgetting for a moment the problem of spelling and actually making words out of 

these letters, how long does it take this Chinese learner of English to master the various 

components of the Enghsh writing system? Maybe a day or two. 

Now consider the American undergraduate who decides to study Chinese. What does it take 

for this person to master the Chinese writing system? There is nothing that corresponds to an 

alphabet, though there are recurring components that make up the characters. How many such 

components are there? Don't ask. As with all such questions about Chinese, the answer is very 

messy and unsatisfymg. It depends on how you define "component" (strokes? radicals?), plus a 

lot of other tedious details. Suffice it to say, the number is quite large, vastly more than the 26 

letters of the Roman alphabet. And how are these components combined to form characters? 

Well, you name it - components to the left of other components, to the right of other 

components, on top of other components, surrounding other components, inside of other 

components - almost anything is possible. And in the process of making these spatial 

accommodations, these components get flattened, stretched, squashed, shortened, and dis toned 

in order to fit in the uniform square space that all characters are supposed to fit into. In other 

words, the components of Chinese characters are arrayed in zwo dimensions, rather than in the 

neat one-dimensional rows of alphabetic writing. 

Okay, so ignoring for the moment the question of elegance, how long does it take a 

Westerner to learn the Chinese writing system so that when confkonted with any new character 

they at least know how to move the pen around in order to produce a reasonable facsimile of that 

character? Again, hard to say, but I would estimate that it takes the average learner several 

months of hard work to get the basics down. Maybe a year or more if they're a klutz who was 

never very good in art class. Meanwhile, their Chinese counterpart learning English has zoomed 

ahead to learn cursive script, with time left over to read Moby Dick, or at least Smnk & White. 

George Kennedy, "A Minimum Vocabulary in Modem Chinese", in Selected Works of George Kennedy, Tien-yi 

Li (ed.), New Haven, 1964, p. 8.

S~hr~ftfest~~h~?!!: &.say irr Honor of John DeFrancis 

This is not exactly big news, I how; the alphabet really is a breeze to learn. Chinese people 

I know who have studied English for a few years can usually write with a handwriting style that 

is almost indistinguishable from that of the average American. Very few Americans, on the other 

hand, ever learn to produce a natural calligraphic hand in Chinese that resembles anythmg but that 

of an awkward Chinese third-grader. If there were nothing else hard about Chinese, the task of 

learning to write characters alone would put it in the rogues' gallery of hard-to-learn languages. 

3. Because the writing system just ain't very phonetic. 

So much for the physical process of writing the characters themselves. What about the sheer 

task of memorizing so many characters? Again, a comparison of English and Chinese is 

instructive. Suppose a Chinese person has just the previous day learned the English word 

"president", and now wants to write it from memory. How to start? Anyone with a year or two 

of English experience is going to have a host of clues and spelling rules-of-thumb, albeit 

imperfect ones, to help them along. The word really couldn't start with anything but "pr", and 

after that a little guesswork aided by visual memory ("Could a 'z' be in there? That's an unusual 

letter, I would have noticed it, I think. Must be an 's' ...") should produce something close to the 

target. Not every foreigner (or native speaker for that matter) has noted or internalized the 

various flawed spelling heuristics of English, of course, but they are at least there to be utilized. 

Now imagine that you, a learner of Chinese, have just the previous day encountered the 

Chinese word for "president" (,g, & zangthg) and want to write it What processes do you go 

through in retrieving the word? Well, very often you just totally forget, with a forgetting that is 

both absolute and perfect in a way few things in this life are. You can repeat the word as often as 

you like; the sound won't give you a clue as to how the character is to be written. After you learn 

a few more characters and get hip to a few more phonetic components, you can do a bit better. 

(''2&tg 

,g, is a phonetic component in some other character, right? ... Song? Zeng? Oh yeah, 

c5g as in c5ngmfng a' .") Of course, the phonetic aspect of some characters is more 

obvious than that of others, but many characters, including some of the most high-frequency 

ones, give no clue at all as to their pronunciation. 

All of this is to say that Chinese is just not very phonetic when compared to English. 

(English, in turn, is less phonetic than a language like Geman or Spanish, but Chinese isn't even 

in the same ballpark.) It is not true, as some people outside the field tend to hnk, that Chinese is 

not phonetic at all, though a perfectly intelligent beginning student could go several months 

without noticing this fact. Just how phonetic the language is a very complex issue. Educated 

opinions range from 25% (Zhao Y~anren)~ to around 66% @eFrancis),* though the latter 

estimate assumes more knowledge of phonetic components than most learners are likely to have. 

One could say that Chinese is phonetic in the way that sex is aerobic: technically so, but in 

practical use not the most salient thing about it. Furthennore, this phonetic aspect of the language 

doesn't really become very useful until you've learned a few hundred characters, and even when 

you've learned two thoosand, the feeble phoneticity of Chinese will never provide you with the 

constant memory prod that the phonetic quality of English does. 

Which means that often you just completely forget how to write a character. Period. If there 

is no obvious semantic clue in the radical, and no helpful phonetic component somewhere in the 

character, you're just sunk. And you're sunk whether your native language is Chinese or not; 

contrary to popular myth, Chinese people are not born with the ability to memorize arbitrary 

squiggles. In fact, one of the most gratifying experiences a foreign student of Chinese can have 

is to see a native speaker come up a complete blank when called upon to write the characters for 

some relatively common word. You feel an enormous sense of vindication and relief to see a 

native speaker experience the exact same difficulty you experience every day. 

Zhao Yuanren, Aspects of Chinese Sociolinguistics. Anwar S . Dil (ed.). Stanford: Stanford University Press, 

1976, p. 92. 

John DeFrancis, The Chinese Language: Fact and Fantasy, p. 109.


This is such a gratifying experience, in fact, that I have actually kept a list of characters that I 

have observed Chinese people forget how to write. (A sick, obsessive activity, I know.) I have 

seen highly literate Chinese people forget how to write certain characters in common words like 

"tin can", "knee", "screwdriver", "snap" (as in "to snap one's fingers"), "elbow", "ginger", 

"cushion", "firecrackeryy, and so on. And when I say "forgetyy, I mean that they often cahnot even 

put the first stroke down on the paper. Can you imagine a well-educated native English speaker 

tot~l2-y forgetting how to write a word like "knee" or "tin can"? Or even a rarely-seen word like 

"scabbard" or "ragamuffin"? No matter how low-frequency the word is, or how unorthodox the 

spelling, the English speaker can always come up with something, simply because there has to be 

some correspondence between sound and spelling. One might forget whether "abracadabray' is 

hyphenated or not, or get the last few letters wrong on "rhinoceros", but even the poorest of 

spellers can make a reasonable stab at almost anythmg. By contrast, often even the most welleducated 

Chinese have no recourse but to throw up their hands and ask someone else in the room 

how to write some particularly elusive character. 

As one mundane example of the advantages of a phonetic writing system, here is one kind of. 

linguistic situation I encountered constantly while I was in France. (Again I use French as my 

canonical example of an "easy" foreign language.) I wake up one morning in Paris and turn on 

the radio. An ad comes on, and I hear the word "amortisseur" several times. "What's an 

amortisseur?" I think to myself, but as I am in a hurry to make an appointment, I forget to look 

the word up in my haste to leave the apartment. A few hours later I'm walking down the street, 

and I read, on a sign, the word "AMORTISSEUR" - the word I heard earlier this morning. 

Beneath the word on the sign is a picture of a shock absorber. Aha! So "amortisseur" means 

"shock absorberyy. And voila! I've learned a new word, quickly and painlessly, all because the 

sound I construct when reading the word is the same as the sound in my head from the radio this 

morning - one reinforces the other. Throughout the next week I see the word again several 

times, and each time I can reconstruct the sound by simply reading the word phonetically - "amor-tis-seur". 

Before long I can retrieve the word easily, use it in conversation, or write it in a 

letter to a friend. And the process of learning a foreign language begins to seem less daunting. 

When I first went to Taiwan for a few months, the situation was quite different. I was 

awash in a sea of characters that were all visually interesting but phonetically mute. I carried 

around a little dictionary to look up unfamiliar characters in, but it's almost impossible to look up 

a character in a Chinese dictionary while walking along a crowded street (more on dictionary 

look-up later), and so I didn't get nearly as much phonetic reinforcement as I got in France. In 

Taiwan I could pass a shop with a sign advertising shock absorbers and never know how to 

pronounce any of the characters unless I first look them up. And eventhen, the next time I pass 

the shop I might have to look the characters up again. And again, and again. The reinforcement 

does not come naturally and easily. 

4. Because you can't cheat by using cognates. 

I remember when I had been studying Chinese very hard for about three years, I had an 

interesting experience. One day I happened to find a Spanish-language newspaper sitting on a 

seat next to me. I picked it up out of curiosity. 'Wmm," I thought to myself. "I've never studied 

Spanish in my life. I wonder how much of this I can understand." At random I picked a short 

article about an airplane crash and started to read. I found I could basically glean, with some 

guesswork, most of the information from the article. The crash took place near Los Angeles. 

186 people were killed. There were no survivors. The plane crashed just one minute after take- 

off. There was nothing on the flight recorder to indicate an critical situation, and the tower was 

unaware of any emergency. The plane had just been serviced three days before and no 

mechanical problems had been found. And so on. After finishing the article I had a sudden 

discouraging realization: Having never studied a day of Spanish, I could read a Spanish 

newspaper more easily than I could a Chinese newspaper afer more than three years of studying 

Chinese. 

What was going on here? Why was this "foreign" language so transparent? The reason was 

obvious: cognates - those helpful words that are just English words with a little foreign make-

Schn$$estschri/: <strong>Essays</strong> irr Homr of John DeFrancis 

up.9 I could read the article because most of the operative words were basically English: 

(C 

"aeropuerto" , "problems rnecdnico" , un minuto", "situcidn critica" , "emergencia", etc. 

Recognizing these words as just English words in disguise is about as difficult as noticing that 

Superman is really Clark Kent without his glasses. That these quasi-English words are easier to 

learn than Chinese characters (which might as well be quasi-Martian) goes without saying. 

Imagine you are a diabetic, and you find yourself in Spain about to go into insulin shock. 

You can rush into a doctor's office, and, with a minimum of Spanish and a couple of pieces of 

guesswork ("diabetes" is just "diabetes" and "insulin" is "imulina" , it turns out), you're saved. 

In China you'd be a goner for sure, unless you happen to have a dictionary with you, and even 

then you would probably pass out while frantically looking for the first character in the word for 

insulin. Which brings me to the next reason why Chinese is so hard. 

5. Because even looking up a word in the dictionary is complicated. 

One of the most unreasonably difficult things about learning Chinese is that merely learning 

how to look up a word in the dictionary is about the equivalent of an entire semester of secreschool. 

When I was in Taiwan, I heard that they sometimes held dictionary look-up contests in 

the junior high schools. Imagine a language where simply looking a word up in the dictionary is 

considered a skill like debate or volleyball! Chinese is not exactly what you would call a userfriendly 

language, but a Chinese dictionary is positively user- hostile. 

Figuring out all the radicals and their variants, plus dealing with the ambiguous characters 

with no obvious radical at all is a stupid, time-consuming chore that slows the learning process 

down by a factor of ten as compared to other languages with a sensible alphabet or the equivalent 

I'd say it took me a good year before I could reliably find in the dictionary any character I might 

encounter. And to this day, I will very occasionally stumble onto a character that I simply can't 

find at aN, even after ten minutes of searching. At such times I raise my hands to the sky, Joblike, 

and consider going into telemarketing. 

Chinese must also be one of the most dictionary-intensive languages on earth. I currently 

have more than twenty Chinese dictionaries of various kinds on my desk, and they all have a 

specific and distinct use. There are dictionaries with simplified characters used on the mainland, 

dictionaries with the traditional characters used in Taiwan and Hong Kong, and dictionaries with 

both. There are dictionaries that use the Wade-Giles romanization, dictionaries that use pinyin, 

and dictionaries that use other more surrealistic romanization methods. The~e are dictionaries of 

classical Chinese particles, dic tio5aries of B eijing dialect, dictionaries of chengylr (four-charac ter 

idioms), dictionaries of xi~hhyu (special allegorical two-part sayings), dictionaries of y&nyz 

(proverbs), dictionaries of Chinese communist terms, dictionaries of Buddhist terms, reverse 

dictionaries ... on and on. An exhaustive hunt for some elusive or problematic lexical item can 

leave one's desk "strewn with dictionaries as numerous as dead soldiers on a battlefield."lO 

For looking up unfamiliar characters there is another method called the four-comer system. 

This method is very fast - rumored to be, in principle, about as fast as alphabetic look-up 

(though I haven't met anyone yet who can hit the winning number each time on the first try). 

Unfortunately, learning this method takes about as much time and practice as learning the Dewey 

decimal system. Plus you are then at the mercy of the few dictionaries that are arranged 

according to the numbering scheme of the four-corner system. Those who have mastered this 

system usually swear by it. The rest of us just swear. 

Charles Hockea reminds me that many of my examples are really instances of loan words, not cognates, but 

rather than take up space dealing with the issue, I will blur the distinction a bit here. There are phonetic loan 

words from English into Chinese, of course, but they are scarce curiosities rather than plentiful semantic 

moorings. 

lo A phrase taken from an article by Victor Mair with the deceptively boring title "The Need for an Alphabetically 

Arranged General Usage Dictionary of Mandarin Chinese: A Review Article of Some Recent Dictionaries and 

Current Lexicographical Projects" (Sino-Platonic Papers, No. 1, February, 1986, Dept. of Oriental Studies, 

University of Pennsylvania). Mair includes a rather hilarious but realistic account of the tortuous steeplechase of 

looking up a low-frequency lexical item in his arsenal of Chinese dictionaries.

Sino-Platonic Pupms, 27 (August 3 1, 199 1 ) 

Another problem with looking up words in the dictionary has to do with the nature of written 

Chinese. In most languages it's pretty obvious where the word boundaries lie - there are spaces 

between the words. If you don't know the word in question, it's usually fairly clear what you 

should look up. (What actually constitutes a word is a very subtle issue, of course, but for my 

purposes here, what I'm saying is basically correct.) In Chinese there are spaces between 

characters, but it takes quite a lot of knowledge of the language and often some genuine sleuth 

work to tell where word boundaries lie; thus it's often trial and error to look up a word. It would 

be as if English were written thus: 

FEAR LESS LY OUT SPOKE N BUT SOME WHAT HUMOR LESS NEW ENG 

LAND BORN LEAD ACT OR GEORGE MICHAEL SON EX PRESS ED OUT 

RAGE TO DAY AT THE STALE MATE BE TWEEN MAN AGE MENT AND 

THE ACT OR 'S UNION BE CAUSE THE STAND OFF HAD SET BACK 

THE TIME TABLE FOR PRO DUC TION OF HIS PLAY, A ONE MAN SHOW 

CASE.THAT WAS HIS FIRST RUN A WAY BROAD WAY BOX OFFICE 

SMASH HIT. "THE FIRST A MEND MENT IS AT IS SUE" HE PRO CLAIM 

ED. "FOR A CENS OR OR AN EDTI' OR TO EDIT OR OTHER WISE BLUE 

PENCIL QUESTION ABLE DIA LOG JUST TO KOW TOW TO RIGHT WING 

BORN AGAIN BIBLE THUMP LNG FRUIT CAKE S IS A DOWN RIGHT DIS 

GRACE." 

Imagine how this difference would compound the dictionary look-up difficulties of a non-native 

speaker of English. The passage is pretty trivial for us to understand, but then we already know 

English. For them it would often be hard to tell where the word boundaries were supposed to be. 

So it is, too, with someone trying to learn Chinese. 

6. Then there's classical Chinese (wenyan wen). 

Forget it. Way too difficult. If you think that after three or four years of study you'll be 

breezing through Confucius and Mencius in the way third-year French students at a comparable 

level are reading Diderot and Voltaire, you're sadly mistaken. There are some westemers who 

can comfortably read classical Chinese, but most of them have a lot of gray hair or at least tenure. 

Unfortunately, classical Chinese pops up everywhere, especially in Chinese paintings and 

character scrolls, and most people will assume anyone literate in Chinese can read it. It's truly 

embarrassing to be out at a Chinese restaurant, and someone asks you to translate some 

characters on a wall hanging. 

"Hey, you speak Chinese. What does this s&oll say?" You look up and see that the 

characters are written in wenyan, and in incomprehensible "grass- style" calligraphy to boot. It 

might as well be an EKG readout of a dying heart patient. 

'Wh, I can make out one or two of the characters, but I couldn't tell you what it says," you 

stammer. "I think it's 'about a phoenix or something." 

"Oh, I thought you knew Chinese," says your friend, returning to their menu. Never mind 

that an honest-to-goodness Chinese person would also just scratch their head and shrug; the face 

that is lost is yours. 

Whereas modem Mandarin is merely perversely hard, classical Chinese is deliberately 

impossible. Here's a secret that sinologists won't tell you: A passage in classical Chinese can be 

understood only if you already know whar the passage says in thefirs? place. This is because 

classical Chinese really consists of several centuries of esoteric anecdotes and in-jokes written in 

a kind of terse, miserly code for dissemination among a small, elite group of intellectually-inbred 

bookworms who already knew the whole literature backwards and forwards, anyway. An 

uninitiated westerner can no more be expected to understand such writing than Confucius 

himself, if transported to the present, could understand the entries in the "personal" section of the 

classified ads that say things like: "Hndsm. SWGM, 24, 160, sks BGM or WGM for gentle 

S&M, mod. bndg., some lthr., twosm or threesm ok, have own equip., wheels, 988-8752 lv. 

mssg. on ans. rnach., no weirdos please."

Schri!esesrschrihriji: Ersqs in Honor of John DeFrancis 

In fairness, it should be said that classical Chinese gets easier the more you attempt it. But 

then so does hitting a hole in one, or swimming the English channel in a straitjacket. 

7. Because there are too many romanization methods and they're all lousy. 

Well, perhaps that's too harsh. But it is true that there are too many of them, and most of 

them were designed either by committee or, worse, by linguists. It is, of course, a very tricky 

task to devise a romanization method; some are better than others, but all involve plenty of 

counterintuitive spellings.11 And if you're serious about a career in Chinese, you'll have to 

grapple with at least four or five of them, not including the bopomofu phonetic symbols used in 

Taiwan. There are probably a dozen or more romanization schemes out there somewhere, most 

of them mercifully obscure and rightfully ignored. There is a standing joke among sinologists 

that one of the fust signs of senility in a China scholar is the compulsion to come up with a new 

romanization method. 

8. Because tonal languages are weird. 

Okay, that's very Anglo-centric, I know it. But I have to mention this problem because it's' 

one of the most common complaints about learning Chinese; and it's one of the aspects of the 

language that Westerners are notoriously bad at. Every person who tackles Chinese at first has a 

little trouble believing this aspect of the language. How is it possible that shhue' means 

\ ..v 

"mathematics" while shkd means "blood transfusion", or that gwpang means "you flatter me" 

while gGj&zg means ''fruit paste"? 

By itself, this property of Chinese would be hard enough; it means that, for us non-native 

speakers, there is this extra, seemingly irrelevant aspect of the sound of a word that you must 

memorize along with the vowels and consonants. But where the real difficulty comes in is when 

you start to really me Chinese to express yourself. You suddenly find yourself straitjacketed - 

when you say the sentence with the intonation that feels natural, the tones come out all wrong. 

For example, if you wish say something like "Hey, that's my water glass you're drinking out 

of!", and you follow your intonational instincts - that is, to put a distinct falling tone on the first 

character of the word for "my" - you will have said a kind of gibberish that may or may not be 

understood. 

Intonation and stress habits are incredibly ingrained and second-nature. With non-tonal 

languages you can basically import, muratis mutandis, your habitual ways of emphasing, 

negating, stressing, and questioning. The results may be somewhat non-native but usually 

understandable. Not so with Chinese, where your intonational contours must always obey the 

tonal constraints of the specific words you've chosen. Chinese speakers, of course, can express 

all of the intonational subtleties available in non-tonal languages - it's just that they do it in a 

way that is somewhat alien to us speakers of non-tonal languages. When you first begin using 

your Chinese to talk about subjects that actually matter to you, you find that it feels somewhat like 

trying to have a passionate argument with your hands tied behind your back - you are suddenly 

robbed of some vital expressive tools you hadn't even been aware of having. 

9. Because east is east and west is west, and the twain have only recently met. 

Language and culture cannot be separated, of course, and one of the main reasons Chinese is 

so difficult for Americans is that our two cultures have been isolated for so long. The reason 

reading French sentences like "Le prksident Bush assure le peuple koweitien que le 

gouvernement amkricain va continuer d &fendre le Koweit contre la menace irakienne," is about 

as hard as deciphering pig Latin is not just because of the deep Indo-European family 

resemblance, but also because the core concepts and cultural assumptions in such utterances stem 

from the same source. We share the same art history, the same music history, the same history 

history - which means that in the head of a French person there is basically the same set of 

I have noticed fiom time to time that the romanization method first used tends to influence one's accent in 

Chinese. It seems to me a Chinese person with a very keen ear could distinguish Americans qdang, say, Wade- 

Giles-accented Chinese fiom pinyin-accented Chinese. 

67


archetypes and the same cultural cast of characters that's in an American's head. We are as 

familiar with Rimbaud as they are with Rarnbo. In fact, compared to the difference between 

China and the U.S., American culture and and French culture seem about as different as Peter 

Pan and Skippy peanut butter. 

Speaking with a Chinese person is usually a different matter. You just can't drop Dickens, 

Tarzan, Jack the Ripper, Goethe, or the Beatles into a conversation and expect to be understood. 

I have a Chinese friend who at one time had read the first translations of Kafka into Chinese, yet 

didn't know who Santa Claus was. And forget about mentioning anything as current as 

Madonna or Teenage Mutant Ninja Turtles; you will get a very, very blank stare. (American 

movies and TV shows, staple entertainment fare in other parts of the world for decades, have 

only recently been allowed into China.) They will know a lot about Nixon, of course, but don't 

be surprised if they tell you with a straight face that he was the greatest American president of the 

twentieth century. 

Similarly, how many Americans other than sinophiles have even a rough idea of the 

chronology of China's dynasties? Has the average history major here ever heard of @n Shi 

Huang Di and his contribution to Chinese culture? How many American music majors have ever 

heard a note of Peking Opera, or would recognize a pipd if they tipped over one? How many 

otherwise literate Americans have heard of Lu Xun, Ba Jin, or even Mozi? 

What this means is that when Americans and Chinese get together, there is often not just a 

language ban-ier, but an immense cultural barrier as well. Of come, this is one of the reasons 

the study of Chinese is so interesting. It is also one of the reasons it is so damn hard. 

Conclusion 

I could go on and on, but I figure if the reader has bothered to read this far, I'm preaching to 

the converted, anyway. Those who have tackled other difficult languages have their own litany 

of horror stories, I'm sure. But I still feel reasonably confident in asserting that, for an average 

American, Chinese is significantly harder to learn than any of the other thirty or so major world 

languages that are usually studied formally at the university level (though Japanese in many ways 

comes close). Not too interesting for linguists, maybe, but something to consider if you've 

decided to better yourself by learning a foreign language, and you're thinking "Gee, Chinese 

looks kinda neat." 

It's pretty hard to quantify a process as complex and multi-faceted as language-learning, but 

one simple metric is to simply estimate the time it takes to master the requisite language-learning 

skills. When you consider all the above-mentioned things a learner of Chinese has to acquire - 

ability to use a dictionary, familiarity with two or three romanization methods, a grasp of 

principles involved in writing characters (both simplified and traditional) - it adds up to an awful 

lot of down time while one is "learning to learn", Chinese. 

How much harder is Chinese? Again, 1'11 use French as my canonical "easy language". 

This is a very rough and intuitive estimate, but I would say that it takes about three times as long 

to reach a level of comfortable fluency in speaking, reading, and writing Chinese as it takes to 

reach a comparable level in French. An average American could probably become reasonably 

fluent in two Romance languages in the time it would take them to reach the same level in 


One could perhaps view learning languages as being similar to learning musical instruments. 

Despite the esoteric glories of the harmonica literature, it's probably safe to say that the piano is a 

lot harder and more time-consuming to learn. To extend the analogy, there is also the fact that we 

are all virtuosos on at least one "instrument" (namely, our native language), and learning 

instruments from the same family is easier than embarking on a completely different instrument 

A Spanish person learning Portugese is comparable to a violinist taking up the viola, whereas an 

American learning Chinese is more like a rock guitarist trying to learn to play an elaborate 30-stop 

three-manual pipe organ. 

Someone once said that learning Chinese is "a five-year lesson in humility". I used to think 

this meant that at the end of five years you will have mastered Chinese and learned humility along 

the way. However, now having studied Chinese for over six years, -1 have concluded that

Schnffestschrifl: Bsay in Honor of John DeFrMcir 

actually the phrase means that after five years your Chinese will stiN be abysmal, but at least you 

will have thoroughly learned humility. 

There is still the awe-inspiring fact that Chinese people manage learn their own language very 

well. Perhaps they are like the gradeschool kids that Baroque performance groups recruit to sing 

Bach cantatas. The story goes that someone in the audience, amazed at hearing such youthful 

cherubs flawlessly singing Bach's uncompromisingly difficult vocal music, asks the choir 

director, "But how are they able to perform such difficult music?" 

"Shh - not so loud!" says the director, "If you don't tell them it's difficult, they never 

know."


Bibliography 

(A longer version of this paper is available through CRCC, Indiana University, 510 N. Fess, 

Bloomington, IN, 47408.) 

Chen, Heqin, (1 928) "Yutiwen yingyong zihui" [Characters used in vernacular literature], 

Shanghai. 

DeFrancis, John (1966) "Why Johnny Can't Read Chinese", Journal of the Chinese Language 

Teachers Association, Vol. 1, No. 1, Feb. 1966, pp. 1-20. 

DeFrancis, John (1984) The Chinese Language: Fact and Fantasy, Honolulu: University of 

Hawaii Press. 

. . DeFrancis, John (1989) Visible Speech: The Diverse Oneness of Writing Systems, Honolulu: 

University of Hawaii Press. 

Kennedy, George (1964) "A Minimum Vocabulary in Modem Chinese", in Selected Works of 

George Kennedy, Tien-yi Li (ed.), New Haven: Far Eastern Publications. 

Mair, Victor (1986) "The Need for an Alphabetically Arranged General Usage Dictionary of 

Mandarin Chinese: A Review Article of Some Recent Dictionaries and Current 

Lexicographical Projects", Sino-Platonic Papers, No. 1, February, 1986 (Dept. of Oriental 

Studies, University of Pennsylvania). 

Zhao, Yuanren, (1972) Aspects of Chinese Sociolinguistics, Anwar S. Dil (ed.), Stanford: 

Stanford University Press.

Schri!esrschn~: <strong>Essays</strong> in Honor of Joltn &Francis 

Ethnolinguistic Notes on the Dungan 

Lisa E. Husmann and William S-Y. Wang 

Proj ect on Linguistic Analysis 

University of California at Berkeley 

We had the good fortune of spending New 

Year's Eve of 1986 with John DeFrancis. That 

memorable evening, in his elegant living room 

overlooking Manoa Valley, we were treated to 

many fascinating slides from his trip through 

northern China. We are delighted that he has 

now written a memoir of that 1935 trip, 

entitled In the Footsteps of Jenqhis Khan. 

Among the peoples he met were the Dungan, 

whose language stirred his curiosity. His 

recollections planted a seed in our minds. 

Three years later, New Year's Eve of 1989, we 

were in Leningrad, on our way to Kirghizstan, 

one of the regions where the Dungans live. 

The following essay, which resulted from our 

brief visit there, owes its inspiration to 

John, and it is fitting that we offer it to 

this volume, to honor this remarkable - 

scholar . 

The population of the Union of Soviet Socialist Republics, 

like that of China, presents a diverse array of ethnic groups. 

Compared to the 56 officially recognized nationalities of the 

People's Republic of China, preliminary results of the 1989 

Soviet census list separate population figures for over 100 

ethnic groups. As in China, language is the main criterion by 

which the Soviet Union distinguishes its national groups. 

Textbooks on the Soviet Union usually list three major language 

families into which the languages of the Soviet population can be 

divided: 1ndo-~uropean, Ural-Altaic, and Paleo-Asiatic. 

Typically, these textbooks fail to mention a fourth language 

family which is spoken in the USSR: Sino-Tibetan. Sino-Tibetan 

is spoken by a relatively small Soviet nationality living near 

the Chinese border: the Dungans ( #?SL)-


Our first contact with the Dungans came about as a series of 

fortunate coincidences which landed us in a Dungan village just 

east of Tokmak in the ~irghiz SSR. (It is hard to say, had we 

tried to plan the contact through official government channels, 

what kind of difficulties we would have encountered). The 

village that we wandered into was in the process of preparing for 

a wedding. Several men were busy slaughtering a cow, and the 

women were scurrying back and forth from courtyard to kitchen, 

adding chopped vegetables into large pots of boiling water in 

which the meat would be made into soup. 

Our attempts to talk to the Dungans in Putonghua seemed to 

generate an immediate sense of kinship, and before we knew it we 

were inside one of their homes, sitting on a kang, and being 

offered tea and lavish trays of dried meats, fruits, and sweets. 

The small room was quickly filled with friendly people, neighbors 

and relatives. Using an odd assortment of broken Russian and 

Putonghua, we spent a memorable afternoon learning about the 

people of the village. 

Although a fair amount has been written on the Dungans, most 

work has been in the form of rather specific or technical 

scholarship. Moreover, a large percentage of the work is 

published in either Russian, Japanese, or Chinese. This short 

paper is an attempt to consolidate into a more readily accessible 

manner, an introduction to the people who call themselves the 

'Dungan'. Even this point--the extent and implications of the 

use of Dungan as a self-reference--is a complex issue. This, as 

well as several other issues explored in this paper, including 

the origins of the Dungan, the formation of their culture and the 

emergence of the names used to refer to them, are not yet fully 

understood and often raise more questions than answers. 

According to preliminary 1989 census results, there are 

today approximately 69,000 Dungans in the Soviet Union (1). They 

live primarily in cities and villages within the Soviet Socialist 

Republics (SSR's) of Kirghizstan, Kazakhstan, and Uzbekstan (2). 

The Dungans are people of the Hui Hui (00 ) or Hui nationality 

who migrated from China to Russia after anti-Manchu revolts of 

1862-1877. Although the Dungans are often described in Western 

sources as 'Chinese Muslims', this term seems particularly 

inappropriate as a reference for the Dungans, at least for those 

who live in the Soviet Union today. The Dungans that we met in 

~irghizstan were particularly sensitive about their Dungan 

identity, and reminded us more than'once that the Dungan language 

is not \Chinese1 (i& $*), but its own separate language. The 

issues of language and self-identity are very interesting in the 

case of the Dungan, and will be discussed later in this paper.

Another interesting point is the perceived relationship of 

the Dungans to the Hui nationality of China. While in Frunze, we 

had the opportunity to spend an evening with a Dungan man, Mr. 

Zhang, from the Kirghiz Academy of Social Sciences. Mr. Zhang is 

in his early forties. His great-grandparents migrated from 

Xinjiang to Russia in 1882. When we asked him, he described the 

nature of the Dungan-Hui relationship for us in the following 

way: 'We are Dungans ... our Hui Hui people live not only in the 

Soviet Union but across the border in China as well." Mr. Zhang 

then, (and presumably other Dungans, at least of his generation), 

acknowledge a common ancestry with the Hui of China, and use the 

term \Hui Huil as an alternate (if not primary) self-reference. 

For this reason, an understanding of the ethnic origins of the 

Dungans begins with a tracking of the origins of the Hui 

nationality in China. The trail leads us far from Mr. Zhang and 

~irghizstan, to the southeastern coastal cities of 7th century 

China. 

651 a.d., the second year of the reign of Tang Gao Zong 

( &*), is cited as the year that Islam was formally introduced 

into China (Ma 1986:182). It was during this period, the middle 

of the 7th century, that considerable numbers of Arab and Persian 

merchants came by ship to China from the Persian Gulf (3). 

Specializing in international trade of exotic commodities such as 

herbs, rhinoceros horns, elephant tusks and gems, these foreign 

merchants began to take up residence in southern China's busy 

commercial ports. Their first settlements were in Guangzhou (4). 

Later settlements were formed in Yangzhou, Quanzhou and Hangzhou 

(Bai, et a1 1964:6). 

Over the centuries, these so-called .\ foreign guests ( $ ) 

or \native-born foreign guests1 ( 2 ) adopted 

monosyllabic surnames that could be wrltten with Chinese 

characters (5). There was also a considerable amount of 

intermarriage with the local Chinese, so that by the 12th century 

these communities had much assimilated into Chinese culture. 

Throughout this long period of assimilation, however, these 

people retained their Islamic religion and a distinctive 

religious culture. They built mosques and maintained separate 

graveyards, in which tombstones were inscribed in Arabic script 

(Bai, et a1 1964: 6) . 

These 'foreign guests1, however, are only one minor source 

of the Hui nationality as we know it today. The term \HuiV does 

not even appear as a name for-these southern communities until 

several centuries later under the influence of the name of 

peoples then living in the north--the major source of today's Hui 

nationality (6) .


In official documents of the Yuan Dynasty (1271-1368), the 

tern 'Hui Huif is first used as a reference to predecessors of 

today's Hui nationality (7). The people described as 'Hui Huit 

in these Yuan Dynasty sources are actually an amalgam of several 

groups, a result of successive waves of migrations in the wake of 

the military campaigns of Jenghis Khan. 

Jenghis Khan began his western campaigns in 1219. By 1258, 

his armies had advanced as far as Baghdad. As his armies moved 

westward, large numbers of conquered peoples migrated to the 

east. Among those that continued eastward as far as China were 

military recruits and prisoners of war, representing a wide 

variety of ethnic backgrounds. The term 'Hui Huit of the Yuan 

official documents refers to the mixture of Persians, Arabians, - . 

and Central Asians (including Uighurs, Tanguts and Khitans) that 

had settled in northern China mainly in the 13th century (8). 

When the first author visited China in 1985, she asked 

various Chinese people why the Chinese Muslims were called the 

'Huit. A common explanation she was given was that because their 

~slamic tradition compelled them 'to returnt to Mecca on 

religious pilgrimages, these people were therefore given the name 

\Huif (in Chinese, meaning 'to returnf). This is an interesting 

piece of folk etymology, but a preliminary look into the ethnic 

history of northern China reveals a more likely origin of the 

term (9). The characters 'Hui Huif were probably first used as a 

phonetic representation of the name of a northern people, most 

likely the 'old Uighurst (see note 7). After the inflow of a 

wide variety of peoples in the 13th and 14th centuries, the term 

\Huit was then used as a reference to an amalgam of several 

different groups. 

Although the point at which these people initiated the term 

'Hui Huif as a self-designation is uncertain, it is generally 

held that a portion of the Hui had adopted a common identity by 

the Ming Dynasty (136801644). This common identity was forged in 

the areas where the Hui were most densely settled: Ningxia, 

Gansu, and Shaanxi, as well as beyond the Yellow River in Shanxi, 

Hebei, and Henan (10). These areas of relatively dense Hui 

settlement correspond to the greater region of the Central 

Plains, the cultural heartland of China. An intriguing 

similarity exists between the Russian name for the Dungan, 

\Zhunyant, and the Chinese term for the Central Plains, 

\Zhongyuant. The possibility of the term \Dungant having been 

derived from the name of the 'Zhongyuanf region that Dungans even 

today speak of as their homeland is an hypothesis that will be 

considered later in this paper.

Schnfjcestschn~: <strong>Essays</strong> in Honor of John DeFrancis 

By the middle of the 18th century, control of the Qing 

Dynasty (1644-1911) in China was in rapid decline. In Central 

Asia, new sectarian religious movements developed within the 

Muslim. communities and challenged Qing supremacy. The conflicts 

created by these new sects, combined with Muslim repression by 

the Qing government, culminated in uprisings against Qing 

authorities in several provinces. Also during this time a new 

leader, Yacob Beg, rose to power in Central Asia. He took 

control of Kucha (1864), Urumqi (1865), and the Ili area of 

northwestern china (1866), and established the region as an 

independent state, the Moslem Emirate of Kashgaria. 

Russia, taking advantage of the political unrest in China, 

moved into Ili in 1871. Using the Muslim disorders in China as 

an excuse for their occupation, the Russians claimed that they 

were protecting their own citizens from Muslim raids, and would 

withdraw from Ili when China succeeded in re-establishing order 

in the region. Ultimate suppression of the Muslim uprisings was 

the achievement of the military commander Zuo Zongtang (11). 

After crushing the rebellion in Gansu and Shaanxi by 1873, Zuo 

Zongtang began to move against Yacob Beg. Less than four years 

later, the whole region northward to the Tian Shan range (except 

for the Kuldja area in Ili) had been secured. By late 1877, Zuo 

Zongtang's campaigns had brought Yacob Beg's rule to an end. 

One result of Zuo Zongtangls campaigns was the displacement 

of a large number of Hui refugees who fled with Bai Yanhu (6 &lf,) 

from northwestern China to the Russian Semirechle in 1877-1888 

(12). Another important outcome of the re-establishment of Qing 

control in the region was that the Russian premise for occupying 

Ili was invalidated. In 1879, China sent a delegation to St. 

Petersburg to ask the Russians to evacuate the territory. After 

failure of this first meeting, a second delegation was sent which 

culminated in the signing of the Treaty of St. Petersburg in 

1881. Under this treaty, almost all of Ili was returned to China 

(13). As control of the land switched hands, the Hui population 

of Ili feared further persecution from the Qing, and again large 

numbers crossed the border into Russia. 

These two migrations are the source of the Dungan 

nationality living in the USSR today: the main wave coming in 

1877-1878 after the fall of Yacob Beg, and smaller groups coming 

between 1881-1884 after the signing of the treaty of St, 

Petersburg. The majority of Dungans that migrated to Russia were 

poor, illiterate peasants. They brought with them to Russia their 

Chinese and Muslim cultures, agricultural techniques (including 

cultivation of rice, opium poppies, and new vegetable types), as 

well as their spoken dialects of northwestern Mandarin (Dyer 

1979:l). Even today, the Dungans continue to live in 

Chinese-style houses, sleep on \kangl (t);?) -style beds, eat

Sino-Plaronic Papers, 27 (August 3 1, 199 1) 

traditional Chinese foods (such as 'manti' and 'lakhmanw), use 

chopsticks, and observe traditional Chinese birth, death, and 

marriage customs (14). 

In an attempt to preserve their national identity, the 

Dungans, like many small emigrant communities, tend to be 

conservative and nationalistic. One of the cultural features 

which has been consciously maintained by the Dungans is their 

language (15). 

As mentioned earlier, the Dungans that we met in Kirghizstan 

were particularly sensitive about their Dungan identity. When, 

at first, we referred to Dungan as a dialect of Chinese, we were 

politely reminded that since the mutual intelligibility between 

Dungan and Chinese is low and since Dungan is written with a 

~yrillic script, it is more appropriate to regard Dungan as a 

separate and independent language. Mr. Zhang presented us with a 

finely printed booklet, which included translations from Tolstoy 

as well as stories by native Dungan authors. We learned that a 

considerable literary tradition is growing around the Dungan 

language. 

The issue as to whether or not Dungan constitutes a separate 

language can of course be debated, since the degree of mutual 

intelligibility between Dungan and its source dialects in 

northwest China has yet to be determined. (Comparing Dungan with 

Putonghua would be missing the point). Also, while the script a 

language uses can have important effects on the development of 

the language, it is not a relevant consideration for determining 

the degree of genetic relatedness. Similarly, the fact that 

Dungan is spoken across a national boundary from its source is 

not a relevant consideration either. Nonetheless, it is a 

significant sociolinguistic fact that the speakers should feel 

that Dungan is now an independent language. 

But Dungan nationalism and conservatism have even deeper 

roots which divide the Dungans into even smaller communities. 

Dungans living only several kilometers apart identify themselves 

as Gansu or Shaanxi Dungans, each preserving their own traditions 

and dialects (Dyer 1979:13-14). On the basis of place of origin, 

the Dungan nationality of the Soviet Union is divided into two 

major and one minor groups. The Dungans living in Kirghiz SSR 

are primarily 'Gansu Dungan', whereas those living in the Chu 

Valley region of the Kazakh SSR are primarily 'Shaanxi Dunganr. 

The third and minor group is the 'Yager ('Iage') or 

'Dungan-Yage'. This group traces its origins to the Chinese 

cities of Lanzhou in Gansu, and Yinzhou in Shaanxi (16). 

The topic of Dungan "homelandsr@ leads us to questions raised 

earlier on the origin of the term 'Dungan'. According to Wixman

Schrififesrschrifi: &say in Honor of John DeFrancis 

(1984:59), "the term Dungan appeared in Sinkiang province in the 

later half of the 18th century and referred to the immigrant 

~hinese from central China who settled in that area." 

Rimsky-Korsakoff (1987:354) clarifies this hypothesis: @'The term 

Tung-kan hui developed in Hsinchiang in the latter part of the 

eighteen century to distinguish from the native Chinese 

Mohammedans the refugees from persecution in the East." 

According to'Rimsky-Korsakoff, the name \Dunganef was firsf 

adopted by the Russians and used incorrectly as a reference to 

all Chinese Muslims: Apparently, this erroneous general usage 

was subsequently adopted by the British, Germans, and others. 

This accounts for what seems to be a preferential usage of the 

term \Dungan1 over the name \Huil by early 20th century European 

explorers in northwestern China. 

As with the term \Hui Huig, the name \Dungan1 retains its 

own folk etymology. In Chinese, the characters that represent 

\Dungant carry the meaning \eastm (%) [tug], and \shieldm (q ) 

[kan]. The initial [k] in the second syllable might have emerged 

under the influence of the velar nasal ending of the first 

syllable. In other words, it is possible that earlier, the 

second syllable was pronounced [an], which in Chinese could mean 

\ shore ( ) . The possibility of such a change is relevant in ' 

considering the folk etymology of the term \Dungang among the 

Dungan themselves. According to Mr. Zhang, Dungans commonly 

believe that the name of their people derives from the fact that 

their original homeland lies beyond the \eastern shores [ 9.3 ] 

of the Yellow River. 

In addition, the hypothesis may be explored for connecting 

the name \Dunganm (east-shield) to \Zhongyuanm (central-plains). 

Orthographically, the question immediately arises on why \zhongl 

(central) should now be written with \dong1 (east), and \yuang 

(plain) should be written with \gang (shield). Was it due to a 

series of confusions that have to do with the immigrants coming 

from the east--perhaps the \gang (shield) deriving from \an1 

(shore), in the manner described above? The confusions could 

have been both geographic and phonetic, as the speech of the 

immigrants confronted that of the native Altaic and other peoples 

in northwestern China. 

Phonologically, there are problems as well. \Zhongm began 

with a plain stop in Old Chinese, and hence had the same initial 

as \ Dong I , ie, , DUAN (a&) category. In the Middle Chinese 

represented in the Qie Yun, however, the initial of \zhongl has 

changed to the ZHI (go) category, while that of \dong1 has 

remained in DUAN. Nonetheless, the plain stop initial in \zhong8 

has been preserved in many of the more conservative dialects of 

the south, notably the Min group. Could it have been preserved

Sirw-Platonic Papers, 27 (August 3 1,199 1 ) 

as well in the speech of the Dungan immigrants? It is suggestive 

in this regard that in the Russian name, 'Zhunyanl, the initial 

is a voiced fricative rather than a plain stop. 

With respect to \ganR, we note that while \ganl (shield) 

begins with the velar stop [k], \an1 (shore) and \yuan1 (plain) 

had an YI (kH) category initial, which has been reconstructed as 

a velar nasal. That this velar nasal has denasalized to a [q] is 

well attested in many modern reflexes. Of special interest 1s 

the reconstruction of a [ 'g] for many YI initial words for the 

ancient northwestern dialects by Lo (1933:24). It seems possible 

that an early [s] in \yuang (plain) became confused with the 

unaspirated [k] in \gant (shield). On a different tack, again we 

may note that the second syllable in the Russian name \Zhunyanl 

corresponds closely with \yuan1 (plain). 

Unfortunately, linguistic material on the northwestern 

dialects is limited, and we have not come upon instances where 

the DUAN initial is preserved in \zhongR (central), nor instances 

where the [g] is preserved in \yuanm (plain). Until such 

instances are found, the hypothesis for connecting \Dungan1 with 

\Zhongyuant must remain at a speculative level. 

Much in the way of ethnolinguistic and ethnogeographic 

research has yet to be done on the Dungans. As demonstrated in 

this paper, preliminary investigation into the issues of 

terminology and ethnic origins of the Dungans brings about more 

questions than answers. It is noteworthy, however, that while a 

considerable amount of confusion continues to surround these 

issues, the Dungans themselves seem quite clear about their 

self-identity. 

Notes : 

(1) (Anderson & Silver 1991). This 1989 population figure 

represents a 33% increase in the Dungan population in the USSR 

since the last soviet census in 1979. Between the years of 1970 

and 1979, the Soviet Dungans showed a similar rate of increase 

(from 39,000 to 52,000). In the eleven years between the 1959 

and 1970 censuses, however, the Dungan population in the USSR 

showed a dramatic increase of 77%. It is reasonable to assume 

that approximately 44% of the growth during this period was due 

to migration from China in the wake of disruption of the Cultural 

Revolution.

Schnfjestschniji: <strong>Essays</strong> in Honor of John Mancis 

(2) Within the Kirghiz SSR, Dungan settlement is concentrated in 

and around the cities of Frunze, Tokmak, Przhevallsk and the 

villages of Yrdyk, Khunchi, Milianfan, Kysyl-Shark, 

Aleksandrovka, etc. In the Kazakh SSR, there are Dungans living 

in and around the cities of Alma-Ata, Dzhambul (with the village 

Dzhalpak-~iube), Panfilov (with the village Chilik), in the 

villages of Karakunuz, and Shor-Tiube, and in the Kurdai district 

of the Dzhambul region. Within the Uzbek S.S.R., Dungans live 

around Tashkent and the cities near Osh in the Ferghana valley. 

According to Rimsky-Korsakoff (1967), many of the villages which 

are entirely Dungan were formed at the time of their migration 

(1887-1884). Other villages, such as Milianfan and Khunchi were 

formed since the Soviet revolution in 1917. 

(3) Because Persia maintained a strong trade connection with 

south China for centuries, it is probable that these merchants 

from the Persian Gulf were not Arabians (as is commonly held), 

but rather were Persians (~ranians) that had earlier been 

converted to Islam (Eberhard 1982:58). Schafer (1963:15) 

similarly proposes a Persian origin for these merchants. 

According to Schafer, they were Shiite Muslims whose main reason 

for settling in China was escape from religious persecution in 

Khorasan (northeastern Iran). 

(4) A tradition preserved by the geographer Marwazi, early in the 

12th century, says that Shiite sectaries fled in 749 and settled 

on an island in a large Chinese river, opposite a port (Hourani 

1951). Schafer (1963) believes that the place being described in 

this tradition is Canton (Guangzhou). 

(5) Some of the characters adopted were already common Chinese 

surnames, for example Ding (J), Bai (61, and Lu (-9). Other 

surnames, however, such as Ma ( .&) , Mu ( a) , HU . (2,) , and Ha (re) 

represent a more obvious phonetic similarity to the foreign names 

from which they were derived. According to Bai, et a1 

(1964:5-6), genealogies of many Hui families in China record the 

original family name from which their abbreviated 'Chinese1 

surnames were derived. 

(6) Apparently, because the southern communities of 'foreign 

guests1 were perceived as being culturally similar to the larger 

13th century Islamic settlements in the north, the name \Huil was 

eventually also applied to them. Only beginning in the Yuan 

Dynasty, therefore, do descriptions ofthe southern communities 

include references such as \ @I piJ $ and \ 8 

('southern-Foreign Hui Huig) (Ma 1986:182). 

%:;,a *C-+ 

(7) Many sources cite Shen Guols Dream Pool <strong>Essays</strong> ( 7- .?ffSa,, 

written circa 1086, as the first recorded reference of the Hui 

Hui people. This is misleading. While the Dream Pool <strong>Essays</strong> do


refer to a people called the 'Hui Huil, it is generally 

recognized that Shen Guo's reference is not to the same group 

that we call Hui today, but instead is in reference to the 

'~ui-hel ( @ 3% ) or 'Hui-hul ( @I$$ ) people. In Western sources 

these groups are sometimes referred to as the Itold Uighur1I (see, 

for example, Eberhard 1982:58). The old Uighur were a Turkic 

people who came to power in Central Asia in the mid-8th century. 

They are believed to be the precursors of the present-day Uighur 

( Ex vd) nationality of China, not of the present-day Hui 

nationality. Unfortunately, this confusion of historic 

terminology has led to confusions in modern designations as well. 

Wixman (1984:59) for example, states that in China the term "Hui 

HuiI1 refers not only to the Hui nationality, but to the Uighur 

nationality as well. This misperception may be based on the 

confusion of historic terms described above. 

(8) This is the explanation of the usage of the term 'Hui Huil in 

Yuan Dynasty documents as given by Bai, et a1 (1964:5-7) and Ma 

(1986:182-183). According to other sources, such as Zhongguo 

Shaoshu Minzu, the general Chinese name for these 13th century 

immigrant groups was \Se Mu Ren I ( $ a L ) , which might be 

translated 'Appearance-to-the-Eye People1, referring to the fact 

that the physical appearance of these people was different than 

that of the Chinese. In he Yuan Dynasty work, 

between Tillinss ( $$&kt($$. ), the author Tao 

lists a total of 31 groups included within the name 'Se Mu Renl; 

'Hui Huil is one of them (Cihai 1967:2417). 

(9) An interesting elaboration of this folk etymology appears in 

a 1932 account of a British missionary, Reverend Andrew. Andrew 

writes: "'Hwei-Hweil means 'to go back upon onels track1, or the 

returners-..In early times, (the) great trail from the Caspian 

Sea through Central Asia to China was a known route, and in the 

early centuries of Christianity is was a well-travelled road. As 

early as the fifth and sixth centuries we know of Arabs who 

penetrated China by this route as well as by sea. We have one 

record of the visit of an Arab embassy to China during the 

lifetime of the Prophet (Mohammed, 570-632). The embassy, which 

landed at Canton, visited the court of the Emperor at Chlang An 

in Shensi, and from there the ambassadors attempted the overland 

journey through Central Asia and Persia to Arabia. In accordance 

with the common usages of the Chinese, the Emperor sent an escort 

to accompany them for several days on their journey, and these 

Arabs, who knew none of the finer terms of courtesy, when they 

begged the escort to return, used the words 'Hwei-Hweif, to 

return, instead of the more polite terms; and from that day to 

this they and their descendants have been known as the Hwei-Hwei, 

the returners" (Andrew 1932 : 89-90) .

kh?7ftf&chrifr: <strong>Essays</strong> in Honor of John DeFrancis 

(10) Although Hui settlement has historically been (and still is) 

scattered throughout China, the most concentrated Hui population 

continues to be in what Barnett refers to as the "Muslim Beltn of 

Gansu, ~ingxia, and Qinghai (Barnett 1963:182). In addition to 

these northern provinces, there is also a considerable Hui 

population in the southwestern province of Yunnan. Most of the 

Yunnan Hui are descendants of troops that the Mongols had brought 

with them into that province. The Mongols conquered Yunnan in 

the 12501s, as part of their military strategy to outflank the 

Song Dynasty in southeastern China. During the Yuan Dynasty, 

rule of Yunnan was handed over to a Muslim from Central Asia, 

under which large numbers of Muslim troops were brought to and 

settled in Yunnan. Hui revolts in northern China in the 19th 

century strongly influenced Hui movements within Yunnan, and 

several Hui perished after an attempt was made in the 1860's to 

create a Muslim Kingdom within Yunnan (Fitzgerald 1972:64-95). 

Although the Hui population of Yunnan dropped considerably in the 

wake of these 19th century uprisings, according to 1982 census 

figures, there are over 438,000 Hui living in Yunnan today, 

making them the 8th largest ethnic minority of that province 

(YNSM 1980: 625) . 

(11) Zuo Zongtang was the first ethnic Han general to take 

command in Xinjiang; his predecessors had all been Manchus 

(Eberhard 1982:61). Zuols earlier post had been as governor of 

the Shaanxi-Gansu region. A by-product of his western campaign 

was the creation of a major road beyond Yumen Guan (the Jade Gate 

Pass), along the northwest passage. The road is said to be over 

3,700 \lit (a Chinese unit of measurement, =1/2 kilometer) in 

length, with willows planted along its sides. Yang Changxun 

( G g t$ ) , who succeeded Zuo in the governorship, celebrated 

Zuo9s accomplishments with the following poem: 

DA JIANG XI ZHENG REN WE1 HUAN 

the general has not yet returned from his western campaign 

HU XIANG ZI DI MAN TIAN SHAN 

Tian Shan range is full of soldiers from Hunan and Hubei 

XIN ZAI YANG LIU SAN QIAN LI 

newly-planted willows stretch for 3,000 li 

YING DE CHUN FENG DU YU GUAN 

win the spring wind crossing Yumen Pass 

(We would like to thank Ovid Tzeng for bringing to our attention 

this historic poem). 

(12) Bai Yanhu was a Moslem chieftain. According to the Zhongguo 

Renmin Dacidian (1964:209), during the last years of the reign of 

Xianfeng (1851-1860), he controlled portions of Xinjiang. During 

the Guangxu reign (1875-1908), he allied with Yacob Beg, and 

fought against the Qing. Bai was finally defeated by Liu Jintang

Sino-Platonic Papers, 27 (Aupt 3 1, 1991) 

( %1'@ % ) , and fled with a group of Dungan followers across the 

~ian Shan into Russian territory. 

\Semirechlet is a Russian term meaning 'between the rivers'. The 

name refers to the land that lies between the Amu Darya (Oxus 

River) and the Syr Darya (Jaxartes River). 

(13) Also under the terms of the 1881 Treaty of St. Petersburg, 

the number of Russian consulates in the area were reduced to two, 

and China was made to pay an indemnity of 9,000,000 rybles. 

Following settlement of the treaty, the whole area of Chinese 

Turkestan was in 1884 incorporated into China as the province of 

Xinjiang, the 'New Frontiert. 

(14) (Rimsky-Korsakoff 1967:355). When we were in Frunze 

(~irghizstan) in 1990, we were urged to try some lttraditional 

Kirghizw food. Two specialties we were told of were 'lakhman' 

and \mantit. Only when our dinner was placed in front of us did 

we realize that what we had ordered was 'lamianl (Jft&~) and 

\mantout (@@)! The \mantouV however, was stuffed with meat, 

in the form of \baoziv (4% 3 ) . (It would be interesting to 

trace how \baozit have become \mantit in Central Asia). Although 

Kirghiz locals and tourist literature describe such foods as 

Ittraditional Kirghiz," Sushanlo (1971:169) refers to these as 

"Dungan dishesm that are popular among the neighboring Kirghiz, 

Kazakh, and Uzbek people. In Chinese, \la1 (8%) of \lamiant is a 

rusheng ( A*$ ) character that ended with -t, and has been 

reduced to a glottal stop in many Mandarin dialects. Apparently,. 

use of the unvoiced velar fricative [x] in place of a glottal 

stop in the word \lakhmant reflects the closest approximation of 

the host language that borrowed the word. 

(15) Dungan became an officially recognized language of the USSR 

in 1929. The Gansu dialect forms the basis of the Dungan 

literary language, which is written in Cyrillic script and taught 

as a first language in local schools. Census data reveal that 

there is little linguistic erosion among the Dungans: in 1979, 

94.8% of the population gave Dungan (ItZhongyuan huafl) as their 

mother tongue (Akiner 1983). 

(16) According to Wixman (1984:60), the Dungan-Yage speak a 

~ingxia-Lanzhou dialect of Mandarin, but share many cultural ties 

with the Uighur nationality. In Russia, they live primarily in 

the villages of Aleksandrovka, Sokuluk, and Chilik. Apparently, 

the name \Yagel carries the pejorative meaning of \refugeet.

References : 

Schn'flfestschrifi: essay^ in Horwr of John DeFruncis 

Akiner, Shirin. 1983. Islamic Peoples of the Soviet Union. 

London. 

Anderson, Barbara & Silver, ~rian. 1991. "Aspects of Soviet 

Ethnic Demography: Tables and Figures." (Presented at UC 

Berkeley, February 6, 1991). 

Andrew, Rev. G. Findlay. 1932. Islam in Northwest China To-day. 

Journal of Royal Central Asian Society. 19:89-100. 

Bai, Shouyi; Han, Daoren; Ding, Yimin et al. 1964. Huihui Minzu 

de Lishi he Xianzhuang. (History and Present Status of the Hui 

Hui Nationality). Minzu Chubanshe. 

Barnett, A. Doak. 1963. China on the Eve of a Communist Takeover. 

1963. New York. 

Cihai. 1967 (2nd edition) . Taiwan. 

Cihai. 1982. Shanghai. 

Dragunov, A. A. 1940. Research in Dungan Grammar. Moscow. (In 

Russian) 

Dyer, S. Rimsky-Korsakoff (= Rimsky-Korsakoff, Svetlana). 1979. 

Soviet Dungan Kolkhozes in the Kirghiz SSR and the Kazakh SSR. 

Oriental Monograph Series no.25. Canberra: Australian National 

University. 

. Eberhard, Wolfram. 1982. China's Minorities: Yesterday and Today. 

Belmont . 

Fitzgerald, C.P. 1972 (1st edition). The Southern Expansion of 

the Chinese People. London. 

Forke, A. 1895. A Comparative Study of Northern Chinese Dialects. 

The China Review. 21.3:181-203. 

Guo, Xiliang. 1986. Hanzi ~uyin Shouce. Beijing. 

Hourani, G.F. 1951. Arab Seafaring in the Indian Ocean in Ancient 

and Early ~edieval Times. Princeton. 

Isayev, M. I. 1977. National Languages in the USSR: Problems and 

Solutions. (Translated from Russian by Paul Medov) . Moscow.

Sim-Platonic Papms, 27 (August 3 1, 199 1) 

Lo, Changpei. 1933. Tang Wudai Xibei Fangyin. Institute of 

History and Philology Monograph 12. Shanghai. 

Ma, Shouqian. 1986. "Hui~u~~ in Zhongguo Dabaike Quanshu: Minzu 

Ce. p.182-186. 

Rimsky-Korsakoff, Svetlana (= Dyer, S. Rimsky-Korsakoff). 1967. 

Soviet Dungan: the Chinese Language of Central Asia. Monumenta 

Serica 26: 352-421. 

Schafer, Edward. 1963. The Golden Peaches of Samarkand. 

University of California Press. 

Shushanlo, M. 1971. The Dungans, an Historical-Ethnographic 

Sketch. Frunze. (In Russian). 

Wexler, Paul. 1980. Zhunyanese (Dungan) as an Islamic and Soviet 

language. Journal of Chinese Linguistics 8.294-304. 

Wiman, Ronald. 1984. The Peoples of the USSR: an Ethnographic 

Handbook. New York. 

Yunnan Shaoshu Minzu (YNSM). 1980. Kunming. 

Zhongguo Shaoshu Minzu (ZSM). 1981. Beijing. 

See also Victor H. Mair, "Implications of the Soviet Dungan Script for Chinese 

Language Reform," Sino-PImonic Papers, 18 (May, 1990), 19 pages.

Schriftfestschri$: Ersqys in Honor of John DeFrancis 

Georgetown University 

In the debate aver the role and efficacy of Chinese characters in the 

writing systems of Ehst Asia, =stern scholars typically are less aware 

of the situation pertaining to Korean. This is unfortunate. Since Koreans 

use Chinese characters in a unique fashion, understanding the nature of 

these conventions .can lead to a better appreciation of how the characters 

function in and of themselves., In addition, virtually every argument that 

has been made for and against Chinese characters by Japanese, Chinese and 

=stern scholars--and some that have not--has been advanced independently 

and prosecuted in great detail by Koreans writing about their, awn language, 

Those of us whose interest lies in the reform of character-based systems 

can learn much from this wide but neglected body of scholarship. 

Of the many Koreans who have addressed writing reform, two in particular 

stand out for their comprehensive treatment of the problem and for their 

passion in pursuing ,it. Chloe ~yEn Bae (1894-1970), father of the allhangul 

movement, taught at Yonsei University and later served as vicepresident 

of Tongdae University. In 1 942, he was arrested by the Japanese 

for activities in support of the Korean language. Upon his release at 

the end of the war, he was made chief of the Ministry of Education's 

Textbook Ompilation Bureau, where he directed the first of several illfated 

attempts to remove Chinese characters from the school curriculum. 

He also served as president of the Hangul Society, a private organization 

headquartered in Seoul which promotes all-hangul writing. Chloe wrote 

more than 40 books and articles on the Korean language. His best known 

mrks, which we cite here, are ~sj a i?i HyYogmy~ng ' (The 

Writing , 1946), andlHangTil man ~sEqi Tii Chuja&(~he Case 

Hangul Only, 1970), a collection of his thoughts on the 

cultural aspects of writing reform. 

Revolution in 

for Writing in 

technical and - 

HG Ung (191 8 - professor of linguistics at Seoul National University 

and president of the Hangul Society, is the best known advocate of allhangul 

writing in Korea today. ~6 has written twelve books and some sixty 

articles since 1956 on a wide range of topics relating to linguistics and ' 

writing reform. The presept essay draws on three of his reformist 

publications, including "~anj a n% P 'ye j itdya Handa" (Chinese Characters 

Must Be Abolished, 1 971 ) , Urimalqwa ~3 3 Naeil a wihay6 (For Our Language 

and Script 'of Tbmrrow, 197$a), and Han3qwa Minjok Munhwa (Hangul and 

the National Culture, 197423). 

Our task here is tmfold: to describe how Koreans use Chinese 

characters, then to show why these two scholars believe that usage to be 

entirely unnecessary.

Sino-Phonic Papers, 27 (August 3 1. 199 1) 

IXvelqxnent of Writing in Korea 

The earliest evidence of writing in the Korean peninsula dates from 

China's establishment of its Lolang colony in northern Korea in 108 B.C. 

Chinese characters were used to record the Chinese language. It was not 

until the 5th century A.D. that Koreans are known to have adopted these 

symbols to elements of their own language. As was later true of Japanese, 

Koreans used the characters in tm distinct ways. They could be used for 

their semantic value to represent indigenous Korean mrds that had meanings 

similar to the character's meaning in Chinese. The reader would look at 

a character and supply an equivalent Korean word, as the Japanese do with 

their - 

kun readings today. The second type involved use of the characters 

for their phonetic value. Korean approximations of the characters ' Chinese 

sounds iere matched with the sounds of Korean words. The characters in 

this way became units of a proto-syllabpry that expressed sound, regardless 

of what the £oms meant originally. These two principles m e used 

concurrently in - Idu (lit. "official readingsf'), the system employed in 

the bureaucracy f m the 7th century onward, and in a number of other hybrid 

systems. 

Many will recognize in this phonetic use of Qlinese characters the 

beginning of a process that led in Japanese and other languages to a 

phonetic script, through which all of a language's words could be written 

with a limited number of symbols, There was movement toward this in Korean 

as well, encouraged by the difficulty Koreans had recording proper names, 

and finding plausible semantic associations between Qlinese characters 

and Korean grammar morphemes. Unlike Japanese, the process was thwarted 

by the large number of syllable types in Korean. As long as the syllable 

remained the basic orthographic unit, a system representing these units 

muld be as unwieldy as the conventions already in place. 

The dilemma was resolved, in principle, with the invention of Korea's 

hangul alphabet in 1446. Based on symbols that bear an actual resemblance 

to the human vocal organs at various points of articulation, hangul was 

designed from the start to represent Korean sounds, and only sounds. Its 

24 basic signs identify the language's vowel and consonant phonemes 

accurately and unambiguously. As with any alphabet, hangul letters are 

canbined to form a model of the word's sound when spoken. In most cases 

the relationship between hangul letter and sound is direct. Elsewfiere 

the spelling reflects what Koreans believe to be a word's underlying 

- - 

phonology, the pronunciation being derivable by a few simple rules. The 

system is practical, sufficient and easy to use. It seems complex only 

because Koreans decided to combine the letters into syllable blocks, in 

deference to the Confucian-dminated court which rpr& that the oukd 

appearance of the Chinese characters be maintained, 

Despite its utility, hangul was unable to replace the Idu script favored 

by the bureaucracy until the end of the 19th century, when both script 

and users were ousted in a series of egalitarian political reforms. For 

a few years after the Kap'o (1894) rebellion, hangul was the official medium

SchnJtJstschriifi: fisay in Honor of John &Francis 

of written cation. However, for reasons as much social as 

linguistic, this brief experiment with all-phonetic writing was supplanted 

by the government I s sanction of another. hybrid system that had since come 

into use, which ccmbined hangul and Chinese characters in the same text. 

Known 'as "mixed hangul-hanja (Chinese character) " writing, it is one of 

two styles endorsed by the South Korean government today, and the sine 

qua non for higher education. It is typically used for documents of 

stricter content, including most scientific and academic works where the 

proportion of Sinitic loanwords is high. The other style, of course, is 

all-hangul, used in novels, popular magazines, the local pages of newspapers 

and most informal types of writing. In North Korea, it is the only style 

used. 

A mitical difference between the mixed hangul-character script, the 

old Idu convention, and Japan's kana--character script is that the last 

two systems can use Chinese characters for indigenous vocabulary, while 

Korea's present system does not. Chinese characters when used in Korean 

today represent Sinitic words only, Moreover, there is no formal 

requirement that these words be in characters, even when using the mixed 

script, Korean writers can and often do just spell the Sinitic mrd out 

in hangul, Amther important difference is that unlike Idu (and modem 

Chinese), characters are never used for their sound value alone, e.g., 

in onomatopoeia and transliterations. The Korean mixed script employs 

a strict division of labor: if the word is Sinitic in origin, it may be 

written in characters. Everything else must be in hangul, 

One result of restricting Chinese characters to Clhinese loanwords is 

that Koreans, as a rule, do not consider the characters their own. Another 

is that the Sinitic words themselves can becoane targets for replac-t 

by indigenous words, real or contrived, in the language purification 

campaigns that periodically surface, Chinese has no tried and universally 

accepted alternative to its character writing system to which its users 

feel emotionally attached. For better or - worse, they. are stuck with the 

characters at present. Japanese, for its part, thomghly assimilated 

the characters by virtue of assigning kun readings. Despite the complexity 

of these associations, the characters are so entrenched in Japanese language 

and culture that -st no me predicts their impending demise. Koreans, 

however, have no such feelings about the characters, and can hardly be 

said to lack a suitable replacement. Why then are they still used, and 

are the justifications for their use valid? 

If utility and tradition are two grounds for the use of a writing 

system, then in the former case at least there are some fairly obvious 

reasons why Koreans would want to abandon characters, Unlike hangul whose 

24 letters designate a finite set of sounds, mese characters represent 

morphemes, the building blocks of words, Hence they number in the 

thousands, So that each unit can be distinguished from others, they are 

also quite complex, containing twelve strokes on average. Because there 

are no predictable relationships between what the characters look like,

Sino-Platonic Papers, 27 (August 3 1, 1 99 1 ) 

sound like and mean, each character and the data associated with it rmst 

be learned individually. Korean high school graduates, after six years 

of study, are v t e d to know 1 ,800 of them. Reading newspapers requires 

2,000 or more. For science and scholarship, the number is still higher. 

Although not as onerous as the task facing Chinese and Japanese students, 

the weight of these numbers alone makes the utilitarian argument for 

characters hard to sustain. 

Both Ch'oe and 6 dutifully cite these statistics and draw the expected 

conclusion: that characters should be replaced by all-hangul writing. 

neir arguments, hcrwever, go beyond that. HE; is convinced the main problem 

with Chinese characters is that they impede the mechanization of writing 

(1971:18; 1974a:41). With so many units, an apparatus of great complexity 

is needed to achieve less utility than what is realized in the West by 

typewriters. ~5 also blames the characters for hangul's squarish shapes. 

Although pleasing aesthetically, the practice requires typists to select 

different keys for the same hangul letter depending on what part of an 

imaginary square the letter occupies, instead of just stringing them 

together serially (1 974a:108). F'rm the reader's point of view, the 

requirement for a square shape forces syllables with many letters into 

the same small space, making them hard to distinguish. Hangul's utility 

is thus reduced by the need to coexist with Chinese characters. ~h'oe 

makes these same general points, adding two remarkable insights which only 

recently made their debut in western scholarship: (1) however advanced 

characterapable mrd processors become, their efficiency will always 

lag behind alphanumeric processors using the same technology on- a script 

with fewer elements (1970:lO). And (2) the cost of the character processors 

will always be greater than those used for hangul, W n g their diffusion 

and putting users at a comparative disadvantage (1970:198). 

Schemes to solve these problems by reducing the number of characters 

win praise from neither of these scholars. &'oe maintains that such 

efforts are doomed from the start, since a list would only encourage the 

literati to vie with each other in damnstrating knowledge of obscure forms. 

Also, how does one determine which characters belong on the list? Would 

it not differ according to meation and interests ( 1946:87)? 6 discovered - 

another flaw in such proposals: limiting the number of characters would 

inmediately produce the ludicrous situation where some Sinitic words 

(usually two syllable campounds) are written half in hangul and half in 

characters (1971 :31). HZ'S final criticism is the most denstating 

indictment of character limitation schemes ever penned. Fbst such projects 

single out "comon use" characters that are frequent enough to justify 

the effort needed to learn them. Wlt if they are used ccmnonly, the mrds 

and morphemes they represent are those least likely to cause confusion 

if written phonetically. Conversqly, if the less cormon morphemes can 

be left to hangul, what justification is there for writing anything in 

characters (1974a:62-3)? 

The question of ambiguity in hangul texts is central to the defense 

of Qlinese characters. Conventional wisdom maintains that since the 

characters identify a word's constituents exactly, the chance of

Schnfrfesestschriji: &says in Honor of John &Francis 

misconstruing one mrd for another is nearly zero. Hangul, on the other 

hand, identifies sounds, Because of the many homonyms and near-homonyms 

in the Sinitic vocabulary, hangul texts are said to be inherently ambiguous 

since there are instances when the reader does not know which word is 

intended. 

There is some truth to this argument, especially as it concems texts 

written in the mixed hangul-character script and translated directly, 

syllable for syllable, into all-hangul. However, such texts were written. 

originally with the understanding that the characters' own redundancy would 

canpemate for the text's "terseness". Materials written as hangul texts, 

by contrast, take the need for more serial redundancy ins account, and 

expand overall context so that homonyms can be readily distinguished, as 

in English and other languages (5, 1974b:218). 

The problem with this remedy is that Chinese characters have let too 

many words into the language that never had to stand on their own 

phonetically. While context will disambiguate many of them, the problem 

is so acute that no tricks effected outside the word boundary will allow 

some mrds to be identified in hangul. But do such words qualify as Korean? 

HG claims they do not (1 974a: 1 04), adding that for all-hangul to succeed, 

writers must give up their habit of using (or making up) obscure expressions 

that are not really words, and are intelligible only in characters. Put 

another way, if hangul had to be used, these ambiguities in time would 

be eliminated, whereas Chinese characters only perpetuate the problem 

(Ch'oe, 1946:65). In the interim, Chloe suggests (1) using indigemus 

Korean substitutes where they can be found, (2) agreeing that in certain 

environments a sound will always depict only one of sever* possible, or 

(3) dcing changes to the shape of the word itself (1970~47). 

aloe sees the homonym "problem" as a reflex of a broader social 

problem, namely, Korean worship of foreign culture. Had it not been for 

Koreans ' sonry habit of revering China and slighting everything indigenous, 

there would have been no massive influx of Sinitic loanwords, and no problem 

with phonetic indistinctiveness (1946:44; 1970:193). Instead, Koreans 

could have maximized use of their own rich stock of morphemes, which have 

mre phonetic shapes and unlike Qlinese can be polysyllabic. At minimum, 

there would be a better balance between the indigenous part, and Sinitic 

part of the lexicon which accounts for 75% of present-day Korean. HE goes 

even f w to claim that this "unnatural" phenomenon hinders the 

development of Korean thought (1974b:124). 

Advocates of all-hangul writing do not reject all Sinitic words, only 

those which, when spoken or written in hangul, cannot be understood given 

a reasonable amount of context (Chloe, 1970:104). Sinitic morphemes will 

continue to be used to form new words. But without characters, the results 

will have to be intelligible phonetically, Pbreover, since the distinction 

between Sinitic and indigenous roots is less visible in hangul, Koreans 

will lose their reluctance to coin new terms from the native stock (HG, 

1971 :26). Pure Korean not only sports a host of "compound words". There 

are also thousands of "derivative words" formed by adding prefixes and

Sim-Platonic Papers, 27 (August 3 1,199 1) 

suffixes to a morpheme's root, The two processes furnish Korean with emugh 

"word-building power" (chot$k) 

(g, 1974a: 52). 

to satisfy most of the language's needs 

C 

If pure-Korean roots have more potential than they are usually credited 

with, Sinitic morphemes have a good deal less. True, they readily combine 

with other Sinitic roots because they are monosyllabic, and identifiable 

through the characters. But these canbinations are not necessarily words, 

if intelligibility in speech is a criterion. Nor is it fair to say that 

because the meaning of a combination can be plausibly reconstructed from 

the manings of its individual mrphmes (depicted through Chinese 

characters), the canbination itself has currency as a word. In Ch'oe's 

view, this whole l@e of reasoning is nonsense. Sinitic dinations do 

nut always refer to the logical sum of their constituent morphemes--whatever 

that is. Rather, their meaning is established by conventim, and it matters 

little how that convention was arrived at (1970:40). 

This applies to forming new mrds. What about learning them? .Here 

again, the notion that Wnese characters allow one $0 identify the meaning 

of a whole through its parts is largely illusory. ' In fact, what hints 

the characters do give can be misleading. Worse still, they can interfere 

with learning, since one is ccmpelled by one's knowledge of the characters 

to supply a -lqical connection -between -the whole and parts, which may not 

exist (Ch'oe, 1970:73), Even where valid connections do exist, what 

relevance do they have for a synchronic user? Words, according to Chloe, 

are used for what they man today (1970:31). Drcept for a few specialists, 

no one thinks about a word's etymology when using it, which is probably 

good, since this information would just clutter our thought processes, 

Reading is another area where the characters enjoy an undeserved 

reputation for an ability to evoke concepts directly without sound, which 

mrds written in hangul supposedly lack. If this claim is based on the 

similarity of symbol to referent, however, it is absurd, because characters 

have undergone numerous changes in shape and meaning. The original 

connections we, in any case, often far-fetched, What supporters of 

Chinese characters really mean is that the meanings of characters are 

obvious (directly accessible) only after repeated use. But this is also 

true of words in hangul, which likewise have fixed shapes, and fixed 

meanings that users learn to access directly (Ch'oe, 1970:54; HE, 1974a:54). 

Assume, however, that the characters, lacking reliable clues to 

pronunciation, are more likely to be processed directly without recourse 

to sound, while hangul lends itself to gecoding simply because so much 

phonetic information is patently available. Then by mixing the tm systems 

together, the reader is forced to shift back and forth from one mde of 

processing to the other, causing difficulty and confusion (Ch'oe, 1970:175). 

Both Chloe and HE concede that reading all-hangul texts can be difficult 

for those who have spent their lives using Chinese characters, This is 

to be expected. Reading the mixed hangul-character script, one becomes 

habituated to lexical and stylistic conventions that differ from what is 

needed for all-hangul, It is inadmissible therefore for the older

SchrifjestschriJi: Ersuqs in Honor of John &Francis 

generation to extrapolate from their awn experiences to claim that no one 

can properly read hangul texts (HZ, 1 974a : 1 3 1 ) . Another reason why older 

Koreans resist all-hangul writing is because having identified a character 

with a given word, they imagine that if the character disappears the word 

cannot exist either. Years of association have made the Ism, for some 

people, conceptually indistinguishable (HE, 1 974a : 4 2 ) . Ch ' oe considers 

a third cause for the intransigence of the older generation, that is, the 

fact that most are bilingual speakers of Japanese, incapable of expressing 

themselves in a way that is truly Korean. When writing, they draw on extra, 

non-Korean resources that are as unintelligible in all-hangul as in speech 

(1970:66). 

Thirty-five years of Japanese rule have bred a mentality that, for 

all its outward protestations, still looks to the Japanese for leadership. 

This applies to language reform as well. If the Japanese have not abandoned 

Chinese characters, and indeed have prospered while using them, is there 

not a lesson for Korea here? Ch'oe finds none, pointing instead to the 

post-war economic developnent of Cermany, which manag& its miracle without 

Chinese characters. If Japan had a serviceable phonetic script, muld 

they not have made even greater progress (Ch'oe, 1970:92-3)? There are 

other reasons why the comparison with Japanese is invalid, beginning with 

what Ch'oe and % both see as the inadequacy of kana, the Japanese phonetic 

syllabary, to function independently as a script. In Ch'oe's opinion, 

kana is - too simple. It does not separate vowels and consonants, and the 

individual symbols are not distinct enough for rapid discrimination (Ch'oe, 

1 970 : 269 ) . Hangul forms are also simple. But because they each represent 

only one phoneme, not two, and are grouped into mrds, they are more easily 

distinguished (~5, 1971 :32). The very feature which enabled Japanese to 

evolve a syllabary of sm 50 signs, i .e., that language's simple phonetic 

structure, is what makes it so much more character4ependent (Ch'oe, 

1946:87). 

A final reason why Chinese characters persist in Japanese is that they 

can take up the burden of two or more syllables, even in their Sinitic 

- on 'readings. In Korean, however, it is a one-for-ne replacement, One 

saves nothing by using characters. HE concedes that an all-hangul text 

may be somewhat longer than a text in the mixed script, because indigenous 

Korean words are occasionally substituted for Sinitic loans that are shorter 

but ambiguous. Since the replacements are Korean mrds, the reader does 

mt feel that the text is unnecessarily long, as do the Japanese who are 

merely substituting a different set of symbols for the same word (~5, 

1971 :32). 

The above pages treat the linguistic arguments for an all-hangul writing 

system, which pertain in a narrow sense to the script's ability to function 

without the aid of Chinese characters. There are broader issues involved, 

Wver, which many Korean consider of overriding concern. We have seen 

that the different countries of northeast Asia use Chinese characters 

differently. Moreover, since the languages themselves are different, users

Sim-Platonic Papers, 27 (August 31, 1991) 

of one language cannot possibly read connected discourse in another no 

matter what units their writing systems share, Thirdly, the shapes of 

the characters themselves have undergone different changes in Japan and - 

China, making their "transitivity" even less viable. These facts 

notwithstanding, the three languages share a large number of Sinitic 

borrowings which, by and large, can be understood by educated readers in 

any of the three languages, provided they are written in characters. Often 

this will enable readers of one language to grasp the essence of headlines, 

titles - or longer passages, For signs and other paralinguistic materials., 

their. transitivity is indisputable. The phenomenon parallels what literate 

native speaJers of Ehglish can accamplish with Rench or Spanish materials, 

for exactly the same reasons. 

If characters are abandoned in Korea, Koreans would lose this marginal- 

ability to decode parts of Chinese and Japanese texts, They would also 

run the risk in the long term of being cut off from new Sinitic terms coined 

in chi^ and Japan (although the Vietnamese experience suggests otherwise), 

More importantly, many Koreans feel they would be isolated culturally. 

This last argument leaves Ch'oe and Ho' stunned, since neither can imagine 

why, in light of the region's history, any Korean would want to identify 

with these neighboring countries. this point, Ch'oe is quite specific: 

Korea's participatian in the so-called Chinese character cultural commmity 

has always been as a juniar member, Do Koreans really need this (1946:71)? 

HS asks if Koreans, one and all, should suffer for the sake of the tourist 

industry (1974a:122), He also wonders why the same people who want Sinitic 

loans written in CZzinese characters do not clamor for English loanwords 

to be written in r-ja and mixed in directly with the hangul and everything 

else (1 974a: 49). More to the point, Korea's foreign contacts are no longer 

limited to East Asia, nor should they be. Koreans need to absorb ideas 

from all over the world, and the writing system they choose should 

facilitate this (ibid), 

The above pertains to the characters' ability to close geographical 

distances, There is also the question of what their abandonment would 

mean for Korea's historical continuity. Cut off from its awn tradition, 

could Korea survive? Ch'oe's and HZ'S approaches to this problem differ 

in emphasis, reflecting their personalities and the times when they wrote. 

ol'oe, as usual, pulls no punches, "We must regard the future as more 

important than the past" (1946:54). Besides the direct benefits to be 

gained by using a more efficient writing system, there are important 

psychological side-effects to writing in all-hangul. European progress 

began only after Latin was abandoned as the medium of written discourse, 

Writing in their own "vulgar" languages, Europeans of various nationalities 

were able to infuse their countrymen with a new vigor that had been stifled 

by the old and . crusty conventions, Thus, the move to all-hangul is more 

than an effort to rid the system of Chinese characters. By decreasing 

dependence on foreign borrowings, the movement fosters attitudes of national 

pride and self-reliance that will spill wer into all areas of society 

(Ql'oe, 1970:234). 

H is more solicitous of what he feels are genuine concerns, but

maintains that this culture could be better conveyed through translations, 

Instead of wasting resources in a gratuitous and futile attempt to teach 

all school children enough characters to read the classics, why not train 

a mqll group of specialists to translate these wrks for everyone's benefit 

(1974a:70). & makes the interesting point that Christianity, despite 

the recentness of its introduction, gained more favor among the comnon 

people than Confucianism, because the Bible was translated into all-hangul 

which everyone understood. Grant that current all-hangul translations 

of classic Korean texts contain sane errors, But this is hardly an 

indictment of the enterprise. If experts have trouble, how can school 

children be expected to understand them in the original (Ch 'oe, 1 970: 1 57) ? 

There are Koreans, in Ch'oe 's view, who cannot shake their belief that 

a education per se means learning Chinese characters ( 1 970 : 1 40 ) . They argue, 

moreover, that primary school children can easily learn two thousand 

characters, since their minds are still .so receptive. Ch'oe sees this 

as a compelling reason - not to waste this opportunity mrizing symbols. 

Time spent teaching characters is time lost from substantial studies, 

Worse, it reduces education to a mechanical'level. Instead of training 

people to think, the character-based curriculum fosters cramming, and a 

predisposition to respond by rote to new situations (1970:87). 

If the worst effects of Chinese characters are felt in education, then 

the solutim to the problem is to be found there, too. Neither Ch'oe nor 

.HE consider themselves revolutionaries, in the sense of one who advocates 

abruptly changing a social convention. Hence neither proposes an outright 

ban m the characters, as was done in the North. Ch'oe reconmends they 

be eliminated from the language slowly by removing them from the mandatory 

educational curricula (1 946:92). I& sees his task as promoting a tradition 

begun more than 500 years ago when hangul was invented, by widening its 

application from literature, which it currently daninates, to all types 

of writing ( I 974a: 60) , 

Recent publications in the United States and abroad attest to renewed 

interest in the functioning of Chinese character-based writing and the 

role these systems play in the material and intellectual lives of societies 

using them. Koreans, being one of the latter, are intimately concerned 

with writing reform, and have published voluminously on both sides of the 

question. Chloe ~$n Bae and H6 Vng are the outstanding proponents of 

all-hangul writing. Their views contrast sharply with those of other 

Koreans who find merit in the mixed hangul-character script, and make 

interesting arguments to support it. Whatever me's disposition to the 

characters may be, examining the Korean data will prove rewarding. 

Notes 

1946 and 1970, Ch6ng1?hsa, Seoul.


1971, in Hangul Hakhoe, Har@l Ol'&yonng 

J 

uro 3 Kil. Seoul. 1974a, 

Kwahaksa, Seoul, 1974b, Kyoyangguksa, Seoul. 

3 Lee Ki Mun, 1977, Geschichte der koreanische Sprache, Seoul, pp, 

52-9, A third usage was concurrent semantic and phonetic assignment, i-em, 

when the characters were used to represent the original Chinese mrds, 

4 

Lee, p. 60. 

p. 28. 

Also, Kontsevicha, L. R- 1979. awnin Ch~n~m- msm, 

Nam Kwang U. 1979. "Hanguk 6ru.n Kyoyuk y&q-uhoe An" (The Plan of 

the Korean Language Education Research Society) . &nun y'ingu, 21 , p. 8. 

"~alggol 3 papgugi". It is unclear whether Woe I s proposal refers 

to changes in the word's pronunciation, or its graphic form, since the 

latter is certainly practicable with present spelling conventions, 

Ch'oe claims that a person does not know the meaning of a word because 

of the characters, Rather, one knows the meanings of characters only 

because me has first learned the meanings of the wrds in which they are 

used (1 946:82), 

Writing before the advent of psycholinguistics, Chloe's terms differ 

from those used today. - m'oe asserted that the characters provide meaning 

primarily, and sound through the meaning. Hangul depicts sound first, 

and meaning through sound,

Schn~festsch?ifi: fisa,vs in Honor of John DeFrancis 

LANGUAGE POLICIES AND LINGUISTIC DIVERGENCE IN THE TWO KOREAS 

Ho-min Sohn, University of Hawaii 

1. The physical insulation and ideological distinction between South (SK) and North Korea (NK) 

since 1945 has given rise to a thick politico-social ddectal division, which is superimposed on the 

long existing historico-geographical dialects.l This linguistic divergence has been accelerated not 

only by the polarized political, ideological, and social differences, but also, more importantly, by 

the different language policies adopted by the two governments. 

The two societies may be summed up, in laymen's terms, as capitalism vs. socialism, 

global dependence vs. self-reliance, free competition vs. tight control, traditionalism vs. 

revolutionalism, openness vs. closedness, liberalism vs. prescriptivism, pluralism vs. uniformity, 

and relative individualism vs. strong collectivism, as they apply respectively to SK and NK. This 

polarization is reflected in the respective educational goals. The Charter of National Education of 

SK stipulates that the basic objective of education is to foster people's way of life that will 

contribute to the development of the nation and to a Renaissance of national culture, while 

developing individuals' fullest potentialities. Under such a liberal atmosphere, Government 

policies and scholars' efforts are effective only to the extent that they are compatible with the needs 

and convenience of the public. 

The main purpose of education in NK is "to bring up the rising generation into steadfast 

revolutionaries who fight for society and people" (Article 39, the Constitution). The 5th Central 

Committee Convention of the Workers Party set the direction of educational policy as 

"revolutionazing all the people" so that they can actively participate in the policy of a communist 

unification of Korea. A policy adopted at the Convention was that language should be used as a 

means of thought-reform and of strengthening the people's consciousness of revolution and class 

This paper is dedicated to Dr. John DeFrancis.


struggle (cf. RCPU 1976). Under a such tightly controlled atmosphere, Kim I1 Song's Teachings 

alone have guided linguistic reality in NK. 

The aim of this paper is to examine the language policies of NK and SK in relation to the 

resultant linguistic divergence. An ovaview of the policies is ma& in Sections 2 (SK) and 3 

(NK); divergent linguistic reality is sketched in 4; and a brief conclusion is given in 5. 

2. The immediate post-liberation years saw Koreans in both Koreas demanding a national 

language that is independent of foreign elements such as Chinese characters and Japanese. - 

loanwords. Both Koreas launched extensive crusades against illiteracy based on hankul (the 

Korean ~lphabet).~ In SK, the Korean Language Society took the lead for this campaign, 

whereas in the NK, Kim I1 Song's 1946 Teachings (kyosi) on the purge of Japanese remnants in 

education and the fight against illiteracy kindled a widespread movement. The subsequent policies 

in the two Koreas have been divergent. 

Let us observe the developments in SK first. Noteworthy is the evolvement of the policies 

toward Chinese characters. The forceful movement of the Korean Imguage Society to eliminate 

characters led the National Assembly to pass the law on the exclusive use of hankul in 1948. 

While schools observed the law, society did not. Repeated Presidential urging for the exclusive 

use of hmkzd in 1956 and 1957 achieved only limited success, such as the hankul-only practice in 

government documents and in street signboards, but the general public and newspapers kept using 

characters. Thus, the Ministry of Education allowed, in 1964, 1,300 common Chinese characters 

to be taught at elementary (600), intermediate (400), and high (300) schools Urged by hah1 

scholars, however, the Ministry again enforced a hankul-only plan as of January 1970, allowing 

no characters in documents and all textbooks at elementary and secondary schools. The ensuing 

situation that even high school graduates could not read newspapers led the Ministry to reinstate 

character education in 1972, allowing 1,800 characters to be taught at elementary and secondary 

* ?he Yale Romanization system is obsmed in tmnsmibing Korean sounds, letters, and expressions. Translations 

of Korean sentences in b s paper are mostly mine. 

.

Schriftfestschnfr: &says in Honor of John DeFrancis 

schools. This practice still obtains at present, although the 1,800 characters do not have a binding 

force on South Korean society. 

As for Romankition, the major issue is what symbols are to be used for individual sounds, 

and how to spell them. This had been a long standing issue in SK, until 1984 when the Ministry 

of Education revised its 1959 system drastically and announced a new system which is based in 

large part on the MaCune-Reischauer system. How to spell loanwords in hankul had also been a 

controversial issue until 1986 when the Ministry announced the current Loanword Spelling 

Conventions. From 1970 the Ministry and scholars made efforts to revise the 1936 Version of 

Standard Speech and the 1933 Hankul Spelling Conventions. As a result, the Ministry announced 

the current Revised Standard Speech Regulations and Hankul Spelling Conventions in January 

1988. 

As for language purification, SK has achieved only limited success, despite the continued 

efforts of the Government, scholars, and language associations. Thus, numerous Sino-Korean 

words are newly coined or being introduced from Sino-Japanese as needs arise and the Korean 

lexicon is inundated with recent English-based loanwords. 

3. NK has launched two stages of language policy with complete success: (a) the policy of 

abolishing the use of Chinese characters and hahl-based literacy movement, to popularize the 

doctrine of socialism by eliminating illiteracy (1945-1966), and @) the policy of "Cultured Speech" - 

(mwunhwae), to standardize Korean based on Pyongyang speech and Kim I1 Song's cwuchey 

(self-reliance) ideology (1966-present)? Success of the fvst stage policies was due to Kim's 

1946 Teachings, as already indicated, and the Government's initiation in 1949 of compulsory 

elementaq education, together with the honlatl-only policy. 

Second stage policies have been implemented according to Kim's two sets of language- 

related Teachings, one in 1964 and the other in 1966. The former presented the basic directions 

Many studies on North dean Lsnguage policies are available, recent ones inclu&ng M. Kim (1985), Chon and 

Choy (1989), and CEH (1990). 

1

Sinu-Platonic Papms, 27 (August 3 1, 199 1) 

.and the latter substantiated them. In the former, Kim brought up eight topics, as summarized 

below:(a) any attempt at script refm should wait until after Korea becomes reunified and Korean 

science and technology become sufficiently advanced; (b) coinage of new words and recovery of 

old words must be based on native elements; (c) use of loanwords must be limited, and spellings 

of proper noun loanwords must be faithful to their original pronunciations; (d) Chinese characters 

must be abolished, but they need be taught for reading purposes only in order to understand South 

Korean publications; (e) words should be spaced properly, and shapes of words should be fured 

after the reunification of Korea; (0 unnecessary Sino-Korean words should be removed from 

dictionaries, and local agencies must be tightly controlled for correct use of words; (g) a 

nationwide campaign should be undertaken for the correct use of the language; and (h) Korean 

language education must be improved and strengthened at all levels of schools. 

In the 1966 Teachings, Kim elaborated upon detailed procedures of refining vocabulary, 

while stipulating the preservation and development of the national characteristics of Korean based 

on the speech of Qongyang. This is the notion of Kim's Cultured Speech. Kim's specific 

directives on vocabulary refinement procedures are: (a) eliminate from dictionaries those Sino- 

Korean words which form synonyms with native words; (b) introduce fine dialectal words into the 

standard lexicon; (c) introduce native words for place names if necessary; (d) coin new native 

words based on native elements; (e) change, as far as possible, Sino-Korean terms of fruits, 

grains, etc. to native words; (f) try to give native names to newborn babies; (g) change new 

loanwords to native words, except technical terms; (h) preserve native-like Sino-Korean words; 

and (i) have the Korean 1 p g e Assessment Committee (kwuke saceng wiwenhoy) control new 

words. Kim restates the need of limited Chinese character teaching to students; calls for the 

training of more linguists to develop cwuchey-oriented Korean; encourages research on script 

reform; and reemphasizes the need for proper spacing, indicating that the current practice allows 

too many spaces.

Sch~iffestschri~: Ersays in Homr of John DeFrancis 

Linguistic theory, policies, planning, and practices in NK are aimed at realizing Kim's two 

sets of Teachings. Dealing with linguistic phenomena and refining Korean must all be done taking 

Kim's mucky ideology into account, i.e., for the people and socialist revolution, rejecting 

toadyism and doctrinalism and safeguarding autonomous and creative positions by developing the 

national characteristics of the language. An unprecedented linguistic reform has resulted, 

encompassing orthography, pronunciation, grammar, vocabulary, meanings and stylistics, and 

usages. 

4. The major areas of linguistic disparity at present between NK and SK may be summed up 

roughly as follows. 

(a) Standard speech and pr~nunci~om: NK takes Pyongyang-based Cultured Speech as standard 

both in pronunciation and spelling. Cultured Speech is defined as "the richly developed national 

language that is formed centering around the revolutionary capital under the leadership of the 

proletarian party that holds the sovereignty during the socialism-constructing period, and that all 

people hold as a standard, because it has been refined revolutionarily and polished culturally to fit 

the proletariat's goals and lifestylew (Cosen Mwunhwae Sacen, A Dictionary of Korem Culncred 

Speech, 1973). SK's standard speech (phyocwune) is defined as "the contemporary Seoul speech 

used by educated people" (Phyocwune Kyuceng, "Standard Speech ~egulations", 1988). 

(b) Word creation: NK has coined some 5,000 lexical items either by nativising Sino-Korean 

words or by creating new words based on native roots, affixes, archaic forms and dialectal 

elements, while maximally limiting the importation of new loanwords. SK has been relatively 

generous in creating or importing Sino-Korean words. Over 10,000 English-based loanwords, 

including such recent loans as 'lame duck', 'incubator', and 'free-lancer', are used in SK. 

(c) Meanings and sryles: While meanings and styles of words and phrases in SK are largely 

neutral, many expressions in NK have metaphorical connotations, orienting the people toward the 

"socialistic revolutionary strugglew. For example, sewulmal 'Seoul Speech' is defined as "the 

speech used in South Korea today which, due to American imperialists and their followers'

Sino-Platonic Papem, 27 (August 3 1, 199 1) 

national language erasure policies, has lost the unique national characteristics of our language and 

is recklessly mixed with Western, Chinese, and Japanese words" (Chong 1981). 'Harvest' is 

kaulkeri centhwu (lit. 'autumncollecting combat*) in NK and kaulketi or cwuw (lit 'autumn- 

collecting') in SK. '15 million ton grain production goal' is rendered as 1,5'00 manthon uy alkok 

bci lul cemlyenghal tey toyhan mokphyo (lit 'goal regarding occupying the hill of 15 million tons 

of grains') in NK, as compared with the neural SK form i,SW manthon uy ymgkok sayngsan 

mkphyo. Norong labor ', tongmwu 'friend', inrnin 'people', etc. have socialist mnnotations. 

(d) Chinese charmers: As already mentioned, characters are taught in NK only for reading South 

Korean publications. In SK, 1,800 characters are taught at elementary and secondary schools. 

Characters are used widely in publications in SK. 

(e) Hankul spelling conve~ons: Both the NK spelling conventions (in Kaycenghan Cosenmal 

Kyupemcip, "the Revised Collection of Korean Norms", 1987) and the SK counterparts (Hankul 

Macchwumpep, "Korean Spelling Conventions", 1988) are modified versions of the 1933 Hankul 

Spelling Conventions (Hankul Macchwumpep Thongdan). This sharing of the source system and 

the fact that both systems follow the same basic principles (e.g. the principles of morphophonernic 

spelling and word- based spacing) have contributed to preventing disastrous divergence. Thus, the 

differences are due mainly to the existence of two standard types of speech and different analyses 

(with regard, for example, to linguistic fossilization) of linguistic phenomena. Spelling divergence 

includes the following aspects: NK's use of horizontal (left-to-right) writing only, and SK's use of 

both horizontal and vertical writings; names of h h l letters (e.g. NK kiuk instead of SK kiyek 

for the letter k); NK's grouping geminate letters (egg. kk, ay) after all basic letters (egg. S a), and 

SK's ordering each geminate after each basic letter; tensified sounds after the suffix -I (e.g. NK -I 

ky vs. SK -I kke 'I promise'); word-initial l and n (e.g. NK lyekra vs. SK yeksu 'history'; NK 

nyenlyeng vs. SK yeniyeng 'age'); other Sino-Korean words (egg. NK pheysway vs. SK 

phyeysway 'closure'; NK hannasan vs. SK hanlasnn 'Mt. Halla'); diphthongs (egg. NK 

mryyessuki vs. SK ttuyessuki 'spacing'), vowel harmony (e.g. NK alwntavo vs. SK alumrowe

Schrifiesestsrihrifi: <strong>Essays</strong> in Horwr of John DeFratzcis 

'pretty and'); epenthetic s (e.g. NK pmoka vs. SK paash 'seaside'); fossilization (e-g. NK 

nepcekkho vs. SK nelpcekkho 'flat nose'; NK ilkwn vs. SK ilkhvun 'worker'); 'standard' 

pronunciations (e.g. NK SOW vs. SK soybki; NK wuley vs. SK wuIoy 'thunder'); loanwords 

(e.g. NK bppwu vs. SK khep 'cup'; NK rneyhim vs. SK meykrikho 'Mexico'); and spacing 

(e.g. NK cohwrkes vs. SK cohun kes 'good thing'). According to Chon and Choy's count in 

Chong (1981) and H. Lee, (1982), 1,400 words have identical pronunciations but different 

spellings and 3;130 words have different pronunciations and different spellings. 

(f) Romanization: NK and SK have different romanization systems. There has been an initial 

Romanization meeting in Europe (in 1989), with both South, North, and Soviet delegates 

participating, in an effort to reach a unified system (Ki-jung Song, personal communication). 

Although this meeting failed to come up with a reasonable agreement, another follow-up meeting is 

reported to be under planning. 

4. In this heated atmosphere for the reunification of Korea, it is timely to consider seriously the 

issue of how to check the progressing linguistic disparities and recover linguistic homogeneity. As 

a first step, it is imperative for Korean linguists from both Koreas and overseas to get together to 

begin discussing the issue of linguistic divergence in general and orthographic problems in 

particular. One serious general problem that contributes to the ever-widening divergence is the 

ongoing cwuchey-oriented language purification movement in NK and the more or less laissez- 

faire policy toward the influx of loanwords in SK. Orthographic problems are relatively free from 

political and ideological sensitivity and thus are conducive to scholarly discussion. Resort to such 

linguistic criteria as simplicity, generality, exhaustiveness, and naturalness, as well as historicity 

and tradition, will lead to the elimination of many existing disparities. For example, as regards the 

problem of vertical and horizontal writings, the SK practice is preferable even kom the perspective 

of the cwuchey-ideology. As far the names of hankul letters, the NK innovation is preferable in 

view of simplicity. Regarding words like NK nepcewchlo and SK nelpcekkho, the former is 

preferable, because the usual pronunciation of nelp is [nel] and 'flat nose' is always pronounced as


[nep.ccek.kho]. Between NK -1 key and SK -1 k k., the latter is preferable, in that this form is 

grammaticalized as a new ending meaning 'speaker's promise'. There is no significant problem 

with the spacing divergence. Chon and Choy (1989) observed that 930 words have different 

pronunciations but identical spellings in NK and SK. This fact suggests a way to eliminate the 

other spelling differences to a great extent, if enough scholarly cooperation is made. 

CEH (Caha emwun hakhoy). 1990. Pw&n uy Cosenehak (Korean Linguistics in North 

Korea). Seoul: Hanshin Co. 

Chon, Su-tae and Hochol Choy. 1989. Nampwukhan Em Pikyo (A Linguistic Comparison 

between South and North Korea). Seoul: Tosechwulphan Nokcin. 

Chong, Sun-Ki, et al. 198 1. Hyenray Cosenmal Socen (A Modern Korean Dictionary). 2nd 

Edition. Pyongyang: Kwahak Paykkwasacen Chwulphansa. 

Kim, Min-su. 1985. Pwan ziy Kwuke Yenkwu (Studies of Korean in North Korea). Seoul: 

Korea University Press. 

Lee, Huy-sung. 1982. Kwuke Tnysocen (A Great Dictionary of Korean). Seoul: Mincwung 

Selim. 

RCPU (Research Center for Peace and Unification). 1976. Socio- Cultural Compa rison bemeen 

South and North Korea. Seoul: Kwangm yengsa

BACKGROUND 

Schnfjests~h ri': Essay in Howr of John &Francis 

Okinawan Writing Systems, Past, Present, and Future 

Leon A. Serafim 

University of Hawaii at Manoa 

The aims of this paper My aims here are to introduce basic information 

about the writing systems of Okinawan since the inception of written records 

in Okinawa, to place them into a typological framework, and to point out their 

differences from and similarities to the Japanese writing system from which 

they are derived. I close by looking at the future of Okinawan writing. 

The Kingdom of the Ryukyus and Japan The Kingdom of the Ryukyus began 

as the state of Chuuzan, which had become a small 'entrepdt trading nation 

. . by the late 1300's. By the early 1500's it had asserted its control over all of 

the Ryukyus. In 1609 it was subjugated by the Satsuma feudatory, had its 

territory north of Okinawa island taken away, and was .made to pay onerous 

taxes. In 1879 it lost its last shred of sovereignty to the newly formed cen- 

tralized Japanese state, and henceforth became Okinawa prefecture. (For 

details see Kerr 1958:60-392.) 

Okinawan and Japanese Okinawan is, broadly speaking, the speech of 

Okinawa island, part of the Northern Ryukyuan language, which is spoken in 

innumerable highly varying dialects, many mutually unintelligible, from 

Okinawa island in the south to Amami Oshima and Kikai islands in the north. 

All of the Ryukyus are well south of Kyushu and northeast of Taiwan. 

I will, however, use the term "Okinawan" here specifically to mean the 

closely related dialects of the Naha-Shuri area, long the cultural and political 

center of the Ryukyu kingdom and now of Okinawa prefecture. 

Northern Ryukyuan constitutes one of probably four Ryukyuan lan- 

guages, the others being Miyako, Yaeyama, and Yonaguni.. All these lan- 

guages are mutually unintelligible, and all are also mutually unintelligible 

with any dialect of Japanese. The commonly held but largely unexamined 

notion that the Ryukyuan languages are dialects of Japanese is one based on 

politics, not on any linguistic criterion. This notion is inadvertently abetted 

by the fact that the Ryukyuan languages are genetically closely related to 

Japanese. Further, it is one-sided, since no Japanese go around saying that 

Tokyo Japanese is a dialect of, say, Okinawan. 

Writing systems and society Previous Okinawan writing systems were de 

facto standards. As far as we know, no central body was laying down the 

rules. There simply was a tradition that was followed. 

There is no standard modern Okinawan writing system, because the 

Okinawan language is being "ignored to death" by those who might save it, 

namely government, media, and educators. If Okinawan is to survive, a de 

jure writing system is needed, one in which a lively written communication 

may once again develop.


OKINAWAN AND THE DE FRANCIS CLASSIFICATION OF WRITING SYSTEMS 

The writing systems of Okinawan all fall within the DeFrancis (1989:58) clas- 

sification scheme under the rubric of either pure syllabic or a mixture of 

pure syllabic and morphosyllabic. DeFrancis gives Japanese (1989:131-143) as 

an example of these, since it has developed two so-called kana syllabaries. 

Yet it has never managed to break free of the thrall of the morphosyllabic 

kanji. That Okinawan should also have had such systems is no accident, 

since it has borrowed and adapted Japanese writing. 

TIME/TYPE-CLASSIFICATION OF OKINAWAN WRITING SYSTEMS 

Archaic age The writing of this age is the earliest available to us, with 

records from 1501 to the first half of the 1600's. The earliest materials are 

royal steles (Tsukada 1968:184-185, 307-308), and the writing system reached 

its zenith (and its end) in the compilation, over a period of nearly a century 

(from 1531 to 1623), of the Omoro sooshi, or book of omoro songs (Ikemiya 

1987a, Hokama and Saigoo 1972). 

Typologically this system is largely pure syllabic, using a syllabary, 

namely hiragana, with only a sprinkling of kanji, to write the Okinawan, then 

in use as a language of narrative and song. While we discern no important 

writing-system distinction between the steles and the Omoro sooshi, the lan- 

guage of the two is slightly different. I assume that the stele language is 

formal narrative language, while that of the Omoro sooshi is the language of 

song, and preserves earlier elements. In addition, Japanese language is fre- 

quently intermixed in the case of the steles, but not in the Omoro sooshi 

(Serafirn 1990). Writing is said to have been introduced from Japan around 

1200. (Sakihara [1987:8] gives it as 1187.) 

Examples of the stele writing system are as follows, with explanations 

immediately below. (Examples are from Serafim 1990 [Tsukada 19681.) 

(1) (kerai-wa- tihe, konomi-yowa- tihe> 

build-HONORIFIC-ing, plan -HONORIFIC-ing 

'building, planning' 

From the Yarazamori Fort stele (1554) 

(2) 

descend-HONORIFIC-ing 

' descending' 

From the Madama-Minato stele (1522) 

(3) (tasikiya-kuki, tui- sasi- yowa- tihe, 

dashikyespike stick-thrust-HONORIFIC-ing 

'hammering in the dashikyawood spike, 

asaka-"ne, to"me-wa- tihe, > 

asakageen stop- HONORIFIC-ing 

placing the asaka wood and the geen reed' 

From the Yarazamori Fort stele (1554) 

(4)

Schriffeslschhrif: Ersay in Honor of John DeFrur~cis 

The angled brackets enclose transliterated writing, with the value of each 

kana given in Kunrei-style romanization. The double apostrophes indicate a 

ditto mark in text. Commas are as in the originals, but I have placed 

hyphens and spaces to aid the eye in analysis of text. 

Examples of the writing system in the Omoro sooshi are as follows 

(Serafim 1990 [Nakahara and Hokama 1967:136A, 270AJ): 

(5) 

Shuri grove castle truly indeed build- HONORIFIC 

'indeed, constructs Shuri castle' 

(6) 

gods-sun SUBJECT protect-HONORIFIC lord-ruler 

'our lord, whom the gods and the sun protectJ 

All examples in earliest surviving texts are done in brushwriting, or in 

an imitation thereof on steles. Voicing marks are frequently omitted, and the 

orthography in many other respects does not match the pronunciation of 

modern Japanese kana. Variations in the spelling of a word show that there 

was not a one-to-one correspondence of spelling to pronunciation. 

I know of no English-language sources for the stele inscriptions other 

than what is discussed here. English-language sources for the Omoro sooshi 

are Sakihara (1987) and Drake (1990), though neither dwells on the writing 

system as such. In addition there are Serafim (1990, 1977, [in preparation]). 

The latter two treat the writing system in detail. Japanese-language sources 

include Nakamoto (1990:783-871), specifically on the writing system, and 

.Hokama & Saigoo (1972) and Nakahara & Hokama (1965, 1967). Many interest- 

ing recent exegeses of omoro have appeared, in a long series by Nakamoto, 

Higa, and Drake (1984-present), and a series recently collected into a book 

(Ikerniya 1987b), to which I have also contributed (Serafim 1987). 

Classical age The writing system of the classical age developed during the 

first few generations under the suzerainty of Satsuma, and was fully formed 

by the 1700's. Typologically it is a mixed kana-kanji system, in that respect 

mimicking the Japanese writing system. By this time well educated Okinawan 

males of the ruling class could read Japanese as well as Okinawan (and 

Chinese). 

The writing system differed from that .of the Archaic age in two impor- 

tant respects: (1) as already noted, this system was a mixed one; (2) the 

spelling conventions for the syllabary portions differed from those of the 

Archaic age of only a few generations before. 

The variety and amount of available texts for the study of the writing 

system of this period are also greater than those for the Archaic period. 

Text types include the kumiodori (dance dramas); ryuuka (Ryukyuan songs); 

and written histories, compiled at the direction of the court. The latter pro- 

vide a bridge, since at least the book that I cite below attempts to use 

archaic orthography for songs, though not always successfully. I give here 

a brief example from the Nakazato kyuuki (Takahashi and Ikemiya [1972:3]), 

where Classical orthography has intruded: 

(7) 

mishooru 

'says/does' 

.


Note the treatment of the equivalent of Archaic 

kaneru mumukwa f uya / !imi yachon n- dan. 

(kaneeru mumukwafuuya / !imi yachoon nn- dan.) 

such happiness TOPIC /dreams-even see-not , 

'I do not see such happiness even in dreams.' 

Also from KKKJ (1963:175A) is the following ryuuka: 

(9) (danziyu kareyosiya / irade sasi- miseru //> 

danju kariyushiya / !iradi sashi-miseru // 

(danju kariyushiya / !iradi sashi-miseeru //) 

truly auspicious / choosing point-HON. // 

'My, how auspicious (the day) that you choose!' 

 

!mi nu tsina turi- ba / kaji ya matumu. 

(!uuni nu tsina turi- ba / kaji ya matumu.) 

HONeship 's rope grasp-when / wind TOP. straightmon 

'You but grasp your boat's line to have the wind come on full.' 

It is an aspect of both styles that suprasegmental distinctions, including 

vowel length, vanish, since these are songs. Thus, the parenthesized 

material. I ignore other differences of the classical and modern language 

here. Spelling varies, as with ... , which is more frequently seen as: 

for example in Kookoo no maki (KKKJ 1963:174B). 

Modern age Since the late 19th century orthography has unraveled sub- 

stantially, yet the general situation is not chaotic. 

While Ryuuka are still written and performed, the way most people see 

Okinawan written most often is as loanwords in a Japanese text, for example 

in Okinawan newspapers or magazines. (Okinawa has an active publishing 

industry.) Thus, people see isolated words in katakana (just as with other 

"foreign" words), with no standard spelling. 

Difficulties include the written differentiation between phonological 

smooth and abrupt onset .of voice at the beginning of words, and between, 

e.g., tu and to, for which the Japanese syllabary is unequipped save through 

a digraphic spelling. Thus the suffix -gutu 'like, as' might be written 

(a carryover from Archaic and Classical spelling) or with a digraph, 

as in (an innovation following similar Standard Japanese innovations), 

though all agree that it should not be written , which would be pro- 

nounced gutsu or the like. The problem with is that then one might 

be at a loss for how to write -gutooru 'which is like/as', since is 

already in use to write tu. (Cf. the Rinken Bando and as 

furigana for -gutu [1990:9]. For more on furigana see below.) This is 

essentially a problem in awareness and use (or non-use) of the Classical con- 

ventions, since one may choose some of the more well known ones, such as

Schnffestschnijt: Bsays in Homr of John DeFrancis 

for tu, but be ignorant of the convention of for too. When only 

individual words are used in an otherwise Japanese text, such problems 

rarely come up, but they will either be dealt with when writing out an 

Okinawan text, or confusion will result. 

First I take up the way isolated Okinawan words are handled in 

Japanese texts, and then I discuss how Okinawan texts are handled. I do 

not pretend to treat all possible types of cases. 

The following example is from Nishirnura (1990), an appreciation of the 

work of the movie director Takamine Goo. Underlining denotes use of 

katakana, and the equal sign denotes the use of a length bar, a common fea 

ture of katakana for showing that the vowel sound corresponding to the 

preceding syllabograph is to be lengthened. Nishimura is following the 

orthography for the movie title used by Takamine himself: 

(11)


Tokyo and widely disseminated. The question then is: What language are 

these items really in? The answer is: It depends on who is reading. Kanji 

may be read in the original Okinawan pronunciation (typically by an 

Okinawan) or in a Japanese equivalent (by either Okinawans or others). If 

read off as Japanese, they may be seen as Okinawan loanwords, or simply as 

specialized Japanese terms. Since such loans are actually loan translations 

(Lea, item-for-item replacements), their status is easy to miss. Here are two 

closely related examples from Kadena (1982): 

(16)

SchnJffes~scschniji: &says irz Horwr of John &Francis 

Note also that 

Sino-Platonic Papers, 27 (August 3 1, 1 99 1) 

(20) + -+ = tu (1988:12) 

+ -+ = ti (1988: 16) 

+ -9 < k ~ = ~ kwa > (1988:26) 

+ -+ or

Schrz!estschnft: &sap in Honor of John DeFrWs 

This has occurred twice before, first in the Archaic system, adapted 

from the Japanese probably in the early part of the Kamakura period 

(1200's), and then again in the Classical system, adapted in the late 1600's. 

Certainly it would not be surprising to see yet a third adaptation, though in 

this case third time is not a charm! 

Such an adaptation is essentially what Funatsu has done, though using 

his modified syllabary. A similar approach could easily be taken using pre- 

sently existing letters to make digraphs. Included in any adaptation project 

will have to be determinations of which kanji may be used in what combina- 

tions (and theref ore in what readings), what constitute correct kan ji-and- 

kana sequences (i.e., okurigana rules), and so on, the very same rules that 

consume so much effort on the part of the Japanese in determining and 

learning what is correct in the Japanese writing system, 

REFERENCES 

DeFrancis, John. 1989. Visible speech: The diverse oneness of writing 

systems. Honolulu: University of Hawaii Press. 

Drake, Christopher. 1990. "A separate perspective: Shamanic songs of the 

Ryukyu kingdom. " Harvard journal of Asia tic studies 50.1:283-333. 

Funatsu Yoshiaki. 1988. Utsukushii Okinawa no kotoba [Okinawa's beautiful 

language]. Tokyo: Gikoosha. 

Hokama Shuzen and Saigoo Nobutsuna, eds. 1972. Omoro sooshi. [Copybook 

of omoros]. Nihon Shisoo Taikei, 18. Tokyo: Iwanami Shoten. 

Ikemiya Masaharu. 1987a. " Omoro sooshi gaisetsu: Hajimete yomu hito no 

tame ni" [An introduction to the Omoro sooshi: For those reading it for 

the first time). In Ikemiya 1987b:3-24. 

-------- , ed. 1987b. Omoro sooshi seika shoo [A selection of the quintes- 

sence of the Ornoro sooshi). Naha: Hirugi-sha. 

Kadena Sootoku. 1982. "Okinawashi sozoro aruki" [A stroil through 

Okinawan history] 10. Shin-Okina wa bungaku 53:130-133. 

Kerr, George H. 1958. Okinawa: The history of an island people. Rut- 

' land/Tokyo: Charles E. Tuttle Company. 

Kina Shookichi ando Champuruuzu. 1989. "Blood line." Japan: Polydor. 

KKKJ. See Kokuritsu Kokugo Kenkyuu jo. 

Kokuritsu Kokugo Kenkyuujo [National Language Research Institute], eds. 

1963. Okinawago jiten [Dictionary of Okinawan]. Kokuritsu Kokugo 

Kenkyuujo shiryooshuu 5 [National Language Research Institute materials 

collection, 51 Tokyo: Ookurashoo Insatsukyoku. 

Nakahara Zenchuu and Hokama Shuzen, eds. 1965. Koohon Omoro sooshi 

[ Omoro sooshi, critical edition 3. ' Tokyo: Kadokawa Shoten. 

-------- , eds. and cornps. 1967. Omoro sooshi jjten soo-sakuin [Dictionary 

and general index of the Omoro sooshil. Tokyo: Kadokawa Shoten. 

Nakamoto Masachie. 1990. Nihon rettoo gengoshi no kenkyuu [A linguistic 

history of the Japanese and Ryukyuan islands {sic}]. Tokyo: 

Taishuukan Shoten.


Nakamoto Masachie, Higa Minoru, and Chris[topher] Drake. 1984-present. 

"Omoro kanshoo: Ryuukyuu koyoo no sekai" [Appreciating omoro: The 

world of the old Ryukyuan songs]. Gekkan gengo [Language monthly] 

1984.10-present. 

Nishimura Ryuu. 1990. "Takamine Goo no eizoo sekai" [The world of 

Takamine Goo's images]. Okinawa taimusu 10.8:[n.p.]. 

Rinken Bando. 1987. "Arigatoo" [Thanks]. Okinawa / Tokyo: Maruteru / 

WAVE. 

-------- . 1990. "Nankuru" [Of its own accord]. Okinawa / Tokyo: Maruteru 

/ WAVE. 

Sakihara, Mitsugu. 1987. A brief history of early Okinawa based on the 

O m o sooshi. Tokyo: Honpo Shoseki Press [distributed by University 

of Hawaii Press]. 

Serafim, Leon A. 1977. "Palatalization in the Omoro sooshi: The 

orthographic and modern dialectal evidence. " Ms. 

-------- . 1987 [1977]. "Shuri mi-tunchi no kirnigimi" [The priestesses of the 

three holy precincts of Shuri]. In Ikemiya 1987b:72-80. Reprinted from 

Aoi umi 64:50-56 (Number 5 of the series "Kenkyuu to kanshoo: Okinawa 

saiko no kayoo-shuu Omoro sooshi"), July 1977. 

-------- . 1990 (19891. "Varieties of 16th-century Okinawan language: Local, 

continental, and Japanese evidence." Paper delivered at Association of 

Asian Studies Annual Meeting, March 1989, Washington, D.C., in back-to- 

back panel, "The Peculiar Organization of the Ryukyu Kingdom," Part 1, 

"The Chuzan Kingdom: Evolution and Organization." Revised ms. 

-------- . 1991. "Prospects for the survival of the Ryukyuan language and 

culture: A language-planning and language-policy perspective. " Paper 

delivered at Association of Asian Studies Annual Meeting, April 1991, New 

Orleans, Louisiana, in panel, "Ryukyu and Japan: The politics of lan- 

guage and culture in a 'homogeneous' statA" Ms. 

-------- . [In preparation]. A window on Okinawan cultural history: The 

orthography and phonology of the Omoro sooshi. Ms, 

Takahashi Toshizoo and Ikemiya Masaharu, eds. 1972 [n.d.]. Kume Nakazato 

kyuuki [Old chronicle of Kume (island), Nakazato (village)]. 

Photographic reproduction of copy in Nakahara collection, University of 

the Ryukyus Library. Privately pubiished by Takahashi and Ikemiya. 

Takamine Goo/Tsuyoshi. 1990. Watakushi no eiga monogatari" [My life as a 

moviemaker]. Ryuukyuu shinpoo 12.1:27. 

Tsukada Seisaku. 1968. (Moji kara mita) Okinawa bunka no shiteki kenkyuu 

[Okinawan cultural-historical studies (as seen through writing)]. Tokyo: 

Kinseisha. 

-.

Schn~esstschnifi: &says iri Horwr of John DeFrancis 

Proposal of a Comparative Study of Language Policies 

and Their Implementation in Singapore, Taiwan, and China (PRC)* 

Robert L. Cheng 

University of Hawaii 

* For John, who encouraged the long-awaited publication of this paper, which was first presented 

at the Asian and Pacific Planning Conference on Language Policy and Economic Opportunity at 

the University of Hawaii on December 22, 1980. Please note that the content and data have not 

been revised, and therefore, may be somewhat dated; however, the general policies of the three 

governments in question have changed little in their basic approaches to language planning. 

At the end, I have suggested some readings of works that have come out after the writing 

of this paper and of which have called my attention.


Despite many differences, there are at least three similarities that are shared by the three 

political entities of Singapore, Taiwan, and the People's Republic of China: the agent responsible 

for planning language policies is ethnically Chinese in each country; the overwhelming majority 

of the population of each country affected by each language policy is ethnically Chinese; and 

each has adopted the policy of promoting Mandarin. 

However, in China and Taiwan, Mandarin is intended to be the lingua franca of different 

linguistic groups of Chinese, as well as of non-Chinese; whereas in Singapore, Mandarin is being 

promoted as the lingua franca for different groups of Chinese only. In Singapore, for 

communication between Chinese and non-Chinese, English is encouraged, even though the 

national language is Malay. In addition to these common features of their respective language 

policies, there are also two features unique to these Chinese communities which may require 

special treatment. These peculiarities are the use of Chinese characters; and the strong influence 

of the traditional Chinese intelligentsia. 

With these similarities and peculiarities in mind, this paper proposes a framework for a 

comparative study of the language policies of these three countries and the way in which their 

implementation has in the past and will in the future affect the economic life of individuals who 

have different degrees of ability in different languages. The areas that I propose for comparison 

are: 

1. Language policy and sociolinguistic changes 

2. Language policy and language changes 

3. Agents (makers and implementers) of language policies and an identification of 

their purposes 

4. Language use and economic activities 

GI JAW POLICY AND SOCIOT mGI JI.mC CHA-N 

Regardless of whether or not there is a language policy, or systematic language planning, 

the sociolinguistic setting (who speaks what language) of a community is always changing. This 

is especially true in communities such as Singapore and Taiwan, where there is intensive contact 

between different languages and a constant influx of outside influences. One task of a language 

policy is to differentiate between desirable and undesirable changes, and to determine a strategy 

to bring about and accelerate the desired changes, while preventing or retarding the pace of 

undesired changes. Such decision-making is always political, because it involves the question 

of who decides what for whom. 

Given the sociolinguistic situation of Singapore, Taiwan, and the Southern Min speaking 

area of China, at the time their governments began to implement their language policies, there 

seemed to be no doubt that Hokkien-- called Taiwanese in Taiwan-- or Southern Min, as it is 

referred to in the southern part of China's Fujian Province, would eventually become the lingua 

franca and even the mother tongue of the residents in these communities. If the natural trend of 

Hokkien becoming the lingua franca of Singapore and Taiwan had been acceptable, the cost of

Schn@festschrif: Ersays in Honor of John DeFruncis 

the consequent language policy would have been minimal. 

We can predict that Hokkien will continue as a natural tendency to be the lingua fianca 

from several factors. Hokkien is spoken by the majority of the population in these areas; and 

the other 'dialects' spoken in these areas, such as Teuchew, Cantonese, and Hakka are much 

closer to Hokkien than Mandarin, since these are all southern dialects of China. (see Table 1) 

Sineawre Taiwan 

Hokkien SM (42%) Taiwanese (SM)(80%) Hokkien 

Teochew SM (22%) Hakka (10%) 

Cantonese (17%) Mandarin (1 0%) 

Hakka ( 7%) 

Hainanese SM (7%) 

others ( 5%) 

Table 1. (SM = South Min or Hokkien) 

Another factor predicting this tendency is that it has been a universal phenomenon for a 

lingua franca to develop in a Chinese community such that speakers of other 'dialects' learn and 

use it for interdialectal communication within the Chinese community. Such is the case of 

Hokkien in the Philippines, Hakka in Tahiti, Zhongshan in Honolulu, and Taishan in most other 

parts of North America. Any governmental policy that goes against this natural trend is very 

costly in terms of its political, economic, and cultural repercussions. 

It will take at least two generations (approximately sixty years) to attain the goal of 

having everyone fluent in Mandarin. It will take much longer for people to actually use 

Mandarin in their daily lives without affecting the normal functions of verbal communication. 

The labor and other resources required for learning Mandarin, a language alien to the majority 

of the community, is beyond calculation. 

During the process of promoting Mandarin, different groups of individuals will develop 

different degrees of competence in Mandarin, depending on their native language, age, type and 

level of education, and their occupation. Different language abilities often result in 

discrimination in the field of economic opportunities; i.e. the opportunity to participate in the 

productive process, to have access to final products, and to increase one's earning capacity. 

These factors will be examined in more detail in the forthcoming discussion. 

After implementing the policy of promoting Mandarin for twenty-five years in Singapore, 

thirty-five years in ~aiwan', and even longer in China; the sociolinguistic goal of increased use 

of Mandarin has been reached, although its success cannot be consider unqualified. While the 

overwhelming majority of productive activities are still carried out in Hokkien and other 

non-Mandarin vernaculars, the younger generation (those under forty in Taiwan, and those under 

thirty in Singapore) have acquired different degrees of ability in Mandarin. It should be kept in 

mind that even though they can speak Mandarin they do not normally do so.


The writer's observation suppoas the general view that Taiwan has been much more 

successful in implementing the policy of promoting Mandarin than either Fujian or Guangdong 

Province in China. Assuming this is true, there are several contributing factors to this situation: 

a. The Taiwan government has better facilities for promoting Mandarin; including 

the school system and mass media such as T.V., radio, newspapers, and magazines. 

b. Convenient transportation and communication make it possible for individuals to 

contact other people beyond their own speech communities. 

c. On average, the level of education is much higher in Taiwan than in China. Since 

the medium of instruction must be Mandarin, people in Taiwan have more years 

of compulsory use of Mandarin. 

d. Taiwan has a higher proportion of 'outsiders', 10-15% who speak Mandarin as 

their main language; and it is significant that the political elite belong to this 

group. In Fujian Rovince, outsiders are far fewer in number and do not 

necessarily have political power. 

e. In Taiwan, because of the higher degree of modernization and the prior experience 

of learning Japanese, the population is more prepared to learn a second language, 

partly to enhance their individual economic opportunities, as well as other 

motivations. 

f. The ideological difference between Taiwan and the PRC seems to play a role in 

the more successful implementation of the policy to promote Mandarin in Taiwan 

as compared to Fujian Province. Taiwan is a competitive society in which people 

strive to climb the social ladder, especially in education. Since Mandarin is a 

necessary tool to success in school, the incentive is there to learn and use 

Mandarin. In China, at least in theory, the equality and dignity of the working 

class is stressed and the reward system in effect over the past two decades did not 

favor individuals who did well in school. Members of the working class who do 

not do well in school cannot be too ill-treated from the communist ideological 

viewpoint. There might even be some motivation to identify oneself with the 

farmers and workers who speak the local vernacular rather than Mandarin. The 

incentive to speak Mandarin might be low if it suggested identification with the 

educated class. 

g. 

Finally, and in my view the most important, the promotion of Mandarin benefits 

those in power in Taiwan much more than it benefits those in power in Fujian 

Province or in Beijing. This was most obvious in the early stages of promoting 

Mandarin, during which local Taiwanese barely spoke any Mandarin, and thus 

hardly ever rose to important positions in government or in government-run

Schnffestschnjf: Ersays in Homr of John DeEraruris 

businesses. Some who already had government positions were subsequently 

removed because of their language background. If Mandarin had not been 

promoted, the mainlanders would have had to learn and use Taiwanese to compete 

with the Taiwanese majority--a drastic disadvantage fiom the viewpoint of those 

holding military and political power. At least in government-related job 

opportunities, the language policy was used as a tool to maintain their privileges. 

The question of what type of Mandarin should be used as the norm in Singapore, Taiwan, 

and China's Hokkien-speaking area, has been barely considered by the respective governments 

of these countries, but it will become more and more important since language constantly 

changes. The rate of language change is especially great when it is learned and used as a second 

language by the overwhelming majority of a community. Even though the schools aim to teach 

Beijing Mandarin, the Mandarin learned and used in Singapore and Taiwan nowadays has a 

distinctively local flavor; which is routinely regarded as Singaporian Mandarin and Taiwanese 

Mandarin. If it will take sixty years for everyone to acquire Singaporian or Taiwanese Mandarin, 

it will take much longer for them to speak Beijing Mandarin, or a Mandarin that will not stand 

out as an overseas brand of the dialect. 

It could be true that a decision on language norm or language change is not as political 

as a decision on sociolinguistic changes; however, even such a decision has political and 

economic consequences. If Beijing Mandarin is taken as the standard, more than 95% of 

Mandarin speakers need to spend some time and energy improving their Mandarin. 

In terms of economic opportunities, the minority of people who have a natural ability to 

imitate and learn the Beijing style of Mandarin would have a decisive advantage. Any new 

emigrant fiom China would be favored in outlying Chinese communities for jobs as radio 

announcers or T.V. actors. Two other areas of government intervention in language change that 

have consequences on job opportunities are (1) the enforced use of simplified characters and (2) 

the elimination of classical elements in writing. If these changes were enforced, the older 

intelligentsia who have established themselves by old-fashioned writing (using non-simplified 

characters and plenty of classical diction) would have to un-learn their own writing habits. They 

would lose their prestige as authorities on Chinese writing. The younger people, on the other 

hand, would no longer need to learn the old writing and, moreover, could now compete with the 

established elderly scholars. 

Another difficulty Taiwan and Singapore have to face is the standardization of new terms. 

Each country has its own methods and conventions of coining new words for new ideas and the 

many terms they now use are different from those used in the PRC. There are also many terms 

that have no counterpart in the PRC because of cultural differences or differences in the pace or 

direction of modernization. The idea of promoting Mandarin either to unify with China or to 

promote trade with China does conflict with the best interests of the people if their language is 

always branded as inferior in spite of their higher standard of living and technological 

advancement, and their distinctive political, social and economical system.


PO- AND D NEEATIDN OF THEIR PURPOSES 

Why is it that certain sociolinguistic or language changes are identified as desirable and 

need to be promoted, while other changes are branded as undesirable and therefore need to be 

prevented or discouraged? More specifically, why should Mandarin be learned and used, while 

the native language of the overwhelming majority of the population be discouraged? It is 

interesting to note how differently these questions are answered by the three governments. he 

Beijing government emphasizes unity, solidarity, and patriotism. The Taibei government 

rationalizes by saying that Taiwan is part of China and Mandarin is the national language of 

China. The Singapore government stresses that Singapore is too small and has to learn Mandariq . 

for international trade. Beneath these proclaimed justifications are complicated motivations that 

authorities have rarely admitted. 

At this point, the question of who participates in language policy formation and its 

implementation becomes very important. People usually 

think of their own interests first; it is the hidden, ulterior motives that more clearly explain the 

drive to promote Mandarin by the respective governments. Critics have pointed out that in 

Taiwan, the Nationalists had used Mandarin to keep the Taiwanese out of important positions in 

government. An all-out promotion of Mandarin was to the advantage of the mainlanders at the 

expense of the Taiwanese. In Singapore, Mandarin seems to have been a second choice. No 

dialect group was particularly happy that their own language was not selected, but neither was 

any group especially unhappy because there was no other group that managed to benefit at their 

expense. When Mandarin was selected as a . official language, every dialect group in Singapore 

was placed at an equal disadvantage. The only people that anticipated some advantage and hence 

supported the selection of Mandarin were those who were associated with Chinese schools. 

During the colonial period, Chinese in Southeast Asia were greatly inspired by the 

establishment of the Republic of China and the idea of China becoming a world power. 

Overseas, Chinese schools were modelled after schools in China imd became centers for 

cultivating Chinese nationalism. The Nationalists supplied teachers and textbooks that taught 

things such as "We are Chinese; we were born in Southeast Asia, but we all love China." It was 

during this period that many Chinese schools switched their media of instruction from Hokkien 

to Mandarin. Because Singapore is today an independent country, the promotion of Mandarin 

has to be justified on grounds other than Chinese nationalism or even ties with China. 

To understand these unexpressed justifications it is important to note that there was a 

group of educated people who had been deeply influenced by Chinese education. This elite had 

learned how to read and write in Mandarin, but had no knowledge of how to do so in their 

mother tongue. Because Mandarin was an official language, they had an advantage over others 

who did not havc this spccial skill. Howevcr, with their native Holrkien, thc actual lingua fianca, 

they had no advantage over others, and had to learn written Hokkien, which had not yet been 

standardized. 

Those political elite, who were educated in English, also saw Hokkien as unattractive, 

because they had attended English schools and had little formal training in expressing 

complicated ideas in Hokkien. Since it was not necessary in school, some barely acquired the 

lingua franca; they could not compete with the masses in Hokkien, but might do so in Mandarin.

Schniffestschri#: Essay in Honor of John DeFrancis 

In China, Mandarin was a reasonable choice, since the majority of the population spoke 

some form of Mandarin. Still, there are many respects in which educated people in Fujian 

Province can benefit themselves by following the policy to promote Mandarin in the Hokkien- 

speaking community. 

In spite of such differences in justifying a similar language policy, it is noteworthy that 

the main rationalization offered in the three countries has been based on a national necessity. 

They all fail to link the role of language policy with the economic life of a modem society. In 

particular they all fail to have any policy on Hokkien or to take a realistic look at what role each 

major language plays in the economy of the whole society and in the economic opportunities of 

the individuals who have different degree of competence in different languages. 

CrE I JSF, AND ECONOMIC 

How is an individual's ability in Mandarin, Hokkien and English related to his job and 

income? Are there striking differences between government and private employment in terms 

of such correlations? What languages are used in the production markets: when a foreman gives 

direction to his men, a carpenter trains his apprentices, a farmer buys his tools or discusses what 

and how to plant, a merchant bargains with his customers, or a banker interviews his loan 

applicants? What languages are used in the consumption market: in barber shops, hotels, trains, 

stores, restaurants, tour buses, taxis, movie theaters, or night clubs? What type of information 

is available (or not available) in a given language, and what section of the population has access 

to such infoxmation? If information concerning such things as how to cut hair, raise cows, or 

weave baskets is generally given orally in Hokkien, can an individual without knowledge of 

Hokkien have opportunities to acquire such skills? Is information of highly advanced technology 

and different types of skills available to people who cannot speak Mandarin or Hokkien? What 

about information concerning weather, marketing and the availability -of equipment and tools? 

How would the availability or unavailability of these types of information in a language affect 

the growth of industry as a whole and individual firms and farmers? A reliable answer to these 

questions can help those who are responsible for language policy and its implementation. It will 

also help researchers assess the effectiveness of these policies and their application. 

There are several ways to show how languages and economic activities are related and 

how a language policy and its implementation have affected such a relationship. I propose the 

following three areas of relationship: language ability and income of individuals; language used 

and types of productive processes vis-a-vis types of consumption of final products; and types 

of information available in different industries and their accessibility to speakers of different 

languages. 

It is hoped that a synchronic study of the correlation between these variables will show 

the relationship between language ability and economic activities. A diachronic comparison of 

these correlations at different points of time within the same community and a comparative study 

across communities will shed some light on the question of how a language policy has had or 

willhave effect on economic opportunities of the individual and on its relations with national 

economic growth.

Sirw-Platonic Papers, 27 (August 3 1,199 1 ) 

4.1 GE AND OF INDMDUU 

It seems reasonable to assume that the greater the number of languages one can speak, 

and especially the better one can speak the most widely used language, the better one can 

function in economic activities and therefore will have more opportunities for better paying jobs. 

If there is a positive correlation between the two variables, we may conclude that there is a fair 

and reasonable reward system when government policy tries to promote Mandarin. In the case 

of Taiwan, government employment does not seem to follow this rule of higher reward for higher 

competence in the most widely spoken language, Higher salary is correlated with Mandarin, 

which as we have seen, is not the native language of the majority of Taiwanese. Proficiency in 

Hokkien or Hakka, the native languages of most people in Taiwan, may be correlated with lower 

salary because lower ranking government employees have more chances to mingle with 

Taiwanese or are Taiwanese themselves. In the private sector, higher salary, as far as I can see, 

is related to proficiency in both Hokkien and Mandarin or Hakka and ~andarin.2 

The discrepancy between the language-salary correlation of the government employee and 

that of the private employee is an indication of the gap between the government and the people. 

If government employees are claimed to be public servants, there shouldn't be a big gap in the 

reward system. If a government wants to be effective, popular and responsive to the people's 

needs, such a wide gap should be taken as a serious warning. 

No matter how powerful and authoritarian a government is, it still does not have the 

facilities powerful and effective enough to force its people to use a certain language. Even 

though the government of Taiwan or Singapore has been very anxious to promote Mandarin, and 

therefore may officially ban the use of "unofficial" languages in public, they have been very 

realistic in allowing the use of unofficial languages in industry and business. People have a 

great deal of fieedom in their choice of occupation and use of language in their economic 

activities. 

It is important to note that in all three areas the overwhelming majority of commercial 

activities are not conducted in Mandarin but in Hokkien. It is also important to note that in the 

consumption market, the use of Mandarin is significantly higher than in the production markets, 

so much so that some farmers in Taiwan describe Hokkien as the language of production and 

Mandarin the language of consumption. To what extent this is true needs to be studied on the 

basis of empirical data. 

As the policy of promoting Mandarin has been most effective in schools, it can be 

expected that people competent in Mandarin will get more jobs that require longer formal 

professional training. It can also be expected that there are jobs that require professional training 

only and there are jobs that require good ability in Hokkien ador Mandarin in order to function 

well. 

Though never officially admitted by the Taiwan authorities, there has been discrimination 

against Taiwanese in governmental hiring and promotion, and in turn, reverse discrimination 

.

Schriftfestschri!: Essq in Horror of John DeFrancis 

against mainlanders in the private sector. TO some extent this has been related to the 

government's language policy which seems to imply it is unnecessary for mainlanders to learn 

Hokkien. Until the early seventies, mainlanders who had not learned Hokkien had to find jobs 

in the military, police, teaching, or enterprises run by the government. Members of the elite who 

chose to leave Taiwan during this period were more often than not mainlanders, rather than 

Taiwanese. How this trend has changed and to what extent the language is a factor affecting that 

change are interesting questions which need to be studied. 

If a language policy affects the equal employment opportunities of the individuals as has 

been noted, it can also affect the economic growth of a nation. This happens when school 

children are taught things related to the Mandarin world, but unrelated to the immediate needs 

of their community. When students are not taught to communicate within industry; or when the 

government makes it harder for non-Mandarin speakers to access, in their dialects, market 

information for particular industries, economic development suffers. In addition, preventing non- 

Mandarin speakers from expressing their views or participating in legislation affecting their 

respective industries in a language they can understand only serves to interfere with the growth 

of their industries. 

I personally spoke with a person whose job was to explain to farmers the nutrition and 

market value of a newly introduced crop. He had slides with directions recorded in Mandarin. 

I asked why the recording could not be in Hokkien. The reply was that for the purpose of 

promoting Mandarin, he was not allowed to record the oral directions in Hokkien. He added, 

however, that written instructions were generally understood and he often used Hokkien when 

explaining in person. 

Written information is especially a problem in these areas. Some'types of infomation on 

technological know-how are more readily available in writing. Once put into writing, using 

written Mandarin, it is easily translated into Hokkien or Hakka. There are types of information - 

that are not easily available in writing, such as manual skills and knowledge about the plants or 

insects peculiar to the local surroundings that has been passed down orally in the vernacular. 

Written information is lacking partly because subject matter of this nature tend to be neglected 

in school education, and partly because students are not trained to write in the native tongues. 

There are many Hokkien words that have no Mandarin equivalents, thus making translation to 

written Mandarin a problem. Compared with Mandarin, Hokkien has a richer vocabulary of 

movement and action, and has richer and more systematic grammatical categories in time 

relations. A study on the correlation between the growth of various industries and the languages 

used in each may reveal something significant about language use and economic strength. 

In this paper I have given a brief background of the language policies in Singapore, 

121


Taiwan, and the Hokkien speaking area of China. It has been shown that given the sociolinguistic 

situations of these Chinese communities, the natural trend is for Hokkien to become the hgua 

fianca, and promoting Mandarin as the lingua franca is at a tremendous expense to both the 

government and the people. I have pointed out that the respective governments have justified 

the promotion of Mandarin on different grounds, but they similarly claim it to be a national 

ne&ity, while neglecting the role that the unofficial languages play in the economic 

opportunities of the individuals and the economic growth of various industries. 

- - 

For the study of the economic consequences of language use and language policies I 

proposed collecting data concerning correlations among the following variables: an individual's 

ability in different languages, an individual's income, the frequency of use of various languages 

in various types of productive processes and consumption, and information available in different 

languages in different industries. I proposed diachronic studies of these conelations at different 

points along the course of implementing the language policies of the respective governments, 

along with a comparative study of these three countries, noting their many significant similarities. 

It is the view of this writer that, after considering the experience of different Chinese 

communities, language policy is the function of political power and the tradition of the Chinese 

intelligentsia, which characteristically uses non-simplified Chinese characters; it has a common 

core of cumculum on the abstract Chinese culture, ambitions for high position in the government, 

disdain for local culture y d language, and is alienated fiom the masses. 

As for my view of a viable language policy, I have argued on other occasions, and 

continue to here, that a bilingual policy that recognizes Hokkien, Hakka and Mandarin as official 

languages is best for Taiwan. 

1.In Taiwan the medium of instruction before 1945 was exclusively 

Japanese, after which Chinese was adopted. During the ten year 

transitional period that followed the defeat of Japan, Hokkien was 

used in elementary schools, with a gradual conversion to Mandarin. 

In Singapore the medium of instruction did not change overnight, 

with the switch from Hokkien to Mandarin occurring gradually. 

There was no sudden termination of English as a medium of 

instruction as had occurred with Japanese in Taiwan. Under the 

current bilingual policy, both English and Mandarin are used as 

media of instruction much more widely than before. 

2. For additional research into the correlation between income and 

language proficiency see: Wescott, K. (1979). "A survey of use of 

English in Hong Kong". Mimeograph. This study shows that high 

income levels correlate positively with high proficiency in the 

"high" language in a model of diglossia. 

Suggested Readings 

Berg, M.E. van den. 1988. Taiwan's sociolinguistic setting. In T. Cheng and S. Huang's The 

Structure of Taiwanese: A Modem Synthesis. Taipei: Crane Publishing Co., Ud. 

122 

*

Schrij&estschriift: &says in Honor of John &Francis 

Jordon, D.K. 1973. Language choice and interethnic relations in Taiwan. 

In Monda Lingvo-Problemo, 5, 35-44. 

Kubler, C. 1985. The Development of Mandarin in Taiwan: A Case Study of Language Contact. 

Taipei: Student Book Co., Ltd. 

Kubler, C. 1988. Code switching between Taiwanese and Mandarin in Taiwan. In T. Cheng and 

S. Huang's The Structure of Taiwanese: A Modem Synthesis. Taipei: Crane Publishing 

Co., Ltd. 

LePage, R.B. 1964. The National Language Question: Linguistic Problems of Newly Independent 

States. New York and London: Oxford University Press. 

Young, Russell. 1989. Language Maintenance and Language Shift Among the Chinese on 

Taiwan. Taipei: Crane Publishing Co., Ltd.

Sino-Platonic Papers, 27 (August 3 1, 1 99 1)

1. Introduction 

Schn'iftfestschn'fl: &says in Honor of John &Francis 

The Topical Function of 

Preverbal Locatives and Temporals in Chinese 

Feng-fu Tsao 

National Tsing Hua University 

Hsinchu, Taiwan 

Locative expressions in Chinese can occur in three different positions as exemplified 

by (1)-(3), while temporal expressions can occur in only two, both preverbal, as 

exemplified by (4) and (5): 

(1) zai Meiguo ta you hen duo pengyou. 

in America he have very many friend 

'In America, he has many friends,' 

(2) ta zai Meiguo you hen duo pengyou.1 

he in America have very many friend 

(i) 'He has many friends in America.' 

(ii) 'Speaking of him, in America, (he) has many friends.' 

(3) nei-fu hua gua zai aiane-shang. 2 

that-CL painting hang on wall-LOC 

(The painting was hanging on the wall.' 

(4) zuotian ta mei lai kan wo. 

yesterday he not come to see me 

(Yesterday, he didn't come to see me.' 

(5) ta zuotian mei lai kan wo. 

he yesterday not come see me 

(i) 

(ii) 

(Be didn't come to see me yesterday.' 

'Speaking of him, yesterday (he) didn't come to see me.'

Sino-Platonic Papers, 27 ( Aup t 3 1, 199 1 ) 

It is generally agreed that postverbal locatives such as the one in (3) should be 

analyzed as complements. I have also presented arguments elsewhere (Tsao 1978, 1979) for 

analyzing sentence-initial temporals and locatives such as those in (1) and 

topics.3 That this analysis is very well-motivated can be seen by extending 

(4) to (4a). 

(1) a. zai Meiguoi 

j 

hen duo pengyou 

in America he have very many friends 

1 j changchang da majiang. 

oft en play mahjong 

'In America he has many friends; (there) (he) often plays mahjong.' 

(4) a. zuotiani ta. meilai kan wo, ta. 

J i -J 

yesterday he not come see me he 

zuo libai qu le. 

do church: service go PART 

'Yesterday he did not come to see me; he went to church.' 

It can easily be shown that the locative expression zai Meiguo 'in America' in (la) and the 

temporal expression zuotian 'yesterday' in (4a) have all the grammatical qualities of a 

primary topic. That is, they occur sentence-ini tially; they are definite in reference;4 they 

extend their domain to more than one clause; and finally, they are in control of 

coreferential NP deletion or pronominalization in their respective chains. 

Semantically, the locative in (la) provides a physical setting for the two comment 

clauses and likewise, in (4a) the temporal expression gives a time frame for the two 

corrmient clauses in the chain. Logically, as Barry (1975) has pointed out, the locative and 

the temporal in (la) and (4a) are "indicators of universe within which events hold true." 

We have thus proved beyond any reasonable doubt that sentence-initial temporals and 

locatives such as those in (la) and (4a) are primary topics. 

The purpose of this paper is to show that nonsentence-initial preverbal locatives 

and temporals such as those in (2) and (5) can in certain contexts pla the role of a topic, 

albeit a non-primary one. In other words, sentences like (2) and (57 are often, taken in 

isolation, subject to two structural analyses as reflected in the two translations of each 

sentence. 

However, in order to provide a general background for the understanding of the 

proposed analysis and our arguments in support of it, we need to digress a little to discuss 

adverbs in Chinese in general.

2. General Remarks on Chinese Adverbs 

Schniftfes~schn~: fisaqs in Honor of John &Francis 

This is certainly no place to go into a detailed discussion of adverbs in Chinese. 

What we would like to do in the following is to concentrate on some aspects that are of 

immediate concern to our topic at hand. Specifically, we would like to take up two 

import ant questions concerning Chinese adverbs, namely, (i) the problem of identifying 

adverbs in Chinese; and, (ii) the placement of different types of adverbs in a multiple 

adverbial construction. 

2.1 The Problem of Identifying Chinese Adverbs 

Just as in many other languages, the adverb in Chinese as a category is an 

. extremely ill-defined cover term for a number of different categories. Tai (1976:393) calls 

it "a wastebasket for a variety of linguistic entities which bear different semantic relations 

to different parts of a sentence." This being the case, it is really difficult to set up criteria 

to identify what adverbs are in Chinese. For instance, Guo (1962) defines an adverb as: "a 

constituent that is placed before a verb or an adjective, but never before a noun to indicate 

degree, scope, time, negation etc." Evidently, there is a catch in the definition in the form 

of "etc.". But even with this vagueness, this definition excludes many linguistic entities 

that other linguists would readily classify as adverbs. Witness (6) and (7). 

(6) minexian-de, ta bu zhidao zhe-jian shi. 

clearly he not know this-CL matter 

'Clearly, he was not aware of the matter.' 

(7) huaner-huan~-zhan~-zhan~ae ta pale jin-lai. 

in:a:flurry he run-ASP enter-come 

'In a flurry, he ran in (toward the speaker).' 

Thus, mingxian-de 'clearly' and huanrr-huang-hane-zhang-de 'in a flurry' can both 

occur before a noun, and yet most linguists would agree to assign them to the category of 

adverbs. 

However, rather than make any attempt to fix up the definition so that it can cover 

all adverbs, which is a task evidently beyond the scope of this section, we would like to 

take up an area which contributes to the difficulty of defining adverbs in Chinese. This 

area, which has a great deal to do with the topic of the present paper, concerns the 

ambivalence of some kinds of expression which occur preverbally. 

Certain expressions in English also exhibit this ambivalence, as Lyons (1977:474) 

points out: 

The difference bet ween certain locative adverbials and place-referring 

nominals is not, in fact, clear-cut in all syntactic positions in English. 

For example, the demonstrative adverbs 'here' and 'there' and the 

demonstrative pronouns 'this' and 'that' are equally appropriate as 

substitutes for 'this placeV/'that place' in an utterance like (This/that 

place is where we agreed to meet.'


n 

Lyons restricts his comment here on locative expressions. Actually, the same commq is 

equally applicable to temporal expressions in some contexts. Examine (8). 

(8) Yesterdav being Sunday, we went to church at about ten. 

Chinese temporal and locative expressions in certain positions also exhibit this 

ambivalence. This is clear when we translate the English sentences mentioned above into 

Chinese. The problem in Chinese, however, is aggravated by a pronounced tendency to 

elide the prepositions in many prepositional phrases. This tendency was very strong in 

archaic Chinese and is still strong in modern standard Chinese. This is exactly the reason 

which prompted Wang Li (1955, 1980) to posit a special category of words called "nominds 

in the relational function, i.e., they have the function of a prepositional phrase but the 

preposition, the governing category, is unexpressed. The following are some of Wang's 

examples. 

(9) Peng-shi zhi zi ban dao er wen yue, 

Peng-shi POSS son half way PART ask say 

'I ' jun jiang he zhi? " 

Lord will where go 

'Pengshi's son during the trip asked, "Where is my Lord going?'' 

(Mozi, 5th c. B.C.) 

(10) &j shu wei da? 

matter which be great 

'Of all the maiters, which is the most important?' 

shi qin wei da. 

serve parents be great 

'To serve one's parents is the most important ,' 

(11) zhe-li bu mai piao. 

(Mencius, 4th c. B.C.) 

this-place not sell ticket 

[i!) 

'(We) don't sell tickets at this placethere.' 

'This place does not sell tickets.' 

(modern standard Chinese) 

b

. . 

Schnfrfts~hchrp: Essay irl Honor of John &Francis 

(12) san ai an ku ai aian mai-le yi-jia 

three thousand dollar money buy-ASP one-CL 

gangqin. 

piano 

'(With) three thousand dollars, (we) bought a piano.' , 

(modern standard Chinese) 

Wang (1980:388-394) correctly remarks that omission of preposition in this type of 

structure was more prevalent in classical Chinese than it is in modern Chinese. He also 

observes that nominals bearing this function are for the most part locative, temporal and 

scope-delimiting expressions.5 Other types of nominals such as instrumentals and 

benefactives do occur, as in (12), but only rarely. 

It is the same consideration which prompted Chinese grammarians (Zhu, 1950; Guo, 

1960; Chao, 1968 and Lu et al., 1981 among others) to analyze the underlined expressions 

in (9)-(12) as nouns at the lexical level, which are then said to have the function of 

adverbial modifiers, or zhuan~~ to use the terminology employed in mainland China, 

syntactically. WhileAhis approach is able to characterize the expressions involved at both 

levels, it fails to explain why in Chinese, but not in English, there are so many nominals 

used to modify verbs (including adjectives). Neither does it explain why most of the 

expressions having this function are temporals and locatives rather than instrumentals and 

benefactives. We will attempt to give an explanation later in the section. 

2.2 Placement of Multiple Adverbials 

When there are several adverbial expressions appearing in a row preverbally in a 

sentence, the most information-wise neutral and unmarked order'seems to be: temporal 

(including those of specific time, duration and frequency) > locative > benefactive > 

manner > instrumental (Chuo, 1987; Li et al., 1983; Zhu, 1959), as exemplified by (13). 

(13) nei-ge lao furen, qunian dongt ian s hichanq 

that-CL old woman 1ast:year winter often 

zai iia-li wei ta erzi renzhen-de 

at home-LOC for her son earnestly with 

gouzhen zhi maoxi an yi .6 

hoo ked:needle knit sweater 

'The old woman often knit sweaters with hooked needles for her son at home 

during the last winter.'

Sino-Platonic Papos, 27 (August 3 1,199 1) 

The above sentence is taken from Chuo (1987), who also discusses in some detail the 

placement of some posit ion-wise versatile adverbs such as 'again', gu yi 'in tent ionally', 

'probablyt7 and 'not'. We feel that it is a very valuable approach to discuss 

these adverbs separately and we will return to the placement of some of these adverbs in 

the next section. 

But before we leave this topic, we would 1ike.to raise a very important question that 

many researchers have taken for granted: Why is there such an order of adverbial 

placement? More specifically, we would like to know whether it is fortuitous that- 

temporals and locatives precede-all others. 

3. Temporals and Locatives as NOR-primary Topics 

To the best of my knowledge, the first linguist who specifically analyzed adverbials 

that occur between the primary topic and the verb as topics is Hockett. He (1958:201-203) 

comments: 

Many Chinese comments consist in terms of a topic and comment so 

that one can have a sentence built up of predications within 

predications, Chinese-box style. 'Wo jintian chengli you shi' freely 'I 

have business in town today' has topic 'wo' '1' and the remainder as 

comment. 'Jintian chengli you shi' 'There is business in town today' 

in turn has topic 'jintian' 'today' and the remainder as comment. 

'chengli you shi' 'There is business in town' consists of topic 'chengli' 

'in town, town's interior' and comment 'you shi' 'there is business.'8 

Chao (1968) also recognizes the existence of non-primary temporal and locative topics, 

although he does not explicitly call them as such. He states' (op. cit. p.534): 

If there are both time and place words as subjects [topics in our 

terms, F.T.], the time word usually though not always precedes the 

place word, as in jintian haishang fenglang hen da. "Today on the sea 

the wind and waves are high." But the main topic is what decides the 

main . subject [the primary topic, F.T.]. For example, women iiali 

jlnnian auonian. keshi aunian meivou. 'In our house, we celebrate the 

New Year this year, but last year we didn't.' where the place word 

jiali is main subject [the primary topic] under which jinnia,n and 

ounian are smaller subjects [non-primary topics] .9 

Neither Hockett nor Chao, however, give any specific argument for this analysis. In 

what follows we would like to present our arguments in its support.

Schnffestschhnfr: &says in Honor of John DeFrancis 

3.1 Placement of the Pause Particles 

One of the grammatical qualities that the primary topic has is that it can be 

followed by one of the pause particles, a (s), & , me and E. The same particles can also 

follow a locative or temporal appearing between the primary topic and the main verb, as 

exemplified by (14) and (15). 

(14) ta zuotian meiyou lai. 

he yesterday PARTnot come 

'Speaking of him, yesterday (he) didn't come. ' 

(15) ta zai Meiguo ya you hen duo pengyou. 

he in America PART have very many friend 

'Speaking of him, in America (he) has many friends.' 

Since a ause particle in Chinese occurs between the topic and the comment part of 

a sentence, (147 and (15) indicate clearly that zuotian Lyesterdayl in (14) and zai Meiguo 

'in America' in (15) are perceived by native speakers as belonging to the topic part of the 

sentence involved. 

3.2 Definiteness in Reference 

Like the primary topic, a temporal or a locative occurring between the primary 

topic and the main verb is definite in reference in most cases as exemplified by (16). 

(16) Li Xiaojie zuotian cheng-li you shi. 

Li Miss yesterday town-LOC have business 

'Speaking of Miss Li, yesterday in town she had business.' 

It is clear that the temporal zuotian 'yesterday' and the locative chengli 'in town' in (16) 

are both definite. There are, however, two minor points that need to be taken care of in 

this connection. First, if you 'EXIST' is analyzed as an indicator showing that the 

following NP is indefinite but specific, i.e., its reference is identifiable to the speaker but 

not to the hearer, then we have to allow for cases where the temporal or locative expression 

involved is indefinite but specific. Compare (17) with (16). 

(17) Li Xiaojie you yi-tian jin cheng lai kan wo. 

Li Miss EXIST one-day enter town come see me 

'Speaking of Miss Li, one day (she) came to town to see me.' 

131


Notice that an indefinite, nonspecific temporal or locative is still not allowed as a 

secondary topic as attested by the ungrarnrnaticali ty of (18). 

(18) *Li Xiaojie yi-tian ya jin cheng lai 

Li Miss one da.y PART enter town come 

kan ~0.10 

see me. 

Notice also that if such an analysis of vou is adopted; then the referential constraint 

on the primary topic will have to be relaxed to allow for'cases of specific NPs as well as 

temporals and locatives. Compare (16) with (16a), (17) with (17a), and (18) with (18a). 

(16) a. you yi-ge ren zuotian cheng-li 

EXIST one-CL person yesterday town-LOC 

you shi. 

have business 

'Someone had business in town yesterday.' 

(17) a. you yi-tian Li Xiaojie jin cheng. 

EXIST one-day Li Miss enter town 

lai kan wo. 

come see me 

'One day Miss Li came to town to see me.' 

(18) a. *yi tian Li Xiaojie jin cheng lai kan wo. 

oneday Li Miss enter town come see me 

As expected, both (16a) and (17a) are grammati'cal while (lsa), in which the indefinite, 

nonspecific temporal secondary topic is fronted to become the primary topic, is not. So 

when the referential constraint is thus revised, it works for both the primary topic and the 

secondary topic played by a temporal or a locative. The parallelism remains intact. 

Second, if the expression involved is a prepositional phrase then the referential 

constraint applies to the NP in the phrase rather than to the whole prepositional phrase. 

This happens only rarely and it happens more often with the locative phrase than with the 

temporal phrase. 

B

S~hnJffeschrifr: &says in Homr of John DeFrancis 

3.3 Tl1e Contrastive Function 

One of the discourse functions of the primary topic is to provide contrast (see Barry, 

1975; Tsm, 1979, Chapter 6). This can be clearly seen in the following examples. 

(19) & bu qu; qu. 

he not go I go 

'(If) he doesn't want to go, I will.' 

(20) fan bu chi le, zai duo he 

rice not eat PART wine still more drink 

yi-dian. 

a-little 

'(As for) rice, we will have no more, but wine, do drink a little more.' 

Likewise, secondary topics such as the second nominal in the double nominative 

construction are often used contrastively as in (21). 

(21) ta -5 zhang de hen hao-kan, - bizi 

he eye grow PART very good-looking nose 

que bu zen-me-yang. 

on: t he: con t rary not so:great. 

'Speaking of him, (his) eyes are very beautiful, (but) (his) nose is just so-so.' 

Now examine the temporals and locatives occurring in the position in question. 

They, too, possess this function, as shown in (22) and (23). 

(22) ta zai Taiwan you hen duo pengyou, zai zhe-li 

he in Taiwan have very many friend in this-LOC 

yi-ge ye meiyou. 

one-CL also not:have 

'In Taiwan he has many friends, (but) in this place he has none.'

Sino-Platonic Papers, 27 (August 3 1. 1991) 

(23) women jia-li iin-nian gU0 nian, 

our house-LOC this-year celebrate New:Year 

keshi au-ni an mei you. l1 

but last-year not :have 

'(In) our house, (we) celebrate the New Year this year, but last year we 

didn't.' 

Thus, it is clear that texnporals and locatives occurring in the position under 

investigation behave like other non-primary topics in having the function of 

contrastiveness just like the primary topic. 

3.4 Placement of Adverbs You and 

In Tsao (1982) the placement of yo^ 'again' and 'also' is used as a test to 

distinguish three const ructions, namely, productive double nominative construction, 

sentences with semi-SP compounds and sentences with frozen SP compounds. The reason 

that the placement of vou 'again' and yg 'also' can provide such a good test is that both 

- 

vou 'again' and 'also' belong to the topic component while what follows them belongs to 

the VP component. This interpretation is in agreement with Chuo's observation (1987) 

about you 'again', which he calls a "repetitive adverb". In his paper he compares sentences 

such as (a) and (b) in (24) and (25). 

(24) a. ta-de pengyou vou zai shane-e libaitian 

he-POSS friend again on last-CL Sunday 

lai zhao ta. 

.come see him 

'His friend came to see him again last Sunday.' 

b. ta-de pengyou zai shang-e libaitian YOU 

he-P O SS friend on last-CL Sunday again 

lai zhao ta. 

come see him 

'Speaking of him, last Sunday his friend came to see him again.'

Schnfrfesestscschrifi: &says in Honor of John DeFrancis 

(25) a. ta yo^ zai xuexiao-li da-le ren. 

he again in school-LOC hit-ASP person 

'He hit a person at school again. ' 

b. ta zaixuexiao-li da-le ren . 

he in school-LOC again hit-ASP person 

'Speaking of him, at school (he) hit a person again.' 

He observes that the difference between the (a) and (b) sentences in each pair lies in the 

"shifting of focus". In the (a) sentences the focus is laid on the adverbial following yo^ 

'again' while in the (b) sentences it is on the verb (1987:137). Since according to our 

interpretation, only what precedes you 'again' can be topic, which normally carries known 

information, the adverbial in the (a) sentences can not be part of the focus in the respective 

sentences. The two observations are, therefore, in agreement. 

With this observation in mind? let us go back to the temporal and locative in 

question. Since they can appear both before and after.^ 'again' and 'also', it is only 

the temporals and locatives that appear before these two adverbs that are secondary topics, 

as those in (24b) and (25b). (24a) and (25a), on the other hand, are single-topic sentences 

with an adverbial modifier. We can easily justify this interpretation by adding another 

comment clause to (a) and (b) sentences in (25) as in (26 a and b). 

(26) a. ta. you zai xuexiao-li da-le ren, 

-1 

he again in school-LOC hit-ASP person 

suoyi bu gan huijia. 

1 

so not dare go:home 

'He hit a person at school again, so (he) dared not go home.' 

b. tai zai xuexiao-li j 

you da-le ren 

he in school-LOC again hit-ASP person 

suoyi 

. bei laoshi chufa le. 

1 J. 

so BE1 teacher punish PART 

'Speaking of him, at school (he) hit a person again, so (he) (at school) 

was punished by the teacher.'


3.5 Domain and Control Properties 

(26b) also shows clearly that temporals and locatives in question can extend their 

domain to more than one clause, a very important property which we have proved that the 

primary topic possesses. However, there is a difference. While a secondary topic can 

extend its domain to more than one clause, it can do that only when the primary topic also 

does so at the same time. A primary topic is evidently not subject to such a restriction. 

Likewise, (26b) shows that the locative or the temporal in question has the control 

property that a primary topic has, i-e., it is in control of the coreferential NP deletion and 

pronominalization in the following clauses in the same chain. But again there is a 

difference. A secondary topic controls the NP deletion and pronominalization only when 

the primary topic does so at the same time. A primary topic is never subject to such a 

constraint. 

. . 

3.6 Similarities to Other kinds of Secondary Topics 

The possessed NP in the double nominative construction often ends up as a 

secondary topic as in (27a). 

(27) a. ta shuxue hen hao. 

he math very good. 

'Speaking of him, (his) math is very good.' 

However, the possessed NP can, in a proper context, be promoted to a primary topic as 

shown in (27b). 

b. shwcue ta hen hao. 

math hevery good 

. 'Speaking of math, he is very good.' 

It has been pointed out that when the possessed NP becomes a primary topic, its 

meaning is somehow changed. It can now only be interpreted in a generic sense. Shuxue in 

(27b), for instance, can only mean 'Speaking of math in general'. It does not denote 'his 

math' as it does in (27a). 

This change of interpretation, however, can be explained in terms of a very general 

rule of topic scope interpretation, which can be roughly stated as (28). 

(28) Theprimarytopic> thesecondary topic> thetertiarytopic ..... 

where "> " means "has a larger scope than"

Schnftfestschij: &says in Honor of John DeFrmcis 

Since a possessed NP is, by definition, only part of the possessor NP, when it becomes the 

primary topic, it cannot retain its original meaning without conflicting with the topic scope 

interpretation rule. Only when it takes on a generic sense is it compatible with the rule 

just mentioned. 

Ths interpretation rule aside, what is shared by the secondary topic played by the 

possessed NP and that played by a temporal or locative expression in questioil is that both 

can be, in a proper context, promoted to become the primary topic. Compare (29) and 

(30) with (27). 

(29) a. ta. zuotian. 

1 J 

lai kan wo le, 

he yesterday come see me PART 

i 

hai dai-zhe tait ai yiqi lai. 

still take-ASP wife together come 

'Speaking of him, yesterday (he) came to see me,(and) (he) brought his 

wife with him. ' 

b. zuotiani shi xingqitian, suoyi i ta 

yesterday be Sunday so he 

lai kan wo le. 

come see me PART 

'Yesterday was Sunday, so (yesterday) he came to see me.' 

(30) a. ta. zai Meiguo. 

1 

3 

you hen duo shiye, 

he in America have very many enterprises 

i 

you hen da-de yingxiangli . 

have very big influence 

'Speaking of him, in America (he) has many enterprises, (and) (there) 

(he) has a great deal of influence.' 

b. zai Meiauoi renren dou dei shou 

in America everybody all must abide:by 

fa? I ta ye bu liwai. 

law he also no exception 

1


'In America, everybody has to abide by the law, (and) (there) he is no 

except ion. ' 

I have also argued elsewhere (Tsao, 1987, 1989a & 1989b) that the &J NP in a 

construction, the compared NPs in a comparative structure and the lian constituent in the 

- lian ... dou/s construction occurring in the postion between the primary topic and the 

verb are all non-primary topics. If non-S-initial preverbal locatives and temporals are 

topics as we have argued, then we would predict that they can occur in all these 

constructions. This prediction is borne out by the following sentences. 

(31) a. @ zai .. - aian~shang wa-le yi-ge dong. 

he at wall-LOG dig-ASP a-CL hole 

'He dug a hole through the wall.' 

b. @ ba aiangshang wa-le yi-ge dong.12 

he BA wall-LOG dig-ASP a-CL hole 

'What he did to the wall was dig a hole through it.' 

(32) a. jx~ xingqitian bu xiuxi. 

he Sunday not rest 

'He doesn't take a rest on Sunday.' 

b. lian xingaitian ye bu xiuxi. 

he including Sunday also not rest 

'He doesn't take a rest even on Sunda,y.' 

(33) a. - ta jintian hen shufu. 

he today very cornfor t able 

'He is feeling well today.' 

b. jintian bi zuot ia.n shufu. 

he today compare yesterday coinfort able 

'He feels better today than he did yesterday.'

SchnftSestschrij: <strong>Essays</strong> in Honor of John &Francis 

(31) is especially interesting as it shows that not only can a11 object NP become a NP, 

as it is generally assumed but also a locative that is a secondary topic. This is a further 

confirmation of our theory that a NP is a non-primary topic. 

In a recent paper (Tsao, Forthcoming), I have presented a number of arguments in 

support of the analysis that treats an important class of clause co~lnectives such as suiran 

'although', yinwei 'because' as occurring in the COMP in the deep structure, as shown in 

(34a), whose occurrence in other positions as shown in (34b) is then accounted for by the 

rule of topi c-raising. 

(34) a. yinwei tai sheng bing suoyi .mei 1 lai. 

because he get sick so not come 

'Because he was sick, he didn't come.' 

b. yinwei slleng bing suoyi lei lai. 

he because get sick so not come 

'Roughly, same as (a). ' 

Notice that if there is such a rule, then, in addition to the primary topic, we will have to 

allow non-primary topics like the fronted object NP and the second nominal in the double 

nominative construction to be raised, as sentences in (35) show. 

(35) a. yinwei ta. nei-ben shu hai mei kan, suoyi 

-1 j 

because he that-CL book yet not read so 

1 

bu zhidao hao-bu-hao. 

not know good-not-good 

'Because he hasn't read the book yet, he doesn't know whether it is good 

or not.' 

b. yinwei nei-ben shu hai mei kan, suoyi 

j 

he because that-CL book yet not read so 

i j 

'Roughly, same as (a).' 


not know good-not-good

Sino-Platonic Pups, 27 (August 3 1, 199 1) 

c. ta. nei-ben shu yinwei hai mei itan, suoyi 

-1 j 

he that-CL book because yet not read so 

1 



not know good-not-good 

If non-S-initial preverbal locatives and temporals can be non-primary topics, as we 

have argued, then we would expect them to allow raising as well. This is indeed the case, 

as exemplified in (36). 

(36) a. yinwei Qi zuotian. sheng bing, suoyi 

J 1 j 

because he yesterday get sick SO 

mei lai kai-llui. 

not come at tend-meeting 

'Because he was sick yesterday, he didn't attend the meeting.' 

b. yinwei zuotian. sheng bing, suoyi . 

J 1 j 

he because yesterday get sick so 

mei lai kai-hui . 

not come at tend-meeting 


c. ta. zuotian. yinwei sheng bing, suoyi 

-1 J 1 j 

he yesterday because get sick so 

mei lai kai-liui . 

not come at tend-meet ing 


Along the same line, I have also argued in the same paper (op. cit.) that in Chinese 

- sihu''seem1, kan-oi-lai 'seem, look', and keneng 'possible' etc. are all raising-predicates 

that allow various topics, temporals and locatives included, to be raised as exemplified in 

the following sentences.13

Schnftestschri!: &sap itr Honor of John DeFratzcis 

(37) a. keneng - ta mingtian na-chan~ Q iu hui shu. 

possible he tomorrow that-CL balkgame will lose 

'It is possible that 11e will lose the game tomorrow.' 

b. Q keneng mintian na-chang aiu hui shu. 

he possible tomorrow that-CL bal1:garne will lose 


c. & rninntian keneng na-ch ane: aiu hui shu. 

he tomorrow possible that-CL bal1:game will lose 


d. ta rningtian na-chane; aiu keneng hui shu. 

he tomorrow t hat-CL bal1:game possible wi 11 lose 

'Roughly, same as (a) .' 

From the above discussion, it is clear that non-S-init ial preverbal temporals and 

locatives do pattern with other non-primary topics in their syntactic behavior in many 

cases and should be analyzed as non-primary topics. 

Finally, we would like to give some examples to show how various topics, primary 

and non-primary, interact to give rise to a variety of sei~tences differing only in the order 

of these topics. 

(38) a. ta-de vaniing zuotian huai le. 

he-POSS eye-glasses yesterday break PART 

'Speaking of his glasses, yesterday (they) broke.' 

b. zuotian ta-de vaniin~ huai le. 

yesterday he-P OSS eyeglasses break PART 

(Yesterday, his glasses broke. ' 

(39) a. jintian hai-shang feng 1ang hen da.14 

today sea-LOC wind waves very big 

'Today on the sea the wind and waves are high.'


b. haishang jintian feng lang hen da. 

sea-LOC today wind waves very big 

(On the sea today the wind and waves are high. 

c. fena lang jintian haishang hen da. 

wind waves today sea-LOC very big 

'Speaking of wind and waves, today on the sea (they) are high.' 

d. fene. lanq hai-shang 5ntia.n hen da. 

wind waves sea-LOC today very big 

(Speaking of wind and waves, on the sea today (they) are high.' 

4 Summary of Arguments and Theoretical Ramifications 

To sum up, we have found on the one hand that temporal and locative expressions 

occurring between the primary topic and the main verb possess all the qualities of a 

primary topic except in some cases the qualities involved have further restriction in the 

case of temporals and locatives. On the other hand, we have also found that the temporal 

and locative in question and the secondary topic in a number of constructions have a great 

deal in common. We have thus proved beyond any reasonable doubt that 

non-sentenceinitial, preverbal locatives and temporals can be non-primary topics. 

This conclusion of ours is further supported by the following two observations. 

First, in our discussion of adverbs in general we have found that universally, locatives and 

temporals have possessed more nominal quality than other kinds of adverbials. This then 

explains why they are easier to become topics for, even though topics are not completely 

restricted to norninals, most of them are, and, other things being equal, the more nominal 

quality a constituent has, the more likely for it to become a topic. This also accounts for 

the fact that Chinese allows far more prepositions in a prepositional phrase, especially 

those expressing time and location, to drop than English does. This is so because Chinese 

is far more topic-oriented than English. 

Second, we have reported the findin s of many linguists that the information-wise 

neutral version of the order of placement o f a multi-adverbial coi~struction is: temporal > 

locative > benefactive > manner > dative > instrumental and we have raised the question 

of why temporals and locatives should come first. We are now in a better position to 

answer the question: Temporals and locatives head the hierarchy because they are, of all 

adverbials, the easiest to become topics. This observation also implies that other types of 

adverbials, though not as commonly as temporals and locatives, can become topics as well. 

This is indeed the case, as can be seen by t.he following examples. 

(40) @ weile ta-de haizi ya chi-le 

j 

hen duo 

he for he-POSS child PARTeat-ASP very much

ku, 

Schrifjcestschrijt: &says in Honor of John DeFrancis 

1 j 

zhe ji nian lao-le 

suffering these few year old-ASP 

xu duo 

very much 

'Speaking of him, for his children, he underweilt much suffering (and) in the 

past few years (he) has become much older.' 

(41) & wu-kuai qian a mai-le nei-jian 

he five-dollar money PARTbuy-ASP. that-CL 

da-yi . 

overcoat 

'Speaking of him, wit 11 five dollars (he) bought the overcoat. ' 

Thus, by positing certain adverbials, inainly temporals and locatives, as 

non-primary topics, we are able to explain these two peculiar phenomena about Chinese 

adverbials very nicely. These two observations can, in this way, be regarded as indirect 

supports for our analysis.


Notes 

An earlier version of this paper was presented at the 1989 International Conference 

on Sino-Tibet an Languages and Linguistics, October 5-8, 1989, University of Hawaii, 

Honolulu, Hawaii. I would like to thank people who attended the conference, especially 

Prof. S. H. Teng of University of Massachusetts and Prof. Jian-ming Lu of University of 

Peking, for comments. Thanks are also due to the National Science Council of Republic of 

China for providing a travel grant to enable me to present the paper. 

1. Depending on whether the locative zai Mei~uo 'in America' is a topic or not, (2) can 

have two different interpretations as shown in (i) and (ii) in the English translation. The 

same comment applies to the temporal zuotian 'yesterday' in (5). We will have more to 

say about this point in our later discussion. 

2. The following symbols and abbreviations are used in giving the English gloss: 

ASP = aspect marker CL = classifier 

POSS = genitive marker PART = particle 

LOC = localizer Rel. Mar.= relative clause marker 

3. For this point, also see Li and Thompson, 1981. 

4. For a brief discussion of the notion of definiteness, see Tsao, 1979, Chapter 5. 

5. A scope-delimiting expression is a term coined by the writer to refer to a 

prepositional phrase or, more commonly in Chineses, a nominal which is used as a topic, 

primary or non-primary, to set a scope within which the following comment is to be 

interpreted. (10) and (i) below are two sentences containing such an expression. 

liu-xue-de shi , 

'i' [%:tgiard to) study:abroad-Rel. Mar. matter 

zhengfu zao gui ding-le banfa 

government long: ago stipulate-ASP regulation 

le. 

PART 

With regard to the matter of studying abroad, the government set up 

regulations long ago. ' 

For more examples of this kind in Classical Chinese see Wang (1980, Chapter 3, Section 

44). 

6. Whether in (13) yong in the phrase yone: gouzhen should be analyzed as a 

preposition i.e. 'with' or a verb i.e., 'use' is a point that we will not take up here, as to deal 

with it will certainly take us too far afield. Suffice it to point out that my present 

hypothesis is that if yong occurs within the scope of a manner adverb as in (i), then it is a 

verb. If it occurs outside the scope of a manner adverb as in (ii), then it is a preposition. 

(i) ta yong bianzi henhen-de da-zhe nei-ge xiaohai. 

he with whip savagely beat-ASP that-CL child 

'He beat the child with a whip savagely.' 

(ii) ta henhen-de yong bianzi da-zhe nei-ge xiaohai. 

he savagely use whip beat-ASP that-CL child 

'He used a whip to beat the child savagely.' 

It follows that we don't quite agree with Chuo in analyzing gong eouzhen in (13) as an

instrumental adverbial. 

SchnfrfestschriJi: Essq in Honor of John DeFrancis 

7. Chuo (1987) regards keneng as a modal adverb. I have argued elsewhere (Tsao, 

Forthcoming) that kenene should be more properly analyzed as a modal verb, equivalent to 

English I' possi ble" . See also Section 3.6 for discussion. 

8. The romanization in the original was in Yale system, which has beell changed to be 

consistent with the system used in this thesis. 

9. Chao's romanization has been changed to agree with the presentation here. 

10. (18) and (18a) in the interpretation under discussion cannot be properly expressed 

in English. That is why no translations are given in those two instances. 

11. (23) appeared earlier in the quotation from Chao that we cited. 

12. As I have observed in the paper "A Topic-Comment Approach to the 

Construction," (Tsao, 1987)) it seems more difficult for temporals to become a & NP. 

However, as (i) shows, it is by no means impossible. 

(i) a. ta yi tian dang san tiail yong. 

he one day regard:as three day use 

'He makes use of a day as if it were three days.' 

b. ta ba yi tian dang san tian yong. 

he BA one day regard:as three day use. 


13. Lin(1989) has argued on independent grounds that weisheme 'why' should be 

generated in the S-initial position, i.e. as the specifier of CP in the most up-to-date GB 

framework, as in (ia . The primary and non-primary topics that occur before weisheme 

'why' as in (b) (c and (d) are the results of topicalization (topic-raising in our 

framework). 

1 

(i) a. weisheme @ zuotian na--cllang c~iu 

why he yesterday that-CL bal1:game 

s hu-le? 

lose-ASP 

'Why did he lose the ball game yesterday?' 

b. & weisheme zuotian na--cham aiu 

he why yesterday that-CL- bal1:game 

s hu-le? 

lose-ASP 


c. zuotian weisheme na,-chann aiu 

he yesterday why that-CL- bal1:game 

s hu-le? 

lose-AS P 

'Roughly, same as (a) .' 

d. zuotian na- hang qiu w ei s heine 

he yesterday that--CL bal1:game why 

shu-le? 

lose-ASP 


14. (39) is also taken from Chao's comment quoted previously.


REFERENCES 

Barry, Roberta. 1975. Topic in Chinese: an overlap of meaning, grammar and 

discourse function. Papers jbom the parasession on finctionalism. Grossman, R. et 

al. eds., Chicago: Chicago Linguistic Society. 

Chao, Y.R. 1968. A grammar of spoken Chinese. Berkeley and Los Angeles: 

University of California Press. 

Chuo, Fred P. 1987. Chinese multiple adverbials. Bramkarnp, A. et al. eds. 

Chinese- Western encounter: studies in linguistics and literature. Taipei: Chinese 

Materials Center Publications. 

CKIP (9 2 a I]\ ) 1989. Guoyude cilei jenn' (An analysis of parts of -. 

speech in Chinese). Rev. ed. Taipei: Computing Center, Academia Sinica. 

Guo, Yizhou. 1960. Fuci, jieci, lianci (Adverbs, prepositions & conjunctions). 

Shanghai: Shanghai Educational Press. (Also in Yuwen hui bian Book 11. Peking: 

Zhongguo yuwen xuehui, 1967.) 

Hockett, Charles F. 1958. A course in modern linguistics. New York: 

Macmillan. 

Lin, R.W. 1989. On adjunct extraction of weisheme and zeme. Unpublished 

M. A. thesis. Hsinchu, Taiwan: National Tsing Hua U niversity. 

Lu, Shu-xiang et al. 1981. Xiandai Hanyv babai ci. Peking: Commercial 

Press. 

Lyons, John. 1977. Semantics. Cambridge: Cambridge University Press. 

Tai, James H-Y. 1975. On two functions of place adverbials in Mandarin 

Chinese. JCL 3, 213, 154-179. 

1976. Semantics and syntax of inner and outer locatives. 

Proceedings of the 1976 mid-America linguistics conference. Brown, R. et al. eds. 

Minneapolis, Minnesota: University of Minnesota Press, pp. 393-401. 

Tsao, Feng-fu. 1979. A jinctional study of topic in Chinese: the first step 

towards discourse analysis. Taipei: Studeilt Book. 

1982. The double nominative construction in hlaildarin Chinese 

Tsing Hua Journal of Chinese Studies. New series 14, 276-297. 

1987. A topic-comment a.pproach to the ha construction. JCL 15, 

1, 1-54. 

1989a. Comparison in Chinese: a topic-comment approach. Tsing 

Hua Journal of Chinese Studies. New series 19, 1, 151-189. 

1989b. Topic and the lian ... dou/ye construction revisited. In 

Tai, James and F. S. Hsueh eds. Functionalism and Chinese spntax. Chinese 

Language Teachers Association Monograph series NO. 1. pp. 24547% 

Forthcoming. Topics and clause connectives in Chinese. To appear 

in Papers in memory of Prof. F.K. Li. Taipei: Institute of History and Philology, 

Academia Sinica. 

Wang, Li, 1955. Zhongguo yufa lilun (Theories of Chinese gramnlar). 

Shanghai: Commercial Press. 

1980. Hanyu shi gao (A history of the Chinese language). Rev. 

ed. Peking: Zhonghua. 

Zhu, Dexi. 1959. Dingyu he zhuangyu Adjectival modifiers and adverbial 

modifiers). Shanghai: Educational Press. Also in Yuwen hui bian. Book 11. Peking: 

Zhongguo yuwen xuehui, 1967).

S~h~ftfts~h?fhrift: Essay in Honor of John &Francis 

YES-NO QUESTIONS IN TAIPEI AND PEKING MANDARIN 

ROBERT M. SANDERS 

I. DEFINITION OF A YES-NO QUESTION 

This paper examines the manner in which yes-no questions are 

expressed in the speech communities of Taipei and Peking. A yes-no 

question here is defined in terms of how one would answer the 

question in English. In other words, if the Chinese question were 

translated into English with the intent of allowing the listener to 

respond in English, the only two options available to that respondent 

would be either "yes" or "no." In Mandarin, then, there exist three 

sentence patterns which are seen to exemplify this type of question. 

They are: 

1. The A-not-A Question 

a. Ni qu-bu-qu ta jia? 

you-go-NEG-go-3rd person-home 

Are you going to his house? 

b. Ni qu ta jia bu qu? 

you-go-3rd person-home-NEG-go 


2. Sentence Intonation + Particle MA 

Ni qu ta jia ma? 

you-go-3rd person-home-PARTICLE 


3. Sentence Intonation Alone 

Ni qu ta jia? 

you-go-3rd person-home 

Are.you going to his house? 

As one can see, in English the only possible response available to the 

listener is either "yes" or "no." In Chinese, on the other hand, the 

situation is not quite that simple. In fact, in two out of the three 

patterns, the respondent has at his disposal more than one option in


how to reply. For example, in the case of all three he can indicate an 

affirmative response by answering with a positive form of the verb, 

as in (4a), or he can indicate a negative response with a negative 

fonn of the verb, as in (4b). 

4a. Wo qu. 

1 st person-go 

I'm going. 

4b. Wo bu qu. 

1 st person-NEG-go 

I'm not going. 

However, as noted by Zhu (1984:203), the intonation + MA and 

intonation-only patterns differ from A-not-A questions in that the 

former two allow for the affirmative and negative responses 

illustrated in (5a) through (5d) below, whereas A-not-A questions do 

not. 

5a. Dui. 

correct 

Correct . 

5b. Shi. 

be 

It is so. 

5c. Bu dui. 

NEG-correct 

Incorrect. 

5d. Bu shi. 

NEG-be 

It is not so. 

The reason provided by Zhu for why A-not-A questions behave 

somewhat differently from intonation + MA and intonation-only 

questions is that A-not-A questions explicitly limit the listener to a 

simple choice between A and its opposite. This type of yes-no 

question, then, is disjunctive . The latter two patterns, on the other 

hand, are somewhat more open-ended, failing to state outright that 

the listener is limited to just those two choices. As discussed by 

McGinnis (1990), however, inspite of Zhu's observations about the

Schn'ftfestschrifi: &say in Honor of John &Francis 

somewhat different syntactic nature of A-not-A questions, there 

exists a body of convincing historical and synchronic evidence which 

argues in favor of lumping all three patterns together under the 

same functional heading. For instance, historical work by Huang 

(1986) and Liu (1988) show that the interrogative particle MA is 

actually derived from the A-not-A pattern illustrated by (lb), 

through a process in which the sentence-final NEG-V constituent was 

first reduced to a bare NEG, and then that NEG was itself phonetically 

reduced to the status of a sentence-final particle. Also, acoustic work 

by Shen (1989) shows that both intonation + particle and intonation- 

only questions (which she calls echo questions) do not differ from 

one another in terms of their repective intonation patterns. Finally, 

in the patterning of the Taipei and Peking data contained within this 

study, it will be seen that there exist good statistical reasons for 

treating the three as if they were semantically equivalent. 

One frequent observation in the literature on Chinese 

interrogatives which argues against categorizing all three patterns 

under the same functional heading involves the issue of speaker 

presupposition at the time which the question is asked. Based on the 

discussion in McGinnis (1990) we shall see that the question of 

speaker presupposition is not necessarily determined by the 

sentence structure itself, but is rather determined by extra-linguistic 

factors. He notes that there is almost universal agreement among 

scholars that when the intonation + MA question or the intonation- 

only question is stated using a negative verb, there exists a strong 

presupposition on the part of the speaker that the correct response 

ought to be given either in the affirmative form illustrated by (4a), 

or in the negative forms illustrated by (5c) and (5d) respectively. 

When such questions contain the non-negated form of the verb, as in 

.(3), on the other hand, there is much less agreement among linguists 

as to the underlying assumptioq of the speaker. For this sentence 

Elliot (1964) expects the response to be either (4a), (5a) or (5b). Li & 

Thompson, on the other hand, expect just the opposite response, i.e. 

either (4b), (5c) or (5d). Resolving this apparant contradiction is 

Tang (1986), who maintains that depending on the context and/or 

the speaker's own assumption at the time of questioning, the 

expected response to (3) may come either from Elliott's set or from Li 

& Thompson's. What is important about Tang's claim for sentences 

like (3) is that knowledge of the speaker's presupposition does NOT 

come from the structure of the sentence, but rather from extra- 

linguistic factors. In other words, it is the context rather than the 

structure itself which determines the neutrality of questions like (3).


In my own work with Chinese informants I have found that without 

the presence of a specific context, most, if not all of these native 

speakers interpret sentences like (3) to contain no presupposition as 

to the expected answer. 

A-not-A questions, on the other hand, are almost universally held 

among linguists to be free of any speaker presupposition, and thus 

are felt to be pragmatically different from the other two patterns. 

McGinnis (1990:65) does however show that at least under some 

specific contexts, it is possible to find an A-not-A question in which 

the speaker does harbor a presupposition as to what the correct 

response ought to be. His example involves a rhetorical question 

which can be asked when the speaker believes that his Chinese 

listener has been behaving in a non-Chinese manner, and thus he 

wishes to chastize that person for his transgression. 

6. Ni shi bu shi Zhongguo ren? 

y ou-be-NEG-be-China-person 

Are you Chinese or not? 

We see then that there really does not exist any strong linguistic 

argument vis-a-vis speaker presupposition to dissuade us from 

treating A-not-A questions as being functionally similar to the other 

two yes-no question forms. This being the case, we can characterize 

these three patterns as forming a continuum from the explicit 

structural yes-no marking of the A-not-A questions, to the more 

impoverished structural marking of the MA particle, to the complete 

lack of syntactic marking found with intonation-only questions. This 

study, then, is primarily concerned with discovering the relative 

frequency at which each of these three patterns are used in 

contextually neutral conditions to ask a yes-no question in Taipei and ' 

in Peking. As an aside, it also touches upon how each speech 

community goes about expressing a strong sense of presupposition. 

In addition to using a negative form of the verb with either the 

intonation + MA or intonation-only questions, there exists in 

Mandarin two other so-called yes-no question patterns which also 

consistently signal a strong sense of speaker presupposition. They 

are the tag question and the use of sentence intonation in 

combination with the sentence-final particle BA. In addition, there 

exist, other sentence-final particles such as A/YA, which depending 

on context, are also in affirmative yes-no questions containing a high 

degree of speaker presupposition. In the statistical analysis found in

Schriftfesestsrifi: <strong>Essays</strong> in Honor of John &Francis 

Figure 2 below, BA is used as a cover term to include all instances of 

any sentence-final particle used in a rhetorical yes-no question. 

7. Tag Question 

a. Ni qu ta jia, shi bu shi? 

you-go-3rd person-home-be-NEG-be 

You're going to his house, right? 

b. Ni qu ta jia, dui bu dui? 

you-go-3rd person-home-corect-NEG-correct 

You're going to his house, correct? 

8. Sentence Intonation + Final Particle BA 

Ni qu ta jia ba. 

you-go-3rd person-home-PART 

I take it that you are going to his home, right? 

Note that both tag and BA questions always contain the 

presupposition that the correct answer should be either (4a), (5a) or 

(5b). Note too that both are formed by suffixing the question-asking 

component to an ordinary statement. Given their common structure, 

in combination with their shared pragmatic character, it is claimed 

that tag and BA questions are in fact functionally equivalent. With 

this claim we can observe the degree to which Taipei' and Peking 

speakers favor either tag or BA questions -in the expression of 

affirmative rhetorical questions. 

11. DATA COLLECTION 

As this is an empirical study, the focus of the data collection was 

on gathering what people actually said rather than on listing what 

they were capable of saying. The reasons for adopting this approach 

to the data collection as opposed to adhering to either the 

psychologically introspective methods of modem Western linguistics 

or the single authentic native speaker approach favored by most 

dialect researchers in China is outlined in Sanders (1991) and will 

not be repeated here. The data for this study comes from fieldwork 

carried out in the Autumn of 1989 in Taipei and Peking. Candid 

audio recordings of natural conversations among small groups of


people. whose relationship to one another was either that of family 

member or close friend. Altogether eight one-hour conversations 

from Taipei and eight one-hour conversations from Peking form the 

data base of this study. One concern which arises once one decides to 

utilize so many independent sources of natural data is whether or 

not such data is consistent across the board in terms of pragmatic 

content. In other words, do these sixteen conversations share a 

similar setting, a similar content, and a similar relationship among 

each set of participants as one goes from one conversation to 

another? In this case the answer to all three concerns is yes, and 

thus comparisons within the data can be made with a high degree of 

confidence. 

In order to understand the significance of the data presented in 

Figures I and 2 below, it is necessary to first understand in at least a 

rudimentary way what a chi-square analysis indicates. For a more 

thorough discussion of this analysis than what is presented here, the 

reader may wish to consult either Butler (1985) or Davis (1990). 

Simply speaking, a chi-square analysis can be thought of as a means 

of determining whether the quantitative differences observed for 

two or more sets of token counts are in fact statistically significant or 

not. Looking at Figures 1 and 2, one finds information on the actual 

number of tokens observed, the expected number of tokens if the 

variables of geographical location and sentence type were 

independent of each other, the chi-square value (a sum of the 

differences between the observed and expected counts), and the 

degrees of freedom (a technical term which is difficult to explain, but 

which indicates whether the chi-square value is great enough to 

signify statistical significance for the data). For the two figures found 

below, only one or two degrees of freedom (df) are observed. 

According to standard statistical practice, for one degree of freedom, 

a chi-square value of 3.8414 or greater indicates statistical 

significance, while for two degree of freedom, the necessary value is 

5.9914 or greater. Therefore we see that the patterns illustrated in 

both of these figures are in fact statistically significant. 

Figure 1 illustrates the number of tokens of each of the three yes- 

no patterns which are found in each of the two speech communities 

when no speaker presupposition exists. In each case, expected 

counts are given directly below observed counts. For example, in the 

.

Schrilifestschnift: <strong>Essays</strong> in Honor of John DeFrancis 

case of A-not-A questions in Taipei, 72 tokens of that pattern were 

counted in the data, whereas one would have expected to find just 

45.65 tokens if Taipei behavior were completely independent of 

Peking behavior, and the occurance of A-not-A questions operated 

independently of the occurance of .the other two patterns. In the 

case of Figure 1, note that although both speech communities contain 

all three patterns in their respective repertoires, in Taiwan the 

obvious choice from among the three is A-not-A, while in Peking 

that pattern is utilized least of all. In fact, if one were to compare 

the ratio of A-not-A questions to the combined sum of intonation + 

MA and intonation-only questions in each speech community, a clear 

difference in linguistic behavior is observed. In Taipei the ratio of 

A-not-A to the other two is about 1:1, while in Peking the ratio is 

about 1:4. Also, note how infrequently Taipei speakers utilize 

intonation alone to express a yes-no question in comparison to 

Peking speakers. In Taipei the ratio of that pattern to the other two 

is about 15, while in Peking it is almost 12. These facts clearly 

indicate that a qualitative difference exists between the way Taipei 

speakers tend to express pragmatically neutral yes-no questions and 

the way Peking speakers do. 

Figure 1. Distribution of Neutral Yes-No Tokens by Pattern and City 

Pattern Taipei Peking Total 

Intonation 22 8 9 11 1 

40.22 70.78 

Total 150 264 414


In Figure 2 we see quite clearly that when Taipei and Peking 

speakers wish to express an affirmative rhetorical question, their 

preferences are not the same. Taipei speakers show a very strong 

inclination to use tag questions, while Peking speakers show a fairly 

even division between the two, and if there is any preference at all, 

it is for BA or some other sentence-final particle. 

Figure 2. Distribution of Affirmative Rhetorical Yes-No Questions 

by Pattern & City 

Pattern Taipei Peking Total 

Total 

ChiSq = 3.972 + 3.928 + 6.299 + 6.229 = 20.429 

IV. CONCLUSIONS 

The data contained within this study convincingly demonstrates 

that although Taipei and Peking speakers essentially share the same 

syntactic repertoire for expressing yes-no questions, the two speech 

communities differ from one another in terms of which form or 

forms they prefer. When the question is pragmatically neutral, 

Taipei speakers prefer the explicit syntactic marking of the A-not-A 

form, while Peking speakers prefer sentence intonation + MA, or 

sentence intonation alone to perform this function. Consistent with 

this dichotemy between expressing a yes-no question either through 

an A-not-A structure or by means of a sentence-final particle, when 

the speaker possesses a strong assumption as to what the proper 

answer ought to be, Taipei speakers once again show a strong 

preference for the disjunctive form while Peking speakers continue 

to favor a sentence-final particle.

Schn$j4esstschn!jf: fisq irr Honor of John &Francis 

Butler, Christopher. 1985. Statistics in Linguistics. New York: 

Basil Blackwell. 

Chao, Y. R. 1968. A Grammar of Spoken Chinese. 

Berke1ey:University of California Press. 

Davis, Lawrence M. 1990. Statistics in Dialectology. Tuscaloosa: 

The University of Alabama Press. . - 

Elliott, Dale E. 1965. "Interrogation in English and Mandarin 

Chinese."Project on Linguistic Analysis 1 156-1 17. 

Huang Guoying. 1986. "'Ma' Ziju Yongfa Chutan" (A Preliminary 

Investigation into the Usage of ma-sentences). Yuyan Yanjiu 

186.2:13'1-135. 

Li, Charles N. and Sandara A. Thompson. 1979. "The Pragmatics of 

Two Types of Yes-No Questions in Mandarin and Its Universal 

Implications." In Papers from the Fifteenth Regional 

Meeting of the Chicago Linguistic Society, pp. 197-206. 

Liu Yuehua. 1988. "Yudiao Shifei Wenju" (Intonation Questions). 

Yuyan Jiaoxue yu Yanjiu 1988.2:25-34. 

McGinnis, Scott G. 1990. A Pragmatic Analysis of Mandarin 

Interrogatives: Data From Modern Taiwan Drama. Ph. D. 

dissertation, The Ohio State University. 

Sanders, Robert M. 1991. "The Expression of Modality in Peking and 

Taipei Mandarin," to appear in Journal of Chinese 

Linguistics. 

Shen, Xiao-nan. 1989. The Prosody of Mandarin Chinese (UC 

Publications in Linguistics 1 18). Bekeley: University of 

California Press.


Tang Ting-chi. 1986. "Syntactic and Pragmatic Constraints on the V- 

not-V Question in Mandarin." Paper presented at the 1986 

Conference on Sino-Tibetan Languages and Linguistics, 

Columbus, Ohio. 

Zhu Dexi. 1982. Yufa Jiangyi (An Outline of Grammar). Shanghai: 

Commercial Press.

Schri$+fesesrscschn$: <strong>Essays</strong> in Horwr of John DeFrancis 

PATRONIZING USES OF THE PARTICLE "maw: 

BUREAUCRATIC CHINESE BIDS FOR DOMINANCE IN 

PERSONAL INTERACTIONS 

Beverly Hong Fincher 

Connecticut College 

Sentence final particles are called mood words in Chinese grammar 

books. Li and Thompson claim their function is to relate an utterance to 

which they are attached to the conversational context in various ways; and 

to indicate how this utterance is to be taken by the listener. Of course, 

many other languages also have sentence final particles whose function is 

similar to that of Chinese. One notable example is Japanese. In contrast to 

Li and Thompson's emphasis on the listener, S. Kuno says that their 

function is to express the speaker's attitude toward the meaning of the 

sentence. 

These two views are thus from different perspectives on the 

communication ch4el. One is from the hearer's point of view and the 

other is from the speaker's point of view. This paper concentrates on one 

particle, ma, and is based on a recorded spoken text taken from Beijing 

Ren. It examines the mood of the speaker and the reactions of the hearers. 

It attempts to probe into why the particle ma occurs with such high 

frequency when bureaucrats speak even though their listeners are repulsed 

by it. Thus, the particle is approached from both ends of the communication 

process. 

157


The particle "ma" in question contrasts minimally with the question 

marker "ma" as in the following: 

1. lai ma ? A= you coming? 

2. lai ma . Do come, for my sake, etc. 

(1 ) is uttered with rising intonation and shorter duration, whereas (2) 

is with falling intonation and longer duration.1 This will be called ma2 

henceforth. 

"m2" typically occurs in casual conversation when status and feelings 

are intricately entangled. Chao defines it as "dogmatic assertionW.2 

Normally it occurs in an expanded context of one kind or another. 

The problem 

I first noticed a problem in interpreting ma2 when examining some 

, , 

spoken texts from Beijing Ren for an Intermediate Chinese class. The text 

that stands out was 'Biaozhun Huat (Standard Talk). Reading it I was struck 

by the frequent use of the particle ma2, and my own feeling of being put 

off by the way the speaker used it to patronize his listeners. When an 

English translation was available, I found that the passage did not give the 

same feeling. The English version gave the impression of ritualized, formal 

speech, of a set way of packaging the bit of information contained in an 

utterance. After examination and comparison of the two texts, it became 

clear that the translator was more interested in the load of cliches which 

wry little in content in the original text and not so much its emotive parts. 

1 Chao, Y.R., A grammar of spoken Chinese, p.800 

2 Ibid. p 801. 

158 

.

SchriJEfestschhr$: <strong>Essays</strong> in Honor of John DeFrmis 

Speaking from the perspective of a translator, the problem is that of 

what to leave in or out. In addition the audience which the translator 

addresses is totally different from that addressed by the original speaker. 

Thus the emotional reaction that a (spokentwritten) text will generate will of 

course be different and may present particularly difficult problems of 

translation. 

The method 

The original Chinese and the translated English texts were given to 

native speakers of the respective languages to determine their immediate 

reaction and their perception of the speakers' attitude and mood in the 

discourse. The informants were 14 native speakers of English (Australian, 

English and American) and 8 'Chinese native speakers from PRC. They 

were all asked to express their immediate reaction to the text and their 

perception of the speaker's attitude, mood and tone of voice. 

The mults 

Among the native speakers of English, reactions varied a great deal 

despite some commonalities which were noted such as "uninformed". and 

"humb1e" as applied to the mood adopted by the speaker. 

Some keywords or phrases used by the English informants are 

indicative of the different reactions: 

"This is more a satire than an actual speech". "Complacency", "makes 

me wonder how much of it is true", "forced optimism" , "deferential", 

"defensive", "slightly aggressive", "formal pleasantries", "approval 

seeking", "doctrinaire", "sycophantic", "insincere", "lackey's voice", "people 

pleasing", "he feels guilty about a poor showing", "apologetic", "talking up 

to his listener", "have the feeling of attending some religious meetings", 

159

Sino-Platonic Papers, 27 (August 3 1. 1991) 

"living by platitudes", "public morality", "confucian sense of group 

responsibilities", "afraid of being an individual in public", "cautious", 

"respectful", "confident", "positive", "optimistic", "gracious", "wordy", 

"inviting flattery", "more than polite deferential". 

"Humble" was used to suggest false humility as well as possibly 

genuine humility, but it is interesting to note that only one informant 

explicitly comments: 

"treating the audience as 'children', patronizing". 

The original Chinese text seemed unambiguous to its readers. By 

contrast with readers' reactions to the English translation, reactions from 

Chinese informants were more consistent, i.e. that it is typical speech of "a 

local cadre", "speech symbol of their status", "their qualifications for having 

been in the revolution", "used by higher or similar ranking officials toward 

lower or similar ranking officials", "children, when imitating cadres, all use 

ma", "a signifier for characterizing conservative cadres portrayed in 

movies", etc. 

Turning from my English or Chinese speaking informants to examples 

of speech of Chinese officials it appears that this last observation is an 

oversimplification. We can observe that not only conservative cadres, but 

also well-known reform-minded high cadres use this speech code liberally. 

For example, in a recorded interview of Hu Yaobang by the journalist Lu 

Keng in 1985, Hu's speech to Chinese students in Japan, Zhao Ziyang's 

press conference after the 1 3 th plenum in 1 987, etc. 

But this confirms that the vast differences between the reactions of 

my two sets of informants can be attributed to the sociolinguistic 

competence of the Chinese speakers among them rather than to linguistic

Schnfrfestschn!: <strong>Essays</strong> in Homr of John DeFrancis 

ability in its simpler sense. Their common knowledge seems to derive from 

extra social and contextual cues. 

The environments in which ma?, - occurs. 

During his interview with Lu Keng, Hu Yaobang used ma2 in the following 

way: 

- 1. after stating obvious fact.: 

(Taiwan) Shizhishang ye shi ge S&g zhzhgE ma. 

(Taiwan) in fact is a local government. 

2. after a set phrase: 

. . . . D6, Mel, d6u mei xing ch6ng t6ngJ;i' de gliojia ma. 

. . . . Gemany, US, (at that time in the 18th century) had not become 

a &ed nation. 

3. after clear folk logic: 

zhei ge pi0 ha5 ma, shi zhhgyi zhi sheng ma. 

This whistle blowing is good, it's the sound of justice. 

4. after a suggestion or an advice, thus, persuasion: 

niv&ng ge MO ma 

you should expose it in the paper. 

5. after a proverb: 

tiashi dili rEnh6 ma. (from Bi2ozhiin Hiia) 

(The success is due to) heavenly timing, profitable locale and 

harmonious human relationships. 

The above uses of ma2 are as an attachment to utterances about 

"obvious" truths. In that respect these utterances are somewhat' like 

proverbs in Chinese. The contexts indicate that the speaker is marshalling 

161

Sino-Platonic Papers, 27 (August 31, 1991) 

social and cultural wisdom to back himself up. He will then be on higher 

ground in his effort to persuade his listeners. 

Contrast with other particles 

If we contrast ma2 with two other closely related sentence final 

particles, a and ne, the meaning becomes more clear. Take an example 

from Xin Fhgxia, the actress and wife of Wii Ziiguibg the Beijing 

playwright. In an interview with a reporter from Central Daily News, 

Taiwan, she describes how during the anti-rightist campaign the party 

functionaries were asking her to divorce her husband after he had been 

classified a rightist. She said: 

Wu bii n6ng he til li, w6 yofi ge h&i ne - I cannot divorce him; 

I have 3 children. 

The particle ne makes the sentence an appeal for involvement from 

the listeners, somewhat like "you see". Her story reveals that xin'F&ngxia 

did not want to reveal her love for her husband. Her use of ne 

"rationalizes" her decision against divorce in terns of responsibility to her 3 

children requiring her husband's support (morally or materially). 

If "new is substituted by "a" then the tone of Xin's observation 

becomes more assertive. 

If it is replaced by ma2, then the meaning changes even more: ethics 

and social morality are introduced with a kind of lecturing tone which 

reminds listeners of their duties, social morales etc. It means: "how can a 

wife like me with 3 children think of divorcing him? You must be out of 

your mind". 

. -

Schriifrfestschny?: fisays in Honor of John &Francis 

ma7 and the power of persuasion 

Returning to Hu Yaobang and Zhao Ziyang, we note a difference 

between these situations and that of a local official. The lower official is 

imitating the speech of his superiors, and it can be taken by his listeners as 

a patronizing pretension of a local bureaucrat. 

We note that in ordinary speech, ma2 may carry the meaning of 

friendly persuasion, as in an expression of paternal concern or similarly in a 

child's plea to an indulgent parent. In short, ma2 operates either up or 

down, in an intimate parent-child relationship as well as in bureaucratic 

relationships. Therefore, the listeners perceive both meanings: intimacy 

(solidarity) and authority (power). These two meanings, though not 

mutually exclusive, are not really complementary with each other. Rather 

the two psychological forces may be in balance, but there is a tension 

between a bid for dominance and a bid for intimacy which can become 

the 

explosive as we have seen in,,June 4 Tiananmen incident. A bid for 

intimacy like that of Hu Yaobang, if prepackaged and propagated through 

the propaganda system without variation becomes patronizing in the ears of 

a listener to a lower bureaucrat. Bureaucrats who want to keep or advance 

their position will adhere rigidly to the sociolinguistic patterns set by their 

superiors. By continuing to use ma2 and ignoring their listeners' reaction to 

its use they convert an appeal for intimacy into an assertion of authority. 

An indicator of solidarity has thus become an indicator of authority, as 

if its charge has been reversed. As a particle of power, ma2 repels just as 

surely as it might attract when it appears as a particle of solidarity.


REFERENCES 

Alleton, Viviane, 198 1, "Final Particles and Expression of Modality in 

Mandarin Chinese". Joumal of ainese Linguistics, 9,9 1 - 1 15. 

Alleton, Viviane, 1986, 'The soded "Rhetorical Interrogation" in 

Mandarin Chinese. Paper presented at the xxxii International 

Congress for Asian and North African Studies, Hamburg. 

Chao, Y.R., 1968, A Grammar of Spoken Chinese, University of California ' ' 

Press. 

Chiao Chien, 1985, "Model Building in the Study of Chinese Strategic 

Behaviour: Some proposals", in proceedings of the conference on 

modernization and Chinese culture, edited by Chiao et al, Chinese 

University of Hong Kong, Hong Kong. 

Huang, Chu-Ren, 1985, "Chinese Sentential Particles: A Study of 

Cliticization", LSA Annual meeting. 

Gumperz, John, 1982, editor, Discourse Strategies, Language and Social 

Identity. Cambridge University Press. 

Huang, Guoying, 1986, "ma"-zi @ y6ngfi chiit211 p. 13 1-135, Yiiylin 

yhjiii, n0.2 vol 11. 

Jenner, WJ-F. and Delia Davin, 1988, translator and editor, "Chinese 

Lives", Pantheon Books, New York. 

Li and Thompson, 198 1, Mandarin Chinese: A Functional Reference 

Grammar, University of California hess.

Schriiffestschn!: Ersays in Honor of John DeFrancis 

Lu Keng, 1985, Hu Yaobang Fangwen Ji, An Interview with Hu Yaobang; 

Pai Shing Monthly, Hong KonglSino Daily Express, New York. 

Shiu, Yu-Ling and Huang, Chu-Ren, 1989, A Unification-based Approach 

to Mandarin Questions, proceedings of Rocling 11. 

White, Jeoffrey W., 1987, "Proverbs and cultural models" p.151-172 in 

Culhual MOdeLs in Language and mought edited by Dorothy Holland 

and Naorni Quinn, Cambridge University Press 1987. 

Zhang XinXin and Sang Ye, Beijing Ren 1986, Shanghai Wenyi Chuban 

She, Shanghai. 

I wish to thank the native speakers of English and Chinese whose reactions 

to the "text" under study constitute part of this paper. Their names are 

listed below alphabetically: 

Richard Barz, Dani Botsman, Marybeth Clark, Anthony Diller, John 

Fincher, Leta Fincher, Drew Gerstle, John Jorgenson, Ann Kumar, Lydian 

Meredith, Virginia Matheson, Helen Musa, Mabel Lee and Anthony Street. 

Chen Jie, Chen Shuchen, He Baogang, Liu Min, Xie Chuanliang, Xie 

Guanghua, Li Xinhua, Bai Yan. 

Huang Guoying of the Central China Institute of Technology in Wuhan 

provided his reaction via correspondence (March 1989). 

I am grateful to Sang Ye, the co-recorder of the original text for discussing 

the original situation with me and to Bill Jemer for reflections on the 

translator's art. 

Thanks are also due to Philip Kuhn for the apt analogy of McDonald's 

Hamburgers to Standard Talk: "It's like McDonald's hamburgers. It is 

standard fare, no variations; thus the cook never makes a mistake".

Schniftfestschrifi: Essa)s in Honor of John DeFrancis 

GENDER AND SEXISM IN CHINESE LANGUAGE AND LITERATURE 

By Angela Jung-Palandri 

Language and literature both reflect and express social attitudes and values. That sexism exists 

in most languages and literatures is not a mere feminist fabrication; it is a fact Were there no 

sexism, there would be no need for feminism. Nowhere, however, is sexism more apparent than 

in the countries of the Near and Far East. This is true even in Communist China. I shall confine 

my discussion to the Chinese language and literature of the past 

Sociologists and anthropologists generally agree that in prehistoric China, communities were 

ruled by matriarchs, as the discovery of Banpo near Xian evinces. Even the Chinese character for 

family name, ring (+$ womposed of the graphs for nii (+) (woman) and sheng (&) (birth or 

life)-attests to the matriarchal origin of the family. For the last six or seven thousand years, 

however, China has been under a patrilineal system, where males have dominated every aspect of 

social and political activity. Chinese literature, beginning with the classic canon, illustrates this 

sexist attitude. In the Book of Changes , Yijing (6 /&, we find that the male is equated with the 

yang principle, symbolized by the sun. It embodies everything that is good and positive, and its 

status is identified with heaven. The female, on the other hand, is equated with the yin principle, 

symbolized by the moon. To it is attributed all that is negative, evil and lowly. In the earliest 

concept of the Chinese myth of creation, these two elemental forces, yin and yang, were on an 

equal footing, as the cosmological emblem (of the Taoists) shows. 

Through later male- biased interpretations that stemmed from a patriarchal order, sexism became 

firmly entrenched. The Confucian commentaries further define the status of the male as the ruler, 

and the female as the ruled. According the the Yijing, man's proper function is in society or the 

world; while the woman's duty is to remain within the household. 

Once a jud ement of sexual roles was formed, sexism was solidified. In the Book of Poetry, 

the Shijing ( $@k ), we see how drastically different the births of sons and daughters are 

regarded: 

Sons shall be born to him: 

They will be put to sleep on couches; 

They wil be clothed in robes; 

Theywill have scepters to play with; 

Daughters shall be born to him: 

They will be put to sleep on the ground; 

They will be clothed with wrappers; 

They will have tiles to play with. 

It will be theirs neither to do good nor to do wrong; 

Only about the spirit and the food will they have to think, 

And to cause no sorrow to their parents. 

(Book IV, Odes vi) 

In the Book of Histov, the Shujing (-@is), women are rarely mentioned; when they are, they 

are blamed for the ruin of the state. Such was the case of Da Ji (4g4tE ) and Baosu ( 

whose alleged evil influence supposedly caused the downfall of Jie of Xia and Zhou o%hang.

Sino-Platonic Papers, 27 (August 3 1,199 1) 

Perhaps it was because of the latent fear of women's influence on men that the ancients established 

rules and regulations and recorded them in the Book of Rites, the Liji (&,a These legitimated 

male domination and put women under male control. Under the guidelines of "the three 

obediences and four virtues"&& @ 4 t -), a woman had to obey her father, her husband, and, after 

the husband's death, even her sons. he status of women was no better than that of a slave. In 

the Analects of Confucius, the Lunyu (% s), the Master is quoted as saying, "Women and 

inferiors (& A) are difficult to handle. If you keep them at a distance, they are resentful; if you 

show intimacy, they become disrespectful." 

+ . 

Sexism is found not only in literature written by men, but in the works of women as well. Ban 

Zhao of Han, daughter of the grand historian Ban Biao, and sister of Ban Gu, was a great scholar 

and historian in her own right. It was she who completed the History of the Early Han that her 

brother had left unfinished. But she is acclaimed not so much for her scholarly contribution to 

history, as she is for her treatise, Lessons for Women, Nii jie &irPi). In this work she spelled out 

in great detail how young women should conduct themselves and serve their in-laws as well as 

their husbands: with humility and submission. Her Lessorzsfor Women helped to perpetuate 

sexism in China for centuries. It so influenced women's education that several works by women 

later were patterned after it For example, both the Female Classic of Filial Piety &**, $&$ ), by 

Chen Miao's wife, nee Zheng of the Tang dynasty, and the Woman's Analects 4 by Song 

Ruohua (x$*), upheld the sexist codes that men had instituted to subjugate women. 

Women's suffering caused by men's repression may account for the pervading sadness in their 

poetry. A famous but ill-fated woman poet of Tang, Yu Xuanji (@ $$Xi), lamented her fate as a 

woman in these lines: 

;g EJ $3 g 65 

tr$3.fKF% 

In a clear spring day clouds and ~I-Z 

peaks fill my field of vision. i- 313 

Elegant ideograms one by one leap 2 t/E';":x$gi2j.?: 

out under my fingers. jji +.:.?~=.+$~rc-. - 

How I hate this chiffon-clad 

body of mine which conceals my poetic talent 

With envy I scan the list heralding 

the successful candidates. 

- -.- I? Y-E 

Zhu Shuzhen (& jkk) of Song Dynasty, whose uncirculated poems were all burned by her 

parents after her untimely death, had in mind the ancient adage that ''lacking literary talent is a 

virtue for women," when she wrote: 

For a woman to dabble in letters is already an offense, _- -k -_ Fe3ii$lq 

Let alone chanting of moonlight and breezes. qp i~ (J?: ;R F- fi- 17, 

Wearing out the inkstones is not to be my lo^ &~~+a+ss 

My virtue lives in breaking needles through embroidering. 2% jq %- &lj z% 

Sexism is perhaps even more revealing in fiction. frequently depicted as immoral 

temptresses or adulteresses. The character Pan who appears in Shuihu zhuan as 

the murderer of her husband, reappears in nymphomaniac. From a male 

perspective, she illustrates how a woman can degenerate when unrestmined by man's moral codes. 

Perhaps to counter this double standard and the degradation of women found in these novels, 

Li Ru-zhen wrote his feminist novel, Jinghuu yuan, and Cao Xueqin his Honglou rneng; both 

assert the superiority of women. But their counteractions could not overcome the overt sexism of 

traditional narratives of the Ming and Qing. 

7y

. Some women novelists or tanci (# 31 ) writers, the authors of Tianyu hua (ma) and 

Zaisheng yuan (PC & &) for instance, believed that the only way their protagonists could compete 

in a man's world or escape becoming man's prey was through male disguise. Indeed, some 

female protagonists in tongsu xiaoshuu &{&,1,j$ and tanci occasionally seem to emerge as 

winners, excelling in literary or military feats by playing male roles. But in the final analysis, they 

could not prevail over their sex limitations set by man and succumb to convention. Ultimately they 

married husbands whom they happily shared with several other wives. 

Turning to the Chinese language, we fmd that the written language, which has preserved 

Chinese civilization and history since their beginning, also betrays a male bias. In a Chinese 

dictionary, even a cursory examination of the listing under the radical 38, nii w), a female sign, 

will reveal several words blatantly derogatory of the female sex. For instance, the word "adultery" 

or "promiscuous fornication" is composed of three female symbols (qe ) pronounced jian; the 

character consisting of two male ideographs with a female s mbol in between, pronounced niao 

(gtB, means "obscenity" or "obscene." The character ji (4& ) meaning "jealousy" is made up of 

two components: "female" and "sickness." One may argue, 'How about the word hao (-&J. ) 

meaning 'goodness or fine,' which is also composed of the female radical?" But the components 

of this character are woman (nu) and son which is the pictograph of a child. The implication is that 

"goodness" links a woman with child. This strongly suggests child bearing or reproduction as the 

primary function of a woman, another sexist attitude. 

While Chinese etymology manifests irrefutable male bias, the language itself is devoid of 

gender consciousness, which to me is a great asset in the modem world where women are gaining 

influence and support in their struggle for equality. All Indo-European languages, except 

Armenian, have grammatical categories of gender. In the Romance and Germanic languages, all 

animals, minerals, and vegetables have genders assigned to them. English fares a little better by 

eliminating grammatical gender, but it retains the natural gender in the third person pronouns and in 

social and professional titles. The third personal pronouns "he" and "she" prove to be the most 

troublesome in our changing society today. Although attempts have been made to replace the all 

inclusive "he" or "his" with "s/heW or "she or he" or "his or her" in a non-gender specific situation, 

the problem remains unsolved, because many people refuse to accept it due to linguistic 

clumsiness. 

No such problem exists in Chinese, because the third personal pronoun is pronounced to for 

both male and female. In the written language, ta qa is composed of the radical ren W 

meaning"hurnan"or "humanity" without gender distinction. However, since the May 4th 

Movement of 1919, under the impact of Western literature and for the purposes of translating 

Western materials, Chinese language reformers such as Hu Shi, devised four written forms for the 

same ta (i.e., the third personal pronoun singular); the ta with ren radical (4 is reserved 

exclusively for "he" or "him. " For the pronoun "she" or "her," the female radical is used instead 

of the original ren @&). The ta with a cow radical (#!5 ) is for neuter gender. In addition, the ta 

( $& ) referring to God or spiritual beings is given the radical shi (n to signify a spiritual quality 

without sexual implication, thus forestalling a recent controversy as to whether God is male or 

female, or both. By the way, the Aztec language Yejua for the third personal pronoun is also nongender 

specitk like the Chinese, because the ancient Aztec religion believed that God or the creator 

was both male and female, two in one. This corresponds to the Chinese myth of creation and the 

ying yang principles I mentioned earlier. 

It is a paradox that in such a strongly sexist society like that of China there should be a nongender 

specific pronoun like ta (4th). If it had a universal application, it would solve the "she/heV 

or "he/sheM dilemma in English. Besides this genderless pronoun, the Chinese have two other


non-sexist traditions which are surprisingly modem in conception. One is that Chinese women 

have always retained their maiden names, a custom not yet widely used in the West, except by a 

few professional women. The other is the use of professional titles. In Chinese, there has never 

been such an anomaly as "Madam Chairman" because the term "chairman" does not bear a gender 

marker of "man" attached to the chair. The Chinese term for "chairperson," zhu*' (3 

be either a man or a woman; and it does not have to be specified, since sex is not an 

Ordinarily the word ren, meaning "person," is a suffix for titles, like shiren (4~) which is for 

male poet or female poet, li~gren (q+a for actors or actresses, and the communist term of airen 

(%A), meaning "loved one," is used for both husband and wife (which I fully approve). Ren 

( h ) means "a human being or a person." Only nanren ($4 is a male person; likewise, niiren 

& fi is a female person. Haoren (@=A) then is a "good person," with no reference to that 

person's sex. The Chinese do accept the natural gender, and only when sex distinction is 

necessary, the word for male or female may be placed before the genderless, personal or 

professional title. 

By not being obsessed with genderization, Chinese seems to be more concerned with the 

human being, the human quality of that being as a whole. This is one step closer to the 

androgynous language envisioned by Mary Ritchie Key, author of MalelFemle Language 

(Metuchen, N.J., Scarecrow Press, 1975). In the last chapter of her book she observes: "If the 

conceptual treatment of human beings moves toward the human being as the higher hierarchy, then 

the language will likewise assume those shapes." An androgynous language is a dynamic 

language that will show neither chauvinism nor bitter grievances, as Key explains: 

An androgynous language will be complementary rather than divisive. It will find balance 

and harmony in its completeness. It will establish an equilibrium in its unity rather than 

invidious separation. It will combine the abstract with the concrete; feeling with logic; 

tenderness with strength; force with graciousness. It will be a balanced tension- 

supporting rather than opposing. it will be exuberant and vibrant, leaving out the weak and 

the brutal. It will move away from the cruel distinctions that have wounded both male and 

female human beings (p. 147). 

Mary Key's perception seems to coincide with the early concept of the dual nature of 

human beings posited in the yinyung principles and manifested in some traces of the Chinese 

language even today. An androgynous language is possible only in an androgynous society in 

both East and West, when both men and women can live in mutual harmony and understanding. If 

language and literature reflect and express social attitudes, they also can have the power to 

influence, to shape, those attitudes and values.

Schrififestschn$- Ersq in Horwr of John &Francis 

A zhezi Anagram Poem of the Song Dynasty 

by 

John Marney 

Oakland University 

4+ 5 d* 

The anthology Tiao xi ji a ;& by the Southern Song poet 

Liu Yizhigl\ -11. (1079-116031 contains a poem identified as a 

zhezigF* verse: 

Sun and moon brighten dawn and dusk. 

From mountain winds mists thence arise. 

Rock and bark split, but remain firm. 

Old trees wither, but do not die. 

That good fellow! when should he come? 

Ideas like double-thousand leagues. 

Ever speaking of intoning the Yellow Crane; 

A thoughtful scholar whose mind ne'er ceases. 

Zhezi is a form of anagram consisting of "dissecting the 

character." However, "dissection" seems here to be the fusion of 

two separate words in each line to form yet a third word in the 

same line. 

Following this technique we find that in line one, "sun" ri 

combines with the next word "moon" yue 4 to form the next word 

"brigbten" ming 0fl . The characters "dawn" chao gfl and "dusk" 

hun 2 also respectively incorporate the "moon" and "sun" 

elements. Line twoAhas "mountain" shan join "winds" feng to 

form "mists" lan . In line three, "stones" shi 6 is juxtaposed 

--P 

with "bark" pi &. to form "split" po . Line four joins "old" 

gu & with "trees" mu to form "withered" ku *$ . 

An irregularity in the scheme in line five reverses the order 

of the elements "good fellow" ke ren 9 h to form "when" he45 . 

"Dissection" is,truly realized in line six, where the third word 

"double" chong 3 splits to form the fourth and fifth characters 

"thousand leagues" qian li $ . In line seven. "ever" yong qc 

is combined with "speak" yan in reverse order to form "intone" - 

yongt*. The first word of the last line, "thoughtful" or 

"ambitious" zhi ,%, , is dissected to form the next two characters 

"scholar" shi * and "mind" xin/i' . 

From the origins of lihe &f$. (parting and meeting) 

anagrammatic verse in second century A.D. Han times, the form 

required that in each line, the anagram elements also provide puninstructions 

for the solution of the riddles. We may thus 

construct an entirely different interpretation: 

[The elements ] sun and moon "brighten" [the characters] 

dawn and dusk. 

[The character] "mist" of itself arises from [the elements] 

mountain and wind. 

[The elements] stone and bark f i r m up [to form the character] 

"smash." 

[The elements] old and tree do not die in [the character] 

"wither. " 

[The character] "when" comes from [the elements meaning]

Sino-Phonic Papers, 27 (August 3 1, 199 1) 

"good fellow. " 

The meaning [of the elements] thousand and leagues is like 

[the character] "doubled up" 

[The character] "intone" eternally speaks as the Yellow 

Crane. 

[The element] scholar [without the element] heart/mind does 

not complete [the character] "thoughtful." 

The homophones or rhyming words that form the anagrammatic 

elements and add complexity to the pun-instructions are 

complemented by obligatory end-rhyme. This follows the abcbdbeb 

sch'eme common to the pentasyllabic-line octet. 

Efficient as good rhyme should be, the rhymewords enhance the 

focus of the poem on the ceaseless continuity of the natural 

universe and of Man's intellectual quest. The first quatrain 

introduces the complementary sun and moon, which combine to 

enlighten the respective elements of their genesis, dawn and dusk. 

M i s t s are spontaneously "born," rhymeword qiB, of the hills and 

winds. Stones and trees (literally, bark) stay firm, though they 

may split. Old trees may wither, but they do not "die," rhymeword 

si , which recalls its complement "born." 

This natural, landscape philosophy is cleverly transferred to 

the persona through the pivotal rhymeword li 2 "leagues, " which 

on the one hand resumes the first quartet concept of infinite 

distances, and on .the other, introduces the far-reaching 

penetration of the good fellow's thoughts in the latter quartet. 

The crucial terminal rhyme, which cements the validity of the 

earlier rhymes, describes the ceaseless activity of his mind wei 

yi & 5 "not yet ended," and elegantly recalls both the matching 

syntax and intent of the first quartet terminal rhyme bu ii 5 a 

"does not die." 

The consistent eremetic Daoist flavor of the sentiment is 

confirmed in the mention of the Yellow Crane. This was a common 

reference in Daoist mythology to the golden crane upon which the 

immortals would traverse the Great Infinity. An early mention 

dccurs in the "Song of the Willow Blossom" Yang hua qu $8 

by the poet-monk Tang Huixiu \g ,$, {& of the late fifth century Liu- 

Song and Southern Qi dynasties, lines from which presage Liu 

Yizhi's ideas: The Yellow Crane northwest goes, 

Carrying my thousand-league hearte2 

Nowadays scholars of the Chinese literary heritage generally 

eschew such compositions as beneath serious consideration. 

Nevertheless, works like this were produced by the very finest of 

China's poets. This verse, by an important Song lyricist, fulfills 

all the demands of poetic form, but in addition to set line- 

length, rhyme scheme, meter, and proper development of the 

content, Liu further incorporates anagram elements within each 

line, and pun-instructions within the anagrams which provide an 

entirely new construction. If today's readers cannot discover the 

high artistry and thou ht here, at least they might enter into the 

fun enjoyed by a consubte writer during his lighter moments. 

8

Schnrfjestschri!: <strong>Essays</strong> in Horwr of John DeFrancis 

Notes 

1. Listed, in Siku tiyao $ . For Liu's biography, 

see Song shi &, (History of the Song) ch. 378, Zhonghua ed., 

pp. 11672-11675. This brief essay is an excerpt from my 

forthcoming book on the history of Chinese anagram and anagram 

verse. The sentiment of this little poem and its philological 

interest seem eminently suited to the celebration of Professor 

DeFrancis ' eightieth birthday festschrift. 

2. For the poet Tang Huixiu, see John Marney, Chiang Yen 

(Boston: G.K. Hall/Twayne, 1981) pp. 115, 126-128. Tang's poetry 

is in Ding Fubao, Quan Han Sanguo Jin Nanbeichao shi (Complete 

Poetry of the Han, Three Kingdoms, Jin, Northern and Southern 

Dynasties) (Taipei:Yiwen yinshuguan, nd.) pp 915-917.

Schniftfestschrifi: <strong>Essays</strong> in Honor of John DeFrancis 

SOME RENARKS ON DIFFERING CORRESPONDENCES IN OLD CEINESE 

ASSUMED TO RSPRESENT DIFFERENT CHINESE DIALECTS 

Nicholas C. Bodman Cornell University 

Professor Emeritus of Linguistics Ithaca, New Ygrk 

This paper was presented first at the 21st International Conference 

on Sino-Tibetan Languages and Linguistics, University of Lund in October, 

1988. It is repeated here virtually unchanged. I am very happy that the 

opportunity has arisen to publish it in the collection of articles commem- 

orating the eightieth birthday of John DeFrancis, a scholar whom allsino- 

logists and linguists respect so highly. I submit tnis with all best wishes. 

Many examples of word families in Old Chinese must reflect dialect 

differences as well as morphologid derivations. Forms found in old 

sources may be labelled as being from a specific location. When this 

occurs, there is usually no corresponding form attributed to another 

locality or to a standard form. Even the words listed in Yang Xiongls 

Fang Yan are not often phonologically related. Of course information on 

their provenience may be given, and is very useful. 

In this paper I limit myself to cases of obvious relationship, and 

particularly cite forms occurring in Sino-Tibetan, with reconstructions 

attested in Tibeto-Burman. With this apprgach one can find instances of 

multiple (usually dual) correspondences between the Tibeto-Burman or 

Sino-Tibetan reconstructed forms and Old Chinese forms. These I 

attribute to different dialect development in Chinese. (I have dealt 

with this phenomenon in my 1980 monograph "Proto-Chinese and Sino-Tibetan, 

.data towards establishing the nature of the relatianshipVpp.!4-199 in Con- 

tributions to Historical Linwistics, Frans van Coettsem & Linda R. Viaugh, Eds., 

Leiden, E. J. Brill.) 

Examples of such a dual development may in my view be found in ST 

final* *-1 which has OC reflexes as -n and -1, and ST initial *sk- which 

has OC reflexes in s- and glottal stop (?-). The most interesting 

examples reflect both these elements in forms reconstructed as ST 

*sk-V--1. One consequence of this procedure is that often one need not 

posit more than one reconstruction and thus reduce the number of 

phonological units that occur in the reconstruction. 

There is no need, in view of the large number of examples, to give 

many illustrations of Proto-Chinese or Sino-Tibetan *-1 which became 

later -n, falling in with earlier *-no However, it is of interest to 

7 -


contrast the two developments when they indicate possible dialect 

variation. I first cite Example 2 from my 1980 monograph. This is then - 

followed by W. South Coblin's item under "joke/laugh, p. 99 of his - A 

Sinologist's Handlist of Sino-Tibetan Lexical Comparisons. 

b.) T styel 'do harm. *skyels, 

play a trick' sjews /sjau- 

hkhyal 'joke, jest ' K. MA 'to laugh, 

ridicule ' 

k yal-ka (Li:*sjagwh) 

rkyal- l a Was) 

**xyial 4 OC * hjian j xjian 'laugh' 

**xya& + -h 4 OC *h jarh > xje- 

' joke, jest ' 

**xyial 

**xyaf 

T 'khyal-ba 'joke, j est1, rkyal-ka kyal-ka 

'joke, jest, trick' 

- stem: 'lchyal 

Certainly from the point of view of present-day Chinese, the first 

example is still in common use while the second, no less valid, is 

however obsolete. Both OC forms here are noted as being from Chu. Example 1, 

however, is noted as from the Odes (%). 

- 

It is interesting to look at my old examples 22 to 27 which were 

given to illustrate a completely different point, e.g. the different 

developments of clusters of *s- and velars. Ex. 22 and .26 show OC -n as 

the outcome of *-1 and Ex. 22 and 27 *-I has developed to *-we (In Ex. 

27 there is a further change where earlier '-1 is dissimilated because 

of the labiovelar initial). 

.2? 

. *skyil, 

(27.) T skyil 'pen up, dam' 7%. 3 jin 13 jtn 4 'dam up' 

123) T. ~kyor 'spoiled. *rkwyrts. 'bad we&. 

w d , Pwjits /3 umi- &n, filth' 

&gelmate '

U L E 

4: 

(24.) T skyerns 'thirst, 

a drink' 

(see also the next ccr:) 

(25) T skom 

skam 

skem 

EXAMPLE 6: 

(26.) T hkhul 

EXAMPLE 7. 

EXAMPLE 9: 

Kanauri 

727.) skupal 

Schr~fesmhnift: &says in Honor of John DeFrancis 

'thirst ' 

'dry' 

Yo dry up; 

km, meagre' 

'subduc, 

subject ' +,,I 

'ex hon, 

enjoin ' 

'to change' 4G 

to shift, 

move' 

'skprn: 

3pm: /3jam: 'to drink' 

*sk~;)ms, 

3 pms 13 jam- 

'give a, dnnk ' 

# *skbrns. 

Xams / x9m- 'emaciated' 

-kbm: I khem: 

F 

*skhhls, 'instrucl, 

xwjans /Xiwan- explain. 

obey, comply 

with' 

*skhwrais, skhupraws, 

xwrajs lxwa- 'trmrforrn, 

change* 

(85.) Proto-Min 

*toi: 'shon' e B 'tol:, 

(Strain A) ton: /man: 'short' 

cp. Lushoi 

tMY 

2 A 'dols, *dows, 

dos / &u- 'bean' 

(the phonetic in 'short' above?) 

(86) Lushai 

ne l 'be flexible' 

L nOl-la-nbl- 'soft, tender' 

I a 

T mnyel 'to tan (of 

hides) 

(= 'soften') 

T mnyen 'flexible, 

supple' 

A *n(r)els, 

*n(r)ews / nau- 'bent wood, 

to bend' 

A +nyel. 

njew / nfjau 'oar (= bent 

wood) ' 

B *n(r)yel 'work leather 

or *n(r)yen, to make it 

n(r)yen: / pjan: smooth and 

soft "' 

(+ -u 10: El Y )Us 

* $51 

& 

foLkwiDg fog B w 

MC Xhja jh: A 

MC khjen:J jtn: (Er Ya) B 

v ~ 1 m 

PC 'khwyal 

PC *khyd. 

OC khwja 

OC khjm 

.f$ MC kbien: thien: (Er Ya) PC +khin:, OC khin: 

EXAMPLE 11: . 

45) T .m-ma 'curdled milk 

# 

13 *d:, lh:, ljuw:' *'wine or 

used as a 

ferment' 

Ijau: / jau: wine must' 

JP fi 

Proro-Tai 

'native beer 

or whiskey' 

'hlau 'wine' 

(tone CI) 

V nfdi 'wine' 

'cyclical sign': 

Ahom do Lil hrau 

PTai 'r- 

" 

'cyclical sign'

Sino-Phonic Papem, 27 (August 3 1,199 1) 

. ' Examples 11 & 12 ha both PC +-u and *-a: 13 & 14 show two 

wLE 12: developents 

cf. hdml 

(*A ml) 

sml 

'to roc, rum 

rancid' 

'become 

putrid' 

'decomposedv 

6 *&I., I~W, ljuw, 

lj au ljw 

jw: 

& 

Ijau 

'rot, decay' 

'stinking 

water plant' 

(*rotten 

smelling) 

, S~rain A 

Strain B 

(2891 % 2 *kal:. 'straw, dried 

*kal:. 'straw of 

kan: I khn: grain' 

. kaw: I kiu: stalk of grain' 

Tf3 '-R r~nd OC -N or -J: 

(?w) T kor 

(STC 350 *kwar 'hole') 

~h~llow, pit 

in ground' 

also - - - 

'hollow tree 

*khwar. 

khwaj I khwa trunk. hollow 

cavity' 

n '& , 

'hole, nest' 

Strain A 

*khwar:. 

khwan: 1 khuAn: 'hole. 

opening' 

EXAMPLE 15: 

PC -L r~nd OC -J (Stroin B): 

$29 1) 7ilgnlny kawaliq 'cooking pol. %$ *kwal. 'kwaw, IShttu rte,l, 

pan' kwaj l kua no text): 

llitrlny kuali 'eanhen 

vessel' 

. Jullu!rrse kuwali 

I, 

n 

'cauldron, 

EXAMPLE 16: 

$4 - 

Proro-Ausrronesian +k(ac)wali * bottle. pot' 

Prolo- Wo + 

1 . 

k30 

(I am indebted to R. Hendon for the Austronesian forms cited here.) 

(299) T mzhil "little bird" ntel: , 

(*m-thyil) ntiu: /tieu: 

Mandaran n.iao: 

*to remove. 44 *scl. 

ckansc' sew / sicu 'eliminate' 

Southern Min ciau: 

'bring up, & *srel:, 

rear' wen: Ian: 'breed. 

bear'

Schn~fesmhnhnfr: fisays in Honor of John DeFrancis 

the phonetic is 5 MC ngj;h 3, and if this is correct, *sngrel: might 

be reconstructed. Note that the Mandarin form is chan, irregular so far 

as the MC initial is concerned. 

EXAMPLE 19: 

635.) L ngrel 'have 

.recourse to 

again' 

L a-ngel 'repetition' 

L nyel 'to repeal' 

tangy-1 

*ngyal, 

nbn / nhn 'gums' 

& *ngdr 

ng jans / ngjfn-3 'steamer 

(double 

also: vessel)' 

'ngyals. . 

ngjans I ngjen- 

(426) 7 hgrul 'to walk. q. 4 *gwnll, pdw. gwju. 'where 

-pass, gwrja /gjwi 3 seved roads 

travel ' meet, 

thoroughfare' 

C427.-) JP khan 

(* khrul) 

'path' 

cp. T Lul 'empty 

('hryul ?) place, 

track, ~ 

furrow, 

t , 

road, way' . 

?'LI 'kwdl:. kwnlw, kwrju:, 

kwja: / kjwi: 3 '... mL ...' 

(432.) Thkhyil 'towind. . $7 *kyil:, 

twist' - kjiw: / kjeu: 4 'twist, plait' 

EXAMPLE 24: 

T hkhyil '(wind, twist), 

water flowing 

to form lakes' 

fs, khyil-thu ( d l ) *puddle' 2 *kwil, kwiw, 

khu-khyil kwe / kiwei 'bole, hollow, 

concave 'z6 

(55) skyil '(to bend). 

dam up water' 

skyil-ding 'small hole 5:: *skwil, skwiw. 'concave, 

(w) filled with %e 13iwei puddle' 

water'

EXAMPLE 26: 


(54 ~kyil 

skyil- 

'LO bend' 

'sitting cross- 7. 2 *skwyil:. 'cross beam in 

khrung legged' swjin: I sjuen: bell frame 

skil-ldir 

(W) 

'handle, ring 

for carrying' 

(support for 

hanging bell)' 

(57) T hkhyil +to wind. #& *kwyil. kwyiw. 'compass, 

twist, (Das:) kwje I kjwie 4 circle' 

LO whirl' 

33 *kulyil, '1.f. potter's 

kwjin 1 kjiuen 4 wheel' 

Examples 25-28 show both glo-1 and sibilant reflexes of original *sk- 

clusters. They also are good examples of *-I> -1. Examples 29, 30, 

and 31 show related words with different OC initials. 

EXAMPLE 31 

(33.) T hgycl *fall. stumble8 

'fall down, 

kill (of 2, 'spryel:, '(bend. bred 

horses) ' *skrje W: , off). premarure 

(*-mel) 'jew:/3jiu:3 btb,LIJ, 

kill r young 

animal ''" 

Sibilant Reflex: 

(48.) skyed 'generate, %. *skyel, 

('skye-d) procreate' sjet / sjat 'name of the 

liarong 

f49.) ka skhiEl '10 take' 

ancestor of 

the Yin- 

Shang 

dynasty' 

('khets 1 khiei- (in other 

meanings) 

Also read: 

Whet I t~hie: 'steal, 

stealthily' 

Glortal Reflex: 

(50. skye 'god, & 'skyep, 'increase, 

p6B) increase, ?jek 13Ak 4 more 

profit. dvanlage, 

benefit' profitable' 

The correspondence of *-9 to OC*-k is sbown..elsewhere (p.135)- The 

phonological correspondences in the last group are regular, but it is 

admittedly speculative to identify the OC in Ex. 48 as an epithet to the

Schriiffesrsch@fl: Essayr in Honor of John DeFrmzcis 

dynastic ancestor meaning "progenitor". Were it not for the tenuous 

nature of Ex. 48, one might surmise that the *s- reflex belonged to the 

"Shang" dialect and the *?- reflex to another dialect, conceivably that 

of the Zhou, but however intriguing, the evidence simply is not 

sufficiently solid. 

(54.) khyil-Chu 

Chu-khyil 

(coil) *puddle' 

" 9 'kwi1, kwiw, 

kwe / kiwei 'hole, hollow. 

concaveyb 

(55.) skyil '(LO bend), 

dam up water' 

skyil-ding 'small hole j& *skwil. skwiw. 'concave, 

(W) filled with %e 13iwei puddle' 

. - 

water' 

(56J rkyil 'to bend' 

skyil- 'sitting cmsc 3. *skwyil:, 'cross beam in 

khrung legged' swjin: / sjuen: bell frame 

skil-ldir 'handle, ring (support for 

(w) for carrying' hanging bell)' 

35: Note the doublet here. 

(57.) T hkhyil *to wind. 4% 'kwyil, kwyiw, 'compass, 

twist, (Das:) kwje / kjwie 4 circle' 

to whirl' 

$9 'kwyil, '1.f. potter's 

kwjin / kjiuen 4 wheel' 

-LE 36: 

(67;) T ma1 'rest, tranquility % .?'snol:, 

('s-?) of mind' hnwaj: / thufi: 'tranquil' 

mnal 'sleep' (1i:hnarx) 

nyal 'lie down. sleep' -6% *snpl, 'to comfort. 

snyol 'lay down. to bed' snhj / swi give repose lo' 

fehis last example shows a morphological relationship, 

not a doublet. This type is easier to identify than 

is t he dialect doublet.

Schrijtfiestschrifi: EFsays in Honor of John DeFrancis 

CAN TAIWANESE RECOGNIZE SIMPLIFIED CIdARACTERS? 

John S. Rohsenow 

The University of Illinois at Chicago 

It is a commonly encountered phenomenon that Chinese from Taiwan state that they cannot 

read the simplified characters now used as the standard forms on the mainland of China. 

DeFrancis (1984: 20 1-202) notes: "...the political separation between Mainland China and 

Taiwan has indeed brought about a sort of linguistic disunity in that the simplified characters 

adopted in the former and the traditional characters retained in the latter have made it difficult if 

not impossible to read materials published on both sides of the Taiwan Strait without special 

training." 

One empirical question then is: Hdw difficult is it for educated readers from Taiwan to 

understand or guess the meanings of standard Mainland simplified characters? In discussing this 

topic with Professor Yin Binyong of the Institute of Applied Linguistics of the Chinese Academy 

of Social Sciences in Beijing. he suggested an experiment in which educated students from 

Taiwan be asked to give the traditional complex forms of a number of Mainland simplified 

characters. 'To that end, we designed a two-part survey (see below) in which Taiwan college 

graduates unfamiliar with Mainland simplified characters were asked lo (1) write the traditional 

farrti character forms corresponding to one hundred standard Mainland simplified jici~rti 

characters given in isolation (Part I), and (2) to read aloud four passages containing such 

simplified characters from a Mainland junior high school textbook (Part 11). The one hundred 

simplified characters chosen are all of high frequency1 and contain simplified characters derived 

or related to the traditional complex forms in a variety of ways to be discussed below. An 

obvious initial hypothesis was that the simplified forms might be easier to understand or guess 

correctly in the context of a running text than in isolation. lnterviewees were presented with the 

two parts of the survey in the order just discussed, i.e. characters in isolation first, and the 

reading passages second. None of the one hundred single characters given in the first part occur 

in the second running text part of the survey. 

The thirty-four interviewees (21 male, 13 female) for this experiment were all post-graduate 

students from Taiwan studying in a large mid-westem U.S. university, all born, raised, educated, 

and graduated from (undergraduate) college or university in Taiwan. Their ages range from 22 - 

to 39 with an average age of 28.6, and their lime spent in the U.S.A. at the time of the survey 

ranged from three months to eight years with a average time of 2.6 years. All born in Taiwan, 

both native Taiwanese (19) and Mainlander (ur~aisltcng) (15) family backgrounds were 

represented. Specifically excluded were any Taiwan students who had visited Mainland China 

or who had had extensive contact with materials written or printed in Mainland simplified 

character^.^ While two of the students were in Mass Communications and Occupational Therapy, 

the hulk of the students were in the "hard" sciences and engineering, both as undergraduates in 

Taiwan and in their present post-graduate studies in the US.; none were Chinese language 

majors in Taiwan. Every effort was taken to insure that the group surveyed would approximate a 

random sample .of educated reader-writers from Taiwan, none of whom had had any significant 

exposure to standard Mainland simplified characters.' 

PROTOCOL: Students from Taiwan were asked to participate in a survey related to 

differences between the Chinese used in Taiwan and the Chinese used in Mainland China, but 

not specifically told in advance that the survey concerned simplified characters.

Sino-Platnnic Papers, 27 (Aupt 3 1, 199 1) 

Each volunteer was interviewed individually and alone by the author, and asked to complete 

both parts of the survey. In Pan I each student was asked to write the traditional complex forms 

of as many of the one hundred simplified characters given as possible. and was encouraged to 

guess if she or he did not know. After completing Part I (no time limit. hut usually within ten to 

twenty minutes), the students were then given the four passages for Part 11, told that they were 

photo-copied from a Mainland junior high school textbook, and asked to sight-read them aloud 

into a tape recorder. Again the interviewees were told to guess at the (pronunication ot) 

characters which they did not recognize. All interaction was conducted in Mandarin Chinese. 

Immediately after all four passages had been read aloud (usually taking about five minutes). the 

Interviewer then went back and asked the reader to clarify orally or in writing which characters 

s/he had intended when it was not clear from context. The interviewees usually completed the 

entire survey in twenty to thirty minutes. 

Part I: Identifying Simplified Characters in Isolation 

1. When asked to give the correct traditional forms of the one hundred standard sinlplified 

characters given in isolation in Part 1 of the survey (and encouraged to guess when unsure), the 

thirty-four respondents were unable to identify correctly an average of 61.53% of the 10U forms, 

with individual student's percentages ranging from only 36% missed to 76% missed. Appendix 1 

below lists the one hundred simplified character forms given in Part I of the survey, followed by 

their traditional standard complex forms as used in Taiwan, plus the number and type of errors 

made by the Taiwan respondents. Note that Appendix 3 lists the 1OO characters from Part I in the 

order of their increasing relative difficulty, from zero (i.e. recognized by all of the Taiwan 

students) to 34 (i.e. rec.ognized by none of the thirty-four students), rather than in the random 

order in which they originally appeared in the queslionaire. 

2. Analysis of Results of Part 1: 

The simplification of Chinese characters carried out in the People's Republic of China in the 

1 950s in many cases either eliminated or codified numerous variant character forms or simplified 

character forms which bad existed in dictionaries or in popular usage for many years, as well as 

extending the various principles upon which characters had already been formed to create new 

forms based on analogous principles. Some of the general principles employed in the simpli- 

fication process (cf. Montanaro 1985: 4-8: Cl~eng 1975) were: 

( 1 ) to adopt as standard some comnionl y used existing simplified forms. or simpler 

antiquated or variant forms still in current use; 

(2) to adopt commonly used cursive or "grass style1' forms, regularized for printing 

purposes; 

(3) to delete redundant parts of characters. or to chose one significant portion of a 

complex character to stand for the whole; 

(4) to employ certain antiquated forms which constitute components of complex 

characters to stand for the whole; 

(5) to substitute homonophous characters for more complex ones when no semantic 

confusion can arise (jir~~ibririn): 

(6) to use certain simpler components to substitute for more complex ones within a 

complex traditional character: 

(7) to extend some of the historical principles of character formation, such as 'picto- 

phonetic' (xingshPng) characters, or 'associative compounds' (huiyi). 

Although no such codifications or simplifications were carried out in Taiwan, and in fact 

were portrayed there as attacks on traditional Chinese culture, nevertheless many of these

Schny?jesstschhriff: Essay in Honor of John DeFrancis 

variants continue lo be used as they have for centuries in informal writing, note-taking by 

sludents. and calligraphic writing. Some of these characters also exist as variant forrns given in 

popular dictionaries in Taiwan, while others have evolved naturally out of fast-flowing, timesaving 

cursive styles of handwriting and calligraphy (e-g. xit~~~.slui; cciushii). Despite the 

attempts of teachers in Taiwan at all levels to stamp out such forrns in formal writing, they 

continue to survive in informal popular usage. 

di 

Obvio~~sly, then. [hose Mainland sim lified characters most easy for Taiwan students to 

recognize will be those (such as \$ for \+ alrC 'to study') which are used in Taiwan, either in 

popular usage, or as dictionary variants. Table I, while by no means exhaustive, gives a number 

of such forms in popular use among students and others in Taiwan. which can be seen to overlap 

with those characters easily recognized by all or by a large majority of the thirty-four Taiwan 

students in Part I of the survey. (See Appendix 1.) 

TABLE I : Some Simplifizd Characters in Popular Use in Taiwan and Their Standard Forms: 

The first rwenly-rwo popularly used forms in Table I explain why the majority of 

participants in the survey could easily identify the standard Mainland simplified forms, in that 

the latter are simp1 y codifications of traditional popular usage and/or variant forms which 

continue to be employed in Taiwan today. Many of the Mainland simplified forms are for all 

practical purpo~ identical to. or differ only to a negligible degree from, their popular 

counterparts used in Taiwan. In addition, because as we have noted another of the bases for PRC 

character simplification in the 1950s was to adopt variant forms as well as popular and 

historically related forms, many of the Mainland standard simplified forms are in fact given as 

variant forms in many of the commonly used dictionaries used in Taiwan, and are thus familiar 

to many educated people. Table 2 lists thirty-four 'variant forms' (yirizi), listed after the 'standard 

forms' (dl2nytizi) in the popularly used dictionary Kuoyii Ribao C'i~licirl in Taiwan, which 

correspond to characters found in Pan I of the survey. 

TABLE 2: Variant Forms Given in the Taiwan Dictionary Guoyii Ribuo Cidicin


Note again that of these thirty-three simplified characters which are given as variants in this 

popular Taiwan dictionary, only nineteen (57%) were correctly identified by more than half of 

the thirty-four respondents. Note also that while many of the variant dictionary forms shown in 

Table 2 correspond to the popular forms shown in Table 1, not all of those popular forms are 

listed as variants in the dictionary (e.g. @), nor are the variant printed forms given in the 

dictionary always reproduced in popular usage. (e.g. 4 for chdng, 'long'). It seems then that 

educated readers in Taiwan have at least some passive familiarity with such dictionary variants. 

even when those forms are not echoed in popular usage, and that this map help to explain their 

ability to identify correctly at least half of those Mainland character simplifications which are 

based on them. Note also that the correspondence between the dictionary variant and the 

Mainland simplified form need not be one hundred percent; for example, while one of the two 

Taiwan dictionary variants for the chracter for rie 'iron' is printed as % , with the full standard 

'gold' or 'metal' radical on the left hand side, participants in the survey were easily able to 

recognize that commonly encountered variant form when printed with the abbreviated 'metal' 

radical && now in standard use in the PRC, probably because such abbreviated component 

forms in fact approximate what they themselves actually write. This reflects the application of 

principles (1) and (2) above in the formation of the standard Mainland simplified forms in the 

1950s simplification, with the adoption of many forms which may be termed c-aoshir kailrlra, 

'regularized cursive' or 'regularized grass' style forms." 

it seems then that simplified characters based upon popular usage which continues to be 

practiced in Taiwan (and other Chinese character using communities) is in fact a more accurate 

predictor of whether Mainland simplified character forms will be recognized than their 

occurance as variants in popular Taiwan dictionaries. Similarly, the principle of merely 

preserving the general configuration (likiio) of one of the traditional forms rather than ado ring 

the traditional form exactly may be seen in comparing the standard Mainland character &for 

yur~ 'salt' with the two variant forms given in the Guoyir Ribao Cidiun ( 5 and .tt ) in which the 

Mainland simplified form seems to be a combination of both. Other highly identifiable characters 

which preserve the general configuration of the orginal character without using a listed variant 

are *$ and $ . We may conclude, then, that the principle of preserving a significantly 

recognizable portion of a character as in the characters $p , & , 9 , 

preserving the general configuration of the original characte both seem on t 

in Appendix I lo be effective in preserving the recognizability of simplified characters by 

Taiwanese, although relatively less so than simply adopting (regularized versions of) existing 

popular variant forms. Similarly, redundant parts of a character are identifiable if abbreviated 

rather than omitted, as in a& and &#$ , as long as significant portions and the general 

configuration of the original are maintained. 

The group which is hardest to examine in a survey such as in Part I, in which characters are 

given in isolation, is of course hotnophone sirhstitlrtes, where one existing traditional character 

was substituted for another more complex form having the same pronunciation. In fact, in the 

four cases in Pan 1 where such forms occurred, many of the respondents simply questioned 

whether those forms ( ;& , y$ , &. , ) were in fact simplified characters at all, even 

when they were not sure of the character's exact meaning. Only in the last case where the 

homophonous character $ is often used in Taiwan in place of the far more common but very 

complex traditional character && did all but five of the respondents correctly identify its 

Mainland usage. 

At the opposite end of the spectrum are those characters which none or nearly none of the 

thirty-four Taiwan respondents were able to identify correctly. As noted above, eleven of the 

thirty-three Mainland simplified forms which also occur in the Glryr, Rihao Cidirrn were 

recognized by less than half of the respondents. (See Table 2.) But the majority of Mainland

Schnfffes~ch n'!: <strong>Essays</strong> in Honor of John &Francis 

simplified characters presenting difficulties for the Taiwan respondents appear to be rather those 

based on principles 3-7 above, that is. simplified characters which were newly o-eated rather 

than merely standardizing characters which were already in popular or dictionary use. 

We may gain further insight into the types of problems encountered by those Taiwan students 

(who on'average were unable to identify more than 61% of the standard Mainland character 

forms) by examining the breakdown of their erroneous answers given in Appendix 1. (Note that 

the total number of errors given equals the sum of the specific errors listed 11llrs those cases 

where the respondents were unable or unwilling to attempt a guess, even when encouraged to do 

so. Simply copying the simplified form was also counted as an error.) 

If we examine those sixty-five out of one hundred characters not correctly idenlified by more 

than one-half of the Taiwan students. sorted into categories according to the above listed 

principles for their simplification, the group presenting the most difficulty were those newly 

created characters which use simpler components to substitute for more complex ones in 

traditional characters (Principle 6: frihao driiri), followed in difficulty by characters created by 

extending the 'pictophonetic' and 'associative compound' principles (#7). and some of those 

created by deleting redundant parts of traditional characters or by chosing one significant 

portion of a traditional characters to stand for the whole (Principle 3). We may also associate 

with category #6 those abbreviated characters which retain some significant components of the 

original characters as well as the overall general configuration ( lhkiro) of the original. As we 

have seen in our discussion of Table 2 above, being an antiquated or variant form (Principle I) 

listed in a commonly used Taiwan dictionary is apparently no guarantee of recognizability. 

Homophones have already been discussed above (Principle 5). Table I groups those standard 

Mainland simplified characters most difficult to recognize in terms of the categories just 

discussed (Some characters are listed in more than one category.) 

TABLE 3: CATEGORIES OF CHARACTERS MOST DIFFICULT FOR TAIWAN 

STUDENTS 

#6: Simplified Components: 4~ . ~ 2 , 

$ ,%a. 

Abbreviated Characters: @ , g, s7 @, p, q---- * - 

#7: Pictophonetic Compounds: g- , $ , c, 

& . 

'&,$h, &, 9, 

a ,y , @ , %k,2+ 

, , 

#3: Redunant Parts Deleted: 4 '@).,a 

, ,$, 3 , x, ,&> 

#3: Use Part for 'Whole: lfF . 2 ,%, 9 ,my 4 , ~. 

#I: Use VariantiAntiquatediCursive Form: 14 . J3 , r/&, 6, $, 5, $ , 

? ?$ D* 

Detailed examination of erroneous forms given by those Taiwan students who were willing to 

guess often reveals t e strategies underlying their rnisperceptions. 

used in Taiwan for @$round' rather than for its homophone @ 'garden', Note and that 

misread on analogy with the commonly used abbreviated form for %&-'true1. 

is popularly 

uy' is clearly 

A 

The general observation about characters in category 6, those which simplify some 

component(s) in a traditional character (with the exception of such commonly used cursive 

radical components such as t for + , k for & . etc.). as well as for the most abbreviated 

@ F

Sino-Platonic Papers, 27 (August 3 1. 1 99 1) 

characters, is that in most cases the Taiwan readers have no unique way of 'retrieving' the deleted 

components which have been simplified or abbreviated, as can be clearly seen from the wide 

variety of guesses for most of this type. The same may be said for those characters which have 

been simplified by ex tending the historical 'pic tophone tic' or 'associative compound' principles to 

create new forms based on those principles ( sin zao de xitzg.~h?~lg zi; xin zao LIC ltiiiyi 2) the 

Taiwan readers have no way to know which one of many homophonous (or near homophonous) 

characters were meant, or in fact that the newly simplified 'phonetic' sum component is in fact 

meant to be a phonetically suggestive component at all. Similarly, how can one know that in an 

'associative compound' such as 'dust', in which the semantically suggestive components are 

'small' and % 'soilf, that these components are here being used solely for their semantic 

properties? Again, if many redundant parts are completely deleted. rather than being replaced 

with simpler components. readers are not able to reconstruct enough of them, as when the 

majority of respondents guess A to be (merely) the simplified form of the less commonly 

encountered character44 'feather'. rather than the correct form 'practice', or do not realize 

which parts have been deleted (taking to he s). or that it is simply a case of parts having been 

deleted, as when 9 is taken as a simplification for9 , rather than for $&? Also, deleting all but 

one central part of a character may not be understood, as with just mentioned. or when both 

and Rare taken to be abbreviations of $. Lastly, as noted a 2 ove. basing a simplification on 

an antiquated. cursive. or variant form is no guarantee of successful recognition. as the cases 

listed under Principle I demonstrate. 

In summary, when asked to identify the traditional forms of standard Mainland simplified 

characters in isolation without any context. unless the simplified form is already in popular use 

or at least based on a fairly well known variant form, readers from Taiwan do not have enough 

information to be able to determine which of several different principles of simplification have 

been applied and often make erroneous guesses, when they are willing to do so at all. 

Part 11: Identifying Simplified Characters in Context. 

In Part 11 of the survey, respondents were asked to read aloud four passages from a junior 

high school reader Yiclr.~;~~: Shiyor~g Kibi.rr, published in Beijing in 1981 .5 (The actual texts are 

reproduced in Appendix I1 below. but with {he simplified characters underlined for reference. 

which was of course not done in the survey version.) The obvious initial hypothesis mentioned 

above, that the reading of simplified characters would be facilitated by context was in fact 

confirmed, especially when characters not understood upon first use were decoded after 

subsequent encounter(s) with additional context(s). This accords with general observations 

concerning the reading process, by which readers make unconscious heuristic predictions based 

on textual redundancy and their native speaker's knowledge of the syntax and usual collocations 

of discourse, as well as their cultural knowledge of the subject matter. Thus, for example, in 

Reading 3, many of the readers in fact said@ 5 for the printed s 5 , unconsciously substituting 

the more common oral form fidng bdi for the written er hai 'two hundred'. In Reading 4, 

although the first encounter with the simplified character >% for Han (dynasty) was not 

recognized by a majority of the readers in the context Xi Hun shi'in the Western Han (dynasty)', 

nevertheless later in the same sentence, in the overwhelmingly familiar context of the historically 

famous name Hull CVu Di, the character was then decoded, so that it was then understood in its 

third mention of $.,>%Dong Han 'Eastern Han (dynasty)'. A more dramatic example of the force 

of context occurred in Reading 3 about astronom 

4 

where many readers could not decode the 

pictophonetic simplified character $& meaning ' 'distant' upon first encounter as an adjective 

in the context id jitli wd~rzen rr211tc yuan, 'it is so ar from us thatt, but the same character was 

unconsciously read correctly in the next to last line in the compound rr~rjrtyira~~jirrS.t 'telescopef, but 

then was not immediately recognized by all of the same readers in the next (final) line as an 

adjective again.

Schrifrestschn'fi: &says in Honor of John DeFrancis 

Appendix 111 records the errors of the Taiwan students in reading each of the four passages 

shown in Appendix 11. Note that the superscripts beside each character indicate the number of 

occurrence, e.g. 6 bd'rneans the second occurrence of the compound tai-vang 'sun' in the text. 

The errors are llsted in order of their appearance in the Reading, each followed by the total 

number of errors, followed by a breakdown of the exact number of each type of error. plus those 

who simply could not guess at all (marked by "0"). The overall number of errors can be seen to 

be small relative to the large number of simplified characters in the readings and to the number 

of students (34). Certain of the more interebiir~g errors in each of the four passages will be 

examined briefly below. 

In Reading 1, note that two of the four respondents who attempted a reading of the coverb 

corrg A).. 'from' chose another grammatically appropriate coverb zai 6 at'. one guessed the 

graphically similar literary cocerb pi 'taking', and the fourth the contextually possible rrreirian 

'every day'. while the remaining twenty -six a1 beit with some hesitation correctly guessed the 

correct reading aided by context. Again, of those ten out of thirty-four students who encountered 

difficulty with the first instance of the simplified character fa 6 in fa(-hlile 'sent out', three 

substituted the verb hti #%'pull out' presumably on the basis of similarity of shape, while the five 

others who guessed erroneously also substituted semantically appropriate verbs. When fa 

occurred again in jZ ~dng 'hurt' in the next to the last line of Reading 1, the choice of verb or 

adverb substituted was again based largely on grammatical and collocational considerations. 

Similarly, note that context usually forced the substitution of another measure word for those 

unable to recognize the simplified form of gi. 'instancelpiece of in zl12gb.~~ 'this' and r~agr 'that1. 

Thus we see that even when erroneous readings were made, those same contextual factors which 

presumably assist the majority of readers in making the correct readings operate to restrict the 

range of possible guesses. 

In Reading 2, similar observations may he made. The various errors made for ling in sinlittg 

'spirit' are all collocations based on coocurrence with the preceding morpheme xirl. The 

misreading of jirn 'military' as ldo 'prison' is obviously based on their graphic similarity, and the 

misreading of hiirliii as hiiwo was explained as being due not only to the similarity of p).and&b 

but also to the idea common in Taiwan propaganda that Kuomingdang agents are 'lying' in wait 

on the Chinese mainland. The collocational force of ke le fdlr 'kowtowed' is so strong that not 

one of the thirty-four respondents misread the simplified character rori 'head' in this context, 

although fourteen later had difficulty recognizing the same character in Reading 3 in a less 

obvious context. The two misreadings of && 'strong' for 'although' were obviously hased on 

graphic similarity, as was the subsequent misreading of for )b. Three readers substituted the 

appropriate particles~@and %g for the simplified 'P;. despite its lack of a D 'mouth' radical. 

Normal collocation explains clrianw2i 'console' being misread as lin~vei 'comfort' and wi.icr+n 

'sympathize'. Lastly, the three substitutes for me /I, in name 'so' are all hased on grammatical 

context. 

In Reading 3, the relative difficulty of yrran 'far' as an independent verb versus its 

recognizability when embedded in the compound noun wangyuhrlgjing 'telescope' has already 

been noted above. As in Reading 1, the second and third occurrences of the coverb cdng 

'from' produced predictable alternative coverbs suitable to the context, but as noted above, roll 

'headlend' is not as recognizable in a less structured context as it was in kt? tori 'kowtow' in 

Reading 2. Context also makes it clear that yi 'one hundred million' is a large number, although 

it was not clear to sixteen of the thirty-four readers which multiple of ten it is. Similarly, context 

made clear the character ji 'extremely' to all but seven readers. Lastly, shu was known 

to all thirty four readers as part of the compound jishir &K'techniquef, but was not clear to five 

readers in the following Reading 4 in the less commonly encountered context xao xhi shir 'paper 

making techniques'.


Reading 4 presented the most difficulty, possibly due the technical nature of the content. At 

the outset of the passage, twenty-eight of the thirty-four readers were unable to decode the 

simplified form rfS, a regularized cursive, or r*aoshir kdihrru form, of the cursive form of shir 

'book'. Although predictably eight of these readers were later aided by the context yi ce slrir 'one 

volume', with the measure word preceding the noun, some readers were still not sure in 

subsequent encounters with the form. (Interestingly, when asked after the entire survey was 

concluded to write the cursive form for shfi 'book', even some of those who wrote the cursive 

form il; apparently did not see the relation between the simplified printed form and what they 

themselves had just written.) The various misreadings for -shfi are clearly based on collocational 

context, as may be seen from Appendix 111. The character Han 'Han (dynasty)' has already been 

discussed above. Aside from those readings obviously based on graphic similarity, those five 

readers who misread the character& in the name Cbi ~un%{h (the inventor of paper) were 

apparently more familiar with the name of a Kuomindang ge6%ral Cai E (fg). We have already 

noted above that five readers who had no difficulty with the character & in the word jishi2 

'technology' in Reading 4 could not correctly decode it in the less familiar context zau zhi shir 

'paper making technology'. In context. - far~liny i:~a ' invent' presented no problems. Dong 

Fangsuo is a famous historical figure, and the character dong 3. 'east' had already been seen in 

the collocation Dong Han 'Eastern Han' above. 

The four reading passages in part 11 altogether contain 1142 characters, of which 305 or 

26.7% are standard simplified characters. [See Appdx. I1 below.] The maximum number of 

potential errors of all thirty-four Taiwan students reading all 305 simplified characters 

incorrectly would be 34 x 305 = 10,370 possible student errors. In fact the total number of all 

student errors for all four passages was only 507 [see Appdx. 1111 or less than 4.9%, meaning that 

the students' overall accuracy of reading of the 305 simplified characters was more than 95%. 

Note again that the 305 simplified characters represent only 26.7% of the total number of 

characters in the four passages, so a 4.889% error rate in reading only 26.7% of the total text 

equals 4.889 x 26.7 or an average of only 1.375%, that is, only slightly over one percent of the 

total number of characters read. 

We may conclude, then, that whatever difficulties educated readers from Taiwan may have in 

recognizing standard Mainland simplified character forms in isolation, Part I1 of this study 

clearly demonstrates that in discourse context that textual redundancy and native speaker reading 

strategies greatly facilitate the decoding of these simplified character forms for such readers, 

albeit sometimes at a level of which they are not consciously aware. Guesses as to unknown 

simplified characters have been seen to be based more on discourse context than on the shapes or 

structure of the characters themselves. An analogy may be drawn to the orthographic reforms 

proposed for English spelling by George Bernard Shaw. in which the letter x would replace tA. y 

would be replaced by i. and the symbol y could then fulfill the function of sh. Without such 

background information, native speaker-readers of contemporary English would presumbably 

have similar difficulties with the following conclusion: wi wud hev a lujikl, kohirnt .spcling in 

j~rs x).rDwurrr xc irrgliy-spikirtg wvt-ld. 

FOOTNOTES: 

1. All of the characters contained in Part I of the survey fall within the basic list of 6763 

standard simplified characters which account for 99.99% of all characters in common 

contemporary use. 

2. For example, two Taiwanese Christians had procured Bibles printed in simplified characters 

in hope of conducting Bible study sessions with their Mainland classmates in the U.S.A., and one 

student called to my attention a handbook of Mainland simplified characters recently published 

in Taiwan (cf. Zhang, n.d.). The majority of students from Taiwan, however, appear to have 

little or no significant contact with their fellow students from the People's Republic. even if they

are in the same academic department or office, nor do they have any exposure to materials 

printed in Mainland simplified characters. Note that since 1980, the overseas edition of the 

people!^ Daifv newspaper (RL:,rr~rirz Rib60 : Hai,rlai Ban). has been deliberately printed in 

traditional characters for distribution to Chinese overseas. 

3 . I wish to express my thanks to Professor Yin Binyong for helping to design the original 

survey as well as for the statistical calculations in Part 11, to Wang Fenghua for collating the raw 

data, and to the members of the Chinese Students' Association of UIC fur their enthusiastic 

cooperation and assistance with this research. 

4. See F.Y. Wang 1958: xx. 

5. The original sources of the four reading passages (see Appdx. Il), which were not identified to 

the Taiwan students, are (Reading 1 :) Ba Jin: Hai Shnng de Riclllr; (Reading 2:) Wri Wei: Wude 

Laoshi; (Reading 3:) Zheng Wengirang: Yiizhorr Li Yolr Xir Shi~trnze?: (Reading 4:) Xiarzy Yi: 

Shrrji cle Bicirrjli. 

Beijing Jingshan Xuexiao Zhongxue Yuwen Zu. (1981) Yrtwrrc: Slliyoug Krbm (Ch~rzlrong Di Si 

Cc). Beijing. 

Cheng, Chin-chuan (1 975) "Directions in Chinese Character Simplification" Jorrt-rial of Chinese 

Linguistics 3: 2 13-220. 

Chinese Language lnsti tu te (1 969) Co~nmorzly Used Simplified Characrct-s (2nd ed.) Hsinchu. 

Taiwan: Fu Jen University. 

DeFrancis, John ( 1984) Thc Chinevir Langiiage: Fact arrd Furzta.sy. I-lonolulu: University of 

Hawaii Press. 

He, Rong (1 975) G'rroytr Rihao Cillian. Taibei: Guoy u Rihao Chubanshe. 

Montanaro, John (1985) Conzples and Sirrzplifirrl Foms of Chinese Chat-actcrs. New Haven: 

Far Eastern Publications. 

Rohsenow, John S. (1986) "The Second Chinese Character Simplification Scheme" 

lnrerrzutiunal Jout-nu1 oj'ilrc Sot-iology of Lurryrruge 59: 73-85. 

Wang, Fang-yu (1958) Chinese C~tr-.si\~e Script: ,411 lrrtr-orlrrcrion to Hnnriwr-ititrg in Clzinese. 

New Haven: Far Eastern Publications. 

Y in Binyong and J. S. Rohsenow ( 199 1) Mori~~rrz Cllitlcsc Chat-ucter-s: A Hurrrll~ook. 

Beijing: Sinolingua Press. 

Zhang Baotai (n.d.) Dulrr Shiyot~g Jiantizi. Taibei: Leiding Yinshua Shiye Youxian Gongsi.

APPENDIX 1 : I00 SIMPLIFIED CHA 

OF RELATIVE INCRE 

34 correct; 

missinolank: (%), 15 @,), 1 

# SIMP. TRAD. ERRORS 

x POIIM FORM (Nu. irf ouch:) 

33 correct: $ 

one each -- 

left blank: jl,! (Zj), 

............................ ~(a 

---------------------- 

4% (is] (1) a& (sic) 

--------------------------------------------------- 

31 correct: #$J$&) 

three blank: 

5 t& (%*) (I) A& $6 

' 4 (eg)(]) 

+ q* 

---------------------------------------------------- 

7 3s (&@) (4) & 

.................................................... 

0 

(EB) (8) 

w 

6 ) 

d = & * 

\9 fl 

(1) 9 Q 

---------------------------------------------------- 

9 I* (fig) (i)4y\%Fl 1% 

Sino-Plaronic Papers, 27 (August 3 1, 199 1) 

.ACTER FORMS TESTED IN PART I IN ORDER 

LSING DIFFICULTY FOR TAIWAN STUDENTS 

-------------------------------------------------------------- 

I 

I # SIMP. TRAD. ERRORS 

i x FORM FORM (No, uf e ~h:) 

276 (h)(4) @ (I) /@ J& g 

-------------------------------------------------------

Schnfifestschhriji: &says in Honor of John DeFrmcis 

APPENDIX 1 (CONTINUED) 

---_---------------------------------------------------------------- 

# SIMP. 'SRAD. ERRORS 

FORM FORM (No. of each:) 

---------------------------------- 

29 &(@)(2)& (I)& t& 

.................................. 

30 ++ (a- (21~4 

.* 6453 

('IF+ 3-h 575 

3 0 5 

30 '13 ($*)(l)~h 

.................................. 

( ) 

31 &l$!) (5) 2- 

(2) & (1) ?& 

31 (31% (2) k?3 

r lP 7 

('14 t* I$1 $, 

f2 * 

31 (4%) (3) g (2) ft !$ 

+7\+lv39+ 

(11 g EJ gp &Nii&$g 

31*(4-)(6)% ('1 %(')% 

31% c*) (Wfi 

.................................. a 

32% (T) (71% (2)* 

g&,s&m ,'SF 

(l)% -q- i& 

......................... 

-+ 

HJ 35 -------------- 

3 3 ( ) 8 (3) (2) 

33%~ 

c$$g-) (2) (1) 4ikW &, $8 (a 

( 4 ) 1 33ff (&)(12)$j!. (2) (1)xm 

&*% lllp~b)(2)$p (I)&&:, 

33 1st (1%) (4)4& (3)l$t. 

(1)IfQ '\$ 7% g4g '1% 

33 5 ($5) (3)3& (2) S (11% 3 t3 & 75 

33k (Zi) (261% 

33k (kt) ( 1 ) b G =p yg a & 

Z$ y .*& q 

3 3 ~ ~ & 1 ( l ) % 

33pt(a)(4)~jfig, ( ~ ) q ; t q e ~ 

3 3 

........................................... 

32&(%1 (2216 (3)& (2)& 3 4 ( 5 ( 3 ) s (2) & 

(1) %2 @ @ & (1)s *&& * 3 & 0 

* 

( 3 ) s (2)& 344 (49 (2) & a (1) 

(1) & 34% g) (2)& (1)3 &q 

32 (3) (18)a 34% (2) ( 3 4 

(rn~9'tigi~ (2)[g (1) # 

3431s (%)(2) ?k (1) gj %% 

34Q(&~)(l0)$fig (4)@ (3) gfi @ (l)Q 

(l)g! ?T 8i 34$iwg7v6) %1; (4)E (l)&$gx 

...................... ----------- -------------------------------------------

Sino-Phonic Papers, 27 (Au~st 3 1, 1 99 1)

Schnftfeschrif: <strong>Essays</strong> in Honor of John DeFrancis

Sirlo-Platonic Papers, 27 (August 3 1, 1 99 1) 

APPENDIX IXk ERRORS IN READING PASSAGES JN PAIT II 

(Line numbers in text shown on left; superscripts indicate number of times of occurrence; arabic 

numbers show number of instances of each enor: 0 = unable to guess.) 

CHARTR. ERRORS: (# OF) 

IN WORD) (0 = NO GUESS) 

I I READING 1 

L 

1 & & ( 1) (See line 3) 

4 0 (2) (See lines 5,s) 

# kIN WORD) (0 = NO GUESS) I 

I READING 2 I 

0 ( I) (See Rdg. 1 Line 7)

' 

Schnjifestschn'fi: Essay in Honor of John &Francis 

(Line numbers in text shown on left; superscripts indicate number of times of occurrence; arabic 

numbers show number of instances of each error; 8 = unable to guess.) 

LN. CHARTR. ERRORS: (# OF) LN. CHARTR. ERRORS: (# OF) 

# (IN WORD) (8 = NO GUESS) # (IN WORD) (O=NOGUESS) 

1 READING 3 

i 1 

17 & @(I) (Seelines8,16) 

READING 4

S~hriff~t~~hn$: <strong>Essays</strong> irz Horwr of John DeFrmis 

SIMPLIFIED CHARACTERS AND THEIR (UN)RELATEDNESS* 


University of Florida 

simplification of the ~hinese characters has been one of the 

main efforts in the modernization of the language. Its success is 

well reflected in the official adoption of the simplified forms in 

Japan and Singapore, though with some modifications.' As a matter 

of fact, even in Hong Kong and Taiwan, simplified versions have 

been in use in hand-writing; they have, however, not been made 

official and therefore do not usually appear in print. All these 

facts ind-icate that simplification is a necessity as long as the 

Chinese characters serve as a medium of communication. . . 

While the above facts do suggest a step toward a written code 

simpler in the form of individual characters, they do not 

necessarily represent an advancement toward a system that is 

internally more consistent among its members. This paper is thus 

an initial exploration of whether the written symbols form a 

logical system in terms of pronunciation and meaning as a result of 

their simplification. It is not my intention, however, to discuss 

the efficiency of the code as an educational tool, which though 

ought to be the ultimate goal of the entire business of 

simplification.' 

1. Principles of Bimplif ication 

Wang (1955:185-187) deduces ten patterns (or modes) of 

simplification from the forms in use. They are as follows: 

for&,$bfor$$, 

(1.1) Part Replacing 

Jf- for&,x for$&, etc. 

(1.2) ~limination of 

etc. .--- 

the Whole: )L for , & for#$,;E forjB, 

Repetition in form: for R , & for & , 

Sign: /a for /&, 55- for 

for % , &for 

Determinative-Phonetic: 

etc. 

Non-Determinative-Phonetic to Determinative- 

,& for@, , etc. 

Homophone as a Sub titute: for#+,/fforla, 

&for&,$ for#,@ forf3j.K for

etc. 

Sine-Platonic Papers, 27 (August 3 1, 1 99 1 ) 

While the aim of Wang's paper is to justify simplification by 

historical development through these patterns, the patterns 

themselves may also have some synchronic significance, which we 

will later discuss, 

2. How Many Are Real ~nnovations? 

Despite the claim (Wang, 1955: 187) that patterns (1.1) - (1.8) 

all involve some kind of 'innovation', there is little evidence 

that innovation is the main force in most of the simplified 

characters under those categories, unless innovation is taken to 

mean deviation from the recorded written form in the history of the 

language. In other words, there are relatively few forms that were 

completely started by the language reformers responsible for the 

movement in order to replace the old complex forms. 

To sustain my claim here, it would be desirable, though not 

quite feasible as an individual project, to pinpoint which 

simplified character was a complete innovation in its strict sense. 

It would, however, be relatively easy based on someone's personal 

experience to see how many are new in comparison with those which 

he had already learned before the official inception of the 

simplification movement. The person chosen for this survey must 

have had at least a high school education before 1954 when 

simplification officially began. He must not have had frequent 

contact with the simplified versions since then. The present 

author happens to roughly meet the cpalif ications .3 But, of course, 

his knowledge of the characters before 1954 is bound to be partial 

and thus does not cover all the existing ones prior to that date. 

A recognition of this shortcoming, however, will only add strength 

to our argument below. 

By my judgment, 268 out of a total of 2,239 in the 1964 

Official General List of Simplified characters4 are quite 

Iunusual1. For lack of a better term, the label gunusual' is here 

used to denote four situations: 

(2.1) The character would be completely unfamilair to me if it 

were not listed, 

(2.2) I would not be able to equate the character to a complex 

one without an appropriate context, 

(2.3) I am likely to give the character a meaning other than 

the one for its corresponding complex form, and 

(2.4) I am familiar with the character, but do not use it in 

my own handwriting. 

In our later discussion, the lunusualf characters will be grouped 

accordingly. These 268 characters thus seem to be qualified for 

the label 'innovation1. 

The above figures easily point to another undeniable fact, 

i.e. in actuality, at least 88% of the officially approved

Schrififestschri$: <strong>Essays</strong> in Honor of John &Francis 

simlified charactes had already been in existence long before the 

movement started. This large number of popular forms of characters 

obviously were later blessed with official sanction to become 

entries on the General List. 

The next questions that we might ask are then: How systematic 

are the characters in terms of their manner of simplification? Is 

there any difference in systematicity between the innovated ones 

and those that were adopted from popular use? 

3. How Systematic Are the simplified Characters? 

The 268 characters, as innovations, might well be expected to 

be quite systematic in the reduction of the numbers of their 

strokes. And, indeed, they are. The following table is a 

breakdown of the number of characters in each mode of 

simplification (as mentioned in Section 2), cross-classified by the 

'unusual1 groups the characters belong to. 

TABLE I: Modes of Simplification for Innovated Characters 

Modes of Sim~lification 

IUnusual1 Groups Part for Simpler/New Homophone Other 

Whole Phon. Siqn - Sub 

1. Completely unfamiliar 

a. Individual Words 18 

b. With Derivatives 14 

2- Unable to Equate to 

Complex Ones 5 

3. Meaning Assigned Other 

Than Intended 3 

4. Familiar But Not Used 

in Handwriting 5 

Total 

The grand total of the characters in the table is only 116, which 

is far short of the above mentioned 268. The reason is that a 

character derivable frpm another is 'not counted, e.g.7 , etc. 

are subsumed under for jr and thus not counted. he 35 

simplifications under 1.b in able I therefore actually account for 

about 150 individual characters, 

From the figures in Table I, it is easy to realize that 

innovated simplified characters are mostly regular and systematic 

except for the 24 under the category 'Other1. These 24 less

Sim-P2arorzic Papers, 27 (August 3 1, 1 W 1) 

regular ones are listed in (3.1) below: 

Among the 24, some are ancient forms, e. g. & and ; some 

from popular forms for other characters, e.g.4 

Still others seem to be semantically motivated, 

and theF Iday or daybreakf in&. lost 

e.g. 9~. 

,& , 13 , , % , 2 , & , 

etc. But all together, they ony constitute less t an 10% of the 

total innovated simplified characters and do not argue against the 

overall regularity of the entire class. 

Turning to those directly adopted from existing forms, which 

are much larger in number, we find it not feasible to do the same 

kind of analysis, i.e. by grouping them into modes of 

simplification and comparing the number of each mode. But the 

overall impression is that many of them are also systematically 

simplified since in many cases what is involved in the 

simplification are the radicals, such as in (3.2). 

(3.2) 4 for s f ?j for$,$for&,s for,& ,% for,%, for 

14 

~ , ~ I ~ ~ 

for* 

~ B,% Y 

for*, etc. 

There is, however, a great amount of confusion among many sets 

of characters. A well-known example is the overuse of the sign 

in more than a half dozen sets of characters, which are not related 

in any sense. These sets are given below in (3.3)-(3.9). 

A question related to this set is: What would ##'and ?$ be? 

(3.7) = g as in &I& (But note that @is not the simplified 

f o m of$@. ) 

\ 

(3.9) % = 6 as in ,& (Note that)& is simplified to )&. ) 

I.

Schnmestschnp: &sap in Honor of John DeFraneis 

This overuse of an arbitrary sign, though contributing nothing to 

inconsistency, is at least suspect in promoting the characters as 

an efficient system of signs. 

We will discuss the simplified characterrs as an internally 

consistent system by giving examples to illustrate their relations 

or non-relations in the next section- 

4. Internal Consistency as a System---How Some Characters Are 

(Un) related 

Chinese speakers are often asked how it is possible to learn 

to read Chinese which consists of thousands of separate symbols. 

The obvious fallacy of the question is that the symbols are 

regarded as independent of each other. As a matter of fact, nobody 

learns each character independently of all others. As most of them 

are related in one way or another, they can be learned by 

association. Relations between characters therefore constitute a 

main cohesive force to make them into an integral whole, i.e. a 

cohesive system. Such relations are mainly of two kinds: semantic 

and phonetic. Many simplificatins are based on semantic relations, 

for example: 

But a much larger number of them are rather based on phonetic 

relations : 

In the following, some examples will be given to show how such 

relations and others are not consistently observed. As a result, 

some confusion is created, at least in matters of rationalization, 

if not in matters of instruction. 

A. The symbol & is used fo$& in all its combinations, thus 

representing not only the pronunciation of [long] as in ( 4 4 , but 

also six other pronunciations as in (4.4). 

(4.4) [gong] : &? ; [xi] : & ; [kan] :%; [chong] : ; 

C pang 1 : z ; lshuangl : 5& 

This is an example of not being able to break away from the 

traditional orthography where the relationship between the 

characters sharing the same sign has already been obliterated. 

B. A similar example concerns the sign to replace8 in all 

instances. Thus, it is used to represent quite a few 

pronunciations while the characters sharing the same sign have no 

semantic relations between them, as given in (4.5).


(4.5) [wan] : 5 ; [mail :ih_ ; 1 , ; [dun]: ,q ; 

[chai] : ZZ 

C. A third example is the symbol ,*, in cases where it 

represents three different pronunciations, as in (4.6) . 

D. A worse case concerns the symbol 2 . Its replacement of 

portions of several different characters creates some similar forms 

which might be expected to have similar, if not identicasl, 

pronunciations. They are, however, pronounced quite differently, 

as shown in (4.7) below: 

(4.7) [yun]: , , [tan]: &Z (=&or@) . 

[ceng] :A ( = ) ; [chang] : (2 = ) ' ( =/+) @ 

t5 

The use of this sign seems to have been adopted from random 

creation by popular practice without serious consideration of its 

consequences. 

E. In quite a few cases, an unnecessary new symbol is created 

or adopted to replce a common complex one in some characters but 

not in others that have the same complex symbol. They are 

illustrated in (4.8) - (4.14) . 

(4.9) Z$'J for $ , but '$3 for and all its derivatives 

J, n f gr all its derivatives except or g 

and 

where the bottom p .rt wou . easi 

identi .th the character + head 

(4.11) for@, but $ itself remains the same 

(4.12) yr f or)'@nd $T f orfg, but* itself remains the same 

(4.13) @ for-& or@ , and)@ for~g, but& itself remains 

the same 

If the cases in (4.8) -(4.10) are justifiable, then there is no 

reason why more than one symbol should not be used in (4.3)-(4.7), 

where different pronunciations call for separate representations in 

spite of the same 'roots' in terms of historical development. In 

(4. 1 1 4 . 14) , if the complex symbols in question are retained 

anyway, then the simplifications only save a few strokes without 

contributing to the consolidation of the relationships between the 

characters. 

F. Some simplified characters with the same phonetic sign have

Schn'$t$iestschrifr: &says in Homr of John DeFrancis 

more than one pronunciation, such as in (4.15) and (4.16). 

, but is pronounced 

(4.16) UF is pronounced [xia] except in LR 0% where it is 

pronounced [he]. 

G. Many phonetic signs are used for dialectal pronunciations 

rather than the Mandarin or Putonghua pronunciations. That is, the 

characters represented by the same phonetic sign happen to have the 

same pronunciation in some dialect but different pronunciations in 

Mandarin or Putonghua. Examples of this nature are given in 

(4.17) - (4.20) below. 

(4.17) (+[jia] for{$;: \ [jie] for?& 

(4.18) [jing] for% : 3% [qiong] for 

(4.19) 84 [yin] for&; [yao] for& 

(4.20) is pronounced [ju], but is used in $E [gui] for* 

H a the complex form,& is represented by two different signs: 

,? andp , as in (4.21) and (4.22). 

This happens perhaps because the four graphs in (4.21) had already 

been used for sometime before 1954. To use the same sign for 

those in (4.22) , howeyer, would dre 

would represent both P [hu] itself where ;$ would 

represent bothfa [hu (which had use for a long . 

time as a shor ened name for Shanghai) 

5. Conclusion 

In this short study, we have shown that a large number (88%) 

of the simplified characters in the official list were adopted from 

those which had already been in popular use before the 

implementation of the simplification movement. It is therefore 

inevitable that the language reformers yielded to the pressure of 

popular practice more than they adhered to principled means of 

simplification. Of the ten patterns (or modes) of simplification 

deduced by Wang (1955) , only the adoption of a new phonetic sign 

(cf. 1.8) and the replacement by a simpler sign (cf. 1.5) could 

potentially add internal consistency to the characters as a system. 

We have thus examined cases where these two modes of simplification 

are involved. To our disappointment, in no case has the 

simplification helped make the characters form a more consistent 

system.


In this initial stage of language reform, attention seems to 

have been centered on the reduction of the number of strokes, It 

is hoped that the problems brought up in this paper will lead to 

more serious discussion about the characters not as separate 

individual entities, but as members of an internally consistent 

system. 

NOTES : 

*An earlier version of this paper was presented at the 

Conference on Linguistic ~odernization and Language-Planning in 

~hinese-Speaking Communities, in Honolulu, Hawaii, September 7-13, 

1983. I am grateful for the comments that I received at the 

Conference, though I didn't necessarily incorporate all of them in 

this version. I would also like to thank the editor of this volume ' 

for his valuable editorial comments. 

1. Confer Chen (1977). 

2. For controversies over simplified characters as an 

educational tool, see Cheng (1977 and 1978); Tzeng, Garro and Hung 

(1977); Tzeng, Hung and Garro (1978); and Leong (1977). 

3. The author has since been visiting PRC and reading 

newspapers and books printed in simplified characters. He perhaps 

would not be as qualified to make such judgments today as he was in 

1983 when this paper was first written. 

4. According to Guangmina Ribao, May 10, 1973, the total 

number of simplified characters in the General List is 2,238. But 

my count has one more. 

5. For a classification of the 268 runusualr characters, see 

~ppendix. 

6. There are other similar cases, such 

$ ; [ail $?& ; [kai]&; [wei]&& ; and [hua] 

. But as they do not involve commonly 

th'e first one in each of the multi-member sets, we will not discuss 

them. 

Chen, Virginia Wei-chien (1977). 'Simplified Characters in the 

People's Republic of China, Japan, Singapore and Taiwan,' JCLTA 

12:1.63-77. 

Cheng, Chin-chuan (1977). 'In Defense of Teaching Simplified 

Characters,'JCL 5.314-341. 

Cheng, Chin-chuan ( 1978) . ' Simplified vs . Complex Characters : 

Socio-Political consideration^,^ JCL 6.272-286.

Jianhuazi Zonqbiao Jianzi < @ ( , $ > (1964). 

Beij ing . 

Leong, C. K. (1977). 'Another View of Complex and Simplified 

Characters,' JCL 5.342-6. 

Liu, William (1977). 'A Rejoinder to "In Defense of Teaching 

Simplified Characters'' by Cheng Chin-chuan,' JCL 5.347-8. 

Tzeng, Ovid J. L., Linda Garro and Daisy L. Hung (1977). 

'Research on Chinese Characters: A Call for Interdisciplinary 

Endeavor,' JCL 5.349-354. 

Tzeng, Ovid 3. L., Daisy L. Hung and Linda Garro (1978). 

'Reading the Chinese Characters: An Information Processing View,' 

- JCL 6.306-314. 

Wang, Xian (1955) 'On the Patterns and History of 

Simplification of Chinese Characters,' 

Reprinted in Xiandai Hanyu Cankao Ziliao < 

Vol 2, pp. 185-192, ed. by Wu Yushu. 

Publications.

Sino-Platonic Papers. 27 (Au,oust 3 1, 199 1) 

APPEND I X 

268 Characters That Are 'Unusual' to the Author 

1. The characters would be completely unfamilar to me if it were not listed in 

the General list: 

A. Individual Characters: 

tie[i / 

a. Part for lihole 8 f o r for m; &for #;$for j$:;-&for gi 

3. for k; for jg;'g for o r 

Pafor for Z@for $;$ for @;>$!for &Zfor and 8 

?+for gi8 for 1; 

$ for g; 

b. Simpler Phonetic Sign (21): '$for etf;fi~for #;gfor J$;&for &o7for 

I#;$? for %;$&.for B;#&for &&for &$j for $f;ikfor $$iiL 

for ~f$&for Ei,&for g;fi for @$%for &;%for &$ifor 

for for Mi+bfor # 

dl Other (9i:fifor i f &%?for PifAior F;i.R.for &@for 

Rif~for 

9; 

6 for 3;~Jj I-. for 

B: Characters with Derivatives or in Combination with Other Signs: 

(14):a 

a, Part for Whole for jf;*for #; 3 for 3 (as int3 I;? for @ 

lPfor #;&for $; P for # (as in@ Ji &for @&for &&for 3 

f o r $9 for @;?for RiKfor 

b. Simpler or New Phonetic Sign 191; for 9; for (as in#Q)iyfor 

8;fifor E; 7 far$ (as inill; f for (as injTl&(as in 

*t and & (as in^?); 1 for f$ (as in@ t; Z for $J (as in

S~hriffsfs~h??@: <strong>Essays</strong> in Honor of John DeFrancis 

The Teaching of Culture and the Culture of Teaching: 

Problems, Challenges, and Opportunities in Language Instruction 

Let me begin this polemical excursion with a few accounts of actual incidents: 

A native-born Chinese teacher of Chinese says to his American student, 

whose pronunciation of Chinese, even after years of instruction, leaves something 

to be desired: ''When you go to China, please don't tell anyone who your teacher 

was. I would be embarrassed." 

* * * * rC 

An American teacher of Chinese, with excellent linguistic skills and fluent 

Chinese, confident in his knowledge of Chinese culture, calls a nativeborn Chinese 

teacher of Chinese, to persuade him to leave the university he's teaching at to come 

work for him. This recruitment call is made to the office where the teacher is 

emvloved, not to his home. The Chinese receiving the call declines the offer 

politely, but he is barely able to suppress gales of laughter at the colossal insensitivity 

of the American-born "expert" on Chinese culture. 

* * * Y * 

An Englishman has a falling out with a long-time Chinese collaborator on 

scholarly and philosophical subjects relating to China. In their acrimonious break- 

up, the Englishman says to the Chinese in exasperation: "You Chinese don't 

understand China at all - only the English understand China." 

1) * IC * 1C 

I relate these stories because they embody certain complexities that need to be 

"sorted out" if we are going to understand what is true and what is false in each 

experience. The interesting thing about each anecdote is it contains both a salient 

truth and an outrageous lie. 

In the first instance, there is no doubt that a native Chinese, upon hearing the 

agonizingly inaccurate accent of an American speaking Chinese, will not think very 

highly of the student's teacher, especially if that teacher were an ethnic Chinese. 

Conversely, he would be inordinately impressed by an American student speaking 

excellent Chinese, particularly if that student's teacher were not Chinese! These 

observations are unarguably -me. What is outrageous about the story is that the 

Chinese teacher of Chinese has abnegated all responsibility as a teacher: given a 

conflict between viewing himself as a Chinese and as a teacher, he opts instinctively 

for his identity as a Chinese. His behavior as a teacher, however, is totally 

reprehensible: he puts his identity as a Chinese above his duties as a teacher. But 

his failure is greater than failing to correct the tones in his student's spoken 


In the second 'instance, the American teacher of Chinese is appropriately 

enterprising - and typically American - in seeking out the most qualified 

instructors for his faculty; he shows a devotion to the field in wanting to improve 

his institution's instructional resources, attracting the best teacher he can find. He 

has certainly been authentic to his own culture by approaching the recruit in a 

forthright and direct manner. In the United States, it is not uncommon to receive

Sirw-Pkrunic Papers, 27 (August 3 1, 199 1) 

"feelers" from prospective employers from a rival university at the offices of the 

very university being raided. In America, "All's fair in love and war." That much 

is true. But what is outrageous is that someone who professes an intimate 

knowledge of Chinese culture should be so ignorant of the mindset of his coveted 

recruit, whose cultural underpinnings are, after all, Chinese. Whatever the interest , 

of the recruit in the proffered opportunity, he will be embarrassed to discuss it on 

the very telephone lines of the institution that he might be in the process of 

"betraying." The interesting thing is that an American would consider an approach * 

to the office open and above board; an approach to the recruit at his home, on the 

other hand, an American would regard as sneaky and underhanded. A Chinese, on 

the other hand, would consider the contact at home appropriately disaeet and 

reassuring, whereas the contact at the office he would find compromising and 

morally incriminating. 

The Englishman in the third story is, of course, right in a sense. There are 

some ways in which a native cannot understand what a non-native understands. 

Unless we are trained linguists, we often cannot explain why certain oddities exist in 

our native language (try explaining to a student of English why it is - as Richard 

Lederer, in his amusing book, Cr English, points out -- that we "park in the 

driveway and drive in the parkway"; why is it that we don't park in the parkway 

and drive in the driveway? Why is it that a person who has "hair" [singular] on his 

head has more hair than-someone who has 'hairs" [plural] on his headj. When 

confronted with such conundrums, we resort to the useful generic non-explanation: 

"I don't know why, but that's just the way it is." It is by now a truism that natives 

become so accustomed to a culture's conventions that they can no longer see its 

peculiarities. But what is outrageous about the Englishman's claim, that only the 

English can understand China (and I can assure you that he is not the only 

Englishman who believes this), is the inference that the Chinese are preternaturally 

ignorant and incapable of understanding their own native culture. To be sure, 

natives who do not reflect on their own culture systematically, who are not trained 

analytically in anthropology, cross-cultural psyd;ology, or linguistics, don't have 

the foggiest idea about their own culture as subject matter. But this would be as 

true of English natives as Chinese natives. If what the Englishman said is true - 

that only the English can understand the Chinese, one might entertain the equally 

outrageous yet plausible daim that "only the Chinese can understand England"! 

What is outrageous about the Englishman's daim is his insistence that a native, by 

virtue of her being native, is incapable of understanding his own culture. I trust I 

don't need to comment on the implied superiority of Englishmen. 

These anecdotes highlight some conflicts and problems that I would like to 

examine reflectively in what is to follow. I want to understand what these 

encounters mean: why they are so upsetting, and what solutions - what insights -- 

one can find to the questions they pose. 

In times past, to find someone to teach a foreign language, the obvious thing 

to do was to hire a foreigner, if you were fortunate enough to find one. If it was to 

teach French, one hired a Frenchman or Frenchwoman, and what one got was a 

native, someone #who knew the language to be taught. His or her authority lay 

-

Schriftes~schri!: <strong>Essays</strong> in Honor of John DeFrancis 

sole1y.k his or her nativity. In the case of rarely taught languages, given the 

shortage of available authentic natives, there was not a great deal of choice. The 

consequences of this all too familiar scenario - for the individuals hired, for the 

students taught, and for the institutions that engaged in these practices - was 

frustration, misunderstanding, failed expectations. The individuals hired were 

treated as "resident foreigners", not as professionals, and they were paid accordingly. 

Subconsciously, institutions asked themselves: how much of an achievement can it 

be to be native? It is not, after all, as difficult as learning to be a surgeon or becoming 

a scientist. Everyone, after all, is native to some culture or other. Why should they 

be paid very much for merely being native? The students taught suffered because, 

even when many of them came to be fond of their often charmingly authentic 

native teachers, most of them somehow failed to internalize the lessons of language 

learning: A great deal of painstaking effort, on the part of student and teacher, was 

wasted. And finally, the institutions that engaged in these practices are . . 

disappointed when they discover they did not get what they were looking for: 

students who could use these languages after their course of instruction. 

The situation is now changing. Although the demand for instructors in 

Chinese, Japanese, and Korean has increased dramatically in the last ten years, that 

demand, in many cases if not all, has been exceeded by the supply of natives 

available to teach these languages. This population is further augmented by the 

sigruhcant numbers of non-native students of these languages who have completed 

their training and who are vying with natives for teaching positions. It is dear that 

merely being native is no longer sufficient - as it was years ago. Even so, 

particularly at the high school level in less populated and less cosmopolitan areas, 

the supply of natives available is not sufficient, and it is difficult for local boards of 

education to attract non-resident, non-native teachers; having spent nearly a decade 

to acquire command over the language, these recent graduates are, understandably, 

reluctant to relocate in these sparsely populated areas, where the opportunity to use 

and to practice their hard-earned language skills, to say nothing of enhancing their 

professional dossier, is minimal. 

It is time, I think, to take stock of the situation. And some of the major issues 

raised by changes in the profession revolve around the question of culture, the 

teaching of culture, and its obverse - a neglected subject that deserves more 

attention: the culture of teaching. 

What is it that is taught when we teach culture? Are we asking the students 

to imitate the teacher? The answer is both yes and no - which is the source of the 

confusion. All language is taught and learned imitatively, because infants have 

neither the intellectual skills nor the memory to learn any other way. So, when we 

teach students a foreign language, we ask them to imitate the way we speak, the way 

we pronounce, the way we gesture. This quite natural dependency on imitation as 

the method of teaching is indispensable, but in teaching culture, imitation can be a 

pitfall: far from being the mode of learning, it can be a hugely inauthentic, 

disconcerting charade. When we teach an American student Chinese, Japanese, or 

Korean, are we asking them to become Chinese, Japanese, or Korean? In the sense 

that we want them to think as if they were Chinese, Japanese, or Korean, our answer

Sino-Plutonic Papers, 27 (August 3 1, 1 99 1 ) 

must be yes: we want them to become like natives of the language they are studying. 

And, quite naturally, native language teachers hold themselves up as models to 

emulate. 

The trouble with this attitude is that some native teachers of a foreign 

language regard themselves as a standard by which to find all their students 

deficient. They acquire what I call the arrogance of the native. Do we really intend 

that American students be mistaken for Chinese, Japanese, or Korean? Do we want rn 

them to give up their American identity in order to become - made over as it were - 

- into Chinese, Japanese, or Korean? The answer is emphatically no, even if it were 

possible. The most successful learners of language do not impersonate a native, 

they are very much themselves; what they have done is to enlarge their identities 

to include another personality; they have "naturalized" a foreign discourse. 

I suspect that much of the resistance to the proficiency movement among 

native teachers of language is that they view it as an assault by non-natives on the 

authenticity of their own culture. The Enghshman in my third anecdote reminds 

us how sensitive we are about our cultural identity. One would not be human if 

one weren't in some sense offended at the prospect of a judgment by a non-native, 

or by some other impersonal measure, as to how "authentic" one's own native 

culture is. But the issue is not to measure whether one is or is not an authentic 

Chinese, Japanese, or Korean: the issue is how effectively an instructor transmits to 

a non-native what he knows about the language which he acquired natively. The 

culture of teaching demands that he put aside his personal cultural identity to 

assess the pedagogical results. Even so, it is difficult for a native teacher to 

appreciate why a non-native might be more successful in transmitting the language A 

and culture to non-native students. That proposition strikes a very tender nerve: 

it's virtually like admitting that "only Englishmen can understand China." This is 

where the confusion between the area of expertise and one's native background is 

mischievous. No one has problems judging between a good or bad teacher of 

physics: it is irrelevant what the native culture of the teacher is, whether Indian or 

English or German. An Indian who is a good teacher of physics could be an 

authentic or inauthentic exponent of Indian culture. The two concerns are totally 

unrelated logically. Good Indians don't necessarily make good physidsts; and good 

physidsts don't necessarily make good Indians. But, when we deal with instructors 

of foreign languages in which he or she is native, there is an inevitable 

psvcholoaical connection. My former colleague, Richard S. Y. Chi, who passed away 

five years ago, was a linguist, a Buddhist scholar, a philosopher, a calligrapher: he 

knew Chinese culture both analytically and intuitively. But, by his own admission, 

he was a poor language teacher. He was secure enough as a Chinese, and as a 

successful scholar, to recognize that fact. He did not confuse his identity as a 

Chinese with his competence as a teacher of Chinese. 

The human and personal response by any native teacher of a language to a 

student who works very hard at learning the language must be encouragement: 

what better encouragement and reward than conferring on that student the highest 

grade? Yet professional responsibility requires that an impersonal judgment be 

made on the student's command of the language. One of the most poignant 

dilemmas for good language teachers is the hard-working but unsuccessful student;

Schn'ftfatsch~iift: &says it1 Horwr of John &Francis 

does one reward the effort? Or does one judge the proficiency? Students somehow 

believe that hard work - whatever the results -- should be rewarded. My own 

response as a teacher is to ate the reality prinaple: not all hard work is recognized, 

nor does all hard work bear fruit. My responsibility is to encourage hard work, but 

unfortunately hard work does not always produce the best results. 

In this regard, I believe that standardized proficiency tests, either the written 

tests or the ACTFL-sponsored oral tests, can free the classroom teacher to use the 

grades more flexibly. I often considered the possibility of suggesting to Institute 

instructors a grade based solely on effort, regardless of progress - or a combination of 

effort and achievement as a basis for a grade. Doubtless, some of you explicitly or 

implicitly employ some such criteria. 

The culture of foreign language teaching involves due consideration of both 

the student's culture and the teacher's; the culture of foreign language teaching 

reminds us that native teachers are not, by the very nature of their nativity, 

inevitably superior as teachers; the culture of foreign' language teaching requires a 

distinction between the native language teacher offering himself as an example of 

the culture and offering himself as the model of his native culture. The culture of 

foreign language teaching regards teaching as an acquired skill, not as a natural 

endowment: one's success as a teacher depends on instinct, skill, and sensitivity, not 

on the place of one's birth. 

The learning of culture does not always depend on the effort invested. I 

how colleagues who know a great deal about Chinese history and Chinese 

literature, who don't have any sense of what Chinese culture means. I can count 

on the fingers of one hand the American scholars of Chinese who truly understand 

the essence of Chinese culture. Clearly, the American teacher of Chinese in my 

second story, the one who called at the office, knew nothing of the essence of 

Chinese culture. The facts of culture - dates, history, names, texts, words - these 

can be enumerated, which is why so many language courses measure levels by the 

number of words "covered" in a semester - whatever that means. But does merely 

memorizing a list of historical facts, and recognizing a number of cultural artifacts, 

constitute functional command of a language or an authentic understanding of its 

traditions? 

As difficult as it is for students to learn these uncommonly taught languages, 

I think we sometimes make it even more difficult. In the East Asian Summer 

Language Institute one year, an instructor accustomed to teaching beginning 

Chinese assumed control of an advanced Chinese course. All summer long, what 

the students learned - to everyone's dismay - was what words and constructions 

covered in First- and Second-year Chinese they should have learned. Over and over 

again, the instructor in the Fourth-year Chinese class would tell her students in 

dismay, ''This construction you should have learned in first- or second-year 

Chinese!" By the end of the summer, those fourth-year students became experts at 

identifying the words and phrases they had failed to learn in beginning Chinese. 

I've often wondered about the perverse burden placed on foreign language students 

- not only do they have to learn the language, but they have to remember which 

words and which constructions were learned in which grades! Even a native 

doesn't have to do that. What that particular fourth-year instructor was teaching

Sino-Phronic Papers, 27 (August 3 1, 199 1) 

was not Chinese culture, but the culture of teaching Chinese. Now, while this may 

be useful knowledge for the teacher, it is totally extraneous knowledge for the 

student. 

In reading the students' evaluation of instruction at EASLI over the years, I 

have been struck - particularly in the early years - by how many students 

complained about the textbooks: it was dear that the time they did not devote to 

learning the language was spent on discovering the deficiencies in the textbooks. 

Indubitably, knowing the faults of each textbook is useful information for the 

teacher, but I cannot see how it can benefit the student. It may be naive, but 

wouldn't the time be better spent learning the language than struggling against. this 

or that textbook or against the teacher who assigned the textbook? Students learning 

a language are presumably not enrolled in a school of education: they want to learn 

the language, not, presumably, how to teach the language. Yet, I daresay that most if 

not all of you have encountered students who are expert in how they should be 

taught languages. Students have wasted more time and concentration critiquing 

the ways of learning the language than in actually learning the language. Ask 

yourself how often you encounter this phenomenon: years after they have taken a 

language course students are voluble about the trials of learning that language; no 

one, however, is eager to use that hard-earned language in functional speech. 

Preposterous and familiar as these occurrences are , they nevertheless point to a 

lacuna in the language-teaching and language-learning process. These expressions 

of frustration, these evidences of failure, reflect a lack of attention on the part of 

instructors to "the culture of teaching." 

What do I mean by "the culture of teaching"? First of all, although there are A 

different traditions and approaches to teaching, I believe that teaching has its own 

culture appropriate to the enterprise - depending on the subject, the setting, the 

environment. The American approach to teaching tends toward the egalitarian; 

the Asian approach to teaching tends toward the authoritarian. The American 

approach to teaching stresses analysis, discrimination, and discursive skills; the 

Asian approach to teaching tends toward rote memory, intuition, and emulation as 

modes of learning. Language teachers must employ both approaches. In the initial 

stages, rote memory, intuition, and emulation are stressed, but the teaching of 

grammar emphasizes analysis, discrimination, and discursive skills. The culture of 

teaching demands that the appropriate approach be used for the subject matter and 

for the student. American students are accustomed to ask why something is before 

they can internalize it; students brought up in Asian educational traditions are 

more emulative and less skeptical. Above all, for the foreign language teacher, the 

culture of teaching distinguishes dearly between one's responsibilities as a teacher 

from one's allegiances as a native. It is no longer enough for a person to be native 

in the foreign language being taught: if it were, there would be over a billion 

quwied teachers of Chinese, at least 120 million qualified teachers of Japanese, and 

more than 45 million qualified teachers of Korean. We must discard the myth that 

the native is the ultimate authority where foreign language teaching is concerned. 

The native & an authentic authority on the culture; he or she is not necessarily an 

authority on the teaching of that culture. Two years ago, the Hokkaido 

International Foundation recognized this distinction when aey sent 17 native 

r

Schriffestschrifi: <strong>Essays</strong> in Honor of John DeFrancis 

Japanese to the United States to learn how to teach Japanese to non-Japanese. 

(Okutsu sensei, I am proud to say, was among this pioneering group.) 

We are now fortunate enough to have good native and non-native teachers : 

that affords us the oppowty of clearly identifying those who are less than 

adequate, native or non-native. The good teacher recognizes not only that he is 

teaching culture, he is also imbued by the culture of teaching. What is remarkable is 

how many native teachers of foreign languages, without formal training, have 

become good teachers. But if there are many good teachers of foreign language who 

are native, it is certainly not true that &l natives of a foreign culture are 

automatically adequate teachers of that language to non-natives. This brings us to 

some of the difficulties and confusions that trouble the profession of foreign 

language teaching: to say that a Frenchman is not a capable teacher of French is not 

to demean his authenticity as a Frenchman. We have no trouble accepting the fact 

that not every American is qualified to teach English; why should it be difficult to 

see that some foreigners are not qualified to teach their native language to nonnatives? 

The awkward thing is that there are some holdovers from previous 

generations who were "dragooned" by history into the profession, for (1) lack of 

something better to do; and (2) for lack of more professionally qualified teachers. 

These individuals are not to blame for the historical circumstance that led to their 

careers as language teachers, nor should they be penalized for the inherent 

misunderstandings among educators and institutions about the true character of 

. language teaching. Furthermore, these very individuals have been victimized by 

the system: they have not been allowed to advance in the profession, their status has 

remained static for decades, and their remuneration has lagged behind their 

colleagues year after year. In a sense, they have been trapped by history. But, if we 

are sympathetic to their plight, we must also ask the level-headed question: what 

would they have been if they had remained in their native culture? The 

outstanding individuals would, of course, have distinguished himself whatever 

the circumstances. Others, however, if truth be known, would hardly have attainedthe 

exalted status of teacher if they had remained in their own countries. Whatever 

compassion one might have for these individuals, one has to recognize that, if the 

profession of teaching foreign languages is to be taken seriously, personal concerns 

cannot enter into evaluative judgments on instructional skills. We no longer have 

the luxury of permissive students who would tolerate deviations from professional 

standards as part of the eccentric charm of instruction in "difficult" languages. 

Students today will no longer be as forgiving as they have been in the past. 

They will not be content merely to have had a pleasant time in a language course, 

and to receive a courtesy A; they will judge the instruction by how much they take 

away from the class and how much they actually are able to use. Yet, there are still 

vestige of previous pedagogical malpractice. I was told of an incident at Middlebury 

some years ago, in which a student of Chinese (she had studied previously at Yale) 

could not keep up with the instruction at the fourth-year level at Middlebury. Far 

from being abashed at her own poor preparation, she vilified the instructor of the 

class, a native-born Chinese, with abusive and obnoxious reminders of the A's that 

she had received from her instructors at Yale. Clearly, this student not only hadn't


learned mu& Chinese, she didn't learn very much about Chinese culture either, 

nor about the traditional respect owed someone older who is also a teacher. 

A similar story occurred in the first year of the East Asian Summer Language 

Institute: a student of Japanese came to tell me, after his first grades were in, that he 

didn't come to Indiana to get a B-. My response was simple: I said, "Oh, I'm so 

sorry, I thought your purpose in coming here was to learn Japanese!" Too many 

students forget what their objective is, which is to learn the language: the grade is 

merely a means that the teacher employs to help the student realize that objective. 

Whenever I encounter grade-mongering students, I am tempted to offer them a tee 

shirt with their transcript printed on it. I wonder how many of these I would sell. 

What does it mean, after all, to get A's in the study of a foreign language and not be 

able to function in that language? Wouldn't we be embarrassed if students who 

received A's in physics or mathematics could not "do" physics or mathematics? 

The tendency to give out A's generously is particularly common among 

native teachers in the older generation, particularly teachers of so-called 

"uncommon" languages like Chinese, Japanese, or Korean. Before we pass 

judgement on this group as academically irresponsible, we must reflect on the 

motivation behind such seeming generosity. There are two reasons behind this 

grade inflation: one practical, one psychological. In the days when the enrollments 

in courses on these rarely taught languages were small, the prosped of an easy A 

was one inducement to counteract the off-putting image of these forbidding 

languages. In a real sense, a liberal sprinkling of A's enhanced the prospects both of 

the same student continmg his studies, and of attracting other students to begin 

study of the language. The psychological reason is equally poignant and 

understandable. For an immigrant to see American students wrestling to acquire 

her native language must be a reassuring experience, especially in the case of such 

visible immigrants like Chinese, Japanese, or Koreans, who are often the victims of 

bigotry and prejudice, who are mocked and derided, and whose native language is 

often satirized in vulgar street-corner imitations. What a balm it must be, how 

consoling, for these immigrants to welcome American students into their classroom 

who, far from insulting them for the way they speak, revere them for their native 

ability in the language? Wouldn't it be difficult for such immigrants to repay the 

interest of such students with any grade less than an A? And what if the 

performance is substandard? The A can still be justified on the basis of the gesture 

made: the student had, after all, the good taste to choose the right language. And 

what of the prospect that the student might embarrass the teacher when he visits 

the country whose language he has been studying assiduously? No matter, the 

natives wouldn't expect an American student to speak their language at all, so any 

even minimal command will be impressive. 

There is another vestige of earlier generations which we must address, and 

that involves the teaching of language and culture as intensely soda1 activities. The 

days when solitary eccentrics picked up a "grammar" and mastered a language - as 

Arthur Waley claims to have done with Japanese in six weeks - are over. The mute - 

language learners - what a Princeton colleague of mine calk "language cripples" - 

will not survive the end of this century. There is a whole generation of s&olars for 

whom Chinese in its noblest form takes the guise of sinology, where an 

I 

-

Schriftfesrschri#: Ersap in Honor of John &Francis 

incomprehensible original text is replaced by an even more incomprehensible and 

unreadable translation. The heirs of this tradition maintain -- as one student of 

Japanese at the Institute once insisted - that no one (including natives) understood 

a sentence unless he had noted comprehensively the etymological history of each 

word in that sentence. I have decided, from anecdotal evidence, that when they 

write the sociology of sinology and of japanology they will discover that the least 

well-adjusted individuals of previous generations gravitated toward the study of 

rare languages. This is symbolized by an unforgettable experience a generation ago 

in a Harvard College Chinese class: the only non-Oriental in the class never bathed, 

never combed his hair, and attended class barefoot. 

Over the years, despite the explicit warnings in the brochure and the 

application forms about the importance of interactive learning, some of these 

students have enrolled at the East Asian Summer Language Institute. I reflect now 

with bemusement on a skit that was performed by the Japanese School a few years 

ago, involving an alien, from Mars, making cutting remarks about Americans and 

about Japanese. Is it surprising that the student who assumed the role of the alien 

was a solitary, anti-social, enthusiast of etymology? I am happy to report that 

students of Chinese, Japanese, and Korean are getting more and more wholesome 

year by year. (At this summer's reception, my wife remarked with some optimism 

how normal the students looked: only one person appeared eccentric. "Who was 

that?" I said: "Describe him." After being told it was someone with a scraggly beard, 

a swarthy complexion, and long hair gathered at the back, I realized she was talking, 

not about a student, but about Kathy's husband.) 

If we are to recognize the teaching of languages as a humanistic profession, 

we must take care to differentiate between our identities as natives and our 

identities as teachers; we must not confuse who we are with what we &. We are 

ethnically Chinese, American, Japanese, Korean, or whatever, but what we do is to 

teach these cultures.* There will be times when loyalty to what we are and what we 

do may be in conflict. But they needn't be if we remember that no one is @ to be 

native, just as no one is paid to breathe: that comes naturally, and is worthy of no 

special notice or reward. (Although frankly I have encountered too many people 

who are cultural nativist snobs: they can't get over the misfortune of others not 

being born into the culture which gave them birth.) What a teacher is paid to do is 

to understand and to respect the culture of teaching. The culture of language 

teaching is especially complex, for the good language teacher is both caring and 

critical, personable but not personal, social and interactive without being frivolous 

and flippant. The native foreign language teacher realizes that the culture he 

teaches has a particular personal significance for him, bred of familiarity, and 

internalized through habit: but while he recognizes the fact that this native 

endowment confers many advantages, it does not constitute supreme authority. 

The non-native teacher is compromised by an inevitable estrangement from the 

* Even here, the situation is more complex than we realize. For some of us do not belong 

ethnically to any one culture, but rather to a combination of cultures: Chinese-American, 

Korean-American, Japanese-American.

Sino-Plutonic Papers, 27 (August 3 1, 1991) 

culture he is teaching, but he has the advantage of seeing the difficulty of acquiring 

another language non-natively. Merely knowing what to teach is merely the 

beginning of the culture of teaching. The native teacher of a foreign language must 

never forget, tempted as she might be on occasion to forget it, that her job is not to t 

create replicas of herself, for an authentic student of another language does not 

pretend to be what he isn't; the non-native teacher must also realize that his job is 

not to produce a pseudoslite of exotic language speakers, which has nothing to do 

with education, and everything to do with pseudo-culture, like I-ching cultists and 

Zen aficionadoes and Mandarin-mongering collectors of orientalia. (I like to 

remind students who are smug about their command of Chinese that there are 

more than a billion Chinese who speak the language more fluently. Perhaps you 

encounter the same arrogance among successful learners of Japanese and Korean.) 

Culture, after all, is neither content - as in history; nor skill - as in mathematics, 

but a dynamic combination of both. Teaching culture is much more complex that 

teaching either history or mathematics, because it disconcerts our very sense of 

selfhood. It confuses what we know with what we are, which is why, in teaching as 

well as learning another language and culture, there must be pride, not arrogance; 

while one cannot afford an inferiority complex, still there must be a sense of 

humility. 

I started off with a verbal mirror image, a clausal palindrome: the teaching of 

culture and the culture of teaching. That reflective doubling seems now symbolic of 

the paradoxes of language instruction. A good language teacher requires a deep 

personal commitment, yet that teacher must not confuse the personal with the 

professional. A good language teacher must spur his student to his best efforts, yet 

he must not hesitate to indicate the same student's deficienaes. A good language 

teacher cannot be judgmental about her students, because each student has the same 

right to quality instruction, yet at the same time she must be utterly fair and 

professional in assessing the progress of each student. 

What I have tried to suggest is that, when we consider the problems and 

challenges of teaching culture, we might give some thought to the demands and the 

complexities of the culture of teaching. 

Eugene Eoyang 

East Asian Summer Language Institute 

Symposium on the Teaching of Culture 

July 14,1990 

Indiana University

Preface 

S~hnffestscch @: Ersays it; Homr of John DeFruncis 

THE CULTURE COMPONENT OF LANGUAGE TEACHING 

Kyoko Hijirida University of Hawaii 

Things Japanese in Hawaii was published in 1973 by the University Press of Hawaii. By 

writing the book, Dr. DeFrancis has con tibuted enormously to the resource materials that Japanese 

language teachers use in their language classrooms. One of the most frequently used materials is 

the article on "Girls' Day and Boys' Day" (DeFrancis, 1973:27-30). It explains everything you 

may want to know about the occasions such as explanations of the mochi shapes and their colors- 

hishi-mochi and kashiwa-mochi. I haveread it every year around the time of the annual events, 

renewing the information for my students in the class. I have cherished Things of Japanese in 

Havaii not only for its usefulness but also for its reminder of my professional development and 

the encouragement provided by Professor DeFrancis. 

It was the summer of 1969 when I came to the Department of East Asian Languages at the 

University of Hawaii as an EPDA Education Professions Development Act) fellow. Dr. 

DeFrancis, as a director of the program, guided the graduate student fellows who majored in 

Teaching Chinese and Japanese in their academic life at the campus. Besides learning language 

teaching skills, the fellows acquired the skill for surveying and reporting culture-related activities in 

the community through participation. We realized that languages and cultures are always preserved 

together among the various ethnic groups in Hawaii and they provide an added attraction for the 

community. Language teaching with culture, one of my continuing themes, has thus stemmed 

from the happy graduate student days spent with kind-hearted professors like Dr. DeFrancis. All 

the highly motivated fellows went to the language teaching field in various locations after receiving 

the graduate training diploma. I was one of the fellows who remained at the department to begin 

language teaching. The 1969 University of Hawaii EPDA program was the beginning of a longterm 

association with Dr. DeFrancis, and the starting point of my professional development with a 

deep awareness of the importance of incoroporating culture in the language class. Twenty years 

after the EPDA training, I am cu~~ently involved in the foreign language teacher education program 

at the College of Education at the university along with my regular language teaching. I have still 

pursued the theme of cultures as one of components of the effective language cllniculum 

With all those years of teaching behind me, I would like to present some of my own thoughts 

from the teacher's perspective about the cultural component of language teaching. 

Curriculum Designs 

Thinking of 'why to teach' (goals and objectives), 'what to teach' 

(teaching materials and content), 'how to teach' (teaching methods), and means of evaluation are 

important components of curricula. The teacher's plans, always with objectives to fuKill student 

needs including the why, the what, and the how and evaluation are essential to effective teaching. 

In planning the teacher must also consider the cognitive, psychomotor, and affective domains for 

students to develop, and how to effectively cover all the three areas in teaching-learning processes 

in the lesson. In foreign language study, culture complements all the three domains. Issues on 

teaching culture in foreign language classes in the area of instructional objectives, content, how to 

teach and evaluate should be discussed by penetrating the threedomain-perspective in education.

Goals and Objectives for Cultural Instruction 

Sirzo-Platonic Papers, 27 (August 3 1, 199 1) 

Leading advocates of teaching culture with foreign languages have identified goals and 

objectives for the culture components (Frances and Howard Nostrand 1970; Howard Nostrand 

1978; H. Ned Seelye 1988; Omaggio 1986). H. Ned Seelye's seven goals are well known and have 

been the basis on which teachers may modify them to best suit their own classrooms. Seelye 

stated the students should be able to demonstrate that they have acquired certain understandings, 

abilities and attitudes in: 

1. The Sense, or Functionality, of Culturally Conditioned Behavior. The student should 

demonstrate an understanding that people act the way they do because they are using the options 

the society allows for satisfying their basic physical and psychological needs. 

2. Interaction of Language and Social Variables. 

The student should demonstrate an understanding that social variables such as age, sex, social 

class, and place of residence affect the way people speak and behave. 

3. Conventional Behavior in Common Situations. 

The student should indicate an understanding of the role convention plays in shaping behavior by 

demonstrating how people act in common mundane and crisis situations in the target culture. 

4. Cultural Connotations of Words and Phrases. 

The student should indicate an awareness that culturally conditioned images are associated with 

even the most common target words and phrases. 

5. Evaluating Statements about a Society. 

The student should demonstrate the ability to evaluate the relative strength of a generality 

concerning the target culture in terms of the amount of evidence substantiating the statement 

6. Researching Another Culture 

The student should show that he or she has developed the skills needed to locate and organize 

information about the target culture b rn the library, the mass media, people, and personal 

observation. 

7. Attitudes toward Other Cultures. 

The student should demonstrate intellectual curiosity about the target culture and empathy toward 

its people. (Seelye 1988:49-58) 

Based on these seven goals, a group of U.S Japanese language teachers who enrolled in the 

course EDCI 64lD, Seminar in Teaching Japanese for M a students, developed their seven 

objectives of teaching culture in the Japanese language class at the high school level. Then we 

asked for a group of teachers in Japan to rank the objectives in order of importance in teaching 

culture with Japanese language. (We plan to later compare the results of the surveys in Japan and 

the US to discover differences and similarities in their ratings.) The following are the results of the 

survey (rank in order of importance). 

The goals and objectives of teaching culture in the Japanese language class are for the student to 

develop:

Schnfifestschrift: <strong>Essays</strong> in Honor of John DeFrancis 

1) proficiency in communication in a culrurally appropriate manner achieved by understanding the 

similarities and differences of both American and Japanese cultures; 

2) an interest and empathetic understanding of value judgment, way of thinking, and lifestyle of 

the Japanese; 

3) an enjoyment and enthusiasm for learning the Japanese language; 

4) an understanding that social variables (such as age, sex, status, etc.) affect language use and 

behavior, 

5) the ability to understand behavioral characteristics of the Japanese in conventional situations; 

6) the ability to evaluate cultural generalities concerning Japanese culture; 

7) an understanding and appreciation of the arts and cultural traditions of the people of Japan (such 

as kabuki, ikebana, calligrphy, etc.). 

The survey results indicate that the teachers in Japan perceive cultural instruction as a 

contribution to develop the affective doman as well as knowledge and skill development. They 

also indicate that small c culture contexts (i.e. cultural behavior, customs, perceptions of reality 

shared by a cultual community) are more popular than big C culture (achievement culture, i.e. 

literature, art, music, etc.) as instructional content of culture. This reminds me of Eleanor Jorden's 

emphasis on so called 'acquired culture' rather than learned culture' in selection of cultural content 

for language teaching. ( In her workshops conducted in March 1991 in Hawaii.) She pointed out 

that the use of language has more impact with the culture unconsciously acquired through being a 

native rather than that learned consciously. 

Cultural Topics 

In order to attain these instructional goals, culture topics should be carefully selected in 

accordance with the language lesson in progress as well as student needs and interests. Favorite 

topics for high school students chosen by the Japanese teacher group in the same survey are 

following areas (order of importance): 

TOPICS EXAMPLES 

1. Daily Lifestyle: Japanese house, shopping, clothes, transportation 

2. School Life: school system, teacher-student relationships, juku 

3. Eating and Drinking: restaurant, food, table manners, kissaten 

4. Language and Culture: keigo/in-group-outgroup, connotation, 

5. Family Life: role expectation: father, mother, children 

6. Life Customs: wedding, funeral, ettiguette, exchanging gfts 

7. Attitude & Values: fomality/rituals, considerationfor others


8. Festivals/Annual Events: New Year's Day, flower- viewing, 

9. Traditional Culture: Kabuki, Tea ceremony, flower arrangement 

10. Business Life: company system and structure, decision making 

11. Amusement: karaoke, pachinko, go, hanahuda, mahjong 

In this selection of the cultural contents, factors such as relevancy to student needs, self to 

family to community, direction of the cumculum development, and the language lesson context are 

reflected. For example, at the high school level, comparison of the school life in both their own 

and the target cultures might be more appealing to the students than those at the college level, 

where they may place a greater interest in honorific expressions or in-group/out-group 

communication modes. 

Instructional materials to teach about these topics are not readily available although some have 

been developed and published in the form of videos, films, books, or pictures. Teachers have 

been making an effort to collect pictures, realia, articles, newspaper ads, songs, etc. 

Therefore, lecture with videos, show and tell, or reading article and discussions are frequently 

used instructional techniques. 

Teaching techniques 

Teachers can make a world of difference in helping students increase empathy for greater 

cultural understanding. Equipped with a rich collection of culture teaching strategies and 

techniques to employ in their classroom, teachers can help to raise the students level of language 

consciousness and proficiency, and internalize language learning through culture. Teachers must 

focus on both appropriate content and learning activities that enable students to assimilate that 

content Activities should encourage them to go beyond facts, so that they begin to perceive and 

experience vicariously the deeper levels of the culture of the speakers of the language (Rivers 1981: 

324). 

The history of Japanese language in a regular school curriculum in the United States is relatively 

short as compared to the longer tradition of European language instruction. Hence, many 

techniques and strategies have been developed for European language culture study. By adapting - 

more of the available techniques, teachers can widen their selections of teaching methods and help 

students internalize the process of language learning through cultural interaction. The following is 

a list of 21 general techniques, jointly collected from various sources in my seminar class, for 

teaching culture. I include brief explanations of their usages. 

TECHNIQUE EXPLANATION 

1. Lecture Teacher presentation of material. 

2. Show and tell Items to share & explain to class. 

3. Demonstration Actual participation in the activity. 

4. Field trips First hand enrichment experience. 

5. Bulletin board Current events, or special occasion displays.

S~hnffe~tsch rifr: <strong>Essays</strong> in Honor of JO~N DeFrancis 

6. Songs and dance Appealing to the young; breathes cultural life. 

7. Role-Play Authentically dramatize cultmil behaviors. 

8. Native informants Visitors who can interact with students. 

9. Cultural assimilators Narrative with multiple choice answers 

illustrating a point of miscommunication. 

1O.Cultural minidramas 3 to 5 brief episodes of a cultural assimilator 

performed orally. 

ll.Culture capsules One minimal difference in culture custom 

accompanied by realia. 

12.Culture clusters About 3 culture capsules integrated into a skit. 

13.Taped interviews Taped speeches of native speakers. 

14.Video tapes 

15.Audiemotor unit 

Provides natural, authentic linguistic exchanges, 

gestures, social distance, or eye contact on film. 

Commands which leicit a physical response from 

the students. 

16.Identifying cultu- Sensitizes students to contrasts and commonalities 

rally conditioned in conventional behavior in their own and the 

behavior target culture. 

17 .Deriving cultural Helps students to associate culturally representa- 

como tations tive images of words. 

18Decreasing stere* Helps students understand the dangers of 

typic perceptions unwarranted generalizations. 

19.Artifact study Discern cultural sigmficance of unfamiliar objects. 

20.Building empathy Learn to explain the behavior and build empathy 

for a culture through knowledge. 

21Authentic reading . Authentic material used in bringing up culture 

materials poings with pre- and post-activities to evoke 

culture awareness 

The common trend to date has been for Japanese language teachers to present culture in a lecture 

format with talks about festivals, and teacher's personal experiences; or show and tell; going on 

field trips; learning songs; studying bulletin board displays; or by showing slides and videos to 

share the Japanese culture. However, there have been some efforts to develop instructional 

materials by adapting some techniques listed above. Pioneering in the application of culture 

capsules in Japanese with simple audio tape recordings are Kazuyoshi Noguchi and Roger A. Van 

Damme 0985); and in cultural assimilators are Hiroko C. Kataoka with Tetsuya Kusumoto 0991).

Sim-Platonic Papers, 27 (August 3 1, 199 1) 

A rich collection of culture teaching techniques can enhance the teacher-studenvstudent-student 

interactive process and help students internalize their language learning. Internalization of language 

and culture is the key to further language proficiency, of which communication is its primary goal. 

I firmly believe that incorporating cultural elements will internalize language learning for better 

retention and increase effective communication, as well as provide motivation for language 

learning. 

In this paper I have presented an overview of my idea and thoughts of the culture component of 

language teaching. As a foreign language teaching professional, I will continue to develop the 

cultural part of the curriculum, in the context of foreign language instruction. It is my dream that I 

will be able to follow the example that Dr. John DeFrancis has shown us, a long life filled with 

work of set purpose, joy, and happiness. 

..

Schri_fifestschri$: Essay in Honor of John &Francis 

BIBLIOGRAPHY 

Crawford-Lange, L., and D. Lange. ''Doing the Unthinkable in the Second Language 

Classroom." In T. Higgs, ed., Teaching For Proficiency, the Organizing Principle. ACTFL 

Foreign Language Education Series, vol. 15. Lincolnwood, Ill.: National Textbook, 1984. 

DeFrancis, John. Things Japanese in Hawaii. Honolulu, Hawaii: The University Press of 

Hawaii, 1973. 

Galloway, Vicki B. "A Design for the Improvement of the Teaching of Culture in Foreign 

Language Classrooms. " ACTFL project proposal, 1985b. 

Kataoka, Hiroko C. with Kusumoto, Tetsuya Japanese Culmral Encounters & How To Handle 

Them. Lincolnwood, Ill.: Passport Books, a division of NTC Publishing Group, 1991. 

Noguchi, Kazuyoshi & Roger A. Van Damme. "Culture Capsules in Japanese (Tapes)." In K. 

Hijirida, etal. ACTFL Selected Listing of Znsbwtional Materials for Elementary and Secondary 

School Program: Japanese. Yonkers, NY: ACFL 1989. 

Nostrand, Howard L. "Empathy for a Second Culture: Motivations and Techniques." In G. A. 

Jarvis, ed., Responding to Nav Realities. ACJrn'Foreign Language Education Series, vol. 5. 

Lincolnwood, 111: National Textbook, 1974. 

Omaggio, Alice C. Teaching Language In Context: Proficiency-Oriented Instruction. Boston, 

Mass.: Heinle & Heinle Publishers, Inc., 1986. 

Rivers, Wilga M. Teaching Foreign-Lnnguage SMs. Chicago and London: The University of 

Chicago Press, 1981. 

Seelye, Ned H. Teaching Culture: Strategies for Intercultural Communication. Lincolnwood, Ill.: 

National Textbook, 1974,1984,1988.

Schnffestschri!: f isq in Honor of John &Francis 

Thinking About Prof-John DeFrancis 

In the early nineteen eighties, I was surprised to 

learn that Professor John DeFrancis was an ardent 

supporter of Chinese language reform, in particular, the 

efforts in Chinese Romanization. Introduced to him by our 

mutual friend Prof. Zhou Youguang, a distinguished Chinese 

expert in the field of Chinese language reform, I visited 

Prof. DeFrancis during one of my transit stays in 

Honolulu. I called Prof. DeFrancis up and was invited to 

stay overnight in his hillside residence near the Manoa 

campus of the University of Hawaii. It was a modest but 

beautiful Japanese-styled two story single family house 

whose slope-facing sides were flanked by a garden, the 

centerpiece of which was a little Japanese rock garden 

with a small pond and perennial flowering plants of pink 

and red colors beside it. They appeared strikingly 

beautiful when contrasted with the prevailing greenness of 

the garden. Everytime I sat gazing out from the computer 

desk in front of a large latticed window facing the rock 

garden, I was mesmerized by the exquisite view and thought 

how lucky it would be to be able to live in such a 

pleasant environment in this age of environmental 

degradation. Later, I learned that Prof. DeFrancis 

himself was the mastermind behind much of the garden 

design and had been spending many hours every week to 

maintain and enhance its beauty; To substantiate my fond 

impressions of the beautiful house and garden, I am 

including herewith a few photos taken recently during my 

last visit on January 19 to 22, 1991. One photo shows the 

house with Professor's car in front of it. Another photo

Sino-Platonic Papers, 27 (Aup t 3 1, 1 99 1 ) 

is the view of the garden from the computer work area. 

Also included is a photo of Prof. DeFrancis sitting before 

a window. 

In my discussions with Professor DeFrancis on the 

problems of Chinese language reform, I was very much 

impressed by his conviction and his keen academic insight 

concerning the desirability of a pinyin script for 

Chinese. He also emphasized that the tone problem has 

already been satisfactorily solved by using diacritical 

marks. He frequently points out that many Chinese tend to 

spend lots of time discussing and debating about the 

merits and demerits of a pinyin script, and the problems 

of homophones or homographs in the Chinese language. He 

believes that it is essential and more fruitful to find- 

out the real problems through practice rather than 

indulging in theoretical discussions. The artificial 

difficulty of " shi shi shi ..." humorously created by 

Professor Zhao Yuanren is rather irrelevant in actual 

language usage. Professor DeFrancis firmly believes that 

whatever problems may have been created for the Chinese 

phonetic script as a result of the long use of the Hanzi 

character system can eventually be overcome. What is 

needed is a will for creating such a script among the 

Chinese people, particularly among their leaders. He is 

disappointed to note that such a political will has been 

very weak, if not virtually absent in China. 

Professor DeFrancis' panoramic knowledge in 

linguistics enabled him to prepare reputable textbooks on 

Chinese for college students. He has also written books

Schn@fiestscchrijl: <strong>Essays</strong> in Howr of John DeFrancis 

about languages and nationnalism, about the nature of the 

Chinese language and the history of the alphabetic writing 

systms. His writings extend beyond academic fields, as he 

recently told me that he had just finished a book about 

his travel adventures in China during the thirties. In 

the book entitled The Chinese Language--Fact and Fantasy 

(published by the U. of Hawaii Press), he eloquently 

refuted the widely held myth about Chinese being an 

ideographic writing system, which could somehow 

communicate meaning directly without resorting to sound, 

thus allegedly making it a suitable candidate for a 

universal script encompassing diverse languages. 

Prof. DeFrancis classifies Chinese as an 

inefficient morphosyllabic script, rooted in the Chinese 

linguistic system. Hanzi not only are an unsuitable choice 

for a universal script, their monosyllabic form does a 

great disservice to the increasingly rich and essentially 

polysyllabic modern Chinese language. Instead of using a 

simple alphabet to build up the required 418 syllables of 

Putonghua, written Chinese requires many thousands of 

characters whose phonetic information has deteriorated 

through more than two thousand years of language change. 

DeFrancis was very much disturbed by the mistaken claim of 

Dr. Logan in his book entitled The Alphabet Effect that 

Chinese is a monosyllabic language. In a recently 

published book entitled Visible Speech, DeFrancis 

corrected Dr. Logan's erroneous assertion, refuted the 

notion of a picture-based ideographic script, and 

discussed in depth the classification of the writing 

systems of the world. Again the emphasis was on the

Sim-Platonic Papers, 27 (August 3 1,199 1) 

misclassification of the Chinese script as an ideographic 

script famous west ern linguists and scholars. Although 

Chinese uses the sane morphemic radical concept 

employed by the Sumerians in their cuneiform script which 

fell into disuse more than three thousand years ago, Prof. 

DeFrancis believes that the two writing systems were 

invented independently. Unlike the Sumerian writing 

system which influenced the developments of other scripts 

the Middle the invent ion the 

Semitic alphabetic system, the Chinese writing system 

remained basically unchanged for more than two thousand 

years. Today, Chinese remains the only completely 

non-alphabetic writing system in the world. 

All these academic debates seem to have a 

practical bearing on the general attitude towards the 

value of the Chinese writing system and the necessity and 

feasibility of its alphabetization. Personal 

communications between Prof. DeFrancis and myself have 

made me aware of his concept that all human languages are 

rooted in the ancient past, but some have evolved while 

others became fossilized, and those which have 

alphabetized are much more efficient. Just as Vietnamese, 

Korean and Japanese have recently (in the historical time 

scale) been alphabetized, Chinese need not be an 

exception. Professor DeFrancis is firmly convinced that it 

is possible to write Chinese alphabetically without Hmzi. 

I also would like to point out the fact that, in view of 

the increasing stability of the spoken languages as a 

result of advances in modern audio communication and 

recording technology, a newly created phonetic script will

enjoy a much better link with the spoken language after 

long lapses of time than was heretofore possible. After 

all, although we cannot hear the voices of our ancestors, 

people a thousand years from now will be able to hear our 

speeches. Judging from the decreasing role played by 

Hanzi in Korean and Japanese, and the fact that Hanzi are 

mainly sound based, albeit ineffectively, the claim of 

Hanzi as a candidate for a universal ideographic script 

is an unfounded myth or fantasy. 

Prof. DeFrancis is a truly remarkable scholar who 

not only has mastered and taught the intricate Chinese 

language for more than 30 years but who has also been able 

to see through the limitations of the Chinese language and 

has been advocating reform for it, so that it can serve 

the Chinese people better. His humanitarian perception of 

the world transcends national interests. He most likely 

believes that what will benefit one fifth of mankind will 

benefit mankind as a whole, and is certainly worth his 

selfless devotion. I sincerely hope that more and more 

Chinese scholars wiil be inspired by his example and 

contribute more and more to the vital mission of 

Romanizing the Chinese language. 

Apollo Wu 

Translator, United Nations 

May 19, 1991

Schri$$estschrifr: Bsap in Homr of John DeFrancis

Sino-Platonic Papers, 27 (August 3 1, 199 1)

SchnfifestschriJE: &says in Honor of John &Francis

Sino-Platonic Papers, 27 (August 3 1, 199 1)

Schn$jfies~schn!j?: &say in Honor of John &Francis

Sch?iftfestschrij?: fisays in Honor of John &Francis

Sino-Platonic Papers, 27 (August 3 1, 1991)

A translation of the quadrisyllabic poem: 

s~h~ilfft~~h~!: 

Ersq in Honor of John &Francis 

On Professor DeFrancis' Eightieth Birthday 

Linguistic authority Young people's teacher 

A cultural bridge Everyone admires 

The (Chinese) year Xinwei Presented by Yang Fu-sen 

A translation of the vernacular poem: 

On Professor DeFrancis' Eightieth Birthday 

You are not only an authority on linguistics, 

You are also young people's teacher; 

For promoting mutual understanding between China and America, 

You have become a cultural bridge. 

199 1 Presented by Yang Fu-sen @chard Fu-sen Yang)

Schriftfestschriji: <strong>Essays</strong> in Honor of JOhn DeFrancis 

Announcement 

B One of Professor DeFrancis' most significant contributions has been his effort to 

make more people aware of the true nature of the Chinese writing system, viz, that it is 

w 

phonetically based and not some arcane, magical set of symbols conveying meaning without 

reference to language (however that might be possible!). In recent years, his two books, 

7%e Chinese Language: Fact and Fantasy and Visible Speech: The Diverse Oneness of 

Writing Systems, have served to clarify Chinese writing in lively prose easily accessible to 

the layman. 

Professor DeFrancis' analysis set me to pondering how students of Mandarin 

Chinese might be systematically helped along the path to familiarity with the various 

phonetic elements that occur fairly frequently in the Chinese graphicon. Eventually I and a 

Chinese colleague developed a computer application, written in Hypercard 2.0 to run on the 

Macintosh, that begins to address this pedagogical problem. Users of the program are 

invited to browse through phonetic families of characters, investigating similarities and 

differences in their pronunciation. 

The program also allows the user access to example words that include the character 

L in question, and the characters' individual Pinyin spellings and definitions are also available 

at the click of the mouse. Characters are not only accessible via their phonetic family, but 

8 also individually though English and Pinyin indexes. The application includes as well an 

"edit" mode that allows the end user to add individual characters or even whole phonetic 

families to the application. Eplon'ng Chinese Phonetics will be available at minimal cost in 

November, 199 1. Anyone interested in acquiring a copy may write to me. 

I and my co-developer are greatly indebted to Professor DeFrancis for the inspiration 

provided by his two books, and for the loan of various materials on the topic of Chinese 

phonetics. 

Stephen Fleming 

Center for Chinese Studies, University of Hawaii at Manoa 

Moore Hall 4 16,1890 East West Road 

Honolulu, HI 96822

Number 

1 

2 

3 

4 

5 

6 

7 

8 

9 

10 

1 1 

12 

13 

Date 

8611 1 

86/12 

87/3 

87/11 

87/12 

88/1 

8811 

8812 

88/12 

89/6 

89/7 

8918 

89/10 

Author 



Andrew Jones 

Hiroshima 



Robert M. Sanders 


Eric A. Havelock 

Vassar College 



Tsung-tung Chang 

Goethe-Universitit 

FmdcfiKt 

Various Authors 

Soho Machida 

Daitoku-ji, Kyoto 

Pratoom Angurarohita 

Chulalongkom University 

Bangkok 

Edward L. Shaughnessy 

University of Chicago 



Jiaosheng Wang 

Shanghai 

Previous Issues 

-Title 

The Need for an Alphabetically 

Anzinged General Usage Dictionary 

of Mandarin Chinese: A Review 

Article of Some Recent Dictionaries 

and Current Lexicographical Projects 

The Poetics of Uncertainty in Early 

Chinese Literature 

A Partial Biblio_mphy for the Study 

of Indian Influence on Chinese 

Popular Literature 

The Four Languages of "Mandarinw 

Chrnese Characters and the 

Greek Alphabet 

Computers and Japanese Literacy : 

Nihonzin no Yormkaki N6yoku 

to Konpyfita 

Indo-European Vocabulary in Old 

Chinese: A New Thesis on the 

Emergence of Chinese Language and 

Civilization in the Late Neolithic Age 

Reviews (I) 

Life and Light, the Infinite: 

A Historical and Philological 

Analysis of the Amida Cult 

Buddhist Influence on the 

Neo-Confucian Concept of 

the Sage 

Western Cultural Innovations 

in China, 1200 B.C. 

The Contributions of T'ang and 

Five Dynasties Transformation 

Texts (pien-wen) to Later Chrnese 

Popular Literature 

The Complete Ci-poems of 

Li Qingzhao: A New English 

Translation 

Pages 

31 

45 

iv, 2 14 

14 

4 

13 

i, 56 

ii, 39 

46 

31 

8 

71 

xii, 122 

Price 

$2.50 

$3.00 

$20.00 

$2.00 

$2.00 

$3.00 

$10.00 

$7.50 

$7.50 

$6.00 

$2.00 

$12.00 

$18.00

Number 

14 

15 

16 

17 

18 

19 

20 

21 

22 

Da.te 

89/12 

9011 

9013 

9014 

90/5 

9016 

901 I( 

9011 

91/3 

Author 

Various Authors 

George Carctona 




Heather Peters 

The University Museum 

Philadelphia 



Bosat Man 

Ndanda 



Philippa Jane Benson 

Carneg ie Me1 lon University 

David Moser 

University of Michigan 

Previous Issues 

Title 

Reviews (TI) 

On Attihides towards Language in 

Ancient Intlia I 

Three Brief <strong>Essays</strong> Conce.ming 

Chinese Tocharistan 

a. The Significance of Dunhuang and 

Turl'an Studies 

b. Early Iranian lrlllue~~ces on Buddhisln 

in Central Asia 

c. History of Chinese Turkistan in the 

Pre-Islamic Period 

Tattooed Faces md Stilt Houses: 

Who Were the Ancient Yue? 

Two Non-Tetragraphc Northern 

S ini t ic Languages 

a. lll~plicatioi~ls of tllr Soviet Dungao 

Script for Cl~incsc Languagc Rcfonn 

13. Who Werc thc Gyitni? 

Backhill I Peking I Beijing 

[The] File for1 tlze Cosmic] Track 

[rrrid Iruliriduall Lhugh[rirre.~.~]; 

lntrorluction and Notes for a 

Translation of the Ma-wang-tui 

Ma.t~uscripts of the Lno Tzir 

Two Cross-Cultural Sttidies on 

Reading Theory 

a. Cross-orthographic Stroop Research: 

One Study in Context 

b. The Composing Process of a 

Bilingual Novice Writer 

Slips of the Tongue and Pen 

in Chinese 

l9 

Pages 

Price 

69 1 17.50 

I $3-50 

28 

45 $8.50

Number 

23 

24 

25 

26 

Date 

9 1/7/4 

9118 

9118 

91/9 

Author 


University of Perlnsylvania 

Previous Issues, cont. 

Davic1A.Utz 


Jean DeBmch 

University of Alberta 

JAO Tsung-i 

Chinese University of 

Hang Kong 

Title 

Tracks of the Tao, Senlantics of' Zen 

hgiiage, Writing, ancl Tradition 

in Ira11 

Linguistic Nationalism: The Cast: 

of Sout hem Min 

Queslions on the Origins of Writing 

Raised by the Silk Road 

Pages 

Checks (United States of America funds only, please) shoiild be made payable to Sino-Pkaror2ic. Pupers and sent 

to the editor at the Department of Oriental Studies 


Philadelphia, PA 19104-6305 USA 

Please include an additional $1.00 for postage and handling for each item ordered. U.S. airrnai 1 $1.50 extra per 

item; overseas airmail $2.50 extra per item. 

10 

24 

22 + 

3 figs. 

10 

Price 

$3.50 

$6.50 

$6.00 

$4.50

Essays on Writing and Language in Honor - Sino-Platonic Papers

Create successful ePaper yourself

Delete template?

Save as template?