Yu Li - The Chinese Writing System
Yu Li - The Chinese Writing System
Yu Li - The Chinese Writing System
System in Asia
The Chinese Writing System in Asia: An Interdisciplinary Perspective integrates a diverse range of disci-
plinary approaches in examining how the Chinese script represents and actively shapes personal and
social identities in and beyond Asia. It is an ideal read for students and scholars interested in a broad
and culturally rich introduction to research on the Chinese writing system. It can also serve as the main
text of an undergraduate course on the subject.
Key features of this volume include:
• Insights from studies of the Chinese writing system in linguistics, script reform and technology,
gender, identity, literature, and the visual arts;
• Examples embedded in inquiries of the cultural history and contemporary society of Asia;
• Rigorous yet accessible discussions of complex concepts and phenomena that assume no prior
knowledge of Asian languages or linguistics;
• Supplementary multimedia materials and resources, including instructional support, available
online.
“This book is a welcome new introduction to the Chinese writing system. I see the strengths of
the book in three areas: its wide scope in the topics it covers, its thoroughness in the treatment
of the topics, and its balance between scholarly discussions of the topics and its attention to
popular interest in the topics.”
— Guohe Zheng, Professor of Japanese,
Ball State University, USA
The Chinese Writing
System in Asia
An Interdisciplinary Perspective
Yu Li
First published 2020
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
and by Routledge
52 Vanderbilt Avenue, New York, NY 10017
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2020 Yu Li
The right of Yu Li to be identified as author of this work has been asserted by her in accordance with
sections 77 and 78 of the Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by
any electronic, mechanical, or other means, now known or hereafter invented, including photocopying
and recording, or in any information storage or retrieval system, without permission in writing from
the publishers.
Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are
used only for identification and explanation without intent to infringe.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
Names: Li, Yu, 1978– author.
Title: The Chinese writing system in Asia : an interdisciplinary
perspective / Yu Li.
Description: Abingdon, Oxon : New York, NY : Routledge, 2020. |
Includes bibliographical references and index.
Identifiers: LCCN 2019025053 (print) | LCCN 2019025054 (ebook) |
ISBN 9781138907317 (hardback) | ISBN 9781138907324 (paperback) |
ISBN 9780429345333 (ebook) | ISBN 9781000698343 (adobe pdf) |
ISBN 9781000698701 (mobi) | ISBN 9781000699067 (epub)
Subjects: LCSH: Chinese characters—Asia.
Classification: LCC PL1171 .L387 2020 (print) | LCC PL1171 (ebook) |
DDC 495.11/1—dc23
LC record available at https://lccn.loc.gov/2019025053
LC ebook record available at https://lccn.loc.gov/2019025054
ISBN: 978-1-138-90731-7 (hbk)
ISBN: 978-1-138-90732-4 (pbk)
ISBN: 978-0-429-34533-3 (ebk)
Preface ix
References 233
Index 238
Preface
The idea of this book came about in the second year of my teaching career. After having
taught a few Chinese language courses, I was eager to create something new to offer my stu-
dents, something that would spark their curiosity, challenge their thinking, and inspire their
creativity. I wanted them to see, as I did at the time and still do now, that learning Chinese
was not just about being able to communicate in the language; it could be and should be, in
no small measure, a journey of intellectual discovery and growth. And so was my experience
researching for and teaching the course Chinese Writing Systems in Asia at Emory for nearly
ten years since then.
Now this book almost feels like a gift to myself: the next time when I teach this course,
I will be able to use it as a main text. I have had to rely on a collection of scholarly writings
that was challenging to present as a coherent body of text in an introductory course. Having a
systematic and accessible set of readings on hand will be a welcome change. I hope it will make
it easier for interested colleagues in the field to offer similar courses as well. Besides what is in
the print edition and the electronic version of the book, I have also made available additional
materials and tips for teaching on the companion website, including exercises and activities
that can be integrated into classwork or assigned as homework.
Although intended as a textbook for a college course, this volume is by no means re-
stricted to a classroom audience. General-interest readers curious or enthusiastic about the
Chinese script will also find the broad-ranging topics interesting and accessible. This book is
designed to be an interdisciplinary introduction to the Chinese writing system. The chapters
integrate studies of the Chinese language, writing, and linguistics into inquiries regarding
script borrowing, writing reform, technology, gender, identity, visual art, and literature, and
they assume no prior knowledge about any of these topics or about Chinese language or lin-
guistics in general. Parts I and II are foundational and will be best read first, but the remaining
ones can be more flexibly sequenced.
It is also with the development of my pedagogical field in mind that I have written this
book. As strongly as ever, I believe that there is much more to be said and done about teaching
Chinese than the worthy endeavor of raising students’ communicative proficiency. Language
and writing are at the core of humanistic flourishing – as we see in this volume, the Chinese
writing system has been and will continue to be part and parcel of the forging and reshaping
of social, cultural, and certainly personal identities – of those who’ve embraced it and those
who’ve abandoned it. Language and culture instructors of Chinese are obligated to bring our
students closer to such understandings, and I hope this book is a step in that direction.
A small step, that is. The subject of Chinese characters is endlessly fascinating, and
I have had to limit the topics to the selected few that there is room to accommodate and
worry about their uneven treatment. Indeed, among the numerous decisions made during
the writing process, I suspect not all were wise, and any true wisdom that does come through
is probably rubbing off the many works I have consulted and to which I am greatly indebted.
x Preface
I also owe a great deal to the many individuals who have helped with this project and
wish to thank them here: my Emory colleagues and friends Juliette Apkarian, Julia Bullock,
Bumyong Choi, Cheryl Crowley, Seth Goss, Sun-chul Kim, Aya McDaniel, Maria Sibau, and
Amanda Wright, who gave me invaluable feedback on various parts of this project; my editor
Kelly Besecke, who patiently read through every word and offered key insights on making
them fit better together; my research assistant Tianqi Wang, who created most of the illus-
trations, helped obtain permissions to use copyrighted images, and organized materials for
the companion website; and my dear friends Ron Janssen, Jin Liu, and Qi Wang as well as my
family in China, for inspiring me and keeping me going. It is to them that I dedicate this book.
PART I
Linguistic preliminaries
1 Foundational concepts
T
HE KEY TO UNDERSTANDING how a writing system works is to understand how
its symbols systematically represent the units of the language it writes. This is often an
intricate matter that involves linguistic concepts of which we may not be intuitively
aware. It is therefore necessary to familiarize ourselves with these concepts first, so that we
will be able to discuss the working mechanisms of writing systems with more clarity and
efficiency. In what follows, we will start with defining and distinguishing between a pair of
terms that provide the disciplinary setting in which we will discuss the other terms: phonetics
and phonology. From there we will look at three analogous pairs of concepts: phoneme and
allophone, morpheme and allomorph, and grapheme and allograph.
meaning) in the English phonology. In French, however, the [θ] sound is normally not used,
so no two words are different from each other solely on the basis of [θ] vs. [s]. If a speaker for
some reason substitutes [θ] for a [s] in a French word, the result may sound odd to a native
French person, but it would not be interpreted as a different word. Thus, in the subconscious
mind of the French speaker, [θ] and [s] are two possible physical realizations of the same
sound category: they are phonetically different, but phonologically the same. By contrast, /θ/
and /s/ are phonologically distinctive in English. This example tells us that in order to be able
to describe the distribution of speech sounds in different languages, we would not only want
to talk about them as concrete entities in the physical world (i.e., phonetics), but would also
need to understand at an abstract level how they relate to one another in a particular language
(i.e., phonology).
alternative forms of -emes. Besides allo-phones, we will also learn about allo-morphs and
allo-graphs in this chapter. As an example for understanding the concept of allophones, think
about how the phoneme /t/ in English may be pronounced differently based on the context.
When it occurs at the beginning of a word, most speakers pronounce it as an aspirated con-
sonant [th], as in tack [thæk]. Aspiration refers to the strong puff of air that accompanies the
production of the consonant, and it is marked as a raised h ([h]) in IPA. When /t/ occurs after
the sound [s] in the same word, by contrast, most English speakers use the unaspirated ver-
sion of the sound, [t], as in stack [stæk]. If you place your hand close in front of your mouth
when saying tack and stack, you should be able to feel a stronger puff of air in tack. Thus, [th]
and [t] are phonetically different sounds. In fact, native speakers of English are fully able to
hear the difference between them.
Allophones are context-dependent, so that the phonological environments in which they
occur do not overlap with each other. In other words, they are in complementary distribution.
In the example of tack and stack, [th] is always at the beginning of a word and never occurs
right after [s] in the same word, while [t] always occurs immediately after [s] and never at the
beginning of a word. Thus, as allophones of the same phoneme /t/ in English, [th] and [t] com-
plement each other in distribution. For this reason, an allophone is also a predictable phonetic
realization of a phoneme as determined by the phonological context: at the beginning of a
word, the English /t/ is always [th], while immediately after [s] within a word, it is always [t].
Companion website
Now that we understand the terms phoneme and allophone as defined in phonology,
we will learn two additional pairs of concepts analogous to them, both of which will be very
useful to us in the discussion of writing systems: morpheme and allomorph, grapheme and
allograph. A morpheme is the smallest combined unit of sound and linguistic meaning or
function in a given language. It is a word-like unit but more rigorously defined than word. In
fact, “word” is notoriously difficult to define in linguistic terms: Is hot dog one word or two
words? What about bookcases? And fun-filled? This is one reason why we need the concept
of a morpheme. It allows us to say for sure how many basic units there are: hot dog contains
two morphemes, hot and dog; bookcases consists of three morphemes, book, case, and the
plural marker -s; and fun-filled also has three morphemes, fun, fill, and the past participle
suffix -ed. Each of these morphemes is a smallest meaningful or functional unit and cannot
be further divided without losing its meaning or function. Indeed, different from phonemes,
morphemes not only involve sounds but also come with meanings or grammatical functions.
In this sense, morphemes are at a linguistic level higher or more complex than phonemes.
Like phonemes, however, a morpheme is also an abstract representation in the speaker’s
mind, and its actual phonetic realizations, or allomorphs, depend on phonological contexts. For
example, the plural morpheme in English is /-z/. It is realized as the voiceless [s] when following
a voiceless consonant. Thus, we have /-s/ in cats [kæts], where [t] is a voiceless consonant. The
plural morpheme /-z/ becomes [-әz] with a short vowel inserted in front when it comes right
after the sounds [s] (as is the final sound in kiss), [z] (buzz), [∫] (wish), [ʒ] (mirage), [tÞ] (church),
Chapter 1 Foundational concepts 5
and [dʒ] (judge). Therefore, we have kisses [-sәz], buzzes [-zәz], wishes [-Þәz], mirages [-ʒәz],
churches [-tÞәz], and judges [-dʒәz]. All elsewhere, the plural morpheme /-z/ is realized simply
as [-z]. The allomorphs [-s], [-әz], and [-z] are context-dependent, they are in complementary
distribution, and we can also say that they do not contrast with each other.
The concepts phoneme, allophone, morpheme, and allomorph apply to speech; in
writing, we may use the terms “grapheme” and “allograph” to think about the roles graphic
symbols play. Graphemes are abstract representations of contrastive graphic categories in
the writer’s mind. In writing English, for example, <m> and <t> are two graphemes that can
distinguish word meaning: <smart> is a different word than <start>. Allographs are actual
realizations of a given grapheme that are determined by context. They are written differently
in different contexts, but we still recognize them as the same letter or character. In the print-
ed text below, the grapheme <m> has two allographs, each in a different typeface:
The Times New Roman {m} and the Helvetica {m} are each consistent with their neighboring
letters in terms of typeface, as is usually the case in printing practice. Here is another exam-
ple: the uppercase {M} and the lowercase {m} are also allographs of the same grapheme <m>.
Based on the convention for writing English, for example, {m} is used at the beginning of a
sentence, while {m} is used elsewhere, unless it is the first letter of a proper name or part of an
acronym, where it should be capitalized:
Notational conventions
Following the notational conventions in linguistics, in this book, phonemes and morphemes
are placed between slashes (e.g., /p/). Allophones and allomorphs use square brackets ([p]).
We use angle brackets (<p>) for graphemes and braces ({p}) for allographs.
Note
1 Transcriptions of speech sounds in the rest of the book are also in IPA. In IPA, each symbol rep-
resents a unique speech sound. The use of IPA allows us to specify and communicate with clarity
what sounds we are referring to. IPA is also designed to be able to transcribe all speech sounds in
human languages, which makes it an indispensable tool for documenting languages not yet written
down. We will not discuss IPA in detail in this book. If you are interested in learning how to tran-
scribe English in IPA, many resources are available on the Internet. The companion website has links
to such resources.
2 What is writing?
Companion website
To fully understand what constitutes writing, let us think further about what is not writing.
Yet, perhaps because many of us today live in a highly literate society, confusion between
the two, especially in using “language” to refer to “writing,” is pervasive. You may have heard
statements like these:
Such statements are linguistically problematic. “Chinese” or “Hebrew” each refers to a lan-
guage.2 The Chinese language is not composed of characters but speech sounds – one does
not need to write or read any characters in order to speak or understand it. Hebrew is clearly
not lacking in vowels, though it is true that vowels are generally not represented in writing
the language. Thus, the first two statements are in fact comments on how the languages are
written rather than on the languages themselves. The third statement may require a little more
explanation. Nǚshū (女书/女書 ‘Women’s Script’) is the name of a script created by and once
used among women in rural Jiangyong county, Hunan province in China. It is not a language
but a writing system designed for writing the local Chinese dialect. We will learn about nǚshū
in detail in Chapter 19 of this book.
Statements like these may reveal a lack of understanding about what constitutes lan-
guage. In cognitive terms, language is a complex system residing in our brain that allows
us to produce and interpret utterances. For our purpose, since we are primarily concerned
with graphic rendering of linguistic utterances, language can be understood specifically as the
speech produced or potentially produced by this system.
Companion website
Indeed, systematicity is one of the fundamental features of writing, and this is why lin-
guists use the term writing system to refer to what we commonly call writing or script. In the
remainder of this book, we will use these three terms interchangeably.
Now let us further examine the ways in which writing is systematic.
Writing is systematic
Writing, first of all, has a systematic relationship to language. A script generally consists of
a set of visual symbols (graphemes) linearly or otherwise arranged to represent the speech
units of a language. For this to work efficiently, a high degree of consistency is required. Each
grapheme in the script usually represents a constant unit of the language, such as a single
phoneme or a single morpheme. For example, the letters <a k i l m> in writing English each
represent a phoneme in the most ordinary cases. In writing Chinese, a character, such as <人
走 文>, corresponds to a morpheme: rén ‘person,’ zǒu ‘to walk,’ or wén ‘writing.’ Of course, the
relationships between a grapheme and a speech unit as exemplified by the writing of English
and Chinese are not the only possible relationships between script and language. In writing
Cherokee or Yí 4 (彝), for example, an individual symbol represents a meaningless syllable.
As another example, the graphemes of the Japanese hiragana (平仮名) and katakana (片仮
名) scripts represent speech units that may be slightly smaller than syllables. The nature of
the relationship between language and script serves as the basis on which linguists categorize
writing systems. For example, Cherokee and Yí scripts are both syllabic writing systems. We
will look at the classification of writing systems in more detail in Chapter 3 of this book. For
now, it is sufficient to understand that the corresponding relationship between a grapheme
and a speech unit is generally constant in that it applies to the entirety of a writing system.
Writing systems in general also have internal organizations of their own. For instance,
English is usually written horizontally from left to right, while Arabic writing is normally
right to left. English letters are arranged sequentially. In writing Korean, however, letters are
combined into blocks of syllables, often with some letters stacked on top of others. In both
English and Korean, words are separated by spaces, but in Chinese they are strung together
with no spaces in between. In English, the first letter of each sentence is usually capitalized,
while in writing German, the first letter of each noun is capitalized regardless of its position
in the sentence.
The internal structure of a writing system has little to do with the language being writ-
ten and is primarily a matter of convention. For this reason, it may change over history. For
example, some ancient Greek texts of the 6th century bce were written in alternate lines of
opposing directions. That is, the text flowed from left to right in one line, and then right to left
in the line below it, and then left to right again below that, and so on. This pattern was likened
to how an ox turned in ploughing a field and was given the name boustrophedon, a word Greek
Chapter 2 What is writing? 9
in origin meaning “to turn like oxen (in plowing).” Chinese text is another example. Today
Chinese is usually written in horizontal lines from left to right and the lines are arranged from
top to bottom on a page, just like in writing English. Before the 20th century, however, most
Chinese texts used a very different arrangement: writing proceeded from the upper right cor-
ner of the page in columns arranged from right to left with each column starting at the top.
This change was the result of Western influence on China.
Notes
Why do we need the concept “mora” after all, if it is so cumbersome to define? It has to
do with the timing and rhythm of the language, in this case Japanese. Apparently, mora is the
timing unit on which the Japanese speech is based. That is to say, each mora is given about the
same amount of time in speaking Japanese. The concept of mora, therefore, is indispensable
for describing and understanding the Japanese sound system. We will talk more about Japa-
nese morae in Chapter 11 of the book. For now, it is sufficient to know that moraic writing
systems are also a type of phonographic script.
Unlike phonographic scripts, which represent speech sounds only, the character-based
writing system for Mandarin Chinese consists of symbols that do involve meaning. A Chinese
character, in most cases, represents the smallest combination of sound and meaning – that is,
a morpheme – in speech. We refer to such a writing system as a morphographic or a mor-
phemic script. More specifically, since a character usually corresponds to a syllable in terms of
speech sounds, we can also say that Chinese uses a morphosyllabic writing system.
Morphographic? Logographic?
You may have heard of another term that is often used to characterize the Chinese script: logogra
phic. Some also refer to Chinese characters as “logographs” or “logograms.” How is “logographic”
14 Part I Linguistic preliminaries
different from “morphographic”? Which adjective offers a more accurate description of the Chinese
writing system?
The “logo-” part of the word “logographic” derives from the Greek lógos, meaning “word.”
A logographic script is thus a writing system in which a grapheme typically represents a word
or a phrase. This is not exactly the case for Chinese characters, however. A great portion of
the words in Modern Chinese contains more than one morpheme (and more than one sylla-
ble in pronunciation) and so must be written using more than one character. In other words,
a large number of Chinese characters represents morphemes that cannot stand alone as
words and must be combined with other morphemes to form words in Modern Chinese. This
is why Chinese writing is not entirely logographic, and “morphographic” is a more accurate
characterization.
So far, we have been assuming that the grapheme-speech relationship for a given pair
of writing system and language is uniform. In the case of English written in the Roman
alphabet, for example, such an assumption would mean that each Roman letter consistently
represents a phoneme in English: <p> would usually be a consonant and <a> would normally
represent a vowel – even though the specific consonant and vowel might not always be the
same ones.1
In reality, however, this grapheme-speech relationship is almost never completely
consistent. English, in particular, is notorious for its pervasive spelling irregularities. One
letter may represent multiple speech sounds, and a single speech sound may be written as
two or more letters. One often cited example, to push this point to a certain extreme, is the
artificial word ghoti as an alternative spelling for the English word “fish.” If we extend the
grapheme-phoneme correspondences in the words listed below to ghoti, we will indeed be
pronouncing it as “fish.” Note that in this form, the letter <i>, which normally represents a
vowel, is part of a spelling combination for a consonant.
Despite such discrepancies, the Roman alphabet in writing English is still, in essence, a
phonemic writing system. The fundamental relationship of “one letter, one sound” still holds
Chapter 3 What kinds of writing systems are there? 15
true. This is especially understandable if we view the subject from a historical perspective.
Many of the spelling irregularities arose from phonological changes in the English language
over time. For instance, why is it that we do not pronounce the <k> in words like knight, knot,
and knowledge yet we spell them with a <k> in writing? In fact, <k> used to be pronounced in
Old English before <n>, and it sounded just like the <k> in kid today. In modern German, a
language genetically related to English, <k> is also still pronounced before <n> in words like
Knecht (/knɛçt/ ‘servant’) and Knoten (/'knoːtn/ ‘knot’).
a language cannot be written using multiple types of scripts. And indeed they already are. You
may remember that Mandarin Chinese is written in both the morphosyllabic character-based
system and in pīnyīn, which uses the phonemic Roman alphabet. Likewise, English may also
be written using non-phonemic systems. On the companion website is a simple exercise that
shows how this may work.
Companion website
Notes
1 The same letter may be pronounced differently in different contexts. For example, <a> in far, bad,
and atop have different pronunciations.
2 Information about the use of ghoti in popular culture comes from the Wikipedia entry Ghoti: https://
en.wikipedia.org/wiki/Ghoti.
3 DeFrancis (1989, pp. 248–249).
4 Pıˉnyıˉn tutorial
B
EFORE DELVING INTO THE character-based writing system for Chinese, it would
be beneficial, if you have not done so, to learn to read and pronounce pῑnyῑn (full name
Hànyǔ Pῑnyῑn 汉语拼音/漢語拼音 ‘sound-spelling system for Chinese language’), the
international standard for Romanizing modern Chinese1 in writing. There are a number of
good reasons why. First of all, as a system of writing for Modern Standard Chinese in its own
right, pῑnyῑn is naturally a subject of study for this book. In Chapter 14, we will be looking
at the history and current uses of this script in more depth and detail. For now, it would be
good to be aware that a substantial number of words in this book are represented in pῑnyῑn. In
fact, pῑnyῑn has been used to represent Chinese lexical items in Western academic literature
and official publications since the 1980s, so chances are that you will encounter pῑnyῑn words
in other books about China published in a Western language. Being able to pronounce such
words, albeit silently in your mind, would allow for a smoother, more confident, and more
enjoyable reading experience.
More broadly, pῑnyῑn is commonly used as a pedagogical aid to learning Chinese by
both native and non-native speakers. It is indispensable for dictionaries, language textbooks,
and children’s books. For many, pῑnyῑn is also the preferred method to input Chinese text on
computers and other electronic devices (a topic we will discuss in Chapter 16). Public signage
in China, especially those signs containing geographical names for streets, around highway
exits, at airports or train stations, or proper names on plaques for shops and restaurants, often
has pῑnyῑn accompanying characters to assist foreign visitors. Figure 4.1 shows the name of a
convenience store, Nóngjiā bǎihuò diàn (农家百货店/農家百貨店 ‘farm family convenience
shop’), run by a local household in a scenic area in rural Héběi (河北 ‘(name of a province in
Northern China)’).
Another reason for learning to read pῑnyῑn is that it can be accomplished even if you do
not yet speak the Chinese language. Chinese syllable shapes are a closed set – that is, there
are a limited variety of permissible syllables – so it is conceivable to learn to pronounce every
possible syllable. If you master pῑnyῑn, you will be able to pronounce passages of Chinese Ro-
manized in this script even if you do not yet understand the textual meaning, and the ability
to do so brings you one step closer to learning the language.
Reading pῑnyῑn involves spelling out syllables by putting together their components. In
fact, pῑn (拼) means “to spell,” and yῑn (音) means “sound,” so the name pῑnyῑn reveals the
script’s working mechanism of “spelling the sounds.” A Chinese syllable usually consists of
two parts, an initial consonant (which may be absent) referred to as an “initial” (shēngmǔ 声
母/聲母 in Chinese) and the remainder known as a “final” (yùnmǔ 韵母/韻母). A final can
be a single vowel, which makes for a “simple final.” It may also contain a main vowel and a
secondary vowel with or without a nasal consonant (i.e., -n, -ng) at the end of the syllable, in
18 Part I Linguistic preliminaries
FIGURE 4.1 Store signage with Chinese characters annotated with pıˉ nyıˉ n
which case the final is referred to as a “compound final.” A syllable typically carries a tone that
spans across its entirety.2
Now, let us get started. We will begin by reading the simple finals first and then move
on to the initials. This is because the initials are usually pronounced with simple finals added
for greater audibility. After that, we will practice reading the compound finals and learn a
few rules about spelling pῑnyῑn in full syllables. Finally, we will learn about tones and practice
pronouncing syllables with tones.
Simple finals
There are six simple finals, as shown in Table 4.1.
Table 4.1 Simple finals
a o e i u ü
Companion website
Listen to the recordings on the companion website, and you will see that four of the simple
finals are pronounced very much the same way the letters are typically pronounced in Western
Chapter 4 Pīnyīn tutorial 19
orthography: a sounds like the a in Dada, o is similar to the o in form when r is not pronounced
(as in British English), i is more or less the same as the i in kiwi, and u the u in truth.
The e in pῑnyῑn may need some special attention. It is not pronounced as in the English
word met, but is rather like the first vowel in ago or above. Thus, he in pῑnyῑn sounds more like
the word her (without the r sound) than he in English.
The vowel ü tends to be more difficult for English speakers, because it represents a sound
not used in English, and its lip-rounding gesture may be particularly unfamiliar to monolin-
gual English speakers. However, it can be easily pronounced by following a few simple steps.
First, say the vowel in kiwi. As you sustain this vowel sound, tighten your lips into a circular
shape while keeping your tongue in the same position. There, you should now end up with
a ü sound. The key here is to round your lips as if you were trying to whistle. You can check
the shape of your lips in a mirror and make sure they are rounded. As you practice, try to
distinguish this sound from u and i, and also distinguish it from the English word you. With
sufficient practice, you will be able to pronounce it accurately.
Companion website
Initials
“Initials” refer to consonants that occur at the beginning of syllables. Much like the Roman
alphabet, pῑnyῑn initials (and finals) are usually arranged in a chart following a conventional
order, as shown in Table 4.2. Syllables in the parentheses annotate how the initials are usually
read. Notice here that simple finals are added to the initials to form full syllables in order to
make the initials easier to pronounce and more audible.
Table 4.2 Initials
b (boˉ) p (poˉ) m (moˉ) f (foˉ)
d (deˉ) t (teˉ) n (neˉ) l (leˉ)
g (geˉ) k (keˉ) h (heˉ)
j (jıˉ ) q (qıˉ ) x (xıˉ )
zh (zhıˉ ) ch (chıˉ ) sh (shıˉ ) r (rıˉ )
z (zıˉ ) c (cıˉ ) s (sıˉ )
[y (yıˉ )] [w (wuˉ)]
Listen to the recordings on the companion website. You will find most of the syllables
straightforward to pronounce. However, in a few places the vowel i is not pronounced as an
English reader may typically imagine: In the fifth and sixth rows it is not pronounced as ee,
but rather as a voiced3 elongation of the consonants zh, ch, sh, r, z, c, and s. For this reason, the
pῑnyῑn syllable shi sounds quite different from the English word she, and si is different from see.
Listen and see if this makes sense to you.
20 Part I Linguistic preliminaries
Companion website
Which of the initials are similar to sounds used in English? Which ones may be new to
you? The first three rows should be relatively easy for native English speakers, as the conso-
nants in these rows are almost identical to those in English. It may become more challenging
as we reach rows four to six, because some of the sounds are not part of the English phonolog-
ical inventory. We will look at these sounds more closely.
The series j, q, and x may pose the greatest challenge for English speakers. These sounds
form a series because they use very similar configuration of the vocal tract. The key gesture
here is to press the middle part of the tongue up against the roof of the mouth,4 and then make
the sound by letting air squeeze out between them, resulting in a great amount of friction. For
j and q, start out with the tongue completely touching the roof of the mouth. For x, the tongue
approximates the roof of the mouth with a slight space in between. The distinction between
j and q is that q is accompanied by a much stronger puff of air or aspiration. J, q, and x are
somewhat similar to the English j(eep), ch(eese), and sh(eep), respectively, but they really have
a different quality. Some say they seem softer, but others feel they are sharper compared to the
English consonants. You can decide for yourself which makes more sense; what is important is
to be able to tell the difference by listening and imitate as closely as you can. The three sounds
also pattern the same way in the Chinese phonology: j, q, and x can only be followed by one of
two vowels, i or ü (written as u after j, q, or x). When they are followed by i, spread your lips in
pronouncing the consonant; when they are followed by ü, round your lips.
Say “eggplant”!
When taking photos, you get a smiling look when your subjects say “cheese.” In Chinese, a popu-
lar word for the same purpose is qiézi (茄子 ‘eggplant’). Can you explain why saying these words
generates a smiley expression?
You may find close equivalents in English for the next series zh, ch, sh, and r. Zh is very
much like the initial and final consonant in judge, ch is similar to the first consonant in chair,
and sh is very close to the English sh in shape. The Chinese r is pronounced like the English
r but without the lip rounding and often with the tip of the tongue curled up.5 Some Chinese
speakers may also pronounce it as something close to the s in treasure.
Now we move on to the next row: z, c, and s. Be careful about z and c – they are pro-
nounced differently from what you may assume from an English perspective. Z is similar to
the final consonants in heads represented by ds. C is very much like the ts sound in cats, so it
is not pronounced like a k. S is the same as the English s.
Chapter 4 Pīnyīn tutorial 21
The bottom row contains two symbols, y and w. They are not in the official chart of
pῑnyῑn initials. However, they occupy the position of initials in writing when the syllables do
not have initials in pronunciation: If the syllable starts with the vowel i, a letter y is added to
the initial position, and when it begins with the vowel u, the letter w is added. In this chart, we
pronounce y as yee and w as woo.
Companion website
Compound finals
The final of a Chinese syllable may contain two or three vowels. There may also be a nasal -n
or -ng following the vowels, or a retroflex6 -r in the case of er. Finals like these are referred to
as “compound finals.” Listen to the recording for Table 4.3 and practice saying the compound
finals. Notice how the vowels a and e may have different qualities in different syllables. This
is because they are affected by the sounds around them. For our purpose, instead of trying to
remember the different pronunciations of these vowels in different contexts, it may be helpful
to simply memorize the pronunciation of the compound finals as a whole.
Companion website
Spelling rules
When writing pῑnyῑn, the letters we put down on paper sometimes do not correspond exactly
to the sounds we pronounce or hear. The discrepancies are mostly by design and are compro-
mises made to render the orthography more systematic and easier to use. The patterns are
quite regular and can be summarized as a few spelling rules.
22 Part I Linguistic preliminaries
When a syllable does not have an initial and starts with the vowel a, o, or e, no initial
needs to be added in writing. For example, ā, áng, ǒu, èn are legitimate syllables in full form.
By contrast, if a syllable starts with the vowel i, u, or ü, an initial must be supplied in
writing.7
Syllables beginning with the i sound must be written with the letter y at the beginning. If
the syllable contains only one vowel i, then it is written with a letter y added to the front: yi (i),
yin (in), ying (ing). If the syllable contains the vowel i followed by another one or two vowels,
then the letter y replaces the letter i: ya (ia), yao (iao), ye (ie), you (iou, spelled as iu elsewhere),
yan (ian), yang (iang), yong (iong).
Likewise, if a syllable begins with the sound u, then in writing it starts with the letter w.
If the syllable contains only one vowel u, then it is written with a letter w added to the front:
wu (u). If the syllable contains the vowel u followed by another one or two vowels, then the
letter w replaces the letter u: wa (ua), wo (uo), wai (uai), wei (uei, spelled as ui elsewhere), wan
(uan), wen (uen, spelled as un elsewhere), wang (uang), weng (ueng).
If a syllable begins with the vowel ü, ü is replaced with yu in writing. When ü is the
only vowel in the syllable, it is written as yu (ü) or yun (ün). Notice that the two dots of the
umlaut ü are taken off following the letter y, but the vowel represented by the yu sequence is
still just ü. If ü is followed by another one or two vowels, it is also replaced by yu: yue (üe),
yuan (üan).
Table 4.4 is a summary of all zero-initial syllables written in p ī ny ī n.
Table 4.4 pelling rules for zero-initial syllables (syllables in parentheses are written as the forms
S
above them)
yi ya ye yao you yan yin yang ying yong
(i) (ia) (ie) (iao) (iou) (ian) (in) (iang) (ing) (iong)
Companion website
The compound finals iou, uei, and uen, when preceded by initials, are abbreviated as
iu, ui, and un, respectively. Thus, liu should be pronounced as /liou/, sui as /suei/, and tun as
/tuen/, for example. As stated earlier, as stand-alone zero-initial syllables, iou, uei, and uen are
written as you, wei, and wen, respectively.
ü is written without the two dots when following the initials j, q, and x (in addition to
when following y as previously discussed). This is because the sound u never occurs after j,
Chapter 4 Pīnyīn tutorial 23
q, and x in Chinese, and leaving out the two dots on ü should not cause any confusion. Just
make sure to round your lips when pronouncing the syllables ju, qu, xu, jue, que, xue, juan,
quan, xuan, and jun, qun, xun. Also be aware that both ü and u can occur after the initials n
and l, so it is necessary to keep the two dots for ü here in order to distinguish nü from nu, and
lü from lu.
The compound final er cannot take any initials. When it is a stand-alone syllable, it is
written simply as er. This syllable may also serve as a suffix used after certain nouns and verbs.
In pronunciation, the suffix usually merges with the syllable in front of it, forming one single
syllable. For this reason, it is written as a single letter r. A few examples are in Table 4.5. You
can listen to a recording of these words on the companion website. Note that this is a case in
which two Chinese characters are used to represent a single syllable.
Although character writing does not indicate word boundaries, pῑnyῑn writing does.8
Thus, a space is placed in between two words, while a multisyllabic word is written with its
syllables strung together. Sometimes it can become ambiguous as to where one syllable starts
and another ends. In such cases, an apostrophe is used to separate the syllables. It is also used
to separate two syllables with conjoining vowels. A few examples are in Table 4.6.
Companion website
Tones
When pitch variations are used to distinguish word (morpheme) meaning in a language, we
refer to them as tones. Chinese makes use of four basic tones. They are simply called “the first
tone,” “the second tone,” “the third tone,” and “the fourth tone.” It is important to be able to
articulate Chinese syllables in correct tones, because changing a syllable from one tone to an-
24 Part I Linguistic preliminaries
other may alter the meaning of a word or morpheme. A classic example used to illustrate such
tonal contrasts is the syllable ma. When pronounced in its high-level first tone, mā may mean
“mother.” In the mid-rising second tone, má may mean “hemp.” The third tone mǎ consists of
a dip followed by a rise in the tonal contour and can mean “horse.” In its high-falling fourth
tone, mà may mean “to scold.” You can hear these examples on the companion website.
Companion website
Does English have tones? English makes use of intonation, but not tones. Both into-
nation and tones are essentially variations in pitch, like changing from one note to another
on the musical scale. However, only when such variations are used to distinguish lexical
meaning do they become tones. Intonation in English may indicate whether the speaker is
making a statement or asking a question – for example, “yes!” with a falling pitch vs. “yes?”
with a rising pitch – but it does not change the fundamental meaning of the word. “Yes” still
means “yes.”
Chinese tones are conventionally described on a five-level scale. If we divide a speak-
er’s natural comfortable pitch range into five levels – 1 through 5, from low to high – we can
represent the contours of the four tones on a diagram as shown in Figure 4.2. The first tone
(5–5) is high and level. It starts at level 5 and remains all the way through at level 5. The
second tone (3–5) is a mid-rising tone, starting at 3 and ending at 5. The third tone (2–1–4)
starts fairly low at 2, dips to 1, and then goes up to 4. The fourth tone (5–1) is a high falling
tone, starting at 5 and finishing at 1. Such tonal forms shown in the diagram are referred
to as citation forms. When a Chinese syllable is pronounced in isolation, its tonal contour
is usually close to the citation form. In continuous speech, however, the actual tonal con-
tours may be different. This is especially true for the third tone (2–1–4), which is usually
realized only as its first half (2–1) in connected speech, so that it starts low and dips to 1
but does not then rise to 4, unless the syllable bearing this tone ends the utterance. That is,
the third tone is usually realized as a low-falling tone. The low-falling tone sounds almost
like a low-level tone. Thus, overall it may be useful to think of the Chinese tonal system as
an almost symmetrical one that consists of a high-level tone (the first tone 1–1), a low-level
tone (the third tone 2–1), a rising tone (the second tone 3–5), and a falling tone (the fourth
tone 5–1).
In writing, tones are marked using diacritics on vowels, for example: mā (first tone), má
(second tone), mǎ (third tone), and mà (fourth tone). What happens when there is more than
one vowel in the syllable? This is where the vowel sequence a-o-e-i-u-ü comes in handy as a
reference tool: the tone mark goes on the vowel that is closer to the beginning of this sequence.
For example, hǎo, tài, mōu, léi, xiè, duì, lüè. There is one exception to the rule, however. For
the compound vowel iu, the tone mark goes on u instead of on i. One way to remember this
exception is that when the compound final consists of the two vowels i and u, regardless which
comes first, the tone mark always falls on the second vowel (e.g., diū and duῑ).
Chapter 4 Pīnyīn tutorial 25
Companion website
Example 4.2 Tones: hǎo, tài, moˉu, léi, xiè, duì, lüè; diuˉ, duıˉ
Exercise 4.5 Four tones in citation form
When two syllables with third tones are right next to each other, the first syllable changes
into the second tone. Thus, nı̌hǎo (你好 “hello”) is actually pronounced as níhǎo even though
it is written as nι̌hǎo. Such tonal changes of syllables based on the tones of adjacent syllables
are referred to in linguistics as tone sandhi. The word sandhi means “fusional change” and is
derived from the Sanskrit word sam.dhi, meaning “joining” or “juncture.”
Companion website
In addition to syllables with the four tones as described earlier, sometimes Chinese syl-
lables may be pronounced briefly and lightly with no fundamental tonal contour. Such syl-
lables are said to have a neutral tone and do not bear any tone marks in writing. All neutral
tones may not be quite the same in pitch value, however, as the pitch value may change based
on what tone the preceding syllable has. After a first-tone syllable, a neutral-tone syllable
usually has a pitch value of 2. After a second-, third-, and fourth-tone syllable, its pitch value
is most often at 3, 4, and 1 respectively. Listen to the examples in Table 4.6 and see if you can
tell the difference.
Now that we have learned the initials, finals, and tones of Chinese syllables, you should
be able to pronounce any given syllables represented in p ī ny ī n. The companion website links
to a chart of all Chinese syllables with audio. You can use it to practice or refer to it for future
study.
Companion website
Notes
only vowel in the syllable, to maintain regularity in orthography, the initial consonant is still added,
thus resulting in yi for i, wu for u, and yu for ü, for example, even though the pronunciation of the
full vowels remains unchanged.
8 Word segmentation is a complex issue in p ī ny ī n orthography because it is relatively difficult to
define word boundaries in Chinese. Although a set of standards was established in as early as the
1980s, they are usually not taught at school, and native writers are often oblivious to these standards.
As a result, it is not rare to see variations in word segmentation in p ī ny ī n writing.
PART II
Writing Chinese
5 The Chinese speech
W
HEN WE TALK ABOUT the native speech of the Chinese people, we often simply
call it “the Chinese language.” But in reality, there is not a singular or unified Chi-
nese language. In its broad sense, “the Chinese language” encompasses hundreds
of varieties,1 much like “the English language” includes Australian, British, and American
English as well as many different regional variations – only much more varied. To understand
“the Chinese language,” therefore, we will need to learn about the different kinds of speech
that fall under the umbrella term “Chinese.” This is what we will do first in this chapter. We
will then take a focused look at the standard variety of the Chinese speech in contemporary
China, which we refer to as Modern Standard Chinese (MSC) or Mandarin.
(an earlier version of) Chinese: monolingual Mandarin speakers and monolingual Canton-
ese speakers usually find it quite impossible to understand each other, albeit both speaking
a variety of Chinese. Clearly, “Chinese” should be more accurately understood as a group of
related languages rather than as a single language. This group also includes Wú (吴), spoken
in Shànghǎ i, Zhèjiāng, and the southern part of Jiāngsū, Mι̌n (闽), spoken in Fújiàn, Hǎinán,
and Táiwān, and a few other languages. We use the term Sinitic to refer to this group, and we
say that Mandarin and Cantonese are both Sinitic languages. Other examples of language
groups with which you may be familiar include Romance languages and Germanic languages.
The relationship between Mandarin, Cantonese, Wú, Mι̌n, and so on is quite like, for instance,
that between Spanish, French, Italian, and Romanian, members of the Romance group.
Although Chinese consists of multiple languages (and dialects), scholarly literature on
Chinese linguistics still commonly uses the term “dialects” rather than “languages” to refer to
the Sinitic speech varieties. This has to do with the convention of the discipline. Linguists in
China, in particular, have long used the term fāngyán (方言 ‘speech of a geographical region’)
to refer to the various speech varieties of Chinese. This term encompasses both the languages
and the dialects of Chinese and places them on equal footing regardless of their mutual intel-
ligibility. For example, Northeast Mandarin, Southwest Mandarin, and Cantonese can all be
referred to as Hànyǔ fāngyán (汉语方言/漢語方言 ‘Chinese regional speech varieties’). This
practice is congruent with the shared history and cultural heritage of the speakers of various
Chinese speech varieties. For this reason, while keeping in mind that translating fāngyán as
“dialect” is not accurate, we will use the term “dialect” in this book to refer to the Chinese
speech varieties.2
How many varieties of Chinese speech are there? Many linguists recognize seven to ten
major dialect groups that are not mutually intelligible. (Within a dialect group, there may also
be mutually unintelligible varieties of speech.) Aside from Mandarin and Yuè (粤 Cantonese),
with which we are by now familiar, there are also Wú (吴), Mι̌n (闽), Kèjiā (客家), Xiāng
(湘), and Gàn (赣). Some linguists also recognize Jìn (晋/晉) as a dialect group separate from
Mandarin, Huī (徽) separate from Wú, and Píng (平) separate from Yuè. Table 5.1 provides an
overview of the dialect groups ranked by the size of their native-speaker population.
The map in Figure 5.1 shows the primary regions of their native speakers. As you
can see, the southeastern area of China is more linguistically diverse than the other parts.
In fact, eight of the ten aforementioned dialect groups, that is, all but Mandarin and Jìn,
find home in this area, while different varieties of Mandarin disperse across Northeastern,
Northwestern, and Southwestern China, and Jìn speakers concentrate in the central area of
Northern China.
Geographical factors contributed to the formation of this pattern. With its dense moun-
tains and rivers, Southeastern China affords much less mobility to its residents compared to
the other regions (especially to the Northern plains). When speakers settled in various pock-
ets of this area, it was more difficult for them to maintain contact, and as a result, their speech
changed in divergent ways. In time, the diverging speech forms became so distinctive that
their speakers could no longer understand each other.
To understand the linguistic diversity in China, it is also important to know that, while
the majority ethnic group (i.e., the Hàn) speaks Sinitic languages, the ethnic minority groups
(shǎ oshù mínzú 少数民族/少數民族) in China mostly speak other languages in addition to
Sinitic. Some examples of these languages are Mongol in Inner Mongolia; Kazakh, Kirghiz,
and Uighur in Northwestern China; Tibetan in Western and Southwestern China; Burmese,
Thai, and Vietnamese in Southwestern China; and Malayo-Polynesian languages in Táiwān.
any meaningful or effective discussion. It obviously may refer to only one variety of Chinese, so
which one should it be? In today’s China, what we call Modern Standard Chinese (MSC) is the
variety most widely promoted and used for public and official purposes, including in govern-
mental administration, education, commerce, and the media, so it may make sense that this is
the variety we refer to when we speak of “the Chinese language.” In this book, we will also use
“Mandarin” as an alternative term, in keeping with the convention in the field, but “Mandarin”
in this case is much more narrowly defined and needs to be differentiated from the “Mandarin”
used as the name of a language group in our earlier discussion. MSC or Mandarin is the speech
variety on which the writing of Chinese is based. It is not only the predominant form of speech
for official use in Mainland China and Táiwān but also one of the four official languages in Sin-
gapore and one of six used by the United Nations. On the Mainland, it has been referred to as
pǔtōnghuà (普通话/普通話 ‘common speech’), a term coined during the language and script
standardization efforts led by the People’s Republic of China (PRC) government in the 1950s.
In Táiwān, guóyǔ (国语/國語 ‘national language’) remains to be the most common term. Yet
another term, huáyǔ (华语/華語 ‘Chinese language’), is used in Singapore.
Historically, the need for a Chinese lingua franca (i.e., common speech) mostly existed
within the circle of government officials, who were required to serve outside of their home re-
gions and needed to be able to communicate with each other. The speech variety that arose to
meet that need was usually that of the nation’s capital, and therefore it changed over time as
the capital of China changed locations. In the Táng (唐) dynasty (618–907), it was the local
speech of Cháng’ān (长安/長安), today’s Xīān (西安) in central China. By the Southern Song
(1127–1276), this role had shifted to the language of then capital Lín’ān (临安/臨安), today’s
Hángzhōu (杭州) in Eastern China. During the last imperial dynasties, the Míng (明; 1403–
1644) and the Qīng (清; 1644–1911), when the capital was located in Běijīng (called Shùn-
tiānfǔ 顺天府/順天府 and later on Jīngshī 京师/京師) in the north, the mandarins, or public
officials, communicated in a speech variety based on the northern dialects, hence the term
“Mandarin” continuing in use today.
A national language for China, however, is a recent establishment. Up until the late 19th
and early 20th centuries, most Chinese people spoke only their local dialects. The largely agrar-
ian population at the time, living a land-bound life, had little need to communicate with those
beyond their local areas. When the country was forced to open its doors to Western colonizers,
the promotion of a standard form of speech became part of an urgent nationalist movement,
and the concept of a “national language” emerged. It took decades for MSC to eventually be-
come the common language of the Chinese in the real sense. Today, it is generally expected that
young Chinese speak MSC in addition to their regional dialects (if they are significantly differ-
ent from MSC), though many of the older generation may still only speak their local dialects.
Now let us take a closer look at some of the linguistic characteristics of Modern Standard
Chinese. In the next chapter, we will discuss the writing of Chinese. Although, in theory, any
kind of writing system may be used to write any language, the script that is most suitable or
efficient for a given language is one that capitalizes on the structural features of the language.
In this case, learning about the syllable structure, tonal system, and the role of homophony in
MSC will be especially helpful for understanding the Chinese writing system.
Syllable structure
A syllable (represented by the Greek letter σ) is a timing unit in language that consists of,
minimally, a vowel. The vowel is referred to as the nucleus of the syllable. A syllable may also
34 Part II Writing Chinese
include an onset, that is, one or more consonants that come before the nucleus, and/or a coda,
that is, one or more consonants following the nucleus. The nucleus and the coda together are
referred to as the rhyme of the syllable. In a language like Chinese, a syllable may also have an
element that applies to its entirety, that is, a tone.
Let’s look at an example of an English syllable in a diagram to get a more concrete sense
of its structure. As shown in Figure 5.2, the monosyllabic word feast can be analyzed as con-
sisting of the onset /f/, nucleus /i/, and coda /st/.
We can use a similar method to diagram the syllable structure of MSC. Traditional Chi-
nese phonology divides the syllable into two parts, initial or shēngmǔ (声母/聲母), which
is the syllable onset, and final or yùnmǔ (韵母/韻母), which corresponds to the rhyme. The
onset can maximally be a single consonant and is optional. The rhyme is required and is min-
imally a single vowel, but it can also contain a sequence of two vowels (i.e., a diphthong). The
rhyme may also have a glide (i.e., -i- /j/, -u- /w/, or -ü- /ɥ/)3 in front of the nucleus and/or a
coda nasal after it. The coda consonant must be one of two nasals, -n /n/ or -ing /ŋ/. Below is
an example of the MSC syllable biān (Figure 5.3).
There are 1,840 possible MSC syllable shapes with tones, but only 1,359 (74%) are ac-
tually used. English has more distinct syllables, because consonant clusters are allowed both
as the syllable onset and the coda. For example, the English word sprints has a sequence of
three consonants (/s/, /p/, /r/) as its onset, and another three (/n/, /t/, /s/) as the coda. Indeed,
compared to the 8,000+ syllable shapes in English, the MSC inventory is much smaller.4
Biángbiáng noodles
Legitimate but unused MSC syllable shapes are sometimes employed for playful use online or
in real life to represent onomatopoeic or dialectical pronunciation. One example is the name of
a popular Shǎnxıˉ noodle dish, biángbiáng noodles. Biáng conforms to the syllabic structure
of MSC, but it is not associated with any morphemes in the language. That is, its meaning is
unclear. Some say that the syllable mimics the sound the noodle-eaters make, and others spec-
ulate that it is in fact the sound of the chef slapping the dough to make the noodles. Biángbiáng
noodles are usually handmade, broad and thick, and are consumed with a generous amount
of hot red chili oil. Originally a poor man’s meal in the country, it is now served in fashionable
urban restaurants.
The popularity of biángbiáng noodles can partly be contributed to its playful name. Not only is
the pronunciation not part of the standard MSC inventory, but the Chinese character that has been
used to write biáng is also nonstandard and is not yet typable on the computer. What is more, with
57 strokes in its traditional form and 43 strokes simplified, it is one of the most complex Chinese
characters ever invented.
Companion website
Tones
Modern Standard Chinese makes use of four contour tones and one so-called neutral tone5 to
distinguish lexical meaning, or word meaning. To be more exact, altering the tone of a Chi-
nese morpheme changes it into a different morpheme. For words that contain only one mor-
pheme, this means changing the meaning of the word. The classical example used to illustrate
this, which you may recall from the pīnyīn tutorial (Chapter 4), is repeated in Table 5.2. The
syllable ma, when realized with different tonal contours, results in different words.
It may be worthwhile to further clarify here the relationship between morphemes and
words in MSC. Although mono-morphemic words are reasonably common, MSC words often
consist of multiple morphemes. In most cases, a syllable with meaning constitutes a morpheme
on its own; however, it may have to combine with at least one other morpheme to fully func-
tion as a word. For example, the morpheme xiào (校) is usually understood to mean “school,”
but it is rarely used independently to represent that meaning; rather, the di-morphemic
word xuéxiào (学校/學校 ‘school’; 学/學 ‘to study’) is preferred. There are many such exam-
ples in MSC.
As described in the p ī ny ī n tutorial, the contour tones are customarily referred to by
their numerical names: first, second, third, or fourth tones. When pronounced in isolation in
their full forms, the pitch contours of syllables bearing these tones can be represented using
numbers indicating pitch level on a five-step scale. On this scale, 1 represents the lowest pitch,
and 5 the highest, as shown in Figure 5.2 of the p ī ny ī n tutorial. In Table 5.2, the numbers in
the parentheses after the tone names refer to such a notational system. For example, the pitch
of the third tone, in its full form, starts out at mid-low (2), dips to low (1), and then rises to
mid-high (4), and thus we can use 214 to represent the third tone.6
Homophony
One characteristic of MSC is the prevalence of homophony, the linguistic phenomenon of
two or more different morphemes sharing the same pronunciation. You may be able to easily
think of examples of homophony in English: won, one; two, too, to; meet, meat; hear, here; their,
there; and red, read. Yet, you will probably find it challenging to compose a hundred-word pas-
sage all in the same syllable. With a stretch, we may come up with something like the below.
Give it a try, and you may find that the size of a homophonous group of words is so small in
English that you quickly run out of material.
In Chinese, however, this may be easier. In fact, homophony is even more prominent
in Classical Chinese than in MSC, so much so that the linguist Chao Yuen Ren (赵元任/趙
元任, 1892–1982) proved it possible to compose a poem in the Classical Chinese style that
tells a developed story, albeit a bizarre one, all using the same syllable. The poem consists of
morphemes pronounced shi in various tones. Table 5.3 shows the poem in p ī ny ī n and in sim-
plified characters. Note that although the poem is composed in the style of Classical Chinese,
the p ī ny ī n representation is based on MSC pronunciation.
Chao made use of about 33 different morphemes in this poem.8 The meaning of these
morphemes and their MSC pronunciation are given in Table 5.4.
As the story goes:
A poet named Shıˉ lived in a stone den and liked to eat lions, and he vowed to eat ten
of them. He used to go to the market in search of lions, and one day at ten o’clock, he
chanced to see ten of them there. Shıˉ killed the lions with arrows and picked up their
bodies, carrying them back to his stone house. His house was dripping with water so
he requested that his servants proceed to dry it. Then he began to try to eat the bodies
of the ten lions. It was only then he realized that these were in fact ten lions made of
stone. Try to explain this incident.9
Homophony exists in MSC to a lesser extent than in Classical Chinese. This is because many
mono-morphemic words in Classical Chinese must be expressed in poly-morphemic (and
therefore polysyllabic) words in MSC. For example, 誓 (shì ‘to vow’) and 狮/獅 (shī ‘lions’)
become 发誓 (fāshì ‘to vow’) and 狮子/獅子 (shīzi ‘lions’) in MSC – no longer homophonous
of each other. For this reason, it would have been more difficult to compose such a poem in
MSC. Nonetheless, homophony remains an important feature of the language.
Notes
1 The expressions “varieties,” “speech varieties,” and “language varieties” are used as neutral terms to
refer to particular kinds of speech. This is to avoid confusion between such concepts as “languages”
and “dialects,” which are defined by relating speech varieties to each other, and are in this sense not
neutral. As a rough analogy, “(speech/language) varieties” is a neutral term for “languages” and
“dialects” just like “(speech) sounds” is a neutral term for “phonemes” or “allophones.”
2 To avoid the confusion caused by the use of the term “dialect” to refer to distinct Chinese languag-
es, linguist Victor Mair (1991) coined the term “topolect” as an English equivalent for the Chi-
nese fāngyán. “Topo-” indicates “region” or “location” as a translation of fāng (‘region, location’)
in fāngyán (“regional speech”). Because this term is not widely used in introductory materials on
Chinese languages and dialects, this book does not adopt this term.
3 Pīnyīn followed by IPA is used here to represent the glides and later to represent the nasals.
4 According to DeFrancis (1977, p. 7), there are some 75 to 100 distinct syllables in Hawaiian, 100 to
300 in Japanese, 1,100 in Korean, 1,300 in Chinese, 4,800 for the Hanoi dialect of Vietnamese and
4,500 for the Saigon dialect, and 8,000 in English.
5 Neutral tones are so named because a syllable that bears such a tone is usually light and brief, and
it can be perceived as “toneless” compared with syllables with full tones. However, it does not mean
that a neutral-tone syllable does not have any pitch value. In fact, depending on the phonological
context – usually the tonal value of the syllable immediately preceding it – a neutral tone can be
realized as different pitches. It becomes a mid-register tone after a high-level (first) tone or a
high-rising (second) tone, a mid-high tone after a low-dipping (third) tone, and a low tone after a
falling (fourth) tone.
6 As noted in the p ī ny ī n tutorial, the full contour of the third tone is not generally realized in con-
nected speech. When a third-tone syllable is immediately followed by at least one other syllable in
the same utterance, it usually stops short as a low dipping tone (21). Speakers would sound highly
stilted if they pronounced third-tone syllables in the full 214 contour in utterance-medial positions.
When a third tone is utterance-final, however, it may be more fully realized.
7 Text adapted from Rogers (2005, p. 30).
8 In the Contemporary Chinese Dictionary, there are 85 shi morphemes represented by distinct
characters.
9 Translation adapted from Rogers (2005, p. 30).
6 Written Chinese
C
LASSICAL CHINESE WAS THE standard written form of Chinese for most of the
imperial Chinese history. It still has a profound impact on the Chinese language today.
In this chapter, we will take a brief look at this once dominant form of Chinese written
language and how it relates to Chinese speech and writing today.
(early 5th century bce) to the end of the Hàn dynasty (220 ce). The language spoken at the
time was what is referred to today as “late Old Chinese.” Although it is unclear if Classical Chi-
nese can simply be understood as late Old Chinese rendered in writing, the common under-
standing is that they were very similar. After the Hàn dynasty and until the early 20th century,
however, the similarity between the written language and the speech decreased as literary and
formal writing continued to be done in the Classical style, while the spoken varieties of the
Chinese language evolved.2
Used loosely, the term “Classical Chinese” may also incorporate writing done after the
Hàn dynasty until the end of the imperial era in the Classical style, the standard style of writ-
ing for roughly 2,000 years. It is remarkable that Classical Chinese came to serve as the written
standard for such a long period of time. How was it able to do so? The primary reason was
that many important literary and philosophical works, for example, the canons of Confucius
(孔子Kǒngzι̌; 551–479 bce) and Mencius (孟子 Mèngzı̌; 372–289 bce) were written in Clas-
sical Chinese. These works were so revered that later writers continued to write in their style.
Another key factor that solidified Classical Chinese as the standard written style was the civil
service system in imperial China, a meritocracy adopted to select able scholars to become
ruling officials. Aspiring candidates were required to study the Confucian canons and write
examinations in Classical Chinese in order to acquire a position in the ruling class. Being
able to read and write in Classical Chinese thus enjoyed official, sanctioned prestige. The civil
service examination system was established around the year 600 during the Suí (隋) dynasty,
and it was not abolished until some 1,300 years later in 1905, shortly before the last dynasty
ended. During this long period of time, although the spoken standard of Chinese continued to
change, Classical Chinese remained the standard written language for official use and serious
literature.
Classical Chinese, or wényán, was by no means the only written Chinese language be-
fore the 20th century. During the Táng (唐; 618–907) and Sòng (宋; 960–1279) dynasties, as
the Chinese speech continued to diverge from the Classical written language, a new form of
written Chinese emerged. By that time, the vernacular varieties of Chinese had become so
distinct from Classical Chinese that understanding the latter would require dedicated educa-
tion. The majority of the Chinese population was thus unable to read Classical text. The new
written language approximated the vernacular varieties of speech and later came to be called
báihuà (白话/白話), meaning “plain speech.”3 Báihuà gained wider popularity as time went
on, and important literary works were written in báihuà during the Míng (明; 1368–1644) and
42 Part II Writing Chinese
Qīng (清; 1644–1912) dynasties. With the last imperial dynasty coming to an end in the early
20th century, progressive intellectuals considered wényán an obstacle to mass education and
literacy and condemned it as a “dead language.” Instead, they advocated for writing in báihuà,
a “living language” and “language of the people” that they refined and upheld as the national
standard. Báihuà became the written language of the progressive education curriculum estab-
lished since then, and it eventually replaced wényán as the official written language.
As is easy to notice, the MSC sentence is considerably longer. This is because monosyl-
labic words in Classical Chinese often (though not always) correspond to bi- or multisyllabic
words in MSC. For this reason, Classical Chinese is in general terser than Chinese written in
MSC. Table 6.1 gives a side-by-side comparison of the content words – nouns, verbs, adjec-
tives, and adverbs – in the example.
Although Classical Chinese is no longer the standard written language, it still exerts
its influence on writing in the Chinese language. Writers often incorporate Classical-style
words or expressions not usually used in speech to make their writing sound more formal.
Because of this practice, there may be a more significant difference between the written
style and the spoken style in Chinese compared with English. The use of Classical-style,
single-syllable content words where multisyllable alternatives are available may give a piece
of writing a Classical flavor and thus makes it sound more proper or official. For example,
in the written abstract of an academic article, to refer to the article itself, it would be more
formal and often more appropriate to write 本文 (běnwén, ‘this article’) than 这篇文章/這
篇文章 (zhè piān wénzhāng, ‘this article’). While giving a presentation about the article,
Chapter 6 Written Chinese 43
Table 6.1 Correspondence between Classical Chinese and MSC content words
Classical Chinese MSC Meaning
(p ī ny ī n) (p ī ny ī n)
子 (Zı̌) 孔子 (Kǒngzı̌) ‘Confucius’
曰 (yuē) 说/說 (shuoˉ) ‘say’
学 /學(xué) 学习/學習(xuéxí) ‘study’
而 (ér) 而且 (érqiě) ‘and also’
时/時 (shí) 时常 /時常(shícháng) ‘often’
习/習 (xí) 复习/複習 (fùxí) ‘review’
不 (bù) 不 (bù) ‘not’
亦 (yì) 也 (yě) ‘also’
悦 (yuè) 使人快乐/使人快樂 (shı̌ rén ‘makes one happy’
kuàilè)
however, the author may speak of it using the latter, longer expression to avoid sounding
too bookish.
Such impact of Classical Chinese on Modern Standard Chinese is possible, as some have
argued, because of the morphographic nature of the Chinese writing system.5 Although the
sounds of Chinese characters have changed, their meanings have remained by and large con-
stant. This connection in character meaning makes Classical Chinese accessible to modern
readers and writers. If Chinese had been written with a phonographic system like the Roman
alphabet, then Classical Chinese words would have been spelled much differently than MSC
words. This is because the phonology of MSC is drastically different from that of late Old Chi-
nese, on which Classical Chinese is based. Classical Chinese pronounced in its Old Chinese
sounds would be incomprehensible to MSC speakers, just as Old English would be incompre-
hensible to today’s English users. If Chinese had been written with an alphabet as English is,
Classical Chinese would have become much less accessible to modern readers and writers.
This is not to say, however, that an average Chinese reader today is able to understand a
Classical Chinese text without much difficulty. Far from it – because MSC and Classical Chi-
nese are greatly different in syntax and lexicon, it requires years of dedicated learning to be able
to read Classical text with proficiency, and probably even more for composing a piece of writ-
ing in that style. The study of Classical Chinese is part of the standard curriculum from third
grade through high school in Mainland China and is required for the college entrance exam,
but even college students often hesitate to claim functional proficiency in Classical Chinese.
Notes
1 The term “triglossia” may be used when three language varieties are involved.
2 The written language used after the Hàn dynasty until the modern period is referred to as “Literary
Chinese,” although this term is often used interchangeably with “Classical Chinese.” This book uses
“Classical Chinese” as a cover term for both the pre-Hàn and post-Hàn periods. When “Literary
Chinese” is used, it is considered interchangeable with “Classical Chinese.”
44 Part II Writing Chinese
3 The term “plain speech” may be misleading: báihuà is a written language and not speech. The lan-
guage being written can be understood as Modern Standard Chinese or Mandarin.
4 The example is given in simplified characters in the main text. In the traditional script, the Classical
Chinese version is “子曰:學而時習之,不亦悅乎?” The MSC version is “孔子說:學習而且
時常復習,不也使人快樂嗎?”
5 See Hannas (1997, pp. 128–129).
7 The Chinese writing system
T
HE CHARACTER-BASED CHINESE SCRIPT is said to have the longest history of
continuous use of all current writing systems.1 Dating back to no later than the ora-
cle-bone script (ca. 1250–1050 bce) of the Shāng (商) dynasty,2 it has remained the sole
official writing system for Chinese for over 3,000 years. During the course of this long history,
major changes took place: its overall inventory of characters grew, the shape of the characters
evolved and underwent multiple reforms and standardizations, and most recently, the layout
structure of Chinese text also altered. The nature of the writing system, however, remained
the same.
To many Western readers first exposed to the Chinese writing system, two characteris-
tics often stand out as vastly different from their native scripts: one, its drawing-like symbols,
and two, the huge inventory of such symbols. Are Chinese characters indeed “little pictures”?
How are they formed? How many of them are there? Are new ones created all the time? Are
all the characters used? How many characters does one need to know? These are some of the
questions we will address in this chapter.
You may have heard of the Chinese writing system being categorized as “logographic.”
What does this term mean? Is this an accurate characterization of the Chinese script? We in
fact discussed this in Chapter 3 as a sidenote, and it may be worth revisiting it here. A logo-
graphic writing system is one in which a grapheme (in this case a logograph or logogram)
represents a word or a phrase. Since in writing MSC, a Chinese character usually represents a
morpheme, and because most of the MSC words consist of two or more morphemes, a mor-
pheme in MSC is oftentimes a smaller unit than a word, and in writing, a character only rep-
resents part of a word. For this reason, it may not be accurate to classify the Chinese writing
system as a logographic script in writing MSC. In comparison, the concept of a logographic
script may be more applicable to Classical Chinese, in which single morphemes much more
often function independently as words.
日 日 rì ‘sun’
水 水 shuǐ ‘water’
人 人 rén ‘person’
馬 马 mǎ ‘horse’
女 女 nüˇ ‘female’
Other examples of modern pictograms may be less picture-like. The oracle-bone form
for “sun” looks like a round object. The modern 日 (rì ‘sun’) character, however, is no longer
round. If not informed of the connection between this form and its more circular ancestor,
one may not be able to recognize it as a picture of the sun at all, and may think that it bears
more resemblance to a window or a ladder.
Regardless, such changes in appearance have not altered the nature of the character,
which was critically determined by the way the character was originally created. 日 (rì ‘sun’),
therefore, is still considered a pictogram. The existence of pictograms such as this one lends an
element of truth to the popular belief that some Chinese characters are “little pictures.”
Companion Website
However, it would be a gross exaggeration to make the blanket statement that Chinese
characters are pictograms, because most of the characters in use today are not pictograms.
Rather, the great majority is semantic-phonetic compounds that combine a meaning ele-
ment and a sound element into a single character. For example, the left side of the character
妈/媽 (mā, ‘mother’) is 女 (nǚ, ‘female, woman’), and the right side is 马/馬 (mǎ, ‘horse’). It is
easy to see here that the 女 (‘female’) element serves as the meaning component of the char-
acter. In this case, its pronunciation as an independent character is irrelevant. 马/馬 (mǎ ),
on the other hand, serves as the sound element, and its meaning is irrelevant. Although the
semantic element and the phonetic element in such a compound character may in themselves
be pictograms when they stand alone as individual characters, as part of a semantic-phonetic
48 Part II Writing Chinese
compound, they lose that identity and together constitute a different type of character. This
is because the linguistic process of forming a semantic-phonetic compound is categorically
different from that of a pictogram, and the outcome is also categorically different.
How are semantic-phonetic compounds created? They usually are the result of two
consecutive processes: extension and differentiation. Extension is the process by which an
existing character is used to represent a morpheme that has a similar pronunciation (pho-
netic extension) or meaning (semantic extension). For example, 馬 (马) initially represents
the morpheme mǎ ‘horse’ in Chinese. Later on, the same character is used to also write the
morpheme mā ‘mother,’ because mā has a similar sound to mǎ ‘horse.’ Thus, by phonetic
extension, the character 馬 (马) now is used to represent two unrelated meanings: “horse”
and “mother.” Here, the use of the symbol is extended purely for its sounds, and its meaning
does not matter, so it is a case of phonetic extension.
The use of an existing symbol, such as a pictogram, to represent a similar-sounding
morpheme unrelated in meaning is referred to in linguistics as the rebus principle. The rebus
principle operates in phonetic extension. Many early writing systems used the rebus principle
to represent abstract ideas with pictograms, as shown in the Chinese 馬 (马) example. For this
principle to work, however, the symbols involved do not have to be pictograms. In modern-
day texting, for example, “c u l8er” representing “see you later” contain three rebuses: “c” for
“see,” “u” for “you,” and the number “8” for “ate” in “later.”
Extension alone does not create new characters. Rather, it introduces new uses of existing
characters. The creation of semantic-phonetic compounds is the result of a differentiation pro-
cess following phonetic or semantic extension. In the case of 马/馬, representing two morphemes,
mǎ ‘horse’ and mā ‘mother,’ with the same character eventually became confusing, and the need
arose to be able to differentiate the morphemes from each other. Thus, a semantic component 女
‘female’ was added to 马/馬 to form a new compound character 妈/媽 mā ‘mother.’ As explained
earlier, in the new compound, 马/馬 (mǎ ) signifies the phonetic value only. After the differen-
tiation process, the stand-alone character马/馬 (mǎ ) represents exclusively the morpheme mǎ
‘horse,’ and 妈/媽 mā represents ‘mother.’ So again, it is the differentiation following the exten-
sion that generates new characters, and these characters are semantic-phonetic compounds.
Table 7.2 shows different types of Chinese characters as percentages of the total char-
acter inventory over time. Before we look at the numbers in detail, take a look at the first
row and note that Chinese characters are traditionally classified into four categories: pic-
tograms (象形字 xiàngxíng zì), abstract pictograms (指事字 zhı̌shì zì), semantic-phonetic
compounds (形声字/形聲字 xíngshēng zì), and semantic-semantic compounds (会意字/會
意字 huìyì zì). Today, however, not all of these categories are accepted as linguistically accu-
rate. We know that pictograms and abstract pictograms are the product of pictography, and
semantic-phonetic compounds result from the extension-differentiation process. The origin
and status of semantic-semantic compounds, however, have been called into question. Some
scholars consider it an “artificial, retrospective category,”6 possibly the result of pedagogi-
cal inventions. They believe that the ostensibly semantic-semantic compounds may in fact
be semantic-phonetic compounds, and because the phonetic components have undergone
substantial phonological changes, their sound value is no longer apparent to MSC speakers.
To make the characters easier to memorize, instructors may have created stories that reinter-
pret the meaning of the character as a sum of its components. Take the character 明 (míng
‘bright’) as an example:
Chapter 7 The Chinese writing system 49
Instructors may say something like “when the sun and the moon are put together, it is very
bright.” Over time, these mnemonic stories may have become popular beliefs about the ori-
gins of the characters.
Now, looking at the data in Table 7.2, what patterns do you observe? As you can see,
along with the increase in the total number of characters, the percentages of pictograms, ab-
stract pictograms, and semantic-semantic compounds decrease. On the other hand, the per-
centage of semantic-phonetic compounds dramatically increases. By the 2nd century ce that
category became about four times as large as all the other categories combined. This trend
continues. By the 18th century, semantic-phonetic compounds were estimated to constitute
about 97% of characters in common use.7
Table 7.2 Different types of characters as percentage of total inventory over time
Period Pictogram Abstract picto- Semantic- Semantic- Number of
% gram semantic phonetic characters
% compound compound
% %
12th–11th 23.9 1.7 34.3 28.9 1,155
century bce
2nd century ce 3.8 1.3 12.3 81.2 9,475
8th century ce 2.5 0.5 3.1 90.0 24,235
Source: Adapted from Rogers (2005, p. 45)
Companion Website
standard because it was the style required by the civil service examination in imperial China.
Another popular style in modern handwriting is the running script (行书/行書 xíngshū), in
which the strokes are more fluid and have a tendency to connect. When the strokes become
even more fluid and connected, and some characters are simplified, the style is known as
the cursive script (草书/草書 cǎoshū). Standard, running, and cursive scripts developed at
around the same time, in the 2nd-–4th centuries bce, and they were all derived from an ear-
lier style called the clerical script (隶书/隶書 lìshū). The clerical style was designed to meet
the demand for increased official documentation, and, indeed, it could be written much faster
than its predecessors, the seal scripts (篆书/篆書 zhuànshū), including great seal (大篆
dàzhuàn) and small seal (小篆 xiǎozhuàn). The seal scripts received their names retrospec-
tively when these styles were no longer used for mainstream writing and were mostly reserved
for seal carving. Before the seal scripts, and, for a long time, the so-called bronze script (金文
j ī nwén) was used. It was the style of inscriptions cast on bronze ceremonial vessels. The ear-
liest archeological evidence we have so far for Chinese writing is carving of characters on ox
bones or turtle under-shells used for divination some 3,000–3,500 years ago. These characters
are collectively referred to as the oracle-bone script (甲骨文 jiǎgǔwén). Figure 7.1 shows a
few examples of the various script styles.
One major principle for classifying Chinese characters is to group them by their
common components. If you have some knowledge about Chinese characters, you may un-
derstand that there are basically two types in terms of their structural composition: one is
composed of a single graphic element, such as 马/馬 (mǎ ‘horse’) and 女 (nǚ ‘female’), and the
other contains more than one element, such as 妈/媽 (mā ‘mother’). Characters of the second
type often have a component that reoccurs in a number of other characters. In the case of
妈/媽, for example, the 女 element is also a part of 姐 (jiě ‘older sister’), 妹 (mèi ‘younger sis-
ter’), 好 (hǎo ‘good’), 姓 (xìng ‘surname’), 婴/嬰 (yīng ‘infant’), 婪 (lán ‘greedy’), and hundreds
of other characters. It seems natural, therefore, to group these characters together under their
shared component 女.
In fact, this is how most Chinese dictionaries index the characters. A modern print
dictionary typically contains a character index divided into some 200 sections, each headed
by the graphic element that characters in that section share. Such elements, therefore, are
referred to as section headers (部首 bùshǒu). Another term for section headers is radicals,
probably because they are seen as graphic roots for generating characters. Characters with the
same radical are grouped in the same section in the character index, as shown in Table 7.3.
Within the same section, characters are arranged by the number of strokes they have,
from the fewest to the greatest. Sections of characters are then arranged based on the number
of strokes of their headers or radicals, also from the fewest to the greatest. The radical system
developed for the Kāngxī Dictionary contained 214 radicals. It has served as a de facto stan-
dard for Chinese dictionaries compiled thereafter. Contemporary dictionaries typically make
use of 180–200 radicals, which is usually the Kāngxī radicals minus one or two dozen that
form characters no longer in common use.
Kāngxī, however, was not the first dictionary that had made use of radicals to orga-
nize Chinese characters. This practice dates back some 1,600 years before to a dictionary
called Shuōwén Jiězì (说文解字/說文解字 lit. ‘explaining graphs and analyzing characters’)
compiled by Xǔ Shèn (许慎/許慎 58–148 ce). Completed in 100 ce during the Eastern Hàn
dynasty, Xǔ Shèn’s dictionary was not the earliest in China, but it was the first to analyze
the structure of Chinese characters in order to provide a rationale behind their organiza-
tion. It is regarded as “a major conceptual innovation in the understanding of the Chinese
writing system,”11 and the radical system made it much easier to look up characters. Most of
the radicals in Shuōwén Jiězì indicate the meaning of the characters, but some represent the
pronunciation.
Chapter 7 The Chinese writing system 53
Companion Website
Layout conventions
The page layout of Chinese text changed in recent history. Up until the early 20th century,
Chinese had typically been written in columns from right to left on a page as a matter of con-
vention. Pages in a book were also arranged to be read from right to left. Thus, the front cover
of a book opened on the left, and the reader turned the pages from left to right as she read
on. Now modern books usually follow the Western convention, with text written horizontally
from left to right on a page and the front covers opening on the right. However, the old format
is sometimes used to invoke a sense of tradition, especially on books about historical subjects,
as a style choice or a marketing device.
Native English speakers learning Chinese may find it challenging to read and write Chi-
nese text because of how characters are arranged on a page. Whether they are written or print-
ed in rows or in columns, Chinese characters are placed in imaginary squares. In other words,
each character is roughly the same size in a square shape. Schoolchildren in China typically
practice writing on grid paper, fitting each character in one square cell, as shown in the top
row of Table 7.4. Beginning students of Chinese who are native English speakers sometimes
make the assumption that characters in the same words need to be written closer together,
and may produce something like what is in the bottom row when they are expected to write
as shown in the top row. Unlike writing in English, however, word boundaries are not marked
in Chinese. Characters are written next to each other with even spacing regardless of whether
they are part of the same word.
今 天 天 气 很 舒 服 。
今天 天气 很 舒服 。
Translation: 今天 jīntiaˉn ‘today’; 天气/天氣 tiaˉnqi ‘weather’; 很 hěn ‘very’; 舒服 shuˉ fu ‘comfortable’.
“The weather today is very comfortable.”
least two reasons. First, Chinese characters represent meaningful units of the language (i.e., mor-
phemes). Although word boundaries are not marked, characters naturally mark morpheme bound-
aries. Proficient readers are familiar with how morphemes combine to form words, so delineating
word boundaries mentally as they read may not be very difficult. In writing English, however, letters
are not meaningful units, and morpheme boundaries are not consistently marked, so visually there
are no clear units of meaning. This reality may interfere and confuse the reader when she tries to
make use of her lexical knowledge to parse the text. For example, identifying words in the following
string of letters may produce the, rear, ear, spa, aces, cat, ate, or, and aries, none of which is in
the intended text “there are no spaces to indicate word boundaries.” As the text gets longer, it gets
even more confusing.
therearenospacestoindicatewordboundaries
A second reason is that the Chinese script contains a much greater variety of symbols. In a
lengthy text, Chinese characters can be much more easily distinguished from each other than
Roman letters. This also aids the reader in making out the words to comprehend the text.
Notes
1 This view is generally accepted. However, disagreement exists among scholars. For example, see the
blog post here: http://languagelog.ldc.upenn.edu/nll/?p=3954.
2 The oracle bones are the earliest archeological evidence for Chinese writing that has been officially
recognized. Inscriptions on artifacts of an even earlier period were discovered between 2003 and
2006 in excavations of the Neolithic-era Liángzhǔ (良渚) relic site near Shanghai. These symbols are
etched on more than 200 fragments of ceramic, stone, jade, wood, ivory, and bone, and are dated
some 1,500 years before the oracle-bone script. Scholars still debate as to whether these markings
should be considered writing in the strict sense. See Tang (2013) for more information.
3 In the word dàren (大人, ‘adult’), the morepheme rén (人, ‘person, people’) is typically pronounced
in a neutral tone in MSC. This is why ren does not carry a tone mark.
4 Wang, Hsu, Tien, and Pomplun (2014)
5 This framework is based on Rogers (2005).
6 Boltz (1996, p. 197); quoted by Rogers (2005, p. 37).
7 DeFrancis (1989, p. 99).
8 Ibid.
9 These standards refer to the National Standards for Chinese Language Courses of Full-time Compul-
sory Education (全日制义务教育语文课程标准), published in 2011.
10 Hue and Hu (2003).
11 Boltz (1993, p. 431).
12 Homophones are ordered by their tonal value in the order of first, second, third, fourth, and neutral
tones.
8 Demythifying the Chinese script
A
LTHOUGH THE CHINESE LANGUAGE (speech) is not necessarily more complex
than any other natural language, it has been commonly perceived as one of the most
difficult to comprehend. An interesting illustration of this point is a “map of incom-
prehensibility,” created by linguist and professor Mark Lieberman based on sayings equivalent
to the English “it’s Greek to me” in a variety of languages. If the language thought to be pro-
hibitively incomprehensible to English speakers is Greek, then what is it to the Greeks? You
guessed right – it is Chinese. The Greeks are not alone, however. It turns out that native speakers
of Dutch, French, Hebrew, Hungarian, Latvian, Lithuanian, Polish, Portuguese, Russian, and
Spanish all point to Chinese when they find that a language is difficult to understand. On the
map of incomprehensibility, although not all arrows are directed to Chinese, Chinese apparently
stands out for being on the receiving end of far more of them than any other language. Now, the
question is, what do the Chinese say when they encounter a language impossible to understand?
Well, instead of pointing to another language, the Chinese look upward at the sky – they say
“it’s heavenly script (天书/天書 tiānshū) to me.” Perhaps even the Chinese themselves cannot
find another worldly script that is more difficult than their own.1 Indeed, probably thanks to
its complex writing system, Chinese has been perceived as one of the most difficult languages.
Whether the Chinese script is the most difficult to comprehend can be a matter of de-
bate. There have been, however, a number of popular myths surrounding the writing sys-
tem. In fact, Victor Mair once commented, “there is probably no subject on earth concerning
which more misinformation is purveyed and more misunderstandings circulated than Chi-
nese characters (漢字, Chinese hanzi, Japanese kanji, Korean hanja) or sinograms.”2 What are
these misunderstandings? If you recall, we have in fact talked about a few of them earlier in
this book: the ideas that Chinese characters are pictures, each Chinese character tells a story,
or speakers of mutually unintelligible Chinese languages can read and understand the same
newspaper. John DeFrancis summarized such misconceptions into six myths: the ideographic
myth, the universality myth, the emulatability myth, the monosyllabic myth, the indispens-
ability myth, and the successfulness myth.3 In this chapter, we will focus on the first two: the
ideographic myth and the universality myth, because they are more closely related to the
topics in Parts II and III of this book.
speech sounds. This is because, just like any other script, the Chinese writing system is a sys-
tematic set of symbols designed to make speech visible. It cannot bypass the language (speech)
it writes to convey meaning and still do so in a fully functional manner. Perhaps because
Sinitic characters have been one of the most salient visual symbols of East Asia and have long
embodied the Western fascination with the mysterious Far East, this insight had evaded spe-
cialists and amateurs alike for centuries in the West. On the other hand, Chinese scholars have
contributed to the dissemination of the ideographic myth as well. It is not rare even till this
day to see the semantic dimension of the characters emphasized in scholarly publications in
Chinese, for example, by labeling the writing system a biaˇ o yì wénzì (表意文字), a “mean-
ing-representing script,” as if the characters have little to do with speech sounds. The dif-
ference between the morpheme-based Chinese writing system and Western phoneme-based
alphabets have been accentuated to such an extent that the Chinese writing system has come
to be perceived as almost the complete opposite of the alphabetic script.
Yet in fact, the notion of an ideographic writing system is in itself a myth. That is, no
fully functioning scripts consist primarily of visual symbols that represent meaning without
referencing speech. To further understand this point, let’s take a step back and think about
a larger question: how do languages represent meaning? Languages work through mutual
agreement among members of their speech communities. People who speak the same lan-
guage and wish to communicate in that language with each other must agree on how it works.
At the morphological level, this means they must agree on the meaning of the individual
morphemes. For example, English speakers must accept that the word books is composed of
the morpheme book that means “book” and the morpheme -s that indicate the plural form
of a noun. At the syntactic level, members of the same speech community must agree on
how morphemes combine to make meaningful sentences. For example, in English one must
say I have books instead of I books have. Now, it is crucial to understand that such agree-
ments are almost always arbitrary. That is, there is little inherent to the speech sounds /book/
that decides that its meaning should be “book.” As a result, such agreements are subject to
change over time. The morpheme book, for example, came from the Proto-Germanic word
*bōk, which meant “beech,” referring to the beechwood tablets or beech trees on which runic
symbols were inscribed. Although a historical connection exits between the two meanings,
“beech” and “book,” the connection has become a covert one and, more to the point, modern
English speakers would not need to know it in order to effectively use the word book. The
same can be said about Chinese characters, including the very small portion of pictographs
and abstract pictographs. The character 目 (mù), for example, was derived from the image of
an eye. The angular shape of the modern character, however, makes that visual connection
rather tenuous and ambiguous: Does it represent a window? A ladder? Or perhaps a highly
abstract plow? What is the basis for choosing one possibility over another? As it turns out,
modern readers simply rely on learning the arbitrary linguistic connection between the char-
acter and the morpheme it represents. That is, they learn that the speech sounds /mù/ in Chi-
nese can mean “eye,” and that when they mean “eye” they are conventionally represented in
writing by the grapheme 目They learn to use the character without having to know anything
about the visual connection between the earlier form of the character and the physical object
on which that form was based. That knowledge is more likely acquired after they have already
learned the character. In this sense, Chinese characters convey meaning by referring to the
Chinese language, and therefore they ideographs.
Chapter 8 Demythifying the Chinese script 59
As you can see, pronunciation only accounts for part of the differences between the
Mandarin and the Cantonese sentences. They also use different words, including grammatical
particles, and different word orders. The morpheme “please” in Mandarin is a single syllable
represented in a single character 请 (qı̌ng), while in Cantonese it is bi-syllabic and written with
two characters 唔該 (m4goi1). In Mandarin, the possessive marker de (的) indicates posses-
sion (‘his’) when placed immediately after the pronoun tā (他 ‘he’), while in Cantonese this
role is fulfilled by the measure word bun2 (本). What may throw the reader off the most is that
the Mandarin word order is “verb – indirect object – direct object,” while in the Cantonese
sentence it is “verb – direct object – indirect object.” This example shows that for the speakers
of one Chinese language to be able to fully and accurately comprehend the written text of an-
other, lexical and syntactic knowledge will be crucial as well as the ability to read and under-
stand the written characters. In other words, one must know the language behind the writing.
Yet it is indeed a common observation that many Cantonese speakers do read the same
newspapers as the Mandarin speakers. They can even read out loud, character by character,
text in the newspaper in the Cantonese pronunciation. How do we understand that? At least
two factors may have contributed to this phenomenon. First of all, educated speakers of Can-
tonese are generally speakers of Mandarin as well. Mandarin or Pǔtōnghuà (普通话/普通話
‘common speech’) is the official language of the People’s Republic of China. Since the 1950s,
the PRC government has made sustained efforts in promoting the common speech through-
out China. Teachers must be certified in Mandarin proficiency and students of non-Mandarin
regions are required to learn to speak, read, and write Mandarin in schools and universities.
In addition, official broadcasting and publications, especially those at the national level, are
done in Mandarin in almost the entirety of China today. Aside from school education, speak-
ers of non-Mandarin Chinese are regularly exposed to Mandarin programming on television,
the Internet, and other venues of social communications. As a result, they become proficient
Mandarin readers and writers essentially by learning it as an additional language. It is no
surprise that they are able to read the newspapers when the great majority of newspapers
published in China are written in Mandarin.
Another factor at play is the existence of cognates among the various Chinese languag-
es. Cognates are words in related languages that have been derived from the same ancestral
form. They are typically similar in meaning but are pronounced somewhat differently. For
instance, gratitude in English and gratitud in Spanish are a pair of cognates. In fact, there are
many cognates between English and other Indo-European languages. The English beer and
the German Bier are another example. Based on the estimates mentioned earlier, the Sinitic
languages share between them on average about 60% of the vocabulary, which may be under-
stood as composed of cognates. In the example we looked at earlier, for instance, shū ‘book’ in
Mandarin and syu1 ‘book’ in Cantonese are such a pair. The existence of many cognates allows
Cantonese speakers to articulate words written in Chinese characters in the pronunciation of
Cantonese, even though it is really Mandarin that has been written. The result of such render-
ing, however, is something of a cross between Mandarin lexicon and syntax and Cantonese
Chapter 8 Demythifying the Chinese script 61
By this point, we see that sharing the character-based Chinese script as a common writ-
ing system does not in itself enable fully functional communication between speakers of mu-
tually unintelligible languages. But may a morphographic writing system like the Chinese
fare better, still, than a phonemic script like the Roman alphabet, in potentially serving as a
universal script? When we think of the languages written in the Roman alphabet in addition
to English – Afrikaans, Estonian, Icelandic, and Turkish, to name just a few – and how little
we may be able to read the text written in these languages if we have not studied them, it may
not be difficult to appreciate the futility of a universal script that is alphabetical. Indeed, its
phonemic nature means that differences in the sounds of the languages are inevitably cap-
tured and reflected in the spelling of the words. This is not the case with Chinese characters.
Although they represent speech sounds just like any other writing system does in the sense
that they make speech visible, the connections between the sounds and the symbols are largely
opaque, and, to a great extent, one must learn these connections by memory. This opaqueness,
however, is the very property that makes it possible to link a character to sounds (morphemes)
62 Part II Writing Chinese
from different languages. For instance, the character 海 (‘sea, ocean’) in 上海 (Shànghǎi) may
not only be read as /hai/ in Mandarin, /hɛ/ in Wú, and /hoi/ in Cantonese, but also /umi/ in
Japanese, /hae/ in Korean, /biển/ in Vietnamese, /sea/ in English, /mar/ in Spanish, and so on.
This idea was perhaps what gave rise to the thought of using the Chinese writing system as a
universal script in the first place. Enticing as it is, we should keep in mind that, as we have seen
in the case of Cantonese and Mandarin, a shared script even between closely related languages
may not work that well.
Notes
1 Be aware that this example confuses “language” and “script,” two distinct linguistic concepts. Al-
though the Chinese writing system may be significantly more complex than other scripts, the Chi-
nese language (speech) is not necessarily so compared to other languages.
2 See the foreword to the book Ideogram by J. Marshall Unger (2004).
3 DeFrancis (1984).
4 Refer to Rogers (2005, p. 45).
5 DeFrancis (1989).
6 John K. Fairbank, quoted by Kraus (1991) on p. 6. The other one of “the two great institutions” in
this quote is the ruling elite.
7 These are IPA symbols.
8 Tone marks are omitted in these examples to reduce complexity, but Mandarin, Wú, and Cantonese
have different tonal systems.
9 DeFrancis (1984, p. 63).
10 Jyut6ping3 is a Romanization system for Cantonese designed by the Linguistic Society of Hong
Kong in 1993. The numerals mark tones.
11 This example is adapted from an example in the “List of diglossic regions” entry on Wikipedia,
http://en.wikipedia.org/wiki/List_of_diglossic_regions.
12 See Snow (2004).
PART III
I
T IS WELL KNOWN that the Chinese script, or hànzì (汉字/漢字 ‘Chinese charac-
ters,’ Japanese kanji, Korean hanja, Vietnamese hán tự), has been used not only for rep-
resenting the Chinese language but also for writing Korean, Japanese, and Vietnamese.
Until recent history and for thousands of years, the character-based script had been the
primary writing system throughout East Asia. This shared script, along with a common
written language in Classical Chinese (aka Literary Chinese) style and a shared lexicon in
historic and modern forms of the languages, served as the foundation for a prevailing East
Asian textual culture among the neighboring cultures of Korea, Japan, Vietnam, and Chi-
na. While maintaining its diverse identities,1 the region was also distinguished by a shared
heritage in Confucian teachings and values, Mahayana Buddhist beliefs and practices,
as well as Chinese legal codes.2 Perhaps due to the visual distinctiveness of hànzì, this re-
gion was sometimes referred to as the “Chinese characters cultural sphere (汉字文化圈/
漢字文化圈 hànzì wénhuà quān).”3 In this chapter, we take a broad overview at the role
of the Chinese writing system in China’s neighboring cultures, and we will more closely
examine how the Chinese script was borrowed to write Korean, Japanese, and Vietnamese in
Chapters 10–12.
FIGURE 9.1 Map of East Asia at the beginning of the 3rd century ce
The Chinese writing system spread to Korea and Japan via the dissemination of lit-
erature written in Classical Chinese, including the Confucian classics and Buddhist sutras.
Koreans became aware of Chinese writing before the 2nd century bce4 – possibly the earliest
outside of China. In the 4th century ce, Confucian classics were taught in Korea and Koreans
were also writing Buddhist scriptures in hanja. Sometime later in the 4th or the 5th century,
Korean scholars went to Japan to teach Classical Chinese text to the Japanese, further spread-
ing the use of the Chinese script.5 By this point, speakers of Korean and Japanese were learning
to read and write Literary Chinese in Chinese characters. In doing so, they might have read
and recited out loud the Classical Chinese texts following local conventions in pronunciation.
In ordinary speech, of course, they used their native languages. Neither the Koreans nor the
Japanese, however, wrote down their native languages before their exposure to Chinese texts.
For a long time, becoming a learned scholar meant learning to read Classical Chinese and to
compose literature in that style using the Chinese script. Classical Chinese, much like Latin in
the European context, served as the common literary language of East Asia.
The situation in Vietnam was somewhat different. Northern Vietnam was ruled by
China for more than a thousand years, from 111 bce to 939 ce, and Classical Chinese was
introduced there in as early as 186 ce as the written language used by the governing Chi-
nese officials.6 The Vietnamese language, on the other hand, was not written on a large scale
until after Vietnam gained independence from China. A Vietnamese script based on Chinese
characters was then developed to represent the Vietnamese vernacular. For official writing
purposes, however, Literary Chinese represented in the Chinese script continued to dominate
until the late 19th century.
Having served as the written language for China, Korea, Japan, and Vietnam for millen-
nia, Classical Chinese encoded in Chinese characters became the foundation of a common
66 Part III Borrowing the Chinese writing system
textual tradition in East Asia. As some scholars argue, from the perspective of the non-
Chinese cultures, Classical Chinese – and the revered body of works that was written in the
language – should not be considered uniquely Chinese. It “was as much Vietnamese, Japanese
or Korean as it is [was] Chinese.”7 This shared textual medium sometimes allowed scholars
from different linguistic backgrounds to communicate in writing, albeit in a limited manner,
when direct verbal exchange could be difficult. Peter Kornicki, for instance, offers one illumi-
nating example of “brush talks”:8 a Korean envoy to China and his Vietnamese counterpart
were able to exchange poems written in Literary Chinese when they met at China’s capital in
the 16th century. Yi Sugwang (李睟光, 1563–1628), the envoy from Korea, wrote:
“The writings of the ancient sages” referred to Confucian classics, including the Classic of
Poetry (诗经/詩經 Shījīng) and the Classic of Documents (书经/書經 Shūjīng), part of what
formed a shared literary and philosophical canon among the East Asian societies.
The linguistic and textual tradition of East Asia was institutionally sustained through
the imperial examination system, the civil service examinations (科举/科舉, Chinese kējǔ,
Korean gwageo, Vietnamese khoa-cử) established in China (605–1905 ce) and adopted in
Korea (788–1894 ce) and Vietnam (1075–1913 ce).9 The examinations functioned as a the-
oretically egalitarian system to select able adult males to the ruling class regardless of their
social status or family pedigree and was considered a progressive meritocracy for most of its
history. To compete in the exams, candidates had to be well-versed in Confucian classics, be
able to compose in the style of Classical Chinese essays and sometimes poetry, and be mas-
ters of calligraphy in at least the standard script (楷书/楷書 kǎishū). Under the examination
system, literacy in Literary Chinese was a basic educational requirement throughout the area.
Owing to the shared literary and philosophical traditions rooted in Confucian and Bud-
dhist texts and the use of the same script, a large number of Chinese words were borrowed
into Korean, Japanese, and Vietnamese (KJV). To prepare for the civil service exams, candi-
dates mastered Chinese words through reciting the texts and composing prose and verses
in Classical Chinese, so it is hardly surprising that they gradually added Chinese vocabu-
lary to their native lexicon. In today’s terms, these borrowed words are referred to as Sino-
Korean, Sino-Japanese, and Sino-Vietnamese words, where the prefix “Sino-” refers to China.
However, they have long been integrated into the vocabulary of the KJV languages so that
native speakers may be oblivious of their Sinitic origin.
There are several sources for Sino-KJV lexical items. Among them, non-Chinese learn-
ers importing words and expressions from Classical Chinese into their native languages may
be the earliest and most important one. Interestingly, for the same target language this process
took place multiple times at different points in history. In the case of Vietnamese, for exam-
ple, specialists distinguish at least two layers of Sinitic loanwords: the Old-Sino-Vietnamese
Chapter 9 Chinese characters in Asia 67
(OSV) layer borrowed during the Eastern Hàn dynasty (25–190 ce) and the more recent
Sino-Vietnamese (SV) layer borrowed during the Táng dynasty (618–907 ce).10 Thus, the
same word may have been borrowed more than once and end up with multiple pronuncia-
tions due to sound change in both the Chinese and the Vietnamese languages. For example,
the morpheme for “shoe,” represented by the character 鞋, was borrowed first as giày from
the Old Chinese *gre11 and then again as hài from the Middle Chinese hɛ. Similarly, the mor-
pheme for “(for a female) to marry to someone,” written as 嫁, first came into OSV as gả from
the Old Chinese *kras and then from the Middle Chinese kæH into SV as giá.12
Another important source of Sino-KJV vocabulary is creation of new words based on
the Classical Chinese model. Some of these words were coined in classical times. They often
represented native concepts that had no counterparts in Chinese: for example, maccha (抹茶
‘powdered green tea’), jūdō (柔道 lit. ‘gentle way,’ a sport of unarmed combat), and haiku (俳
句 a traditional form of Japanese poetry) in Japanese. Sometimes, they represented concepts
that did also exist in Chinese, such as jujeonja (酒煎子 ‘kettle, teakettle’) or pyeonji (便紙
‘letter’) in Korean.
Many Classical-Chinese-style new words were created in the modern era to trans-
late Western concepts that were not available in either Chinese or the native language.
For instance, linh mu.c (靈牧 ‘pastor’) in Vietnamese was coined by combining the mor-
phemes for “soul (靈)” and “shepherd (牧).” Creation of this kind was particularly prolific
during the Meiji period (1868–1912) in Japanese, and, interestingly, many compounds
coined at that time were not much later back-borrowed and thoroughly integrated into
Modern Standard Chinese. Examples included diànhuà (电话/電話 ‘telephone’: ‘electric’
+ ‘speech’) from the Japanese denwa, kēxué (科学/科學 ‘science’: ‘classification’ + ‘study’)
from kagaku, and shèhuì (社会/社會 ‘society’: ‘community’ + ‘meet’) from shakai. Some
were imported into Korean, for instance, yeonghwa (映畫 ‘movie’: ‘shine’ + ‘picture’) from
Japanese eiga (映画13) and gongjang (工場 ‘factory’) from kōjō (工場), or into Vietnam-
ese. The Chinese themselves also translated Western works and created new words in the
process, some of which were then borrowed into the other languages. For example, the
word dàishùxué (代数学/代數學 ‘algebra’: ‘replace’ + ‘number’ + ‘study’) invented by a
Chinese mathematician became dai sūgaku in Japanese, dae suhak in Korean, and đa.i số
ho.c in Vietnamese.14
As you can see, the creation and borrowing of Chinese or Chinese-style words among
the East Asian cultures were dynamic and multi-directional. Often multiple compounds
were created around the same time for the same concept and would coexist until a win-
ner emerged. Though sometimes the winner differed between languages,15 often the same
form prevailed. As a result of the cross-borrowing, a substantial (though limited) common
lexicon may be observed in the modern forms of the languages. It is estimated that about
60% of Korean words are considered Sino-Korean.16 Similarly in Japanese about 60% of the
words in a typical modern dictionary are Chinese in origin.17 Sino-Vietnamese words may
account for about one-third of the Vietnamese vocabulary when the register is informal,
and in formal text, that percentage goes up to also about 60%.18 Table 9.1 shows a few exam-
ples from Chinese, Korean, and Japanese. It is remarkable that a shared vocabulary, limited
as it is, could exist on such a scale among four languages that are, as we will discuss further
in this chapter, not genetically related. A shared written language encoded in a common
writing system may have made it possible.
68 Part III Borrowing the Chinese writing system
Why borrow?
You may wonder why the people speaking Korean, Japanese, and Vietnamese borrowed a
writing system to record their language instead of creating their own script. To begin with,
the borrowing of writing systems is very common, and inventing a writing system from
scratch is, in fact, quite rare. Writing was invented independently only about three times in
history: some 5,000 years ago by the Sumerians, about 3,500 years ago by the Chinese, and
more than 2,000 years ago by the Mayans. Since ancient times, as cultures came into contact,
the idea of writing also spread, and more and more languages came to be written. In almost
all cases, however, an existing writing system or a derived form was adopted to record the
new language.19 The Roman or Latin alphabet, for example, was the outcome of a series of
borrowings: the Romans borrowed it from the Etruscans in Italy, who borrowed their script
from the Greeks. The Greek alphabet came from the Semitic writing system of the Egyptians,
which was inspired by the script of the Sumerians in Mesopotamia.20 Chinese may be the only
modern language the writing of which did not involve borrowing from a pre-existing script
to some extent.21
Another factor that led to the adoption of the Chinese script in writing other East Asia
languages was the perceived prestige of the Chinese culture. When Korea and Japan came into
contact with China and for the 1,500 years after that, the powerful and sophisticated Middle
Kingdom (中国/中國 Zhōngguó ‘China’: 中 ‘center, middle’; 国/國 ‘country, kingdom’) was
regarded the center of the world and was held in absolute high esteem. Educated Koreans, Jap-
anese, and Vietnamese absorbed Classical Chinese texts and wrote in that style using Chinese
Chapter 9 Chinese characters in Asia 69
characters before their native languages were written down. Therefore, when the time came to
write their own languages, the Chinese script was adopted as a natural choice.
As we discussed in Chapter 3, theoretically speaking, any language can be written using
any writing system. That is, a language is not inherently associated with a particular (type) of
writing system. The pairing between a language and its conventional script is precisely that –
conventional. This is because, unlike language, for which “native speakers” exist, a writing
system does not exactly have “native writers” (when we used this term, we used it to mean
writers who were native speakers). Language (speech) is naturally acquired and the acquisi-
tion outcome is biologically conditioned. Beyond the first dozen years or so – termed by lin-
guists as the critical period – it becomes very difficult for humans to achieve full proficiency
in a new language. But learning to write is a different matter. A learners’ potential proficiency
is not critically conditioned by age, and anyone with reasonable general competence should be
able to fully learn a new script given the appropriate amount of time and effort.
This is not to say, however, that the conventions we have for writing various languages
are insignificant; quite the contrary, as we will see later in this book, the choice of one writing
system over another often carries with it hefty cultural meaning and may signify the sociopo-
litical identity of an institution or the writers. In this sense, we may feel that the conventional
writing system of a language is more “native” than other alternatives. It is also in this sense
that we use the term “native script” to refer to the writing system customarily employed to
write a language.
Table 9.2 ommon strategies for borrowing Chinese characters to write Korean, Japanese, or Viet-
C
namese22
Strategy Examples
Source (Chinese) Target (K/J/V)
(1) Sound-only [C] 五 wǔ ‘five’ [K] 五 o ‘(a verb ending)’
borrowing
[C] 春 chuˉn ‘spring’ [K] 春 pom ‘spring’
(2) Meaning-only
[C] 心xˉın ‘heart’ [J] 心kokoro ‘heart’
borrowing
[C] 海 hǎi ‘sea’ [J] 海 umi ‘sea’
[C] 春 chuˉn ‘spring’ [K] 春 ch‘un ‘spring’
(3) Sound-meaning
[C] 心xīn ‘heart’ [J] 心shin ‘heart’
borrowing
[C] 海 hǎi ‘sea’ [J] 海 kai ‘sea’
70 Part III Borrowing the Chinese writing system
Sound-only borrowing disregards the meaning of the characters in Chinese and focuses
on their Chinese pronunciation instead to represent morphemes that have similar pronunci-
ations in the target language. In the Korean example, the character 五’s meaning in Chinese
is irrelevant. The reason why it was chosen to represent the Korean grammatical marker is
because its pronunciation is close to the Korean morpheme. Meaning-only borrowing uses
Chinese characters to represent native morphemes of the same or similar meaning. The pro-
nunciation of the character follows that of the native morpheme, which is usually very differ-
ent from the Chinese pronunciation and may even have different numbers of syllables. The
Chinese morphemes are monosyllabic, while the Korean or Japanese counterparts are often
multisyllabic. For instance, written as 海, the Chinese ‘sea,’ hǎi, has only one syllable, while the
Japanese morpheme umi has two syllables. Finally, for sound-meaning borrowing, essentially
an entire Chinese morpheme is imported into the native or target language. For example, the
morpheme ‘heart’ in Chinese is pronounced as xīn in Modern Standard Chinese and written
as 心. This morpheme is borrowed into Japanese, pronounced as shin, and written as the same
character.
Now let’s learn a couple of new terms. They will be important for us to know before we
continue the discussion of borrowing strategies. You may have noticed that, as the result of
borrowing from the source language (Chinese), the same character may have multiple pro-
nunciations, or “readings,” in the target language (Korean, Japanese, or Vietnamese). This
is because the same character may represent multiple morphemes: the Chinese morpheme
that has been borrowed into the target language and the native morpheme that has the same
or similar meaning. When the character is read as the Chinese morpheme (e.g., 心 as shin in
Japanese), it is referred to as on (音 ‘sound’) reading. When it is read as the native morpheme
(e.g., 心 as kokoro), it is called kun (訓 ‘meaning, interpretation’) reading. These terms are
used in Japanese linguistics to talk about the various readings of kanji. Here we borrow them
to discuss the strategies used in adopting the Chinese script to write Korean and Vietnamese
as well.
More specifically, the three strategies outlined in Table 9.2 may be referred to in these
terms. First of all, we can call the sound-only borrowing phonetic on. Since the sound is
borrowed from Chinese, and the character is read based on its pronunciation in Chinese, the
reading is considered on reading (instead of kun reading). The character 五 as a verb ending in
Japanese is pronounced as o, similar to the Chinese pronunciation of this character. Although
the pronunciation of this character in Chinese has changed, one can still detect the similarity
between the modern Japanese reading o and the modern Chinese pronunciation wǔ. Then
why do we refer to this kind of borrowing as “phonetic” on? This is because only the sound is
borrowed in this case and the meaning is not. When the meaning is borrowed together with
the sound, we use a different term: semantic on refers to sound-meaning borrowing. In this
case, both the pronunciation and the meaning of the Chinese morpheme are borrowed into
the target language. As an example, the character 春 ‘spring’ in Korean can be read as ch‘un, a
pronunciation based on Chinese. It is in fact very similar to the modern Chinese pronuncia-
tion of the same character, chūn (春 ‘spring’). Thus, this borrowing of 春 ‘spring’ from Chinese
into Korean constitutes a semantic on. The difference between phonetic on vs. semantic on
is in whether the character is borrowed purely for its phonetic value (phonetic on, sound-
only borrowing) or whether the meaning of the morpheme is imported as well (semantic on,
sound-meaning borrowing). The third strategy, meaning-only borrowing, may be referred to
as a semantic kun. In this case, the character is read not in its Chinese or source pronunciation,
Chapter 9 Chinese characters in Asia 71
but in its native or target pronunciation. Its meaning is consistent with the character’s meaning
in Chinese. For instance, 海 ‘sea’ in Japanese, as a semantic kun, is pronounced as umi, a native
Japanese morpheme that is very different in sound from the Chinese morpheme hǎi (‘sea’).
Is there also phonetic kun, you may ask? Yes, there is. A phonetic kun involves a character
borrowed to represent a morpheme in the target language that is not apparently connected to
the source language in either sound or meaning. This is usually the accumulative effect of two
or more borrowing processes. For example, the character 加 in Modern Standard Chinese is
pronounced as jiā and has the meaning “to add” or “more.” The character was first borrowed
into Korean via semantic on and was read as ka with the same meaning. Then it was used to rep-
resent the native morpheme do ‘more,’ making it a semantic kun, losing its Chinese sound. After
that its usage extended again to writing the verb ending -hadoni ‘do so and’ due to sound simi-
larity, shedding its Chinese meaning. At this point, 加 has transformed into a phonetic kun and
no longer has any apparent connection to the Chinese original in either sound or meaning.23
If all this may seem a little too abstract, let us take a step back and do an exercise with
English. Our task, let us assume, is to write the English utterance /the flowers are pretty/ using
Chinese characters. We use // to indicate that it is the speech that is being represented. So, what
is it like to record this utterance using Chinese characters? There may be a range of possibilities.
First of all, we can use the first strategy (i.e., sound-only borrowing) to write the entire
sentence. To do this, we just need to choose characters that have similar sounds in Chinese
and use them to represent the English. This should be relatively straightforward. Because
many characters have the same pronunciation, here we choose characters that are neutral or
positive in meaning, as is generally the practice representing English proper nouns in Chinese.
The pronunciation of the English utterance remains exactly the same, but now it is written in
Chinese characters. These characters are read as phonetic on, because their pronunciation
in writing this sentence is close to their original Chinese pronunciation. The meaning of the
characters is irrelevant. This is the first approach.
Another approach is to write the English words, where possible, through meaning-only
borrowing of Chinese characters. This is not difficult to do for /flowers/ and /pretty/, where
we can simply use the Chinese equivalents: 花 (huā ‘flowers’) and 漂亮 (piàoliang ‘pretty’).
So, instead of writing four characters福拉沃玆 for /flowers/ we now only need one character
花, but we keep the pronunciation of /flowers/. So now we’ve imported the character and its
meaning but changed its pronunciation to accord with the spoken English word /flowers/.
Similarly, /pretty/ is now written as 漂亮. However, there are no equivalents of the words
/the/ or /are/ in Chinese, so we cannot use this strategy on these words, and they will have to
be written phonetically. So we have:
Written as: 仄 花 阿 漂亮
Pronounced as: /the/ /flowers/ /are/ /pretty/
Chinese sounds: ze hua a piaoliang
English meaning: ‘the flowers are pretty’
72 Part III Borrowing the Chinese writing system
Now let us take it one step further: we can borrow both the meaning and the pronunci-
ation of Chinese characters. This requires importing whole words from Chinese into English.
This concept is probably familiar to you, because there are a good number of Chinese loan-
words in English that you likely know well. For example, wonton, fengshui, and ginseng come
from Cantonese, Mandarin, and Southern Mı̌n,25 respectively. These words are borrowed into
English so that their meaning and pronunciation in English are the same as or are close ap-
proximations of their meaning and pronunciation in Chinese. The difference in the current
task is that, since we are borrowing the Chinese writing system to write English, we will also
write the words in Chinese characters, i.e. 云吞/雲吞 (wonton), 风水/風水 (fengshui), and 人
参/人參 (ginseng), and read them as, respectively, /wonton/, /fengshui/, and /ginseng/. Let’s
try applying this strategy to /flower/: /huā/ (花 ‘flower’) and write it using the same character
as we do for Chinese and pronounce it more or less similar to what we do in Mandarin. Since
English does not use tones, we will drop the tones.
Written as: 仄 花 阿 漂亮
Pronounced as: /the/ /hua/ /are/ /pretty/
Chinese sounds: ze hua a piaoliang
English meaning: ‘the flowers are pretty’
Now we have incorporated all three strategies in the same sentence. 花 (/hua/ ‘flowers’)
is a semantic on because we’ve borrowed both the sound and the meaning of the character.
漂亮 (/pretty/ ‘pretty’) is a semantic kun, because we’ve borrowed only the meaning and kept
the English pronunciation. Finally, 仄 (/the/ ‘the’) and 阿 (/are/ ‘are’) are both phonetic on,
because we’ve borrowed only the pronunciation and not the original meaning.26
Again, because Chinese does not have equivalents of the words /the/ and /are/, we can-
not borrow them from Chinese. In fact, foreign loanwords in a language are almost never
words like /the/ or /are/. These words are what we refer to as function words – words that
serve certain grammatical functions – a category that includes articles (e.g., the), pronouns
(e.g., she, he, or they), and auxiliary verbs (e.g., are).27 Loanwords are almost always words like
/flower/ and /pretty/, or content words – words with concrete meaning, such as nouns, main
verbs, adjectives, and adverbs. This is because function words are a closed class. New function
words are almost never created, so there is little “room” for importing words into this cate-
gory. On the other hand, new members of content words are frequently added to a language.
Companion Website
that these languages are all related to Chinese. We know that, as previously discussed, this un-
derstanding is far from the truth. Chinese and Vietnamese are not related to each other or from
the same language family. Chinese is part of the Sino-Tibetan family, which includes Tibetan,
Burmese, and more than 450 other languages. Vietnamese is in the Austroasiatic family togeth-
er with over 1,200 other languages such as Khmer and Mon. The genetic relations of Korean
and Japanese are more controversial, but most linguists today consider them unrelated to each
other28 or to Chinese.
Not being genetically related, however, does not mean that the languages do not share
any structural or functional features. Indeed, it may be difficult to imagine two languages
that are completely different from each other, given the limited typological variations shared
among the vast number of languages. In the case of Korean, Japanese, Vietnamese, and Chi-
nese, there are important common linguistic features between them as well as significant dif-
ferences. In adapting the Chinese script to representing the speech of the other languages,
therefore, both divergences and parallels may be observed. Linguistic challenges exist in all of
these cases, but some of the challenges are similar and others are not. Here we will focus on
three linguistic aspects that matter a great deal in the consideration of writing: word order,
verb form, and morpheme shape. We will see that in all three aspects, Chinese and Vietnam-
ese resemble each other, and Korean and Japanese are similar to each other yet both different
from Chinese. This means that using the Chinese writing system to record Vietnamese is a
relatively more straightforward process than adapting it to writing Korean and Japanese, while
between the latter cases there are significant parallels in the kinds of strategies employed to
make the writing work.
First of all, in terms of word order, Chinese and Vietnamese are largely SVO29
(subject-verb-object) while Korean and Japanese are SOV. In the beginning of their contact
with China, the Koreans and the Japanese studied Chinese text and learned to read and write
Classical Chinese in Chinese characters, much like today’s students learning to read and write
in a foreign language. For a long period of time, all writing was done in Classical Chinese
following the Chinese word order, and the native languages were not recorded in writing.
Chinese texts composed in Literary Chinese in China were brought to Korea and Japan and
carefully studied by scholars. To facilitate this learning, auxiliary written symbols were added
to the original texts so that an educated Korean and Japanese scholar would be able to read
them as Korean or Japanese. This kind of marking system in Korean is called gugyeol (口訣)
‘oral formulae’ and its Japanese counterpart is called kanbun (漢文) ‘Chinese text’. Gradually,
the Chinese script was adapted to writing the native languages as well, and writing in Chinese
characters shifted into the SOV word order. In Korea, this kind of writing was called idu ‘cleri-
cal reading’, and for Japanese, it occurred when man’yōgana (万葉仮名 Man’yō ‘ten thousand
leaves’, the name of an ancient Japanese poetry anthology written using man’yōgana; -gana
‘borrowed names’), the script that first used Chinese characters to write Japanese, was adopt-
ed. We will take a closer look at each of the above scripts when we talk about writing Korean
and writing Japanese in the next two chapters.
When it comes to how verbs change form, the four languages also divide into two
camps: Chinese and Vietnamese are the so-called isolating languages. Their verbs and nouns
generally do not change forms. That is, they do not inflect or take on grammatical affixes to
indicate grammatical functions such as tense (past, present, future), number (singular vs. plu-
ral), or case (normative when a noun is the subject, accusative when the noun is the direct ob-
ject, etc.). By contrast, Korean and Japanese are agglutinative languages, in which words are
74 Part III Borrowing the Chinese writing system
formed by attaching suffixes to stems. The suffixes are meaningful units, or morphemes, that
indicate grammatical functions. Thus, a word in Korean or Japanese may contain a signifi-
cantly greater number of morphemes than in Chinese and Vietnamese. While this was not a
challenge that writers of Vietnamese needed to deal with, those who used Chinese characters
to write Korean or Japanese must come up with ways to represent the grammatical suffixes
that did not exist in Chinese. In both languages, these elements were first represented by full
Chinese characters used as phonetic symbols, then by simplified forms of the characters, and
eventually by symbols of scripts later designed to write specifically Korean or Japanese. For
Korean historically and Japanese until this day, a mixed script has been used in which the verb
stem is often written in hanja (Korean) or kanji (Japanese), and the inflectional suffixes are in
the Korean hangeul letters or the Japanese kana symbols.
Another linguistic distinction that deserves our attention is in the shape or size of the
morpheme in terms of the number of syllables. Chinese and Vietnamese morphemes are gen-
erally single syllables or monosyllabic, but the native morphemes of Korean and Japanese are
usually polysyllabic. In writing Chinese, as we learned in Part II, the majority of characters
represent monosyllabic morphemes. This one-character-one-syllable relation translated eas-
ily into Vietnamese. For writing Korean or Japanese, however, the difference in morpheme
shape introduced additional complexity. Chinese characters sometimes represented Chinese
morphemes imported into the language, which were almost always monosyllabic, while oth-
er times they represented native morphemes, which were typically polysyllabic. As a result,
there was no consistent symbol-syllable relation as there was for Chinese and Vietnamese. In
the case of Korean, when a mixed script was later adopted in which Sino-Korean words were
written in hanja and native Korean words were rendered in hangeul, a one-character-one-
syllable relationship emerged for hanja characters. For Japanese, however, kanji continued to
represent both Sino-Japanese and native Japanese words when kana was incorporated to form
a mixed script, so there continues to be varied relationships between kanji characters and the
number of syllables they may represent.
Interestingly, the one-on-one relationship between character and syllable had a lasting
impact on Vietnamese and Korean writing even after the Chinese script had been complete-
ly abandoned in writing these languages. Since the 1940s, Vietnamese has used the Roman
alphabet as its official script just like English or French. Unlike English or French, however,
Vietnamese is written morpheme by morpheme – much as Chinese characters used to be
written – rather than word by word. Words are segmented into syllables (morphemes) and
spaces are placed between syllables (morphemes) even if they belong to the same word. In the
case of Korean, this one-on-one symbol-syllable correspondence is in a sense maintained as
well. Hangeul letters are written in syllable blocks resembling Chinese characters instead of
being arranged linearly as in writing English. Each syllable block occupies a squarish space
and may contain multiple letters arranged along both the horizontal and the vertical dimen-
sions. Unlike writing Vietnamese in the Roman alphabet, however, Korean written in hangeul
does mark word boundaries with spaces or punctuation marks.
Today, Chinese characters have been largely disused in the writing of the other lan-
guages. Although they are still used in the mixed Japanese script, their numbers have been
significantly limited. In writing Vietnamese, the Chinese script has been completely replaced
by the Roman alphabet. The use of hanja in North Korea has been abolished since 1949, while
in South Korea hanja has been by and large abandoned in day-to-day writing since the 1970s.
Chapter 9 Chinese characters in Asia 75
Nonetheless, it is both interesting and instructional to learn about the borrowing of the Chi-
nese script in writing these languages. Building on the overview presented in this chapter, the
following chapters will delve into this topic in more detail.
Notes
A
S WE LEARNED IN the previous chapter, Korean was the first non-Chinese language
that the Chinese writing system was adapted to write. It is likely that Koreans were
exposed to Chinese writing as early as the 2nd century bce,1 and by the 3rd to the 4th
centuries ce educated elites were already using the Chinese script to write Korean.2 Since then
and until the Joseon dynasty (朝鮮1393–1897), Chinese characters, or hanja (漢字), had been
used in a variety of ways in traditional Korea, both to write the Korean language, as in the idu3
(吏讀, ‘clerical reading’) script and the hyangchal (鄕札 ‘vernacular letters’) script, and to an-
notate Classical Chinese texts using a system called gugyeol (口訣 ‘oral formulae’). The Korean
alphabet, hangeul (한글han ‘great,’ geul ‘writing’), was created in the 15th century, but Chinese
characters continued to dominate the written culture for hundreds of years after that. A mixed
script of hanja and hangeul emerged in the modern era. After a gradual transition since the
mid- to late 20th century, today’s Korean writing is almost exclusively done in hangeul. In this
chapter, we will take a closer look at each of the ways of writing mentioned above following
roughly their order of historical occurrence.
language or phonology, the Yale system may be more challenging as a tool for approximating Ko-
rean pronunciation.
Below are how한글 ‘Korean writing’ and한자 ‘Chinese characters’ are Romanized in the various
systems:
RR: hangeul, hanja
MR: han’gŭl, hancha
Yale: hankul, hanca
It is worth noting that the han in hangeul is unrelated to the han in hanja. The former is a native
Korean morpheme that means ‘great, right’ or represents the Korean identity, while the latter is
from the Chinese morpheme 漢 (MSC: hàn) meaning “China” or “Chinese.”
‘Even though a woman breaks the law of the seven valid reasons for divorce, there exist
three exceptions. If one expels his wife despite the exceptional cases, he will be demoted
by two grades and his wife will be sent to her husband.’
To a Chinese reader, the Sino-Korean words are readily recognizable based on the En-
glish translation: for example, 犯 ‘violate,’ 三 ‘three,’ 不 ‘not,’ 減 ‘to reduce,’ and 婦女 ‘woman’
all carry the same basic meaning as in Modern Standard Chinese. These words have become
part of the Korean lexicon.
However, the text as a whole may not be easily comprehensible to the Chinese reader.
There are a few reasons for this difficulty. First, as you can tell from this example, idu writing
was in the Korean word order. The object of a verb precedes the verb rather than following
it as in Chinese: for example, “woman-return” means “return (the) woman.” Second, besides
the Sino-Korean words, the text also contains native Korean words and grammatical suffixes.
If the Chinese reader is unversed in the Korean language, she or he may not be able to un-
derstand the meaning or the function of these elements, even if they are written in Chinese
characters. The characters are used purely for their phonetic values, and their Chinese mean-
ing is irrelevant to the meaning of the text. For instance, the Chinese meanings of 必 (in Chi-
nese: bì ‘must’) and 于 (yú ‘at’) may not add up to the meaning of the word 必于 in idu (pilok
‘although’). Such characters may confuse a Chinese reader interpreting them as meaningful
graphs.
紫布 岩乎 过希
tolpwoy pahwoy kos-oy
azelea-[bu] rock-[hu] side-LOC
‘by the side of the azalea crag’
80 Part III Borrowing the Chinese writing system
The second row in the example provides Romanization based on the Korean spoken at the
time. For Chinese characters used as phonetic on, their pronunciation in Modern Standard
Chinese ([bu] and [hu]) is given in the third row as a reference. The first two characters (紫
布) represent the bi-syllabic native Korean noun tolpwoy ‘azelea.’ Each syllable is represented
by one character. The first character (紫, in MSC: zı̌ ‘purple, violet’) is related in meaning to
‘azelea’ and is read in the native Korean sound tolp. The second character (布, in MSC: bù
‘cloth’) is used to represent the second syllable of the word by phonetic similarity to the Ko-
rean syllable woy, and its meaning in Chinese (‘cloth’) is irrelevant. The same pattern applies
to the second noun pahwoy (‘crag’). Such “semantic kun + phonetic on” combinations are a
salient feature of how Chinese characters are used in writing hyangka.16
天 地 之間 萬物 之中 厓
heaven earth between all things among at [ey]
唯 人 伊 最 貴 爲尼
only man SUBJ. [i] most noble does, and so. . . [honi]
所 貴乎 人 者 隱
what is noble man TOPIC TOPIC [nun]
以 其 有 五倫 也 羅
because of his have Five Human Relationships EMPHATIC is [la]
‘In the multitude of the myriad things midst heaven and earth (at that place), man (he) is the
most noble (and so): What is noble in man (it) is his possession of the Five Human Relation-
ships (it is).’
The gugyeol characters are phonetic on borrowed for their sound values. If we take these
characters away, the text will be no different from the original Chinese text. Based on such
markings, however, proficient gugyeol readers would be able to render the annotated text into
grammatical Korean. This practice resembled the Japanese kanbun (漢文 ‘Chinese writing’)
tradition that we will discuss in the next chapter.
As this practice developed, some 30 frequently used gugyeol characters evolved into
simplified graphs that were either fragments or shorthand shapes of the original characters.18
Chapter 10 Writing Korean 81
For example, the character 厓 is simplified into 厂, 伊 into 亻, 爲 into丷, and 尼 into 匕. The
simpler forms allow the graphs to be written in a size smaller than the original text, making it
easier to differentiate the Chinese words and the Korean grammatical morphemes. Below are
the first two lines of the example above written in this manner.
天地之間萬物之中厂 唯人亻最貴丷匕
The simplified gugyeol is similar in form and function to katakana in Japanese, which
also represent grammatical morphemes amidst Chinese characters, a role later taken over
by hiragana. These are not the only parallels between Korean and Japanese in how they have
adapted Chinese characters to fulfill their writing needs. We will see a few more in the next
chapter when we discuss the Japanese scripts in detail.
Hangeul
Hangeul (한글 han ‘great,’ geul ‘writing’) is the official script used to write Korean today in both
North Korea, where the script is referred to as josongul (조선글 ‘Korean writing’), and South
Korea. The creation of hangeul is attributed to King Sejong the Great (世宗大王1397–
1450), who ruled as the fourth monarch of the Joseon dynasty (reign 1418–1450). Hangeul
is an alphabetic writing system designed with careful attention to the sound system of the
Korean language at the time. Consisting of no more than 40 letters,19 it is much easier to
learn and use than the Chinese script or its Korean adaptations. During King Sejong’s time,
only elite members of the society could afford an education to learn hanja, the Chinese-
character script. It is said that the king invented the script so that all his people would be
able to easily and accurately read and write the Korean language. The work to design han-
geul was completed in 1443,20 and in 1446, King Sejong introduced the alphabetical script
in a publication titled Hunminjeongeum (訓民正音), The correct sounds for the instruction
of the people. Its preface reads:
Because the speech of this country is different from that of China, it [the spoken lan-
guage] does not match the [Chinese] characters. Therefore, even if the ignorant want
to communicate, many of them in the end cannot state their concerns. Saddened by
this, I have [had] 28 letters newly made. It is my wish that all the people may easily
learn these letters and that [they] be convenient for daily use.
However, the alphabet met with strong objections from the educated elite. To them, writing
simple letters was for women and children, the uneducated of the population. Some even
went so far as to claim that only “barbarians like Japanese and Mongolians possess their own
scripts,” and for Koreans, there was no need for such a “mean, vulgar, useless” script.21 Despite
King Sejong’s efforts to popularize hangeul, it took almost 500 years for it to eventually re-
place hanja as the predominant script in actual use. During that time, hangeul was primarily
used by women and authors of popular fiction. Writing for official purposes was done almost
exclusively in Chinese characters and remained so until the early 20th century, and until that
time hanja, and along with it the Chinese culture, maintained its high status in Korea.22
82 Part III Borrowing the Chinese writing system
Although hangeul appears to have been inspired by other writing systems, including the
Phags-pa script used to write Mongolian (as well as Chinese) and the Chinese seal script, it
is different from both of them in fundamental ways and is generally considered an original
creation. The design of the hangeul alphabet is based on a sophisticated understanding of
15th-century Korean phonology.23 If you recall, in traditional Chinese phonology, a syllable
is divided into two parts, an initial (consonant) and a final (e.g., vowel, vowel plus nasal). By
contrast, King Sejong’s analysis divided a Korean syllable into three parts: an initial conso-
nant, a medial vowel, and a final consonant. This analytical approach to understanding the
Korean syllable structure paved the way for the creation of an alphabet. Furthermore, what
makes hangeul an interesting alphabet is that the design of its consonant and vowel letters
follows certain scientific or philosophical rationales. Let us take a closer look.
The consonant graphemes are not a collection of discrete letters but form a system built
upon five basic letters. These basic symbols are shaped in such a way to suggest the articulatory
configurations in the speaker’s mouth (Figure 10.1). The bilabial /m/ is written with ㅁ, which
looks like a mouth with two lips. ㅅ, a supposedly tooth-like character, represents /s/, a dental
consonant that involves the front of the tongue approximating the upper teeth. The pronuncia-
tion of /n/ requires that the tip of the tongue touch the bony protruding structure right behind
the upper teeth, a gesture represented with the letter ㄴ. The /g/ sound involves the back of the
tongue raised to touch the back portion of the roof of the mouth, and the letter ㄱ supposedly
represents this gesture. The consonant /ng/ is written with a circle depicting the open throat.
This symbol is now also used to indicate the lack of an initial consonant in a syllable.
FIGURE 10.1 Five basic hangeul consonant shapes suggesting phonetic rationale
Chapter 10 Writing Korean 83
On the basis of these five symbols, strokes are added to derive additional consonant let-
ters. For example, adding one stroke above ㄴ <n> generates the letter ㄷ <d>. The consonants
/n/ and /d/ are closely related to each other because they are pronounced at the same place of
articulation involving the same gesture of the tongue. Then adding another stroke gives us the
letter ㅌ <t>, representing the aspirated version of the /d/ sound. Doubling the ㄷ symbol pro-
duces ㄸ <dd>, the letter for the tense version of the /d/ sound. Such derivative relationships
between consonant letters are shown in Table 10.1.
Hangeul vowels are said to have been designed based on a set of neo-Confucian no-
tions – heaven, earth, and man – that were understood as the three “great powers of the uni-
verse or the three germinants of Chinese philosophy.”25 The horizontal line represents earth,
the essence of yīn (阴/陰 ‘shade, overt, negative, opposite of yang’), the point (or short hor-
izontal or vertical line) represents the sun in heaven, the essence of yáng (阳/陽 ‘sun, overt,
positive, opposite of yīn’), and the vertical stroke symbolizes human beings who are mediators
between heaven and earth.
Like the consonant letters, vowel letters also form a system. The six basic vowel letters
are created by adding a short horizontal or vertical line to a long horizontal or vertical line, as
shown below:
Then, two or more basic vowel letters combine to form compound vowel letters. For example,
adding a short horizontal or vertical line changesㅏ<a> toㅑ<ya>, ㅓ<eo> to ㅕ <yeo>, ㅜ
<u> toㅠ <yu>, andㅗ <o> to ㅛ <yo>.
An important feature of hangeul, as mentioned previously, is that it is written in syllable
blocks and is probably the only alphabetical script written this way. Instead of being written
linearly as they are in English, hangeul letters are arranged in a square, two-dimensional space
to form a combination of letters that represents a syllable. The letters within the same syllable
block are read left to right and top to bottom. For example, the word hangeul (한글) has two
syllables and is thus written in two syllable blocks. The results of such arrangement are squar-
ish letter combinations reminiscent of Chinese characters, each of which also corresponds to
a syllable. The visual resemblance between hangeul and hànzì (the Chinese script) is indeed
prominent. Make no mistake, however – they are not the same kind of writing system. Han-
geul is phonemic, as is by definition for all alphabetic scripts, and, as you may recall from Part
84 Part III Borrowing the Chinese writing system
II, hànzì is morphemic (or morphographic). Each hangeul letter corresponds to a Korean
phoneme, while each Chinese character represents a morpheme in writing Chinese. It is the
basic linguistic unit a grapheme represents, not the spatial arrangement of graphemes, that
determines the nature of a script.
Syllables, however, are the units in which proficient hangeul users read and write. Ko-
rean syllables may be in the shape of V, CV, CVC or CVCC. (“V” represents “vowel” and “C”
represents “consonant.”) Each hangeul syllable block consists of at least a C and a V. When the
vowel is not preceded by a consonant in speech, in writing the letter ㅇ /ng/ is used to stand in
as a null consonant. See Table 10.2 for a few examples of Korean syllable shapes. The Korean
syllable structure is more complex than that of Modern Standard Chinese, which makes use
of about 1,359 different syllable shapes, and even more so than Japanese, which has only 113
distinct morae. (Recall that in Chapter 1 we learned that a mora is similar to but not the same
as a syllable. It is a phonological unit intermediate between a phoneme and a syllable, and it is
the basic unit for the Japanese kana: each kana symbol represents one mora.) There are 3,000
to 4,000 distinctive Korean syllables.26
Hangeul’s status has risen since the mid- to late 19th century as Korea underwent reforms
in favor of Western-style modernization. Gradually, a mixed script emerged with hanja repre-
senting Sino-Korean words and hangeul representing native Korean words and grammatical
morphemes. This pattern was very similar to, though not as complex as, the modern-day
mixed script of Japanese, which we will discuss in the next chapter. Hangeul is now the pre-
dominant script used in both Koreas. In North Korea, hangeul became the sole script for
official writing when hanja was abolished in 1949. The use of hanja in South Korea, although
not completely abandoned, has declined dramatically since the 1950s and, by the 1980s, hanja
practically disappeared from public media such as newspapers. In both Koreas, however, a
limited number of hanja is still taught in schools. For example, high school students in South
Korea are generally expected to learn about 1,800 characters before they graduate. Some lin-
guists argue that hanja should continue to be taught and used for the variety of advantages it
offers to learning and communication.28 It will be interesting to see if hanja will rise again as
elements in the Korean script in the foreseeable future.
Notes
W
HEN JAPAN CAME INTO contact with China, the Japanese language was gen-
erally not recorded in writing. In around the early 4th century,1 Chinese written
text – and along with it the idea of writing – was first introduced via Korea to Ja-
pan. The Japanese first learned to read and write in Chinese, then adapted the Chinese writing
system to recording their own language, and eventually created additional systems of writing
to supplement, facilitate, and in certain cases replace the use of Chinese characters. Today,
Japanese writing makes use of a mixture of four scripts: kanji (漢字2 ‘Chinese characters’),
hiragana (平仮名 ‘simple borrowed characters’), katakana (片仮名 ‘partial borrowed charac-
ters’), and rōmaji (Roman letters)3 and is considered one of the most complex in the world. In
this chapter, we will look at how Japanese writing came to be the way it is.
Kanbun
To further understand how the Japanese language came to be written in Chinese characters,
it is useful to learn about kanbun (漢文 ‘Chinese text/literature/writing’), Classical Chinese
text annotated to facilitate Japanese speakers’ reading of the text as Japanese. Although ini-
tially a method to help Japanese speakers understand Classical Chinese material, kanbun also
became a style of writing in Classical Japanese using Chinese characters. Educated Japanese
not only studied literature brought in from China but also composed poetry and wrote official
documents in Classical Chinese in the kanbun style. A great amount of Japanese literature was
written in kanbun over thousands of years until as recently as the 20th century.
How does reading kanbun work? Let us look at an example. In Table 11.1, the kan-
bun text is written vertically on the right. Because Chinese text was traditionally arranged
in vertical columns read from right to left, kanbun also used this arrangement. To facilitate
the reading process, an annotation system consisting of kaeriten (返り点 ‘return marks’) was
developed. These were symbols added to the text to indicate the order in which the characters
should be read. The kaeriten were written to the lower left of the characters in a size smaller
than the main text. The text in this example is the first sentence of a story from the book Hán
Fēizı̌ (韩非子/韓非子) of the Warring States Period (mid-3rd century bce). In English, this
sentence means, “there was a Chǔ person selling shields and spears.” Chǔ (楚) was the name
of one of the seven major Warring States before China was united in 221 bce.
On the surface, the text appears no different from ordinary Classical Chinese text. In
order to read it as Japanese, the reader must make adjustments to account for the difference
in word order between Chinese and Japanese. In highly simplified terms, word order in a
Chinese sentence is typically subject – verb – object, while in Japanese it is usually subject –
object – verb. The reader would rearrange the characters into the Japanese order, read them
in Japanese pronunciation, and add grammatical suffixes where needed. The outcome would
essentially be a translation of the Classical Chinese text in Classical Japanese with a certain
kanbun flavor.
Let us try to read the kanbun text in the example. We start from the very top, 楚人 (So
jin, ‘Chǔ person’). There is no marking of any kind for these two characters, so we simply read
them and move on. The next character 有 (yū, ‘there exists’) is marked with a 下 (ge, ‘down,
bottom’). This tells us that we will need to move this character to below where the mark 上
(jō, ‘up, top’) is, in this case at the very end of the sentence. So, we skip 有 and move on to the
next character 鬻 (iku, ‘to sell’). This character has a number二 (ni, ‘two’) attached to it, which
means it cannot be read until the character marked with the number 一 (ichi, ‘one’) has been
read. So, we skip 鬻 as well. The character following it, 盾 (jun, ‘shield’), is unmarked, so we
go ahead and read this one. After that, 與 (yo, ‘and’) is marked again. レ (re, ‘reverse mark’)
indicates that the two characters adjacent to it switch positions with each other. Therefore, we
read 矛 (mu, ‘spear’) next, and then go back to 與. After 與 we bring in 鬻, as this is the po-
sition marked with the number 一. Then we read 者 (sha, ‘nominalizer’), and finally the verb
有 (yū, ‘there exists’). There. As you see, it is quite a laborious process. But at the same time it
is also quite fascinating.
Companion Website
The existence of kanbun shows us that, in the case of Japanese, the borrowing of a writ-
ing system, as we have seen, is not merely a matter of linguistic encoding of the native speech
in a “non-native” script. In fact, the importing of the Chinese script took place amidst exten-
sive and deep learning of the Chinese culture, including its language, literature, philosophy,
and so on.4 Thus, the borrowing of the Chinese script was a by-product of the sustained con-
tact between cultures and it facilitated rigorous integration of one culture into another.
Kanji
The use of Chinese characters, or kanji, in writing Japanese traces its origin to as early as the
4th century ce. Between the 4th and the 7th centuries, a large number of Koreans and Chinese
emigrated to Japan. Besides introducing Chinese text to Japan, including Confucian classics
and Buddhist writing, the scholars among them also tutored the Japanese people in the arts,
science, and technology of China. By the 7th to the 8th centuries, Japan’s educated elites,
including aristocrats, officials, and Buddhist monks, had integrated many aspects of the Chi-
nese culture into Japanese life, including the use of Chinese characters. For instance, one of
90 Part III Borrowing the Chinese writing system
the early examples of writing in kanji was a 17-article constitution drafted by Prince Shōtoku
(574–622 ce) in the year 604.5 For hundreds of years since the initial use of kanji and until
kana was created in the 9th century, the character-based writing system had remained the sole
script for Japanese, and it still serves an important role in Japanese writing today.6
How many kanji are used in writing Japanese? Up until the Meiji period (1868–1912),
although the kana scripts had been in existence for centuries, official and scholarly writing
was almost exclusively done in kanji. Since then, government-sponsored script reforms have
transformed Japanese writing from an all-kanji script to a mixed script. From 1879 to 1968,
the percentage of kanji used in newspapers declined from 90% to 60% of all written graphs.7
Based on data collected between 1941 and 1981, the number of different kanji ranged around
3,000 for magazines and newspapers and around 5,000 for Japanese literature.8 In general,
the 2,000 most frequently used kanji account for about 99% of all kanji occurrences, and up
to another 3,000 account for the remaining 1%.9 Table 11.2 and Table 11.3 offer information
from various official lists of kanji decreed for common use and elementary school instruction
from the end of World War II to the beginning of the 21st century. On one hand, the Japanese
government made efforts to limit the use of kanji. On the other hand, as these tables show, the
number of kanji officially sanctioned for basic literacy has somewhat increased.
Chinese characters in Japanese may have multiple pronunciations. This is because the
same kanji may represent both the borrowed Chinese morpheme and the indigenous Japanese
counterpart. Official documents in Japan were written in the kanbun style for thousands of
years, from the late Heian period (794–1185) until after World War II. Not surprisingly, it
made a lasting impact on the Japanese language. As discussed in the overview, a large number
of Chinese words were borrowed into the Japanese lexicon – it has been estimated that about
60% of Japanese vocabulary is Chinese in origin.12 Oftentimes, the Chinese loanword and
the native Japanese word are both preserved in the language and associated with the same
kanji. In other words, the same character may represent multiple morphemes and have mul-
tiple pronunciations. However, their uses are usually not interchangeable. The reader must
know which to use given a specific context. The Chinese reading typically appears in a com-
pound, while the Japanese reading usually occurs on its own. For example, 山 (in Chinese:
shān ‘mountain’) is read as san as part of the proper noun 富士山 Fujisan ‘Mount Fuji’ and the
Chapter 11 Writing Japanese 91
names of other mountains that are well known. But it is read as the native word yama when
referring to a generic mountain.
Recall that in the overview chapter we have learned a pair of linguistic terms to help us
talk about the different readings of kanji. When the character is read with pronunciation in
its source language (Chinese), it is called on-reading (音読み onyomi ‘sound reading’). If it is
read with the pronunciation in the target language (e.g., Japanese), it is called kun-reading (
訓読み kunyomi ‘meaning reading’). Furthermore, depending on whether the original mean-
ing of the character is kept or not, the reading can be regarded as either semantic or phonet-
ic. Taken together, there can be four different types of readings: semantic on, semantic kun,
phonetic on, and phonetic kun. We already looked at a few examples for these concepts in the
overview chapter. Let us review them with a few additional examples:
First, semantic on is when the whole Chinese morpheme – both the meaning and the sound –
has been borrowed into Japanese. For example, the word 着手 (着 ‘to put’; 手 ‘hand’) means
“to start” in both Chinese and Japanese. Its pronunciation in modern Japanese, chakushu, is
reminiscent of how it sounds in modern Chinese, zhuóshǒu. The similarity in both meaning
and sound is a clue that the entire word was borrowed from Chinese into Japanese. Thus, the
着 in the Japanese word 着手 has a semantic on reading.
When only the meaning is borrowed but the kanji is read with its native Japanese pro-
nunciation, this reading is referred to as semantic kun. The character 着 (ki ‘to wear’) in the
second example 着物 (kimono ‘clothing’; 物 mono ‘thing’) has a semantic kun reading. On
one hand, its Chinese meaning has been preserved in Japanese: 着 (zhuó) in Chinese means
“to wear” as well. On the other hand, the character’s Chinese pronunciation is not adopted in
Japanese, and it is pronounced as the native Japanese morpheme ki.
Phonetic on refers to the reading of kanji based on its Chinese pronunciation when
the meaning of the character in Chinese is irrelevant to its meaning in Japanese. In the third
example, the character 天 in Japanese is pronounced as ten, akin to its modern Chinese pro-
nunciation tiān. Yet its Japanese meaning “oven” has little to do with its Chinese meaning
“sky.” Apparently, in this case, the character 天 was borrowed purely for its phonetic value to
represent a native Japanese morpheme.
Finally, a Chinese character may be used in Japanese to represent a morpheme irrelevant
to its Chinese original in either meaning or sound. In this case, the reading is referred to as
phonetic kun. In the last example, the kanji 張 (haru ‘spring season’) appears to be unrelated
to the character in Chinese in either pronunciation (MSC zhāng) or meaning (‘to stretch’).
How did it come to be used this way? It was likely a two-step process. First, 張 (‘stretch’) was
used to represent the synonymous morpheme haru (‘stretch’) in Japanese. At that point, it was
92 Part III Borrowing the Chinese writing system
a case of semantic kun. Then, the character was used to represent the Japanese morpheme
haru ‘spring (season)’ that happened to have the same pronunciation. The new meaning it
took on was no longer related to its original meaning in Chinese. As a result, the character lost
its connection to the Chinese original both in sound and in meaning and came to be read as
a phonetic kun.
As we have seen, the appropriate reading of a kanji is determined by context. The same
kanji may be read in its Chinese pronunciation or its Japanese pronunciation. This makes
reading kanji sometimes a very complex matter.
To make matters worse, Chinese words were borrowed into Japanese not all at once, but
during different historical periods and from varied dialectal sources. This resulted in three
major types of on-readings: Go’on (呉音 ‘sound of the Chinese state Wú’), Kan’on (漢音 ‘sound
of Chinese’), and Tō-Sō’on (唐宋音 ‘sound of Táng and Sòng dynasties’). Go’on borrowing
occurred during the 5th and the 6th centuries from a Southern dialect of China’s Six Dynas-
ties period, when Buddhist works were introduced to Japan via Korea. Kan’on was brought to
Japan during the 7th and the 8th centuries and reflected the pronunciations of Chinese in the
cities Cháng’ān (長安) and Luòyáng (洛阳/洛陽) of the Táng dynasty. Finally, Tō-Sō’on came
to Japanese from southern China along with Zen Buddhism during the 14th century, when
China was in the Yuán or Míng dynasty.
As a result of this complicated borrowing history, although most kanji have just one
on-reading, some may have as many as three. For example, the character 明 in today’s Manda-
rin is pronounced as míng and has several meanings: (1) ‘bright, light, clear’; (2) ‘immediately
following this year/day’; (3) ‘as in Míng dynasty.’ In Japanese, the kanji 明may be read in its
Go’on, Kan’on or Tō-Sō’on depending on what word it is in, as shown in Table 11.4.
In addition, the same kanji may also have a number of kun-readings, that is, read with
its Japanese pronunciations. On-readings often occur in compounds, while kun-readings are
usually stand-alone morphemes or words. What distinguishes kun-readings in writing is the
addition of some type of inflectional ending written in hiragana. Let’s look at the kanji 明
again as an example (Table 11.5).
The reading of the kanji 明, at least superficially, may correspond to a- (1, 2, 7, 8, 9, 10),
aka- (3, 4, 5), or aki- (6). There is much morpheme-specific information the reader or writer
must know in order to pronounce or write the morpheme correctly.
Companion Website
Man’yoˉgana
The Japanese term kana (仮名) refers to the moraic scripts14 used to write the Japanese lan-
guage historically and in modern times. Kana comes from karina, in which na means “names,
letters,” while kari may have a range of meanings including “borrowed,” “fake,” “temporary,”
“unofficial,” and “irregular.”15 Thus, kana may mean “borrowed names,” because in the earliest
kana script, kanji or Chinese characters were “borrowed” as sound symbols to represent the
Japanese speech. Kana may also mean “fake names,” or “unofficial letters” as opposed to the
kanji script that was already in place and deemed the authentic and official system of writing
for Japanese at the time.
Man’yōgana (万葉仮名) was considered the earliest method of writing that made use of
kanji as sound graphs to represent Japanese. Man’yō literally means “ten thousand leaves.” This
poetic name was derived from the Man’yōshū (万葉集), Collection of Ten Thousand Leaves,
the famous anthology of approximately 4,500 Japanese poems composed during the Nara
period (710–794 ce). It is the earliest known anthology of Japanese poetry. The poems in
this collection were recorded in writing using a complex combination of methods. In many
instances, Chinese characters were used for their phonetic values regardless of their meaning.
For example, the character 天 (t’ien in Middle Chinese) were used to represent the Old Japa-
nese mora te, and 安 (ân in MC) for a.16 This kind of writing came to be known as man’yōgana
(Man’yō + kana). Some scholars believe that the Korean hyangchal, which also used Chinese
characters primarily as sound symbols, had an impact on man’yōgana.17
94 Part III Borrowing the Chinese writing system
When man’yōgana was in use, it was up to the writers to choose which kanji to pick to
represent a given mora. Since different characters could be pronounced the same, multiple
characters might be used to represent the same mora. By the end of the 8th century, more than
970 kanji characters were in use to write the fewer than 90 morae of the Japanese language at
the time.18 For example, over 40 different kanji were used to represent the mora shi and over
30 characters were used for ka.19 Gradually, however, each mora came to be written with a
preferred Chinese character, and over time these preferred characters became simplified.
Man’yōgana was the linguistic precursor for both hiragana and katakana. As the practice
of using Chinese characters to represent Japanese morae became more common, the character
forms also became increasingly simplified. Simplification took two directions: hiragana sym-
bols were derived from the cursive forms of their source man’yōgana characters, and katakana
symbols were formed by taking pieces from the source characters, which were often but not
always the same characters as the source of the corresponding hiragana symbols. For example,
the five vowels below are usually the first five morae represented in a kana chart of a modern
Japanese textbook. The source characters for u and o are the same for the hiragana and the ka-
takana symbols; for a, i, and e, however, they are different characters. In fact, the hiragana and
katakana signs were not standardized until the year 1900, when Japan’s Ministry of Education
enacted regulations on kana forms and usage.20
Hiragana た ち つ て と な に ぬ ね の は ひ ふ へ ほ
Katakana タ チ ツ テ ト ナ ニ ヌ ネ ノ ハ ヒ フ ヘ ホ
Rōmaji ta chi tsu te to na ni nu ne no ha hi fu he ho
Hiragana ま み む め も や ゆ よ ら り る れ ろ
Katakana マ ミ ム メ モ ヤ ユ ヨ ラ リ ル レ ロ
Rōmaji ma mi mu me mo ya yu yo ra ri ru re ro
Hiragana わ を ん
Katakana ワ ヲ ン
Rōmaji wa wo n
The nature of the kana scripts may need some clarification. Hiragana and katakana are
often referred to as Japanese syllabaries, but this categorization is not entirely accurate. In
Chapter 3 of this book we learned that a syllabary is a writing system in which a grapheme
consistently represents a syllable. In kana, however, a grapheme does not always correspond
to a syllable. One obvious example we can see in Table 11.7 is んor ンrepresenting a single
consonant /n/, which does not fully constitute a syllable. We will see more examples shortly
for how one kana symbol may not correspond with one Japanese syllable. To more accurately
understand the nature of the kana scripts, we will need to reintroduce the concept “mora” that
we touched on in Chapter 3. If you recall, a mora is a phonological unit intermediate between
a phoneme and a syllable. In Japanese, it is most of the times a syllable, but it can also be a
phoneme (i.e., a single sound).
What exactly constitutes a mora in Japanese? To answer this question, let us first look
at the phonemic inventory of Japanese – what consonants and vowels the Japanese language
uses. Japanese makes use of five vowels /a, i, u, e, o/ and 14 consonants /p, t, k, b, d, g, s, h, z, j,
r, m, n, w/. In addition, there are two kinds of consonants usually represented with the capital
letters /N/ and /Q/. /N/ is a nasal sound that usually occurs at the end of a syllable. Depending
on what comes after it, the phonetic value of /N/ is different: it is [n] before /n, t, d/, [m] before
96 Part III Borrowing the Chinese writing system
/m, p, b/, and [ŋ] (sounds like the English ng) before /k, g/. /Q/ is used to represent the first
segment of a double consonant, which includes kk, tt, ss, and pp. Vowels in Japanese can be
either short or long, sometimes affecting the meaning of the word, for example kite (‘Come!’)
vs. kıˉte (‘Listen!’), and sometimes not, for instance sayonara or sayōnara (‘goodbye’).22 A long
vowel can also be called a geminate vowel. These consonants and vowels combine to form
Japanese syllables.
To understand Japanese morae, we will also need to understand the syllable structure of
the Japanese language. The vowels and consonants cannot randomly combine to form sylla-
bles and must follow certain patterns. For example, the 14 regular consonants can only occur
at the beginning of a syllable and not at the end. /N/ and /Q/, however, occur at the end of
a syllable. Consonant clusters, such as pl, dw, str that we use in English, are not allowed in
Japanese. These restrictions mean that the Japanese syllable structure is relatively simple. We
can summarize its possibilities in Table 11.9, which also offers examples. First of all, a syllable
in Japanese can be any of the five single vowels, short (V) or long (VV), or a combination of
vowels (V1V2) called diphthongs. It can also be any of these varieties of vowels preceded by
a consonant (CV, CVV, CV1V2), sometimes with a [y]23 sound in between (CyV, CyVV). Fur-
thermore, it can be a short vowel followed by a /N/ or a /Q/ without a consonant preceding
it (VN, VQ), with a consonant preceding it (CVN, CVQ), or with a Cy sequence preceding it
(CyVN, CyVQ).
Now we are ready to come back to our question of what constitutes a mora – a mora is a
timing unit equivalent to the time required to pronounce a short syllable in Japanese speech.
It can be any of the following: V (note that VV or V1V2 counts as two morae), CV, CyV, or a
syllable-final N or Q. Notably here, N or Q each counts as a mora. Therefore, unlike a syllable,
which requires minimally a vowel, a mora can correspond to just one consonant. All the above
segments or combinations receive one mora in speech. That is to say that pronouncing them
Chapter 11 Writing Japanese 97
psychologically takes about the same amount of time in Japanese, which can be thought of as
one beat. For example, the word /nippon/ (‘Japan’) has two syllables and four morae. The word
/ichi/ (‘one’) also has two syllables, but it only has two morae. (Syllable boundaries are indicat-
ed using dots, and mora boundaries are marked with dashes.) Although these two words have
the same number of syllables (and thus take about the same amount of time to say in English),
it takes significantly more time to say /nippon/, which receives four beats, than /ichi/, which
has two beats, in Japanese.24
Companion Website
Mora is an important concept not only for describing the timing of Japanese speech but
also for characterizing the nature of the kana scripts. You may have noticed that, in Table 11.9,
although all examples are one syllable long, they are not all written in one kana symbol, and
some require two or more kana graphs. This is because every kana graph does not correspond
to one syllable – it basically represents one mora. (Note that a smallerょ/ャ (yo) is used when
the [y] sound occurs between a C and a V, but it does not receive a mora on its own. A smaller
っ/ッ(tsu) is used to represent the first segment of a geminate consonant, that is, a silent beat,
which does take up one mora.) Let us look at two more examples below.
Okāsan has five morae and is written with five kana graphs. Kinō has three morae and
is written with three kana symbols. If they were written using a syllabic script, then because
o.kaa.san has three syllables, it would be represented with three graphemes, and by the same
token, ki.noo would be represented by two graphemes. In sum, the kana scripts are more ac-
curately described as moraic rather than syllabic writing systems.
A mixed script
Using Chinese characters to write Japanese requires overcoming substantial linguistic chal-
lenges. We already touched upon some of the major differences between Japanese and
98 Part III Borrowing the Chinese writing system
Chinese in Chapters 1 and 9. Here we will examine them in a little more detail. The two lan-
guages differ in how words and sentences are formed. Japanese verbs conjugate depending on
whether it is in the polite form, negative form, perfective form, and so forth. Chinese verbs,
however, generally occur in the same form. Japanese also makes use of a variety of particles
that do not have counterparts in Chinese. Some of them may be attached to nouns to indicate
their grammatical role: whether they are the topic (-wa は) or the subject (-ga が) of the sen-
tence, or whether they are the direct object (-o を), the indirect object (-ni に), the location
(-de で), or the direction of the action (-ni に), and so on. Others can be attached to sentences
to form questions (-ka か) or to soften the tone of a request (-ne ね), and so forth. Another ma-
jor difference is in sentential word order, as you may recall from the overview chapter. For these
reasons, writing Japanese using the Chinese script is not always a straightforward process.
Let’s look at two examples to get a better sense of the syntactic differences between
Chinese and Japanese and to see how the mixed script works. To express “go to the library to
meet with Xiao Wang,” the word order in Chinese is largely the same as in English, while in
Japanese, the object comes before the verb, and the verb comes last. The same difference can
be observed for the second example, “I do not eat fish.” In both examples, nouns, pronouns,
and verb stems are written in kanji, while particles and verbal suffixes are in hiragana.
Writing in the Chinese script was adapted to the Japanese language over time, and
changes were made to accommodate the major differences between Japanese and Chinese.
First, the Japanese spoken word order was gradually adopted into Japanese writing, and even-
tually, additional scripts were created to make writing and reading easier. To write the suffixes
and particles that did not exist in Chinese, early on, Chinese characters were borrowed purely
for their phonetic value. These characters were usually written smaller and were added to the
main characters of the text. Today, Chinese characters are no longer used in this capacity. In
their place, hiragana is used instead, as shown in the examples above.
In the mixed script today, kanji, hiragana, and katakana usually play different roles.
Although the number of kanji is officially limited to a little over 2,000, kanji is still an indis-
pensable and arguably the most important component.26 Hiragana and katakana are used
Chapter 11 Writing Japanese 99
for different purposes and functionalities. Hiragana is used to write grammatical particles,
inflectional endings, and sometimes whole words. It is also often used to indicate the pronun-
ciation of kanji in publications such as newspapers and magazines. Katakana was formerly the
script for representing grammatical particles and suffixes. Today, it is used for non-Chinese
loanwords, specialized vocabulary, emphasis, and onomatopoeia. Aside from these linguistic
factors that regulate – to a great extent – the choice of script, there are also socio-cultural
stereotypes that predisposes native Japanese writers to using one script over another. We will
discuss this point in detail in Chapter 20. For now, the story below may serve to illustrate how
script choice is deeply rooted in convention. Switching to a new script may indicate a change
in the connotation of a word even though the literal meaning remains intact.
Fukushima
The Toˉhoku earthquake and tsunami on March 11, 2011, led to a nuclear disaster in the Fukushi-
ma I Nuclear Power Plant in Japan that involved hydrogen-air chemical explosions, nuclear melt-
downs, and the release of radioactive material. Fukushima, the name of the coastal prefecture
(located to the north of Tokyo on the island of Honshu) or the name of the prefecture’s capital
city, is customarily written in kanji as 福島. Since the nuclear accident, it was sometimes written
in katakana, フクシマ, to distinguish it as the site of a disaster instead of a prefecture or a city.
This practice had its historical precedence. After the August 1945 nuclear bombings, “Hiroshi-
ma (広島, ヒロシマ)” and “Nagasaki (長崎, ナガサキ)” were written in katakana instead of kanji
when referred to as the bombing sites27 or used to represent “nuclear victimhood.”28
Furthermore, the word hibakusha may serve as an example for script flexibility that allowed
associations to be made between nuclear contamination in Fukushima and nuclear victimhood
in Hiroshima and Nagasaki. The term hibakusha (被爆者) originally referred to survivors of the
World War II atomic explosions. Written in katakana, it ambiguously represented 被爆者 ‘atomic
bomb victims’ and 被曝者 ‘people exposed to radiation.’ The two words were pronounced exactly
the same, as baku could either mean 爆 ‘explosion’ or 曝 ‘exposure.’29 In March 2012, the first an-
niversary of the Fukushima disaster, about 2,000 protesters gathered in Hiroshima. In their slogan
“No More Hiroshimas, No More Fukushimas, No More Hibakushas,” Hiroshima, Fukushima, and
hibakusha were all written in katakana, precisely with the intention to bring together victims of the
atomic bombs and those of the nuclear plant accident.30
Although useful as means to distinguish the locations as sites of disastrous events from names
of cities or prefectures in their normal sense, such katakana rendering might also come to be
associated with the fear, shame, and anxiety people experienced in the aftermath of the disasters.
Perhaps for this reason, a year after the nuclear plant accident, voices on the Internet called for the
abolishment of using katakana to write “Fukushima.”31 To Japanese readers and writers, restoring
the writing of the name to its original kanji script might represent a successful and complete res-
toration of the contaminated place to its original, pristine condition.
Notes
1 This time period is based on Rogers (2005, p. 56), but it varies depending on the sources. For
example, Taylor and Taylor (2014) dates the introduction of Chinese writing to Japan to after the late
4th century, because according to them the Koreans did not begin to use Chinese characters until in
100 Part III Borrowing the Chinese writing system
the 3rd century (p. 273). Yamada (1991) suggests that kanji was first introduced to Japan in the 5th
century (p. 141), and Gottlieb (1995) dates it to the 6th century (p. 3).
2 Here the characters 漢字 are Japanese kanji, so the simplified Chinese characters used in China do
not apply. Although some kanji characters are also simplified, they may not be the same as the sim-
plified characters in China.
3 Or, at least three scripts – kanji and the two kana – are required. The use of roˉmaji is comparatively
marginal. Some scholars (e.g., Taylor and Taylor (1995)) also consider Arabic numerals as a fifth
script. Since its use is mostly limited to representing numbers and is not normally used to write
non-numerical speech, it is not considered a full script in this book.
4 Taylor and Taylor (2014, p. 272).
5 Ibid., pp. 272–273.
6 Ibid., p. 274.
7 Kaiho and Nomura (1983), quoted by Taylor and Taylor (2014) on p. 274.
8 See Taylor and Taylor (2014, p. 274).
9 Ibid.
10 Adapted from ibid., p. 275, Table 16–2.
11 Ibid.
12 Shibatani (1990) notes that this percentage is as of 1859 (p. 142). This means it does not include the
new compounds created based on Chinese characters during modernization in the Meiji period
(1868–1912).
13 Poem 529 of the Man’yōshū. Seeley (1991, p. 50).
14 See Chapter 3 for the definition of mora and moraic writing system.
15 Taylor and Taylor (2014, p. 285).
16 These examples are taken from Seeley (1991, p. 50). Transcriptions follow what is in this book:
Middle Chinese uses the reconstruction given in B. Karlgren’s Grammata Serica Recensa, and Old
Japanese uses the modern Hepburn Romanization.
17 For example, King (1996, p. 218).
18 Joshi and Aaron (2006, p. 483).
19 Taylor and Taylor (2014, p. 285).
20 Smith (1996, p. 212).
21 Sugimoto and Levin (2000, p. 137).
22 This example is taken from Taylor and Taylor (2014, p. 258).
23 Here, the y is not an IPA symbol, but an English letter, representing the sound /j/ in IPA. Note that
in IPA, the letter y is also used, and /y/ represents a vowel.
24 Theoretically a word with four morae takes about twice as much time to say as a word with two mo-
rae. A native speaker of Japanese I interviewed, however, reported that for her, -n would take about
half the time as a CV mora and C(C) would take about one third of the time as a CV mora. By this
calculation, Ni-p-po-n is about 2.8 times the length of a CV mora and i-chi is twice as long. Thus,
although Nippon is significantly longer than ichi, it is not quite twice as long.
25 Usually speakers leave out the topic unless it is unclear. It is kept here for comparison purposes.
26 Taylor and Taylor (2014) say that “kanji is the most complex and at the same time the most import-
ant” (p. 272).
27 Brannigan (2015, p. 156).
28 Seaton (2016, p. 347).
29 Ibid.
30 Penney (2012).
31 Tahara (2012).
12 Writing Vietnamese
L
IKE KOREAN AND JAPANESE, the writing of Vietnamese has had a fascinating his-
tory. It involves at least three different scripts: chũ’ nho (𡦂儒 ‘Confucian script’) refers
to the Chinese writing system, chũ’ nôm (𡦂喃 ‘southern script’) contains Vietnamese
characters created based on the Chinese script, and quốc ngũ’ (國語 ‘national script’) is an
adaptation of the Roman alphabet. Among the languages of the Chinese-character cultur-
al sphere, Vietnamese is the only one that has completely abandoned the Sinitic script to
be written with the Roman alphabet in modern times. In this chapter, we will look at how
Vietnamese writing transitioned from one script to another with a focus on the contexts of
these transitions. We will see that, linguistic factors aside, the changes in writing were closely
intertwined with the political history of Vietnam and the national identity of the Vietnamese.
The use of the Chinese writing system persisted in Vietnam even after its independence
from China. The Chinese rule in Vietnam ended in 939 ce, soon after the collapse of the Táng
dynasty. During more than 700 years of self-rule and until Vietnam was taken under colonial
governance by France in the mid-19th century, Classical Chinese remained the administrative
and scholarly language. Chũ’ nho was eventually made the official writing of Vietnam in 1174
ce. As in the Chinese tradition, Vietnamese scholars recorded history and composed poetry
using chũ’ nho. In addition, chũ’ nho was employed to write a wide range of popular literature,
including oral tales of myths and legends and stories concerning everyday life.3
The sustained use of chũ’ nho was testimony to the further flourishing of the Confucian
culture in independent Vietnam. Although no longer subjugated to Chinese control in polit-
ical terms, the cultural influence of China took root and continued to develop, and Vietnam
continued to look to China for models of governance. For example, an examination system
similar to the Chinese civil service examination was established for selecting officials. In 1075,
the first examination on Confucian learning was held. By the 15th century, Confucian culture
had gradually spread into southern Vietnam. The use of Chinese characters further expanded
to writing for such mundane purposes as billing and invoicing in commercial transactions. By
this point, however, chũ’ nho was no longer the sole written language and script. A modified
script based on the Chinese writing system came into existence and began to be used to write
the Vietnamese language.
were Chinese in origin. For example, the character 才 (cái ‘talent’) was used to write the Viet-
namese morpheme tài with the same meaning.8 This was a case of semantic on. On the other
hand, a borrowed character might be used as a sound symbol for a Vietnamese morpheme
that had a different meaning. An example is the character 半 (in MSC: bàn), which in Chinese
meant ‘half,’ used for the Vietnamese morpheme bán ‘to sell.’ Another example is the use of the
Chinese characters 布 (bù ‘cloth’) and 盖 (gài ‘cover’) together for the archaic Vietnamese term
bố cái ‘father and mother.’9 The sounds come from Chinese yet the meaning is irrelevant. Thus,
these were cases of phonetic on and were similar to uses of Chinese characters in the Korean
idu (‘clerical reading’) and hyangchal (‘vernacular letters’) scripts and the Japanese man’yōgana.
The idea of using Chinese characters as phonetic symbols in representing another lan-
guage was likely not new to Vietnamese speakers. As Buddhism spread to Vietnam from the
2nd century ce, speakers of Vietnamese had been exposed to Buddhist scriptures encoded in
Chinese characters. The Sinitic symbols were used as phonetic representations for words with
origin in Sanskrit and other foreign languages, so the idea of using Chinese-like characters to
transcribe speech had not been a strange one. This might have been another factor that con-
tributed to the creation of nôm characters.
Fascinatingly, however, a large number of new characters – both modified Chinese
symbols and newly invented characters – were employed to phonetically transcribe native
Vietnamese words. Based on their method of creation, invented nôm characters consisted of
two major types: one, composite creations that combined two Chinese symbols into single
characters, and, two, partial or simplified Chinese characters that functioned as independent
nôm characters. Among the first type, the most numerous subtype was semantic-phonetic
compounds created following the Chinese principle of character formation, with one com-
ponent taken from Chinese for its semantic value and the other component representing the
pronunciation. For example, the character 至, 典 represented the Vietnamese word đê´n ‘to
arrive (at).’ It contained on the left-hand side a Chinese character 至 (zhì), meaning “to
arrive (at)” (its pronunciation is irrelevant), and on the right-hand side the character 典
(diǎn), representing the pronunciation (its meaning is irrelevant). Another example was the
character 媄, pronounced as me. for “mother” in Vietnamese. This pronunciation was repre-
sented by the component on the right, 美, which in Chinese had a similar pronunciation, and
it is pronounced měi in Modern Standard Chinese. The meaning of this component in Chi-
nese (‘beautiful’) was irrelevant. The part on the left, 女 (in Chinese: nǚ, ‘female’), indicated
the meaning of the whole character.10
A smaller group was semantic-semantic composite characters. The nôm character 𡗶
contained in the upper part a Chinese character 天, meaning “heaven,” and in the lower part
a Chinese character 上, meaning “above.” The combination represented the Vietnamese word
trǒi, meaning “heaven,” but there was no hint of the pronunciation in the formation of the
character. Another example was the composite character 𠅎 that represented mât ‘to lose’ in
Vietnamese. It was composed of two Chinese characters, 亡 (wáng) and 失 (shī), both with the
same meaning, “to lose,” in Chinese.11
There were also phonetic-phonetic compounds composed of two sound components.
In fact, one reason Vietnamese speakers created new compound characters was because con-
sonant clusters – strings of two or more consonants (e.g., /str/ in /strong/) – existed in the
Vietnamese language at the time. For example, the characters 巴and 賴 are put together with
巴 (/ba/) on top of 賴 (/lái/) to represent the morpheme /blái/ ‘fruit,’ and 巨 (/cự/) above 朗
(/lặng/) to write the morpheme /klặng/ ‘silence.’12
104 Part III Borrowing the Chinese writing system
A second type of new character created as part of the chũ’ nôm script was simplified
Chinese symbols. Often only parts of the original characters were preserved. For example, the
top component of the Chinese character 爲 (wéi, ‘to do’), that is, 爫, was taken on its own to
represent the Vietnamese morpheme làm ‘to do, to make.’ The bottom portion of the Chinese
character 衣 (yī ‘clothing’), i.e., 衣, was used to represent ây ‘that, those.’ Another possibility
was that the nôm character changed the original character by deleting some strokes while add-
ing others at the same time. For instance, the character for la. ‘strange’ came from the Chinese
character 羅 luó ‘net.’13
Although promotion of the chũ’ nôm script was limited for most of its history, it received
support from several Vietnamese emperors. One of them was Trần Nhân Tông (1258–1308),
who used chũ’ nôm during his reign (1279–1293) to explain his ordinances to the people.14
Another royal supporter of chũ’ nôm writing was Hồ Quý Ly (胡季犛, 1336–1407). When he
was the emperor (1400–1407), he endorsed efforts to replace chũ’ nho with chũ’ nôm as the of-
ficial writing of Vietnam.15 Unfortunately, the reform did not take off, because Chinese forces
of the Míng dynasty invaded in 1407 and ended his rule. Despite the promotion of chũ’ nôm
by some among the ruling elites, the script never gained enough prestige to replace chũ’ nho
as the script for formal writing except during very brief periods. That role continued to be
fulfilled by chũ’ nho, as it remained the script of choice for official and scholarly purposes until
the beginning of the 20th century.
The use of chũ’ nôm for creative purposes continued to expand, however, and left behind
hundreds of literary examples.18 It was used to record folk literature and came to be associated
with Vietnamese cultural independence. In the 18th century, when chũ’ nôm writing reached
its peak, many Vietnamese writers and poets used this script to compose their works.
Chũ’ nôm and chũ’ nho remained in use until the arrival of French missionaries and
colonizers in the 19th century, when the promotion of a drastically different script came to
dominate the writing of Vietnamese and eventually replaced the Chinese writing system to
become the national script.
The use of Romanized writing for Vietnamese was rather limited in the beginning. For
about 200 years after its invention – that is, until the military conquest of Vietnam by France
in the mid-19th century – it was restricted to the Catholic community. Even within this group,
its use was very limited, possibly mainly as a tool for foreign missionaries to learn the lan-
guage. Furthermore, the Catholic community occupied a marginal position in Vietnamese
society. Its converts were commonly the most economically deprived and were viewed by the
Confucian literati as “the quintessential social parvenus, outcast devotees of a foreign god,
subversive and untrustworthy.”22 The Romanized writing system associated with the Catholic
converts were possibly perceived with equally aversive disdain, and chũ’ nho continued to be
the dominant form of writing as Confucian ideology remained the social mainstream.
So, how did the Roman alphabet eventually become the national script of Vietnam?
It had to do with both the colonial policy of the French, who occupied Vietnam for nearly
100 years since the 1850s, and the nationalist movement of modern Vietnam that led to the
nation’s independence in 1945. In what follows, we will see how, within a period of less than
100 years, quốc ngũ’ transformed from a form of writing imposed by the French colonizers
to a script that reflected cultural independence from France as well as from China. It will be
good to keep in mind that the general linguistic landscape for most of this period consisted of
four combinations of languages and writing systems: Chinese written in chũ’ nho characters,
Vietnamese written in the Chinese-based chũ nôm script, Vietnamese written with the Roman
alphabet, and French written in Roman letters.
Under French colonial rules, the administration sought to eliminate chũ’ nho and chũ
nôm to be rid of any Chinese influences. Replacing the Chinese writing system with the Ro-
man alphabet was an important component of the French colonialist policy, because the col-
onizers recognized that the security of French dominance fundamentally rested on a shift
in the cultural identity of the Vietnamese people from one that was “Chinese-inspired (or
Chinese-reactive) towards a new colonial identity loyal to France.”23 The extant use of Chinese
(chũ’ nho) or Chinese-like (chũ nôm) characters was perceived as an almost impenetrable
obstacle to this endeavor. The Sinitic script was too complex and represented a culture that
was too alien for the colonial administration to find ways to assimilate. Abandonment of the
Chinese writing system was thus considered the only way to remove the barrier to bringing
the Vietnamese closer under subjugation by the French.
Chapter 12 Writing Vietnamese 107
In the beginning, Romanized Vietnamese promoted by the French was seen by the
Vietnamese people as symbolic of French colonial control, while the traditional chũ’ nho and
chũ nôm styles of writing represented resistance against French rule. French promotion of
Romanization during this period was mostly in the colony in the south, where the occupa-
tion started. In northern Vietnam, Chinese writing maintained its prestige. In fact, chũ’ nho
remained the high-status means of written communication overall and continued to be used
for serious purposes during the late 19th century, including for anti-colonialism materials
advocating for the Vietnamese national identity. The nôm script for vernacular Vietnamese,
despite past endorsement by some of the ruling elites, largely continued to be seen as vulgar
in the first half century of French rule.
The transition in the Vietnamese perception of quốc ngũ’ happened amidst the ambiv-
alent attitude of the French colonizers towards the Romanized writing. On one hand, the
colonizers made use of the Western script as a tool to dissociate the educated elites from tra-
ditional Confucian learning encoded in chũ’ nho. Elementary schools were set up in the major
centers of the colony where indigenous students would learn to read and write Romanized
Vietnamese. French language schools, often staffed by Catholic priests, were established in
various locations of the colony. At the same time, as Romanized writing received broad pro-
motion, the learning of the Chinese script was discouraged. In 1865, the termination of the
Chinese-based examination system in the colony dealt “a final deathblow”25 to Sinitic writing
in the French colony. In 1910 quốc ngũ’ was made the script for all public documents.
On the other hand, the colonizers were wary of raising the literacy rate of the Vietnam-
ese masses, for fear that they would become more difficult to subjugate. The very case of how
the term quốc ngũ’ came to be used to refer to Romanized Vietnamese serves to illustrate the
situation. According to John DeFrancis, the term quốc ngũ’ in the sense of “national language”
108 Part III Borrowing the Chinese writing system
rather than “national script” was initially used to refer to the Vietnamese language as opposed
to Chinese long before the European missionaries arrived. It was later used by Vietnamese
Catholic converts to refer to the Romanized writing of their own language. The French colo-
nizers adopted this usage, possibly not entirely “conscious of its possible nationalistic impli-
cations relative to France”26 at the time. When they became uneasy with the association of the
“national language” with Vietnamese – a language other than French – it was already too late
to do anything about the usage.
Notes
4 Ibid., p. 21.
5 Nguyen (1984, p. 1).
6 Taylor (1983, p. 206); DeFrancis (1977, p. 23).
7 DeFrancis (1977, pp. 24–25).
8 This example comes from ibid., p. 24; also see Taylor (1983, p. 206).
9 See DeFrancis (1977, pp. 21–23).
10 See Rogers (2005, p. 76), Table 4.27 for a similar example, in which 美 and 母 form a composite
character for me. ‘mother.’
11 These two examples are both from ibid., Table 4.27.
12 See Nguyen (1984, pp. 10–11).
13 These three examples are all from Rogers (2005, p. 76), Table 4.27.
14 Ibid.
15 Le and O’Harrow (2007, p. 421).
16 DeFrancis (1997, p. 23).
17 Ibid.
18 Le and O’Harrow (2007, p. 421).
19 DeFrancis (1977, pp. 57–58).
20 Nguyen (1984, p. 6).
21 DeFrancis (1977, p. 51).
22 Le and O’Harrow (2007, p. 423).
23 Ibid.
24 Quoted by DeFrancis (1977) on p. 77 from a quote in Bouchot (1925, pp. 36–37).
25 DeFrancis (1977, pp. 79–81).
26 Ibid., p. 87.
27 Ibid., pp. 82–84.
28 Le and O’Harrow (2007, p. 425).
PART IV
T
HE CHINESE SCRIPT IS widely regarded as one of the most difficult in use. In China’s
recent history, embroiled in a national crisis of survival, the issue of writing became
in itself a matter of life or death. The author Lǔ Xùn (鲁迅/魯迅, 1881–1936), a lead-
ing figure of the New Culture Movement (新文化运动/新文化運動 Xīn Wénhuà Yùndòng
1917–1921), cautioned with all sincerity: “Rid China of Chinese characters lest China dies!”1
Lǔ Xùn was not the only one who saw Chinese characters as a serious impediment to learning
and literacy, technological advancement, and overall modernization of the nation. A series of
reform proposals were put forward from the late 19th to the mid-20th centuries, some more
successful than others, arousing heated debates that have reverberated to this day. Meanwhile,
rapid development in computer technology altered the competitive landscape of Chinese
characters versus alphabetical writing once and again, posing new questions while offering
new answers to the debates of writing reform. This chapter and the following few provide
an overview of the script reform proposals and some of the changes that ensued during this
time. The emphases are on phonetic writing (Chapter 13) – especially p ī ny ī n (Chapter 14) –
character simplification (Chapter 15), and a few issues concerning the impact of computer
technology on writing the Chinese language (Chapter 16).
As you probably already know, p ī ny ī n is currently the global standard for writing Chi-
nese using the Roman alphabet and has been since 1982. P ī ny ī n was an outcome of the script
reform sponsored by the People’s Republic of China (PRC) government in the 1950s–1980s.
It was by no means, however, the first or only phonetic script that had been proposed or im-
plemented for Chinese. To proselytize their religion among Chinese-speaking populations,
Western missionaries devised Romanization systems for transcribing Mandarin in as early
as the 16th century. About two and a half centuries later, with successive military and diplo-
matic conquests by foreign powers aimed to colonize China, the desire to write Chinese in a
simpler script arose both in and outside of the country. Chinese intellectuals, eager to save the
nation from foreign invasions, embraced the notion of phonetic writing by devising hundreds
of phonographic scripts. Meanwhile, Western scholars designed Romanization systems that
went on to serve as the standard in international communications for nearly a century. In this
chapter, we will trace the major historical precedencies that led up to the eventual creation of
p ī ny ī n. We will then take a closer look at p ī ny ī n in the next chapter.
Luó Míngjiān 1543–1607), Matteo Ricci (利玛窦/利瑪竇 Lì Mǎdòu 1552–1610), and Nicolas
Trigault (金尼阁/金尼閣 Jīn Nígé 1577–1628) made especially significant contributions. Rug-
gieri and Ricci were pioneers in the endeavors to overcome the lingual-cultural hurdle the
missionaries encountered in proselytizing Catholicism. They worked together in Macau (澳
门/澳門 Àomén) and Zhàoqìng (肇庆/肇慶) and, during 1583–1588, compiled a Portuguese-
Chinese dictionary to aid their work. A typical page in this dictionary consisted of a few col-
umns: the first column was word entries in Portuguese arranged in the alphabetical order,2
and the second column was Romanization of the Chinese words listed in the third column.3
On the first few pages, there was also a fourth column with the word entries in Italian.
This dictionary adopted Ruggieri’s Romanization system that used 16th-century Italian
and Portuguese orthography to represent Mandarin syllables of the same era. Table 13.1 and
Table 13.2 offer us a few examples.4 Note that the sound system of 16th-century Mandarin was
different from today’s Mandarin,5 so the specific speech sounds represented in the second and
third columns may not always be the same.
missionaries use this system in their writing.7 Unfortunately, the dictionary went missing and
is yet to be found.8 Towards the end of his career, in 1605, Ricci published in Beijing a work
called Xīzì Qíjì (西字奇迹/西字奇蹟The Miracle of Western Letters). It was a booklet of six
pages containing three biblical stories he had written in Classical Chinese and handwritten in
Chinese characters. The Chinese text was annotated using Ruggieri’s Romanization system9 to
assist Western readers with pronunciation.
Twenty years after Ricci’s Xīzì Qíjì was published, another missionary, Nicolas Trigault,
produced a three-volume dictionary titled Xī Rú Ěrmù Zī (西儒耳目资/西儒耳目資 Aid to
the Eyes and Ears of Western Literati), which was essentially a guide to the pronunciation of
Chinese charactrs. The Romanization system Trigault used was based on Ricci’s system in the
Chinese-Portuguese Dictionary.
Chinese intellectuals before Ricci’s time were not completely uncritical of their difficult
native script. As far back as in the Sòng dynasty, the historian Zhèng Qiáo (郑樵/鄭樵 1104–
1162) already noted the simplicity of an alphabetical script while studying Sanskrit and made
comments on the cumbersomeness of the Chinese writing in comparison.13 Fāng Yιˇzhì (方以
智 1611–1671), a scholar-official in the late Míng to early Qīng dynasty also pointed out the
advantages of Western alphabetical writing.14 He observed that Western scripts represented
words based on sounds, and the sounds, in turn, represented the meaning of the words. Thus,
there existed a straightforward one-on-one relationship between words, sounds, and meaning.
Chinese characters, on the other hand, were confusing because of “their interchangeability and
borrowing.”15 That is, for semantic-phonetic compounds, the same sound element may indicate
completely different pronunciations in different characters, and the semantic component only
vaguely suggests the meaning of the character, if at all. Many characters take on multiple, un-
related meanings through phonological extension. Therefore, the relationship between sound,
meaning, and the written graph is a lot more entangled and complex than alphabetical writing.
The missionaries’ earliest attempts at phonetic scripts for Chinese were aimed primarily
at helping Western scholars learn to read and pronounce Chinese characters and were not in-
tended in any way to reform the Chinese script for the Chinese. Whatever impact they had on
how the Chinese viewed their own writing system was probably momentary and minimal.16 It
would take at least another 200 years – and, perhaps more importantly, an imminent crisis of
the nation’s survival brought on by the attacks of Western powers – for the phoneticization of
Chinese writing to become a major issue of debate.
it possible for every Chinese person to be able to read and write. The complicated traditional
Chinese script, in their eyes, put the Chinese people at a great disadvantage compared to those
in Europe, America, and Japan. Lú Zhuàngzhāng (卢戇章/盧戇章 1854–1928) observed that
anyone above the age of ten in Europe and America, even those in remote and isolated areas,
was able to read.19 In stark contrast, as Cài Xīyǒng (蔡锡勇/蔡錫勇1847–1897) lamented, in
China, even those scholars who spent their entire lives reading books were not able to learn
all the Chinese characters.20 The extremely low literacy rate among the Chinese, in their mind,
was caused at least in part by the fact that Chinese characters were too numerous and too
complex. What was more, the Chinese writing system could also be confusing – the same
sound was usually represented by multiple characters, and the same character often had mul-
tiple readings. The much more straightforward relationship between language and writing in
alphabetical scripts was what the reformers wished to bring to the writing of Chinese.
Another dimension that made acquiring literacy in Chinese drastically more difficult
was the stylistic difference between speech and writing. For the Chinese, being able to write
did not simply mean being able to transcribe in characters or letters what one said; rather, one
would have to write in the literary style, i.e. Classical Chinese (文言 wényán). As explained
in Chapter 5, Classical Chinese was essentially a different language from Modern Standard
Chinese. Thus, to become literate in China around the turn of the 20th century, one would
not only have to master a complex and demanding script, but also have to learn an ancient
language specifically reserved for writing.
To save China meant to modernize China, and to modernize China required the devel-
opment of science and technology. The reformers believed that writing Chinese in a phonetic
script would not only greatly improve basic literacy of the Chinese population, but would also
open up communicative channels for Chinese scientists to join international exchanges for
Chapter 13 Phonetic writing before pīnyīn 117
the advancement of scientific learning and technological innovation. The traditional Chinese
script, on the other hand, presented an enormous hurdle to this prospect. We will not go into
details here on this topic, but we will return in Chapter 16 to look at the challenges presented
by the Sinitic script to the development of modern information technology in China.
If the practical motivation for phoneticizing the Chinese script was to promote literacy
and education and to develop science and technology, then the ideological motivation was to
establish democracy in creating a modern China. For millennia, reading and writing Chinese
characters had been the privilege of the educated few. The Chinese script, together with Clas-
sical Chinese, represented to the reform-minded scholars and politicians a non-democratic
mode of written communication that had long deprived common Chinese of their basic rights
to participate in society as citizens. It was considered outdated and incompatible with the
modernization of China. By contrast, phonetic scripts, which were much easier to learn and
to use, embodied the democratic ideal they were pursuing. Speaking of the phonetic scripts
of other nations, Wáng Zhào (王照 1859–1933) expressly linked writing systems to the insti-
tution of political equality.
[A]lthough the scripts of the other nations are shallow, each of the people throughout
those nations are thoroughly conversant with them because language and script are
consistent. Their letters are simple and convenient and, for even the dullest youths,
the age they can speak is the age when they become conversant with writing. There-
fore, it is the birth right of all to become specialists in the matters that are conveyed
by their writing. Daily, they strive to refine themselves and make progress, no matter
whether intelligent or stupid, noble or common, young or old, man or woman.22
Companion Website
The various proposals to phoneticize Chinese writing about a century ago may appear
farfetched ideas from today’s perspective. After all, to save China from its perils, what was
required was much, much more than a simpler writing system. To build a strong nation with
cultured citizens, the Chinese people must tremendously expand access to educational re-
sources, an endeavor that would require profound changes of the Chinese society in both
political and economic terms. It may be difficult to imagine that the reformers did not un-
derstand this; yet it is perhaps not difficult to understand that, in searching for solutions to
China’s problems, what stood out to the reformers were the fundamental distinctions between
China and the West. To them, the difference between the complex Chinese writing system and
the simpler Roman alphabet had come to symbolize the disparity in general national strength
between China and the more industrialized foreign invaders. In comparison to those nations
using phonetic scripts, China was seen as impoverished, weak, and illiterate as much as the
118 Part IV Reforming the Chinese script
Chinese characters were seen as complicated, cumbersome, and impossible to learn. In this
sense, the fact that many reform-minded intellectuals pinned their hopes of saving China on
revolutionizing the writing system perhaps spoke louder about the cultural significance of
the character-based Chinese script than any transformative power that reforming it had been
presumed to possess.
As the phoneticization movement went on, Romanization became the most popu-
lar choice among phonographic writing schemes. Scripts making use of character radicals or
strokes maintained a degree of “Chineseness” in their appearance, but they lacked the interna-
tional currency of the Roman alphabet, a script widely used in the Western world. Shorthand-
style scripts did not gain much traction, having neither a Chinese appearance nor ready
intelligibility for Western users. Eventually, the Latin alphabet became the winning script,
and Romanization systems developed in this era paved the way for the design and adoption
of pīnyīn half a century later. Besides Lú Zhuàngzhāng’s Qiēyīn Zìmǔ, two other well-known
Latin scripts were Zhū Wénxióng’s (朱文熊1883–1961) Jiāngsū Xīn Zìmǔ (江苏新字母/江蘇
新字母 ‘new alphabet for Jiāngsū speech’) published in 1906 and Liú Mèngyáng’s (刘孟扬/劉孟
揚 1877–1943) Zhoˉngguó Yīnbiāo Zìshū (中国音标字书/中國音標字書 ‘Chinese sound-mark
script’) two years later. Zhū designed the script to write the Chinese speech of Sūzhoˉu (a city in
Jiāngsū), his native language. However, he intended to one day modify the script to write North-
ern Mandarin.31 This task was fulfilled by Liú’s script, which was designed for Běijīng Mandarin.
One improvement both of these scripts had over Lú Zhuàngzhāng’s earlier system was that they
did not create new letterforms and therefore used a significantly smaller set of symbols.32
斯 Zhái Lῑsῑ 1845–1935), who later on perfected the system. The origin of the system was in
Mandarin Textbook (官话课本/官話課本 Guānhuà Kèběn), the first Chinese language text-
book in English, that Wade published in 1869 in his Collection of Language for Self Teaching
(语言自迩集/語言自邇集 Yǔyán Zì’ěr Jí). Giles then revised and completed the system in
his Chinese-English Dictionary of 1892. Both men were British diplomats in China, and a
Romanization system of Mandarin Chinese based on the Běijīng dialect was necessary for
their work. Because of Britain’s global dominance, the system became widely used in English
publications, including reference books and academic writing, and served as the de facto stan-
dard for transcribing Chinese for most of the 20th century. It was replaced by p ī ny ī n in 1979
in Mainland China but continued to be used in Táiwān for personal names, place names, and
street names.
Table 13.4 hinese city names in the Postal Spelling System, Wade-Giles, and p ī ny ī n without tone
C
marks
Postal Wade-Giles Pı̄nyı̄n Characters
Peking Pei-ching Beijing 北京
Nanking Nan-ching Nanjing 南京
Tientsin Tʻien-chin Tianjin 天津
Chungking Chung-chʻing Chongqing 重庆
Tsinan Chi-nan Ji’nan 濟南
Tsingtao Chʻing-tao Qingdao 青岛
122 Part IV Reforming the Chinese script
Aside from the Wade-Giles system and the various other Romanization conventions
represented in the Postal Spelling System, Yale Romanization (耶鲁方案/耶魯方案Yēlǔ
Fāng’àn) was another system developed by Westerners to primarily serve the need of West-
erners. In fact, the term “Yale Romanization” encompasses Romanization of four major East
Asian languages: Mandarin, Cantonese, Japanese, and Korean, each developed by different
authors. Yale Mandarin was designed by George A. Kennedy (1901–1960), a Sinologist at
Yale University, in 1943 to teach Chinese to American troops to be deployed to China during
World War II. At the time, of course, Wade-Giles was the international standard for Man-
darin Romanization. The Yale system, however, was designed to be more intuitive as a lan-
guage-learning tool, because its spelling was more closely based on the English pronuncia-
tion of Roman letter combinations. For example, the p ī ny ī n syllables zhi, chi, shi, and ri are
spelled as jr, chr, shr, and r in Yale, much easier to pronounce for English speakers than the
Wade-Giles chih, ch‘ih, shih, and jih. Table 13.5 gives these and a few additional examples. In
addition, Yale Mandarin dropped the left apostrophe (‘) used to represent aspiration in Wade-
Giles; instead, it made use of p ī ny ī n-like pairs, distinguishing between p-b, t-d, and k-g. This
also made it more user-friendly.
Table 13.5 Yale, Wade-Giles, and pıˉnyıˉn systems compared (tones not marked)
Yale Wade-Giles Pı̄nyı̄n Characters
jr chih zhi 知
chr ch‘ih chi 池
shr shih shi 是
r jih ri 日
ji chi ji 及
chi ch‘i qi 七
xyi hsi xi 息
Tones in Bopomofo are marked using diacritics similar to those in pīnyīn. First tone is
not marked, but second (´), third (ˇ), fourth (`), and neutral (˙) tones are. The diacritics are
placed to the upper right of the rhyme. For example, Zhoˉngguó wénzì (中国文字/中國文字
‘Chinese script’) in Bopomofo is written as:
GR in pıˉnyıˉn
GR is in fact used along with pıˉnyıˉn in one particular case. The Chinese provincial names Shaˉnxıˉ
(山西) and Shǎnxıˉ (陕西/陜西) become indistinguishable in standard pıˉnyıˉn spelling when tone
marks are omitted as they often are in Western text: both will be Shanxi. To solve this problem, GR
spelling is adopted for this pair. The first tone Shaˉn maintains the original spelling Shan, while the
third tone Shǎn is written as Shaan with a double vowel.
Latinxua movement occurred on the national scale. Organizations appeared across China to
promote and teach the script.41 Within a few years, Latinxua systems for as many as 14 Chi-
nese dialects (languages) were created.42 In Northwestern China, an area under Communist
control, the government adopted Latinxua as the official script.43 By 1955, when the Latinxua
movement finally ended, about 20,000 people had learned the script through more than 300
organizations,44 and over 300 publications with half a million copies had been produced.45 Al-
though neither of the aforementioned goals was eventually realized, Latinxua, together with
Bopomofo and GR, to say the least, served as the linguistic foundation for the development of
a later Romanization system, pīnyīn.46
Notes
A
S WE LEARNED IN the previous chapter, Chinese reformers proposed and implement-
ed a variety of phonographic scripts from the 1890s to the 1950s. Eager to promote lit-
eracy in a mostly illiterate nation, they saw the phoneticization of Chinese writing as a
doorway to broader access to education necessary for a stronger China. None of the proposals,
however, was able to take hold in a country overwhelmed by foreign invasions, internal divi-
sions, and a civil war. The founding of the People’s Republic of China (PRC) in 1949 eventually
made it possible to place writing reform once again on the national agenda. Sponsored by the
PRC government, efforts began in the same year to study, plan, and carry out official changes to
Chinese writing and continued well into the 1980s. This reform had two major aspects: popu-
larization of pıˉnyıˉn (拼音) and simplification of Chinese characters. In this chapter, we will fo-
cus on pıˉnyıˉn, and we will discuss script simplification in the next chapter. If you took the brief
tutorial on reading pıˉnyıˉn in Part I of this book, you will have already become familiar with
the system and may also be able to rely on it to pronounce Chinese syllables and words. In this
chapter, we will learn more about pıˉnyıˉn’s history, current uses, and spelling rules. First of all,
however, let us consider what pıˉnyıˉn is and what it is designed to write.
What is pıˉ ny ıˉ n?
Pıˉnyıˉn is a phonemic writing system that uses the Roman alphabet to represent Modern Stan-
dard Chinese (MSC), or pǔtoˉnghuà (普通话/普通話 ‘common speech’). Pıˉn (拼) means “to
spell,” and yıˉn (音) means “sounds,” referring to “speech sounds,” so pıˉnyıˉn literally means
“spell (speech) sounds.” The script makes use of 25 of the 26 Roman letters – v is not used –
plus ü, and it marks tones with diacritics (e.g., mā, má, mǎ and mà).
The term pıˉnyıˉn may have slightly different meanings in English than in Chinese. In
English, it is used as a proper name to refer specifically to the phonemic writing system
designed for MSC under the PRC government. Used in this sense, it is an abbreviated form for
Hànyǔ pıˉnyıˉn fāng’àn (汉语拼音方案/漢語拼音方案 ‘Chinese sound-spelling scheme’) or
Hànyǔ pıˉnyıˉn. In Chinese, however, pıˉnyıˉn can also be used in a general sense to mean “spell-
ing (speech) sounds,” or, in another term that is familiar to us, “phonographic.” For example,
pıˉnyıˉn wénzì (拼音文字; 文字 wénzì ‘writing, script’) can be understood as “phonographic
scripts.”
What does pıˉnyıˉn write? It is, in this regard, very different from Latinxua, the
Romanization system we have previously discussed. As you may recall, Latinxua was intended
not only to replace Chinese characters but also to write a range of Chinese languages. It was
designed to enable autonomous Romanization of any variety of Chinese dialect and had hence
128 Part IV Reforming the Chinese script
underrepresented certain dialect-specific features. For instance, tones were not marked in
Latinxua because dialectical variations, to a great extent, manifested in tonal differences.
Latinxua provided the linguistic basis on which p ī ny ī n was created.1 However, p ī ny ī n was
designed for writing MSC only, and, at least for the foreseeable future, it is not used for writing
regional dialects (languages).
Although the p ī ny ī n script, in linguistic terms, is a full-fledged writing system, its sta-
tus is secondary to the character-based script: it has been primarily used as a pronunciation
guide, for example, in printed Chinese text, on street signs, and in Chinese-language class-
rooms. It is also used to transcribe Mandarin speech or to represent Chinese in Western texts.
According to the linguist Zhōu Yǒuguāng (周有光 1906–2017), who oversaw the development
work of p ī ny ī n, the script was not meant to replace the character-based writing system, and
“that was made clear when p ī ny ī n was promulgated.”2
The status of pıˉnyıˉn, however, remained an issue of debate as the desire for a phono-
graphic script rose and fell. In the early stages of research on natural language processing
aided by computers, the vast and unsorted set of characters in the Chinese writing system
presented enormous challenges to the digitization and processing of Chinese text. Frustrated
developers saw pıˉnyıˉn as a solution, and this led them to propose having two full scripts for
Chinese, hànzì (汉字/漢字 ‘Chinese characters’) and pıˉnyıˉn, elevating the latter to a position
on par with conventional characters. The idea or phenomena that a language is written in
two distinctive scripts used for complementary purposes is referred to as digraphia.8 Despite
the advocacy by researchers as well as some linguists and language educators for pıˉnyıˉn to
become a full script alongside Chinese characters, it never achieved this status in official
terms. Further development in digital technology has also made it easier to process Chinese
text on computers. On the other hand, pıˉnyıˉn has become indispensable in mediating the
input of characters on computers, cell phones, and other modern electronic devices, leading
some linguists to consider hànzì and pıˉnyıˉn already in a digraphic relationship.9
Chapter 14 Pīnyīn 129
History of pıˉnyıˉn
Motivations for creating pıˉnyıˉn came from, first of all, the PRC government’s desire to establish
a national and international standard for phonographic writing of Chinese. Under the
Republic government (1912–1949), the official phonographic script was Bopomofo, a char-
acter-inspired system that continued to be used after the founding of the PRC and had been
in use for almost four decades by the time the work on pıˉnyıˉn started. To establish a phono-
graphic script for Chinese as an international standard, the PRC government came to rec-
ognize that Bopomofo needed to be replaced by a Romanization system because of the much
greater international currency of the Roman alphabet. As we have learned in the last chap-
ter, a variety of Romanization scripts was already in existence at the time to transcribe Chi-
nese: most notably, Wade-Giles created by British scholars, Gwoyeu Romatzyh (GR) developed
under the Nationalist government, and Latinxua sponsored by the Communist establishment.
However, each of these systems had major shortcomings. For example, Wade-Giles employed di-
acritics that its users often mistakenly omitted, GR had complicated spelling rules, and Latinxua
did not mark tones. The coexistence of these systems also caused much confusion. Adding to the
chaos was the fact that personal names might be Romanized differently in different languages. For
example, the name of Zhōu Ēnlái (周恩来/周恩來 1898–1976), the first premier of the PRC, was
represented as “Chou Enlai” in English and “Tchou Enlai” in French.10 Clearly, there needed to be a
new, unified Romanization system to reduce confusion and aid communication across languages.
Another motivation for a new and improved Romanization system was to facilitate the
PRC government’s goal to promulgate Modern Standard Chinese or pǔtōnghuà as the stan-
dard speech for the nation. The new Romanized script was viewed as a necessary tool to help
with the learning of pǔtōnghuà pronunciation, and thus, the script must be tailored to tran-
scribing the sounds of pǔtōnghuà. Pıˉnyıˉn was designed with this in mind. Since its creation, it
has been used as a pedagogical tool in teaching pǔtōnghuà to both native and foreign learners.
The work on a new phonographic script began as soon as the PRC government formed
in 1949. Up until 1954, the effort was mainly on devising a script based on the strokes of Chi-
nese characters. In 1955, the PRC government charged a team of about 20 specialists under the
Committee of Chinese Writing Reform (中国文字改革委员会/中國文字改革委員會Zhōng-
guó Wénzì Gǎigé Wěiyuánhuì) with the task of developing a Romanization system for Chinese.
It was not until later that year, however, that the decision was made to adopt the Roman alpha-
bet. Headed by Zhōu Yǒuguāng, the team of specialists began to develop what eventually came
to be known as pıˉnyıˉn. The process took as long as three years – a “full-time job” in Zhōu’s
words.11 Besides studying and comparing past Romanization systems, they considered hundreds
of new proposals from the general public. About 1,700 phonographic writing schemes, some
using the Roman alphabet, some not, were submitted to the committee.12 In 1958, pıˉnyıˉn was
published and became the official Romanization system in Mainland China.
new system be based on Chinese characters. The committee charged with writing reform, after
intensive deliberations, came up with six proposals, four based on Chinese characters, one on the
Cyrillic script, and one on the Latin alphabet. A heated debate ensued. Máo eventually supported
the adoption of the Roman alphabet. Premier Zhoˉu Ēnlái, in his speech in 1958, officially endorsed
pıˉnyıˉn and concluded that “the adoption of the Latin alphabet will . . . not harm the patriotism of
our people,”13 ending the debate on script choice.
In the decades after pıˉnyıˉn was declared the official Romanization system of the PRC,
it made strides in becoming the international standard. In 1979, the United Nations passed a
resolution to adopt pıˉnyıˉn as the official system in transcribing Chinese personal and place
names. Since that year, an increasing number of publishers in the West have switched to
using pıˉnyıˉn.14 In 1982, the International Organization for Standardization recognized pıˉnyıˉn
as the standard Romanization for Chinese (ISO 7098). In Táiwān, a variety of Romaniza-
tion systems, including Wade-Giles, GR, and the later-designed Tongyong Pinyin (通用拼音
Toˉngyòng pıˉnyıˉn, 2002–2008), had been in concurrent use, before pıˉnyıˉn was adopted in 2009
as the official Romanization system.
Chinese president’s name appears as Xi Jinping (习近平/習近平) in news media, and the cap-
ital city of China, as you know, is spelled Beijing (北京).
Pıˉnyıˉn input methods are probably the most commonly used for entering Chinese
characters on digital devices using a standard keyboard. Most educated speakers of Chinese
today have learned pıˉnyıˉn, so using this kind of input methods requires minimal practice.
After a pıˉnyıˉn syllable has been entered, the user is presented with a list of characters sharing
the same pronunciation and makes a choice by selecting the desired character or entering
a corresponding number on the keyboard. The process is not as straightforward as input-
ting pure pıˉnyıˉn text or English, but with continued improvements, such as the addition of
context-based adaptive prompts, experienced users can input Chinese-character text with
adequate efficiency for day-to-day communicative purposes.
Spelling rules
Can pıˉnyıˉn replace Chinese characters, after all? Sociocultural considerations aside, for pıˉn-
yıˉn to be used as a fully functioning script, it needs to overcome a number of linguistic chal-
lenges. Among these, word boundaries and homophony are the most significant. In 1988,
Zhōu Yǒuguāng and his committee published an officially approved set of spelling rules for
pıˉnyıˉn.19 With these rules in place, some argued, pıˉnyıˉn was in a good position to tackle the
challenges and become “a full-fledged script.”20
Unlike reading and understanding text in Chinese characters, reading pıˉnyıˉn requires
the marking of word boundaries for easier comprehension. Words are natural units of complete
132 Part IV Reforming the Chinese script
meaning in a language, so it helps readers understand the text when words are segmented. In
Chinese character-based writing, word boundaries are typically not marked. Characters are
placed right next to each other no matter how long the sentence may be. Readers must parse
the text into words as they read based on their vocabulary knowledge and what they know
about the context. This is workable because characters vary a great deal in shape and each
character represents meaning in addition to sounds. For example, the short sentence 今天
星期六 (Jıˉntiān xıˉngqıˉliù, ‘Today is Saturday’) may contain two words to most readers: 今
天 (jıˉntiān, ‘today’) and 星期六 (Xıˉngqıˉliù, ‘Saturday’). The five symbols all appear different
from each other, and readers have the knowledge that 今天 (jıˉntiān, ‘today’) is a word and 星
期六 (xıˉngqıˉliù, ‘Saturday’) is another word. They also know from experience that 天星 (tiān
xıˉng) is not a word and neither is 期六 (qıˉ liù). Thus, drawing the word boundaries as 今/天
星/期六 (jıˉn tiānxıˉng qıˉliù) will not be correct. Marking word boundaries is imperative in
pıˉnyıˉn writing, because Roman letters repeat each other much more frequently and individual
letters do not represent meaning. For example, although reading and understanding “jıˉn-
tiānxıˉngqıˉliù” is not impossible, it requires significantly more effort than “jıˉntiān xıˉngqıˉliù.”
Imagine reading a sentence much longer than this in English. Reading Chinese in pıˉnyıˉn
without word boundaries would probably be just as difficult.
The primary role of the pıˉnyıˉn orthography rules is to standardize word boundaries in
pıˉnyıˉn writing. They aim to ensure that, on one hand, syllables of separate words are separated,
and on the other hand, syllables of the same words are written together. However, the spelling
rules currently are neither taught at school nor strictly enforced, so many may be oblivious of
their existence. For example, visitors in China may see signs such as in Figure 14.1, in which
p ī ny ī n syllables are written completely separately with a space in between very two syllables.
This practice likely mirrors the writing convention of syllable-based Chinese characters, but
in this case it may obscure the distinction between proper names and general names and
make the sign more difficult to understand. Based on pˉı nyˉı n orthography, the place name
Tong Hai (通海) should be written as one word as Tonghai, because both syllables are part of a
proper name, the name of a place. Da Hui Cun (大回村, Dàhuí Cūn, ‘Dahui Village,’ lit. ‘Great
Muslim Village’) should be written as Dahui Cun, because Dahui is a proper name and Cun
(村 cūn ‘village’) is a general noun.21 Another reason why p ī ny ī n spelling rules are often not
put into practice may be that word segmentation may not always be intuitive. In fact, scholars
have found that being accustomed to the character-based system makes Chinese users lack a
sense of “word.”22 The p ī ny ī n annotation for the smaller characters in the sign, for instance,
may be segmented into Chama Gudao (Majia Dayuan) or Chama Gu Dao (Majia Da Yuan),
depending on whether we consider Gudao (古道 gǔdào ‘ancient path’) and Dayuan (大院
dàyuàn ‘big yard’) as one or two words each in themselves.
Another challenge that has been cited for pˉı nyˉı n in considering its potential as an in-
dependent writing system is the prevalence of homophony in the Chinese language. A great
number of words in Chinese share the same pronunciation. Written in characters, they appear
different; in pıˉnyıˉn, they are not distinct from each other. For example:
Some point out that homophony may not be as big a problem as many think.23 In most
cases, contextual information should provide sufficient help needed to disambiguate homoph-
onous words. After all, if people have little difficulty understanding each other in speech, then
there is no reason why they should have major problems differentiating homophones in read-
ing. Occasionally when confusion arises, interlocutors in a face-to-face conversation could
ask questions to clarify for each other; a reader of a book written in p ī ny ī n, however, would
have little recourse. To solve this problem, some proposed adding a silent one-letter suffix to
differentiate homophonous morphemes or words. The suffixes function somewhat like radi-
cals in Chinese characters.24 For example:
The “w” represented the wáng (王) radical, and the “m” represented the mù (木) radical, each
present in the respective characters. ZhŌu Yǒuguāng objected to this proposal on the ground
that this would turn p ī ny ī n into an annotation system for Chinese characters rather than
a writing system for the Chinese language, which would go against the basic principles of
p ī ny ī n.25
Finally, the spelling rules also stipulate the use of punctuation in p ī ny ī n. Overall, the us-
age of punctuation marks in p ī ny ī n is similar to in English text. This means a few differences
exist between punctuating p ī ny ī n writing and text in Chinese characters: p ī ny ī n uses a dot (‘.’)
to represent a period, while character-based writing uses a circle (‘。’). Character writing uses
the symbol ‘、’ to indicate a pause shorter than a comma, as in making a list; p ī ny ī n does not
use this symbol and uses commas instead for such pauses. At the end of a line, a multisyllabic
134 Part IV Reforming the Chinese script
word may break and continue at the beginning of the next line. As in English, a hyphen is
required to indicate that the word is not yet finished, and the hyphen must be placed between
syllables.26
Companion Website
As a final note: the official pıˉnyıˉn orthography rules adopt the principle of following
the original. That is, the original Romanized forms are taken as the standard. This is also the
principle used by the UN’s Conference on Standardization of Place Names. By this principle,
any foreign name will have only one written form: the Romanization used in its language of
origin.
Notes
22 Taylor and Taylor (2014, p. 26) cites a study by Hoosain (1991) that finds that Hong Kong mid-
dle-school students were not able to proficiently segment a Chinese text into words.
23 See Zhou (2003).
24 Ibid., p. 125.
25 Ibid., pp. 110, 125.
26 Ibid., p. 120.
15 Simplification of Chinese characters
A
LONG WITH EFFORTS TO design and popularize a new Romanization system
(pıˉnyıˉn) for Modern Standard Chinese, the PRC government implemented policies
to simplify the character-based script. After work that spanned more than three de-
cades (1953–1986), the result was what is commonly known in Chinese as jiǎ ntı̌ zì (简体
字/簡體字 ‘simple-style characters’). The simplified script is used today in Mainland China,
Singapore, and Malaysia and is widely taught in the Chinese-language curricula in the West,
while in Táiwān, Hong Kong, and Macau, the traditional (pre-simplification) script, referred
to as fántı̌ zì (繁体字/繁體字 ‘complex-style characters’) in Chinese, remains the official style.
Script simplification under the PRC government incorporated work along two major
dimensions: reducing the number of strokes within characters and downsizing and standard-
izing the overall inventory of characters. In this chapter, we will discuss both aspects with a
focus on the process and outcome of the PRC’s script reform. This reform was by no means,
however, the first effort by the Chinese to systematically simplify the writing system in the
modern era. In fact, earlier experimentation laid the groundwork for the implementation of
this reform. We will begin by taking a look at some of these attempts.
1887–1939) brought forward a bill to the Republic government calling for adopting simplified
characters.5 He pointed out that, for thousands of years, Chinese characters had constantly
been reduced to simpler forms, and he advocated using such forms that were already in popu-
lar use as the accepted standard for writing Chinese.6 Furthermore, he proposed to create new
simplified characters: “existent simplified characters are not many, and we should . . . create
many more simplified characters . . . so that the thousands of most frequently used characters,
aside from those that have few strokes to start with, will all have simple forms.”7
With support from Chiang Kai-shek (蒋介石/蔣介石 1887–1975), president of the Re-
public of China (ROC), Qián in 1934 compiled a collection of about 2,400 simplified characters
from a variety of sources: folk characters in popular use, folk characters used in Classical novels,
characters simplified in cursive and running scripts, and variant characters in early dictionaries
138 Part IV Reforming the Chinese script
and stone inscriptions.10 Based on Qián’s work, in 1935, the Nationalist government in Nánjīng11
put forth an official First Set of Simplified Characters (第一批简体字表/第一批檢體字表Dìyıˉ
Pıˉ Jiǎntı̌ Zì Biǎo), which contained 324 folk characters. Due to strong objection from some con-
servative members of the government,12 however, this list was abolished the next year. Nonethe-
less, research on character simplification received significant attention among the educational
circles,13 and reference materials on a simplified script were published, including a Simplified
Character Dictionary (简体字典/簡體字典 Jiǎntı̌ Zìdiǎn) containing 4,445 characters, a Col-
lection of Commonly Used Simplified Characters (常用简字表/常用簡字表 Chángyòng Jiǎnzì
Biǎo) with 3,150 characters,14 and, in 1937, a collection called Simplified Characters: the First List
(简体字表第一表/簡體字表第一表 Jiǎntı̌zì Biǎo: Dì Yī Biǎo) with 1,700 characters.15 Later
that year, the Second Sino-Japanese War16 (1937–1945) broke out.
Liberated characters
During the Second Sino-Japanese War, efforts to simplify the Chinese script continued in areas con-
trolled by the Communist Party, the so-called liberated areas (解放区/解放區jiěfàng quˉ ). A large
number of simplified characters were used in books, newspapers, and other printed materials cir-
culated within the jiěfàng quˉ. Below are a few examples (traditional characters in parentheses):17
拥 (擁 yōng), 护 (護 hù), 卫 (衛 wèi), 胜 (勝 shèng), 运 (運 yùn)
动 (動 dòng), 艺 (藝 yì), 习 (習 xí), 团 (團 tuán)
These characters all have their origin in calligraphy works by the Táng dynasty master Yán
Zhēnqıˉng (颜真卿 709–785). As the war ended, the use of the jiěfàng quˉ characters spread to the
rest of the country and gained the nickname jiěfàng zì (解放字), “liberated characters.”18
After the founding of the PRC in 1949, the ROC government in Táiwān eventually de-
cided to maintain the traditional script that continued to be used till this day. This decision,
however, appeared to have been based more on political than linguistic concerns. In fact,
in 1952, Chiang Kai-shek reopened the subject of character simplification and a committee
was set up to conduct research on this topic. The initiative had significant support from the
public,19 but the idea again met with strong objection from some conservative members of the
legislature.20 By the time when the PRC government on the Mainland had successfully imple-
mented their script reform, the issue of character simplification quickly became politicized for
Táiwān. The Republic government decided to maintain the traditional script, taking a position
clearly opposing that of the Mainland government.
tionalists and the Communists, to become highly politicized. Chiang Kai-shek and his government
changed their attitude, as the Taiwanese scholar Lin An-wu (林安梧 1957–) explained:
Because our political opponent promoted the simplified style, we [had no choice
but to] advocate the traditional script. In philosophical terms, we were on the opposing
side of the enemy, the “passive” side. We were the slaves of the powerful “active” side
and were forced to take the opposing position.
Chiang Kai-shek stopped advocating for character simplification. In fact, anyone who dared to
do so would have been labeled a “red hat” and considered a collaborator of the Communists.21
FIGURE 15.1 Four variants of the character 车 in oracle bone and bronze inscriptions
Source: Adapted from Seybolt and Chiang (1979, p. 8)
140 Part IV Reforming the Chinese script
Initiatives to reform the Chinese script in the modern era continued the historical trend
to simplify the character forms. As you may recall from the previous chapter, the PRC govern-
ment began its work on writing reform as soon as it was officially formed. In October 1949,
with Máo Zédōng’s support, the Chinese Writing Reform Association (中国文字改革协会/
中國文字改革協會 Zhoˉngguó Wénzì Gǎigé Xiéhuì) was established. Its initial focus was to
devise a phonographic script for Chinese. However, script simplification soon became one
of the key components of the overall writing reform, and new committees were set up within
the government to head the relevant research and development efforts. Eventually, in 1954,
the Committee for Chinese Writing Reform (中国文字改革委员会/中國文字改革委員會
Zhoˉngguó Wénzì Gǎ igé Wěiyuánhuì) was created under direct supervision of the State Council
(国务院/國務院 Guówùyuàn) as an organ to oversee the various aspects of script simplifica-
tion. In 1985, it was renamed to State Language and Writing Work Committee (Guójiā Yǔyán
Wénzì Goˉngzuò Wěiyuánhuì 国家语言文字工作委员会/國家語言文字工作委員會). The
change of name signaled the end of major reform work, though the Committee remained
directly under the State Council. Finally, in 1994, it became part of the Ministry of Education.
Although the central government of the PRC was the chief promoter and architect of
the writing reform, simplification of the Chinese characters, at least in its early stage, was not
entirely a top-down endeavor. The Committee actively sought and considered input from the
general public, who also took it seriously to participate in the process. In 1955, the Committee
issued its Plan for Simplifying Chinese Characters (Draft) (汉字简化方案草案/漢字簡化方案
草案 Hànzì Jiǎnhuà Fāng’àn Cǎoàn) and invited feedback from various groups and agencies.
Within half a year, it received more than 5,000 letters or written opinions, and about 200,000
people of varied educational levels took part in the discussion.24 In 1956, the State Council
published the revised Plan for Simplifying Chinese Characters (汉字简化方案/漢字簡化方案
Hànzì Jiǎnhuà Fāng’àn).
Political debate
The reform efforts started off among heated debate and intense controversy characterized by
ˉ
the political sensitivity of that time. One prominent scholar and critic of the reform, Chén Mèngjia
(陈梦家/陳夢家 1911–1966), commented, “[People] have opposing views but do not dare express
them.” Verbal exchanges between those involved were deeply emotional. Because of his criticism,
Chén Mèngjiaˉ was charged with “portraying the new China as a cold, dense forest that is dark and
without the light of day” and accused of “appealing only to emotions in attacking people, of fear-
ing disorder and hating Marxism.” Another critic, Guaˉ n Xıˉ (关西/關西), expressed his concern by
saying that “[even Chairman Mao] who was not a specialist in language and writing studies [had
been deceived by] the vigorous propaganda that the masses unanimously support writing reform.”
His view was criticized as a “rude accusation of Chairman Mao. . . [that] exceeds the limits of what
people will endure.”25 Some critics of the script reform were labeled “Rightists,” and their views
were silenced in the Anti-Rightist Campaign by the end of 1957. In January 1958, Premier Zho ˉu
Ēnlái (周恩来) gave a speech, “The immediate tasks in writing reform (当前文字改革的任务/當前
文字改革的任務 Daˉngqián wénzì gǎigé de rènwù),” officially endorsing the simplification of the
Chinese script. The debate was over.
Chapter 15 Simplification of Chinese characters 141
To what extent are the simplified characters simplified? The 1956 Plan consisted of 515
simplified characters and 54 simplified radicals.26 Before simplification, the 515 characters
averaged 16.8 strokes per character; after simplification, they averaged 8.16 strokes per char-
acter, a 49% reduction.27 Of course, this effect will be less pronounced when measured in a
regular text that contains not all simplified characters. In an article of 2,258 characters, for ex-
ample, it was found that simplified characters reduced the total number of strokes by 22.6%.28
This percentage may also change depending on the type of the text measured. An alternative
measure that can be informative is to look at the 2,000 most frequently used characters. For
these characters, simplification reduced the average number of strokes per character from
11.2 to 9.8 by 12.5%.29
How different is the “simplified script” from the “traditional script”? This question may
not be easy to answer. The 1956 Plan contained 54 simplified radicals that would generate a
great number of characters simplified by analogy, but it was not clear which characters specif-
ically should be simplified in that manner. For example, the radical 廣 (guǎng) was simplified
as 广, and by analogy, the characters 礦, 曠, and 鄺 (all pronounced kuàng) would be simpli-
fied as 矿, 旷, and 邝. Were there other characters containing 廣, such as 兤 (huǎng) and 櫎
(huǎng), that should also be simplified? Questions like this did not have clear answers, because
the work to further spell out the simplified character forms was not done until about a decade
later. In 1964, the Committee published a General Table of Simplified Characters (简化字总
表/簡化字總表 Jiǎnhuàzì Zǒngbiǎo) on the basis of the 1956 Plan, and “the simplification of
Chinese characters became an accomplished fact.”30 The General Table was reissued with slight
revisions in 1986 and contained a total of 2,235 simplified characters (two less than the 1964
publication) divided into three categories. Table 15.1 provides a few examples (traditional
characters in parentheses) for each of them.
Table 15.1 Examples from the 1986 General Table of Simplified Characters
Chart Examples
Chart 1: 350 independent simplified characters 办 (辦), 标 (標), 才 (纔), 奋 (奮), 护 (護), 烂
that cannot be used as combinatory elements to (爛), 灭 (滅), 台 (台, 臺, 檯, 颱)
form other characters
Chart 2: 132 independent characters and 14 爱 (愛), 尔 (爾), 丰 (豐), 几 (幾), 门 (門), 亲
radicals that can also be used as elements to (親), 义 (義), 与 (與)
form other simplified characters 讠(言), 饣(食), 纟(糹), 钅(釒)
Chart 3: 1,753 characters simplified by analogy 迩 (邇), 弥 (彌), 祢 (禰), 玺 (璽), 猕 (獼)
based on the second chart 讥 (譏), 叽 (嘰), 饥 (饑), 机 (機), 玑 (璣)
Do the 2,000-plus simplified characters in the 1964/1986 General Table represent the
difference between the simplified and the traditional scripts? The answer may be yes and no.
The General Table, though much more elaborate than the 1956 Plan, was still not an exhaus-
tive list of all simplified characters. In particular, Chart 3 provided characters in simplified
forms only within the coverage of the Xıˉnhuá Zìdiǎ n (新华字典/新華字典 New China Dic-
tionary), 1962 edition, which contained about 8,000 characters. Given that the total number
of Chinese characters might be in the range of 80,000 to 100,000, there was little doubt that
more simplified characters existed beyond those in the General Table if we were to apply the
simplification mechanisms consistently to the entire inventory of Chinese characters. On the
other hand, the 8,000 or so characters in the Xıˉnhuá Zìdiǎn were sufficient for most reading
142 Part IV Reforming the Chinese script
and writing purposes in Modern Standard Chinese. Beyond that, a great portion of charac-
ters was variant or archaic forms found mostly in works of Classical Chinese. Simplification
of these characters might not serve great practical purpose. For instance, in the aforemen-
tioned example, 兤 (huǎng) and 櫎 (huǎng) were archaic variants of modern-day characters
晃 (huǎng ‘bright’) and 幌 (huǎng ‘curtain’). Their simplified forms, based on the Plan or the
General Table, should contain the 广 instead of the 廣 component. However, such forms
did not seem to exist.31 For written communication in Modern Standard Chinese, the 2,235
simplified characters in the General Table might be all that were necessary and might well
represent the difference between the simplified and the traditional script. If we take the 8,000
as the base, then this distinction is at about 27%.
After the 1956 Plan and between the 1964 General Table and its 1986 reissuance, the
Committee issued a Second Plan for Chinese Character Simplification (Draft) (第二次汉字简
化方案草案/第二次漢字簡化方案草案 Dì’èrcì Hànzì Jiǎ nhuà Fāng’àn Cǎ o’àn) in 1977. The
draft contained a total of 853 characters organized in the same categories of the 1964 Gener-
al Table. This plan did not receive popular support, however. The general opinion was that
the characters had been oversimplified so as to lose visual connections with their traditional
counterparts and to appear difficult to distinguish from each other in reading. The Second
Plan was eventually abolished in 1986, soon after which the 1964 General Table was reissued
to reaffirm its status as the standard set of simplified characters.
One major criticism against character simplification, from the time when script reform
was proposed until this day, was the potential loss of the “cultural history” encoded in the char-
acters. An example critics often cite is the character 愛 (ài) for “love.” Its simplified counterpart
爱 (ài), as they point out, no longer contains the component 心 (xīn) that represents “heart.”
“Heartless love” thus becomes one of the representative cases for the “lack of logic” in script
simplification. Supporters of the reform point to the fact that the character 爱 is not a modern
invention; rather, it is a folk character that has been in use for thousands of years before the
script reform, and the reform simply gives it an official status. In fact, most of the simplified
characters either have or are related to a character that has historical roots. Among the 2,235
simplified characters of the General Table, 521 are basic simplified characters, and the rest is
simplified by analogy from the 521 basic forms. About 79% of the 521 basic characters are
historical forms, and only about 21% are created after the founding of the PRC (Table 15.2).
Important sources of simplified characters also include folk characters and characters
written in the cursive style. See Tables 15.3 and 15.4 for a few examples.
How many Chinese characters are there? If you recall, we considered this question in
Chapter 7. Adding to the earlier discussion, a look at the number of characters recorded in
Chinese dictionaries (Table 15.5) reveals that the size of the character set has expanded dra-
matically through history. The largest dictionary contains more than 100,000 characters – a
mind-boggling number if we consider that alphabetical writing systems typically contain 20
to 50 symbols or even the fact that only 3,000 to 4,000 characters are needed for basic literacy.
This excessiveness may lie in a few factors. First of all, the nature of the writing system
determines that the character set is an open one. Chinese characters represent morphemes.
Because the number of morphemes is infinite, the number of characters, in theory, is also
limitless. When new ideas or concepts occurred, new characters were often created in order
to represent them. Another reason was the lack of standardization. It was not uncommon for
a character used in different geographical areas or in different time periods to have somewhat
different forms. These graphic variants of the same character, called yìtı̌ zì (异体字/異體字
‘variant style character’), make up for a significant portion of the character inventory, and
some characters may have multiple yìtı̌ zı̌. In Shuōwén Jiězì (100 ce), for example, about 89%
were distinctive characters, and the remaining 11% were variant forms. Forty percent of the
characters in the Kāngxıˉ Dictionary (1716) were yìtı̌ zì.34 In the modern dictionary Cíhǎi (1979
edition), the percentage of variant forms recorded was 27%.35 Yìtı̌ Zì Zìdiǎn, the dictionary that
has the greatest number of characters in Table 15.5, is in fact a dictionary specialized in doc-
umenting variant forms. Seventy percent of characters collected in its sixth edition are yìtı̌ zì.
The large number of characters in active use today may also be a result of the variation in
register employed in speech and in writing. Although we say that the Chinese writing system
represents the speech of Modern Standard Chinese, in writing, the delineation between Mod-
ern Standard Chinese and Classical Chinese can be rather fuzzy. Expressions from Classical
Chinese are often mixed in modern text to make it sound more formal. As a result, words that
are particular to Classical Chinese are also part of today’s Chinese lexicon and written sym-
bols used to represent such words, as well, are part of the character inventory.
It is often desirable and, in some cases, necessary to put a limit on the number of char-
acters. First of all, there are psychological constraints as to how many characters human users
Chapter 15 Simplification of Chinese characters 145
may realistically store for active recall from memory. This number has been suggested to be
somewhere between 7,000 to 8,000.36 Limiting the number of characters used is also feasible,
since not all characters are used with equal frequency (Table 15.6). In fact, in non-specialized
publications, the 1,000 most frequently used characters constitute about 90% of all the charac-
ters used. Adding the next 1,400 most frequently used characters would increase this coverage
rate to 99%. This means that if the reader only knows the 2,400 high-frequency characters,
she would encounter one new character in about every hundred on average in non-technical
reading. Consistent with these numbers, criterion for achieving literacy in Chinese is set at
about 2,000 characters, and elementary school graduates are expected to know about 2,500
characters.37
Computer processing of Chinese text also makes it necessary to limit the number of
characters used. The list of characters provided to computers must be a finite one. The Basic
Collection of Character Coding Graphs in Exchange of Information (信息交换汉字标码字符
集基本集/信息交換漢字標碼字符集基本集 Xìnxī Jiāohuàn Hànzì Biāomǎ Zìfú Jí Jıˉběn Jí,
GB 2312–80) published in Mainland China in 1981 contained 6,763 characters, and the Stan-
dardized Character Exchange Code for General Use (通用汉字标准交换码/通用漢字標準交
換碼 Tōngyòng Hànzì Biāozhǔn Jiāohuàn Mǎ ) published in Táiwān two years later included
13,051 characters. These numbers have been increasing, however. Unicode 10.0, the current
computing industry standard for handling text, defines a total of 87,882 “CJK unified charac-
ters” (characters shared in representing Chinese, Japanese, and Korean).39
Official efforts were made to limit the number of characters in use. As part of the script
reform, such efforts mainly consisted of eliminating variant characters. After the National
Conference on Script Reform in 1955 (全国文字改革会议/全國文字改革會議 Quánguó
Wénzì Gǎigé Huìyì), the government published a list of 810 characters with variants and pro-
claimed that the variants, 1,053 in total, be eliminated. For instance, the character 并 (bìng)
was chosen to be the standard form, and its variants 並, 竝, and 倂 were abolished. The char-
acter 杯 (bēi) replaced its variant forms 盃 and 桮. The character 酬 (chóu) was chosen as the
standard to replace its variants 酧, 詶, and 醻.40 Elimination of the variant forms was the first
major step of the script reform and a step taken before the character forms were simplified.
After the script reform, more recent efforts were to standardize the most frequently
used portion of the character set. In 1988, the Mainland promulgated the List of Generally
Used Characters of Modern Chinese (现代汉语通用字表/現代漢語通用字表 Xiàndài Hànyǔ
Tōngyòng Zìbiǎ o), which contained 7,000 characters, the most frequent 3,500 of which consti-
tuted the List of Frequently Used Characters of Modern Chinese (现代汉语常用字表/現代漢
146 Part IV Reforming the Chinese script
語常用字表 Xiàndài Hànyǔ Chángyòng Zìbiǎ o).41 In 2013, the PRC government, on the basis
of previous work, published a new List of Standardized Characters for General Use (规范汉字
表/規範漢字表 Guīfàn Hànzì Biǎo). This list contains 8,105 characters, including 3,500 most
frequently used ones for elementary and literacy education, 3,000 frequently used characters
for publishing, lexicography, and information processing, and 1,605 less frequent characters
for personal names, place names, science and technological terminology, and study of Classi-
cal Chinese at elementary and secondary schools.42
International impact
Since the 1970s, the simplified script has gained increasing prominence in Chinese-speaking
regions outside of China and in other areas of the world. In 1971, the People’s Republic of
China replaced the Republic of China (Táiwān) as the official representative of China at the
United Nations (UN). The simplified script has since then served as the UN’s official writing
system for Chinese. Singapore implemented script simplification of its own in 1969, but in
1974, the country officially adopted the PRC’s version of simplified script (which was different
from the original Singaporean version). Similarly, Malaysia also adopted the simplified char-
acters of the PRC in the early 1980s. Today, the traditional script is still used as the only official
Chinese script in Táiwān, Hong Kong, and Macau. Schools in these regions use traditional
characters exclusively. Schools in Singapore and Malaysia use the simplified characters exclu-
sively though the use of traditional characters is not officially discouraged. Literacy rates of
regions using the Chinese script, traditional or simplified, are consistently high: as of the late
2010s, adult (15+ years) literacy rates in Mainland China, Hong Kong, Macau, and Táiwān are
96%–99%, and youth (15–24 years) literacy rates are all above 99%.44 Chinese as an Additional
Language curricula in the United States has gone through a transition from the traditional to
the simplified script, and almost all of them now teach the simplified script with some offering
the traditional script as an option to students.45
The language and script reform in the PRC from the 1950s to the 1980s reshaped the
linguistic reality in Mainland China. Today, most educated Chinese speak pǔtōnghuà (i.e.,
Mandarin or Modern Standard Chinese) in addition to their local dialects,46 read and write
simplified Chinese characters (and many are also proficient in reading the traditional script),
and use pıˉnyıˉn in their day-to-day life. Mandarin, the simplified script, and pıˉnyıˉn are also
taught to and used by an increasing number of Chinese learners outside of China. Although
Chapter 15 Simplification of Chinese characters 147
there was much debate on the Romanization and simplification of the Chinese script earlier
in the 20th century, as we have learned in the previous chapters, that debate appears to have
been largely settled.
Notes
I
N RECENT HISTORY, THE Chinese writing system was perceived as a major impedi-
ment to the modernization of the nation when most of the Chinese citizens were unable to
read and write. In an earlier chapter, we reviewed a few of the multitude of proposals put
forth during the late 19th to the mid-20th centuries for writing Chinese using alternative, pho-
nographic scripts. These proposals signified a crisis of the Chinese script against the backdrop
of a nation struggling to remake herself in an increasingly globalized world. The low literacy
rate, however, was not the only concern the script reformers had during that time. Another
major challenge was the difficulty the character-based writing system created in developing
and using modern technology to acquire, organize, process, transmit, retrieve, and translate
through machines or electronic devices information coded in the Chinese script. In this chap-
ter, we will, without getting into the technological details, discuss two of these challenges and
the debates around them: ordering information and inputting Chinese text on computers.
Ordering information
Have you ever thought about how textual information in the English language can be ordered?
A good example to consider is the conventional dictionary. To English users, looking up a
word in a physical dictionary is a fairly straightforward process. Word entries are typically
arranged in the alphabetical order a–z: words with the first letter a come before words with
the first letter b; among words with the initial letter a, those with the second letter a come
before those with the second letter b; and so on. To find a word, one simply follows this logic
to locate the string of letters that corresponds to the first letter, and then the second, the third,
until one finds the whole word. The knowledge of the alphabetical order, for native readers, is
usually learned in childhood – perhaps with the aid of a little song – and is generally a pain-
less and insignificant process. When applied to sorting words, the alphabetical order ensures
that the sequencing in the outcome is strict and consistent. That is, any two words are reliably
sorted in the same order under the alphabetical rule regardless of who is doing the sorting or
for what purpose.
Companion Website
The same cannot always be said, however, for looking up for a character in a Chinese
dictionary. You may recall that we have already talked about this in Chapter 7, when we first
introduced you to the Chinese writing system. Here we will elaborate a little more on this
topic.
First of all, it is important to note again that most modern Chinese dictionaries list char-
acters in the alphabetical order based on the p ī ny ī n notation of the characters’ pronunciation.
If you know how a character is pronounced, then looking it up is almost as straightforward
as consulting an English dictionary. However, there may be a couple of complications: one,
the same syllable may occur in various tones. Characters sharing the same p ī ny ī n spelling
are usually arranged in the order of neutral (e.g., ma), first (mā), second (má), third (mǎ ),
and fourth (mà) tones. So, if you know which tone the target character has, it will be easier to
locate it. Another complication is that the same syllable bearing the same tone may be repre-
sented by more than one (and as many as 167!1) characters. These characters cannot be differ-
entiated by their sounds and therefore must be arranged based on their forms. In particular,
it is the number of strokes in the characters that counts: characters with fewer strokes come
before those with more strokes. For example, 马 玛 码 蚂 are all pronounced mǎ . They are
likely arranged in this order in relation to each other because 马 (3) has the fewest strokes, 玛
(7) has more, 码 (8) even more, and 蚂 (9) the greatest number of them all. If you know what
the character looks like, counting the strokes will help you get to it faster. Otherwise, you will
need to patiently go through the list of homophones until you find the character that matches
the one you are looking for.
The process gets much more complicated when you only know what a character looks
like and wish to find out its pronunciation and meaning by looking it up in a dictionary. In
this scenario, all we have to rely on is the graphic form of the character. As we discussed in
Chapter 7, we must understand how characters are categorized in a dictionary based on sec-
tion headers (部首bùshǒu) or radicals, that is, recurring elements in characters that are often
(but not always) indicative of the characters’ rough semantic category. Table 16.1 shows a few
more examples.
Let us review the process of locating a character based on its graphic form in a con-
ventional dictionary. Suppose we want to find out the pronunciation of the first character in
Table 16.1, 河. We will need to first locate its section header in the radical index. The radical
index is a list of section headers arranged based on their number of strokes. Radicals with
fewer strokes come before radicals that have more strokes. In this case, the radical for the
character 河 is 氵, which has three strokes. We can find this radical in the three-stroke section
of the radical index. The radical index leads us to a page number that points to the氵section
of the character index. The character index is a list of characters arranged by radical. Here we
see a list of all characters that have the 氵radical. These characters are arranged by the num-
ber of strokes of the whole character lest the number of strokes of the radical. In our case, if
Chapter 16 Writing and technology in modern China 151
we take氵out of 河, the remaining part 可 has five strokes. Thus, we look for the five-stroke
subsection within the氵section of the character index. Here, we most likely see more than one
character: 沫, 泄, 沽 . . . until we find 河, with a page number attached to it. We turn to that
page of the dictionary and will be able to locate the entry 河 hé ‘river.’
As you see, the radical indexing method is much more cumbersome than alphabetical
indexing. What’s more, it is not guaranteed that any given character can be located this way.
There are characters that do not contain a radical that is easy to identify or is among the list
of the 200 or so section headers. For these characters, the dictionary may have a supplemen-
tal list in which the characters are arranged solely by the number of strokes. For example,
the Contemporary Chinese Dictionary, 5th edition,2 has such a list with 491 characters in it,
including both single-component characters (e.g., 丁 dīng, 正 zhèng, and 或 huò) and com-
pound characters (e.g., 就 jiù, 靠 kào, and 赢 yíng).
It has been argued that the challenge in ordering information that the Chinese writ-
ing system poses may hold the answer to the well-known “Needham Question.”5 The British
Sinologist Joseph Needham (李约瑟/李約瑟 1900–1995), after carefully studying and docu-
menting the history of Chinese science and technology and its influence on the West, asks:
“Why, then, did modern science, as opposed to ancient and medieval science, develop only
in the Western world?”6 He deems it a puzzle that modern science did not develop in China
despite the country’s sustained lead over the Western world in most of the major areas of sci-
ence and technology from ancient times up to the recent two or three centuries.7 Scholars’ re-
sponses to this question, for the most part, have come from considerations of social, political,
and economic factors.8 However, a few scholars have also specifically argued for the impact
of the Chinese writing system.9 The logic in their arguments seems to, at least in part, hinge
on the observation that the Chinese script has made it difficult to order information and has
therefore inhibited the development of taxonomy or scientific classification.10 For example, “to
understand which of the two types of script, alphabetic or ideographic, is best suited to classi-
fication, one needs only compare Western and Chinese dictionaries.”11 In traditional Chinese
encyclopedias as well, because of the lack of an organizing mechanism such as alphabetization,
152 Part IV Reforming the Chinese script
information is often grouped into complex categories from celestial phenomena to terrestrial
matters that do not aid well swift and accurate retrieval of information.12 Arguments have also
been made for the concreteness of Chinese characters13 and the shallowness in their repre-
sentation of speech.14 All in all, the Chinese script is thought to be less conducive to abstract
theoretical thinking than the alphabetical writing system. Whether such arguments can be
substantiated with sufficient empirical evidence may remain a question. Nonetheless, the dif-
ficulties Chinese characters presented in the early stage of information technology were no
doubt significant.
Companion Website
are then assigned letters or numerals to be keyed in on a standard keyboard. The Wǔbı̌
(五笔/五筆, short for五笔字型/五筆字型 Wǔbı̌ Zìxíng ‘five strokes’) method developed
on the Mainland and Cāngjié (仓颉/倉頡 ‘(name of the mythical figure credited for the
invention of the Chinese script’) used in Táiwān and Hong Kong were among the most
popular examples of this type. This approach relied solely on the graphic information of
the characters, and it is particularly useful when the typist did not know how a character
was pronounced. In the 1990s, when personal computers were not yet readily accessible
in Mainland China, authors often took their manuscripts to professional typists and paid
a fee at a per-thousand-character rate to get them processed. Invented in 1983, Wǔbı̌ was
one of the most commonly used methods by typists for inputting simplified characters.
It was a configurative scheme that delegated characters to five zones on the QWERTY
keyboard based on the type of their initial strokes: horizontal (一), vertical (丨), down-
left slant (丿), down-right slant (乀), and bent (𠃍). Each character was input by hitting a
series of one to four keys. Professional typists favored this method because, compared to
the phonetic input methods at the time, it had very low code redundancy – that is, a giv-
en string of keys almost always corresponded to a unique character – and allowed them
to type at a high speed. The downside, however, was that it required significant training
to develop high-level proficiency.
The most popular type of input methods today uses the sound-based approach. The
p ī ny ī n method in Mainland China and the zhùyīn (注音, i.e., Bopomofo) method in Táiwān
are both of this kind. These days, most everyone who writes in Chinese on a computer uses
this kind of approach. On the Mainland, for example, students usually learn p ī ny ī n in elemen-
tary school, so most computer users can start inputting p ī ny ī n without learning any additional
code. P ī ny ī n-based input software is also the most readily available – these days they come
preinstalled on Windows or Mac operating systems. In fact, p ī ny ī n became the PRC’s stan-
dard for Romanized writing of Chinese around the same time as experimentation started with
computer inputting of Chinese characters, and it has been the basis of the phonetic approach
from the get-go.
Over the years, the accuracy and speed of p ī ny ī n input methods have improved as the
technology becomes more sophisticated. First of all, replacing the syllable-based conversion
mechanism with a word-based system solved the homophone problem. As previously men-
tioned, one major drawback of p ī ny ī n-based input methods is their relatively high code
redundancy. This is because homophones abound in the Chinese language. For example,
inputting the p ī ny ī n syllable ma22 on a Macintosh computer using its native p ī ny ī n input
method for simplified Chinese characters23 brings out 18 characters that share the same
pronunciation (tones not considered). The user will then need to pick the intended char-
acter out of this list. As one may imagine, inputting Chinese text syllable by syllable could
be a rather inefficient process. In a word-based system, instead of inputting a single p ī ny ī n
syllable to get a single character, users input whole words that contextualize and disambig-
uate homophonic characters. As a result, accuracy and speed both significantly improve.
This can be accomplished because only a small minority of words in modern Chinese is
monosyllabic – about 2,500 in a list of 60,000 word entries, based on one study.24 By the
early 1990s, most p ī ny ī n-character conversion systems have made use of the word-based
mechanism.25 This could be considered a new stage in the development of computer input
methods for Chinese.
154 Part IV Reforming the Chinese script
Since then, pıˉnyıˉn input methods have significantly evolved. Today, the most popular
input applications among average users support sentence-length pıˉnyıˉn parsing. The previous
generation of word-based conversion was effective in distinguishing homophonic syllables
situated within polysyllabic words, but it was still helpless when the words consisted of single
syllables. This problem is solved when contextual information at the sentence level is taken
into account. As an example, let us look at how the standard pıˉnyıˉn input method on a Mac-
intosh computer26 handles the three de – 的, 地, and 得 – in Chinese (Table 16.2). Pronounced
exactly the same, these three particles are represented with different characters in writing
because they have different grammatical functions: roughly speaking, 的 connects a noun and
its preceding modifier, 地 is used between a verb and its modifier, while 得 connects a verb
(or an adjective functioning as a verb) and the complement27 following it, all within the same
phrase.
When we type one of the above phrases, the computer defaults to the first de (的) until
more information is given from the word following 高兴/高興 (gāoxìng ‘happy, happily’).
At that point, the computer application adjusts accordingly to provide the correct charac-
ter form. The writer can monitor the result during typing and may still need to revise a few
characters here and there – “such an intermediate stage of checking and revising will always
exist.”28 However, the overall accuracy has reached above 97%.29
Chapter 16 Writing and technology in modern China 155
In terms of efficiency, some have pointed out that typing Chinese on a computer using
pıˉnyıˉn has become faster than English thanks to the development of increasingly sophisticat-
ed input software.30 One particularly helpful feature of such software allows users to key in a
sentence by only typing the first letter of each pıˉnyıˉn syllable. The letter combination usually
is sufficient information for the application to “figure out” the entire sentence. For example, to
type 我们明天一起吃饭好吗/我們明天一起吃飯好嗎 (wǒmen míngtiān yìqı̌ chıˉfàn hǎoma
‘how about we have dinner together tomorrow’), one may simply type wmmtyqcfhm. That is,
seven letters instead of 29. And the software is intelligent enough to put together the correct
sentence. This kind of “predictive text,” however, is not currently an option for typing English
on a QWERTY keyboard. The “what you type is what you get” approach for the English al-
phabet since the beginning of computers has remained unchanged, and the typing speed of
English text has accordingly remained unchanged. On the other hand, many will probably
agree that Chinese input methods have developed in such a way that the initial hurdles posed
by the Chinese writing system have been largely overcome.
Notes
1 One hundred sixty-seven different characters are pronounced yi based on the Contemporary Chi-
nese Dictionary (Xiàndài hànyǔ cídiǎn 现代汉语词典/現代漢語詞典), 5th edition, published in
2007 by the Commercial Press in Beijing.
2 Chinese Academy of Social Sciences (2007).
3 Mair (1991).
4 Ibid., p. 1.
5 For example, Logan (1986); Bodde (1991).
6 Needham (1956, p. 214).
7 Lin (1995, p. 270).
8 For example, Needham (2013) and Lin (1995).
9 Logan (1986, pp. 47–48, 54).
10 Bodde (1991), cited by Hannas (2003, p. 257).
11 Logan (1986, p. 57). Note that the Chinese writing system is referred to as “ideographic,” but we
know that this is not an accurate characterization. See Chapter 8.
12 Mair (1991, p. 1).
13 Logan (1986, p. 55).
14 Hannas (2003, pp. 244–262).
15 Liu (1991).
16 The term “information processing,” when applied to computer science, refers to the use of algo-
rithms to transform data on computers.
17 Mair (1991, p. 1).
18 Sometimes Vietnamese is also included in this set, thus “CJKV Unified Ideographs.” The set in-
corporates traditional and simplified characters for writing Chinese, kanji for Japanese, hanjia for
Korean, chũ’ nôm for Vietnamese, as well as phonographic scripts for these languages.
19 GB stands for 国标/國標 guóbiāo, a short form for 国家标准/國家標準 guójiā biāozhǔn ‘national
standard.’
20 Liu (1991, p. 10).
21 Mair (1991, p. 5).
22 Tones are usually not marked when inputting Chinese characters via pīnyīn.
23 The computer runs the Mac OS High Sierra operating system, version 10.13.2.
156 Part IV Reforming the Chinese script
T
O UNDERSTAND THE ROLE of handwriting in the construction of the Chinese per-
sonhood, it is useful to evoke a metaphor Yen (2005) suggests: the writing of Chinese
characters, to the Chinese people, is like “a secondary face.”1 This concept can be il-
lustrated by a Chinese saying, rén rú qí zì, zì rú qí rén. Rén (人) means “person, human, per-
sonhood,” rú (如) “to resemble,” qí (其) “his, her, its,” and zì (字) “handwriting.” Thus, “one’s
personhood resembles one’s handwriting, and one’s handwriting resembles one’s personhood.”
That is to say, Chinese people tend to expect the inner traits of a person to align with the
qualities seen in her handwriting, and vice versa. “It takes no more than one written character
to reveal one’s heart,”2 said the Táng dynasty calligrapher Zhāng Huáiguàn (张怀瓘/張懷瓘).
人 如 其 字, 字 如 其 人
rén rú qí zì, zì rú qí rén
Furthermore, handwriting as a secondary face is “the layer of oneself that comes directly
into contact with the world.”3 In other words, one’s handwriting is the part of the person that
is the most easily seen or often the only part visible to others. For this reason, for instance,
candidates of the civil service exams placed a premium on beautiful and elegant calligraphy,
hoping to create a favorable impression on exam viewers. This tradition has carried over to
today’s college entrance tests and still propels student candidates to strive for graceful and ma-
ture, or at least neat, handwriting on their test papers. During the author’s high school years in
the 1990s, her teacher constantly urged the class to practice writing characters in their spare
time, because she believed that good handwriting was crucial to creating a positive impres-
sion and that a positive impression on the test markers could sometimes make the difference
between a first-tier and a second-tier university in admissions decisions. Even though the
students understood those to be extreme cases, they did not dare to take the teacher’s advice
lightly. They practiced when they could, critiqued each other’s styles, and took great care to
write well on their test papers.
If one’s pǐn is high, then every dot and stroke he writes is teemed with an air of purity
(清qīng), unbending will (刚/剛 gaˉng), elegance (雅 yǎ) and uprightness (正 zhèng).
Chapter 17 Handwriting and personhood 159
If one’s pǐn is low, even if his work looks impressive with deceptive vigor and strength,
his untamed and flamboyant violence can nevertheless be detected.6
Being a moral person was the prerequisite for being a recognized calligrapher. As the
late Míng and early Qıˉng calligrapher Fù Shān (傅山 1607–1684) said, zuòzì xiān zuòrén, “to
write good calligraphy, one must first be a good person.”
作 字 先 做 人
zuò zì xiān zuò rén
For this reason, good calligraphy did not necessarily mean beautiful calligraphy. Fù Shān
famously stated that “rather than clever, beautiful, deft, and affected, I prefer being awkward,
ugly, disconnected, and straightforward.”7 This preference for honest, inner beauty above pre-
tentious, superficial beauty is deeply rooted in traditional Chinese culture.
The concept of pı̌n describes not only the moral quality of a person, but also the quality
of calligraphy works. In calligraphy pı̌n is not necessarily beautiful execution of the lines and
dots or skillful arrangement of the characters on the writing surface. Rather, it is a general
quality similar to those that define a moral person – purity, elegance, and uprightness. Not
surprisingly, such a quality is often believed to derive from the calligrapher. If the calligrapher
is a noble person, then his work most likely possesses this quality in the perception of the
viewers. If the calligrapher lacks moral character or strength, then the calligraphy she produc-
es will not be recognized as superior works. A telling example is the infamous high-ranking
official of the Northern Sòng dynasty, Cài Jīng (蔡京 1047–1126 ce). A highly skilled calligra-
pher, Cài Jīng was at one time regarded one of the four great calligraphy masters of his time,
collectively known as Sū Huáng Mı̌ Cài (苏轼/蘇軾 Sū Shì, 黄庭坚/黃庭堅 Huáng Tíngjiān,
米芾MǏ Fú, and at first 蔡京 Cài J ī ng). He proved, however, a treacherous official whose cor-
rupt deeds contributed to the downfall of the Northern Sòng. For this reason, his name was
later replaced by that of Cài Xiāng (蔡襄 1012–1067 ce), a scholar-official known for his up-
rightness and integrity. Since then, when people speak of Sū Huáng Mı̌ Cài, the family name
Cài has changed its association to Cài Xiāng.8
Appreciating works of calligraphy inherently requires identifying oneself with the moral
values of the calligrapher and therefore says much about one’s own moral preferences. This is
also why students of calligraphy are carefully selective about which calligrapher’s models to
emulate. They invariably select calligraphers with high moral standing, especially those with
unwavering loyalty to their nation.9 One such example is the famed Táng dynasty calligrapher
Yán Zhēnqīng (颜真卿 709–785), who served the Xuánzōng emperor (玄宗 685–762) and
his successors. As a loyal governor, he fought the revolting forces during the Ān ShǏ rebellion
(安史之乱/安史之亂 Ān Shı̌ Zhī Luàn). As minister of law, he was outspoken against corrupt
high-ranking officials, resulting in him being repeatedly demoted. His uprightness was hailed
by commoners, and his standard-script calligraphy of what is widely known as the “Yán style”
(颜体/顏體 Yántı̌) is said to possess the same quality of unbending honesty. To this day, Yán
Zhēnqīng’s calligraphy works in the regular script are some of the most popular options cal-
ligraphy students choose to use as models.
Copying the characters on paper is believed to be inseparable from absorbing the
moral vigor of the calligrapher. Thus, no matter how accomplished a calligrapher is in
his art, if his pı̌n, or moral character, is questionable, his works will not receive the same
160 Part V Identity and gender in writing Chinese
respect and will be shunned by most calligraphy students. The infamous chancellor of
the Sòng dynasty, Qín Huì (秦桧/秦檜1090–1155), for example, was perhaps as good a calligra-
pher in terms of skills as the most respected masters. Because of his role in the political execution
of the loyal military general Yuè Fēi (岳飞/岳飛 1103–1142), however, he is regarded a traitor
against the Hàn Chinese people. As a result of his poor moral reputation, few students imitate
his calligraphic style. The calligraphy of Yuè Fēi, on the other hand, is highly respected until this
day, even though its artistic accomplishment is considered secondary to that of Qín Huì.10
Companion Website
Similarly, being highly controlled is another desired characteristic in both brush tech-
niques and a noble person’s temperament. The calligraphy brush is light in weight, but ma-
neuvering of the brush should preferably make it appear a very heavy instrument. Brush
movement should be slow and sluggish rather than swift or light. Fast and careless brush
movements may result in strokes that are blatant, excessive, or showy, which are all disfavored
qualities of a refined person.
Being humble, reserved, and controlled are not the only desired qualities shared by tra-
ditional Chinese calligraphy techniques and the concept of the noble person from the Con-
fucian perspective. Table 17.1 summarizes a list of such qualities in matching pairs. These
parallels illustrate further the close connection between handwriting and personhood in the
Chinese tradition. Keep this in mind when you practice calligraphy, as it is understanding as
such that gives the act of brush writing its cultural significance – when you understand the
ideal person calligraphers aim to cultivate, you can more fully appreciate the meaning of cal-
ligraphy as you pick up a brush to participate in this time-honored tradition.
Notes
I
T MIGHT BE SOMEWHAT odd to consider that a system of writing could in some way
register gender relations. We’ve defined writing as the graphic representation of language
(speech) in a specific and systematic manner. If a writing system encodes sexism, then at
what linguistic level does this encoding take place? In other words, what linguistic unit of the
writing system is inherently sexist? Only when we can demonstrate the inherence of sexism in a
script can we claim that it is, in any degree, sexist. This is because we may otherwise fall into the
trap of confusing writing with language and inaccurately attribute the gender bias in speech to
the writing system. Indeed, we need to clearly distinguish between language (speech) and writ-
ing in this case. Sexist words, phrases, and statements belong to the realm of speech rather than
writing, even though they may be represented in graphic marks when we encounter them. For
example, in reading printed text we may come across the highly offensive word “bitch” used to
refer to a female person. This may trigger in our mind a connection between the combination
of the letters <b-i-t-c-h> and the highly sexist and offensive notion in this term. However, the
true bearer of the sexist idea is the English word (think speech) [bItʃ],1 and the letter combina-
tion <b-i-t-c-h> just happens to be the way this word is represented visually in writing. There
is little that is inherent about the letters, or their combination, that makes them sexist. For this
reason, it would be illogical to say that the Roman alphabet is sexist.
So, what does it mean to say that a writing system is sexist? It means that the script itself
has built-in gender biases. Such biases can be shown to exist without evoking sexist expres-
sions in the language that the script is customarily used to write; instead, evidence comes from
intrinsic elements of the script. In the case of the Chinese script, we can look at individual
characters to determine whether there is gender-based bias in the derivation or construction
of their forms.2 We will do that in the latter part of this chapter. First, let us take a brief look at
gender bias against women in China.
millennia, governed how Chinese people thought and behaved. The Confucian social struc-
ture, in particular its patriarchal family hierarchy and the corresponding marginalization of
women in the public sphere, has long been criticized as a major source of sexism in the Chi-
nese society.3 Some often-quoted examples come from the Confucian Classics4, the prereq-
uisite, foundational literature for men to enter the ruling class in imperial China. Although
writings in the Classics do not uniformly depict women as inferior, unworthy, or evil, scholars
point to passages in these books that apparently do.5 For example, in the Book of Odes (诗经/
詩經 Shıˉjıˉng), one reads:
When a baby boy was born he was laid on the bed and given jade to play with, and
when a baby girl was born she was laid on the floor and given a tile to play with.6
A clever man builds a city wall/A clever woman overthrows it/Beautiful is the clev-
er woman, but she is an owl, a hooting owl/A woman with a long tongue, she is a
promoter of evil/Disorder is not sent down from Heaven, it is produced by women/
Those who cannot be instructed or taught are women and eunuchs. . . . And there-
fore the women have no public service. They have to abide by their silkworm work
and their weaving.7
In the Book of Documents (尚书/尚書 Shàngshū), the term pìnjıˉ sıˉchén (牝鸡司晨/牝雞
司晨), meaning “the hen (instead of the rooster) announcing dawn,” is used to caution against
women playing any role in the public sphere. This is because, as one scholar expounds, the
belief was that “if women are [were] entrusted with tasks involving contact with the outside,
they will [would] cause disorder and confusion in the Empire, harm and bring shame on the
Imperial Court, and sully sun and moon.”8
The patriarchy of imperial China allowed little space for women in the public domain.
From low-ranking officials to the emperor high above, it was almost exclusively men who
controlled and directly participated in governing and operating the state. Women were not
only considered unfit for such responsibilities, but were also largely deprived of the education
that would have prepared them for formal roles outside the domestic realm. This system kept
women out of social positions that would put them on par with men and perpetuated the
oppression and marginalization of women.
In pre-modern China, women’s “proper” place was in the family: “They have to abide
by their silkworm work and their weaving.”9 However, this did not mean that women
enjoyed higher status than men in the household. On the contrary, Confucian teaching
required women to abide by “three obediences and four virtues” (三从四德/三從四德 sāncóng
sìdé).10 A good woman was to obey her father before she was married (wèijià cóngfù 未嫁从
父/未嫁從父), her husband after getting married (jìjià cóngfū 既嫁从夫/既嫁從夫), and her
sons after her husband died (fūsı̌ cóngzı̌ 夫死从子/夫死從子). A virtuous woman must also
possess feminine morality (fùdé 妇德/婦德), display proper speech (fùyán 妇言/婦言), main-
tain a modest manner and appearance (fùróng 妇容/婦容), and work diligently (fùgōng 妇功/
婦功).11 These moral standards relegated women to a status inferior to the men around them
in the domestic realm.
Chapter 18 Sexism in the Chinese writing system 165
Sobbing in a BMW
Increased commodification in the Chinese culture has also had a negative impact on how Chinese
women are perceived as well as on how they perceive themselves.20 For example, young women
increasingly place more emphasis on material wealth when choosing potential marriage partners.
One infamous incidence that occurred in the popular dating show If You Are the One (非诚勿扰/非
誠勿擾 Fēi Chéng Wù Rǎo) in 2010 illustrates this point well. One of the female participants, when
asked by an unemployed male guest if she might enjoy bike riding with him, responded that she
166 Part V Identity and gender in writing Chinese
“would rather be sobbing inside a BMW than smiling on the back of a bicycle.”21 The video went
viral on the Internet, and the episode has become an infamous example for the rise of materialism
and shift in the values of Chinese women coming of age in recent decades.
Sexism against the female also manifests in compound characters that have 女 as the
radical (the semantic element). As some scholars point out,24 among the hundreds of charac-
ters of this kind, a large number denote negative attributes that often have to do with a person’s
moral character or inner quality. For instance:
When such characters express positive meanings, they are usually complimentary of women’s
physical appearance. For example:
Companion Website
Do these compound characters encoding gender bias serve to confirm and reinforce
gender inequality in the Chinese society? Does the use of these characters cause a writer or
reader to be more prone to sexist beliefs or practices? With perhaps the exception of discus-
sions among scholars, the issue of sexist characters does not seem to have sparked a lot of
attention among users of the Chinese writing system, and there has not been much talk of
a formal script reform based on this issue. From a linguistic perspective, however, this lack
of attention may be understandable. Users of Chinese characters do not need to consciously
analyze the sexist connotation within the characters in order to appropriately use them – and
indeed they do not seem to. This is because a writing system, in its essence, is a set of symbols
(graphemes) used to represent speech, and the same graphemes can be used to write and con-
vey messages that are either sexist or not. The meaning of the messages derives primarily from
the speech being written rather than the internal structure of the graphemes used to record
that speech. Sexist characters may be used to fight for the protection of women’s rights just
as well as they may be used to demean and devalue women. Although some of the Chinese
characters are evidently sexist in construction, the continued existence of gender inequality in
China, however, may have little to do with that reality.
168 Part V Identity and gender in writing Chinese
Notes
N
ǙSHŪ (女书/女書), LITERALLY “WOMEN’S SCRIPT,” is a writing system invented
and used by unschooled, rural women in southern Húnán (湖南) province to write the
local Chinese dialect.1 It is unknown when exactly the script was invented, but many
believe that it had been in use for at least hundreds of years until sometime in the 20th century.2
Girls and women in the area nowadays receive education in the standard Chinese script, so
nǚshū is no longer employed for regular writing purposes. Nonetheless, the story of nǚshū is
an illuminating case of ingenuity, endurance, and mutual support among women. For many
generations since nǚshū’s invention, the women educated themselves in the script and used it to
compose letters and poems, record songs and stories, write autobiographies, and so on. Much
of the nǚshū writing was shared among female friends on various social occasions. In this man-
ner, nǚshū served as the vehicle of life-sustaining literary activities among peasant women in a
male-dominant culture. It continues to capture the fascination of researchers and students in
linguistics, anthropology, and other related disciplines. As an introduction to the nǚshū script,
this chapter aims to answer a few of the basic questions regarding this remarkable legacy.
FIGURE 19.1 Location of Jiāngyǒng (江永) county. The map on the left shows the Yangtze River
(长江/長江 Chángjiāng) in China and the location of Húnán province to the south of the river. The
Yangtze River is usually regarded as the demarcation between northern and southern China. The map
on the right shows the location of Jiāngyǒng (江永) county in Húnán province.
perhaps a handful of researchers, most of the men seemed to have dismissed writing in nǚshū
as merely “something that women did” and therefore unworthy of their serious attention.6
Such contemptuous attitudes could be seen from the names some men gave to the script – for
example, “ant graphs” or “mosquito crawls,” hinting at the long and slender strokes commonly
used in the script – and the fact that many did not know what to call the script at all.7 Scholars
who noticed the nǚshū script either thought of it as an inferior variant of hànzì or found it
difficult to bring to the attention of a wider research community. For instance, in 1954, ZhŌu
Shuòyí (周硕沂/周碩沂), a local researcher at the Jiāngyǒng County Office of Cultural Affairs,
reported the script to the National Science Academy in Beijing but received little response.8
Nǚshū did not survive unscathed the social movements accompanied by radical ideo-
logical changes in mid- to late 20th-century China. In the 1950s, the script was suspected
of being a secret code used for subversive purposes, and its use was prohibited.9 During the
Cultural Revolution (1966–1976), the script was regarded as part of the “four olds (四旧/四舊
Sì Jiù)” – old culture, old ideas, old habits, and old customs – and was banned along with the
cultural conventions and practices that had provided the foundation for its application. Most
of the texts were destroyed, and those who knew the script stopped using it or teaching it to
others. By the 1980s, there were no more than a few elderly women who were still able to read
and write nǚshū. The last proficient user of the script passed away in 2004.
Although nǚshū was in obscurity to the broader world for most of its history, it is by no
means “a secret script,” as it was incorrectly accused to be in the 1950s, or as popular media
sometimes romantically portray it to be much later.10 Most of the texts written in nǚshū were
lyrics for ballads to be performed in public, for instance, at weddings and festivals, where the
display of letters or other documents written in nǚshū was sometimes a matter of pride.11 Men
were not excluded from these occasions. The fact that the script was little known was perhaps
Chapter 19 Nüˇshuˉ 171
more indicative of the lowly status of the women writers than any intention to keep the wom-
en’s communication private.
Companion Website
Table 19.1 Corresponding symbols in nüˇshuˉ and hànzì. The same nüˇshuˉ symbol may correspond
to multiple hànzì characters.
Nüˇshuˉ graph Hànzì in the traditional script Pronunciation and meaning in
Modern Standard Chinese
包 baˉo ‘to wrap, bag’
胞 baˉo ‘womb’
白 bái ‘white’
柏 bǎi ‘cypress’
As you can see from these examples, if we assume that the nǚshū graphs have been
derived from hànzì in form, then we can see that they tend to tilt the hànzì characters clock-
wise by about 45 degrees, resulting in diamond-shaped symbols. The highest point of the
symbols is on the upper right, and the lowest point, the lower left. A single graph may contain
1 to 20 strokes. Unlike hànzì characters, nǚshū graphs consist of few horizontal or vertical
172 Part V Identity and gender in writing Chinese
FIGURE 19.2 Part of a poem in nüˇshuˉ glossed with hànzì characters. Hànzì symbols are supplied to
the left of the corresponding nüˇshuˉ graphs.
strokes, and most strokes are “slants,” stretching either from the upper right to the lower left
or from the upper left to the lower right. Other basic strokes of nǚshū include circles, round
dots, and curves, all of which are less commonly used in hànzì. In writing nǚshū, small, thin,
and uniform strokes are considered a beautiful hand.14 Although almost all the extant nǚshū
documents are written in ink with a brush, the strokes are linear and fine with little variation
in width, which may indicate that the script did not originate in brush writing.15 It has been
Chapter 19 Nüˇshuˉ 173
suggested that nǚshū stroke forms are derived from embroidery.16 Indeed, embroidery fea-
tured significantly in the lives of the nǚshū writers, and the women often incorporated nǚshū
characters into the design of their embroidery work.
Learning to read nǚshū may not be a straightforward process. Although the relationship
between sound and script is predominantly one graph for one syllable, there is a significant
amount of overlap: the same graph can represent several similar, yet sometimes unrelated, syl-
lables, and the same syllable may occur in a number of variant graphic forms. Interpretation
of nǚshū texts, therefore, relies to a great extent on contextual knowledge and familiarity with
their themes and vocabulary. What does help is that almost all the nǚshū texts are written in
verse with fixed sentence lengths: mostly seven syllables, and some five. In fact, many of them
are meant to be chanted in a certain melody following the oral tradition of the local culture,
as it often happened at routine gatherings where female friends did their spinning, weaving,
or embroidery together. The singing perhaps relieved some of the boredom in the repetitive
tasks,22 but there is more to it. What were some of the themes of the ballads written in nǚshū?
What were the social functions that they served? We will discuss these questions in the next
section.
as Endo (1999) explained, this area was blessed with mild climate, ample rain, and fertile soil,
and it was thus ideally conditioned for agricultural productivity. Generally, men alone were able to
handle the labor required to produce sufficient crops for the whole family. Women’s participation
in farm work was not required. Instead, they were expected to perfect such womanly crafts as
weaving, sewing, and embroidery as well as to do other domestic chores. Men were the livelihood
earners, and women took care of the men. Such division of labor served as the economic basis for
discrimination between genders.
Before marriage
Women in the Jiāngyǒng region conducted their lives according to a variety of local customs
and rituals. One such important custom was for unrelated young girls to become “sworn sis-
ters,” that is, close female friends who were mutually supportive and whose relationship was
expected to last for life. Sworn-sisterhood was sometimes prearranged by the families of the
girls but often grew out of frequent companionship among friends in day-to-day activities.
Girls in the area began to take on a significant share of housework when they reached age ten
or so. Many kinds of household chores were done together with village girls of about the same
age.25 A group of young female friends would also spend time together doing embroidery and
singing songs almost on a daily basis. In this group, regular companions of the same age – and
sometimes one or two years apart – might decide to become sworn sisters.26 A girl could have
more than one sworn sister. Before they each got married, sworn sisters would spend a lot of
time together, visiting the local temple to attend festivals, sleeping over at each other’s houses,
etc. This period of devoted friendship before marriage was generally regarded the happiest
time in a girl’s life.27
Nǚshū writing was indispensable to the development of sworn-sisterhood. At the outset
of the relationship, the girls would exchange letters written in nǚshū on folded fans (Fig-
ure 19.3) or silk handkerchiefs to acknowledge the sisterhood. As the relationship continued,
more letters could be written and exchanged inviting each other to visit, responding to the
invitation, or deciding on a time to meet and go to festivals to the local temple. At the temple,
they would deposit their prayers, recorded in nǚshū characters, to the goddesses. Besides fans
and handkerchiefs, letters could also be written on rectangular sheets of red paper. Such letters
were meant to be read by the addressees only and were preferably delivered by female carri-
ers.28 On almost every day, sworn sisters and their friends would sing songs together while
doing needlework. The lyrics of the songs were written down in nǚshū. Along with reading
lyrics and singing songs, the girls would practice together writing in nǚshū, helping each other
learn. Such teaching and learning were informal and most likely unsystematic but were also
motivated and earnest – it has been proposed that the impetus for women to create nǚshū was
so that sworn sisters would be able to communicate after they had married and moved away
from each other.29
arranged by parents and matchmakers, which could mean that the young women had never
met their husbands before the wedding. Getting married could also mean moving away from
their home villages, making it very difficult to spend time again with their sworn sisters and
other friends. Due to women’s low status in the married household as daughters-in-law or
wives, it was common that they were treated as not much more than free labor to do chores
around the house or even abused by their husbands. In the event that their husbands died or
abandoned them, if they had grown sons to support them, they would be very lucky; other-
wise, life would become even harder.
Such harsh realities of the married life make lamenting women’s misfortune a predom-
inant theme of texts written in the nǚshū script. In fact, plaintiveness set the basic tone of the
local marriage custom on the bride side. About a month before the wedding was to take place,
sworn sisters and other female friends would gather at the house of the bride-to-be and sing
songs for about three days. These songs, recorded in nǚshū, were lyrical accounts of women’s
grievance over the doom and gloom of marriages. Singing was often accompanied by tears
and served as a channel for women to openly lament and vent against their parents for mar-
rying them off and against the miseries they would have to suffer after getting married.30 At
the end of the singing and crying, the bride would be escorted to the groom’s house where the
wedding ceremony would take place.
One important type of writing in nǚshū is related to the local wedding custom. On the
third day of the marriage, a member of the bride’s family would be sent to bring the bride back
to her home village. She would be presented by female friends or relatives an important ritual
text called the “third-day book” (三朝书/三朝書 sānzhāo shū).31 It was a cloth-bound book
with songs or letters written in nǚshū in the first few pages followed by blank pages that the
bride could fill in later on. The book was usually prepared by the bride’s sworn sister or moth-
er. The content of the writing had a few recurring themes: friends would lament the painful
separation forced upon them, relatives would urge the bride to be an obedient and pleasing
Chapter 19 Nüˇshuˉ 177
daughter-in-law, and everyone would advise her to accept the new circumstances as there
were no alternatives and wish her happiness in the marriage.32 The book was meticulously
made, often adorned with embroidery on the cover or colorful paint on an inside page.
Through the ups and downs of their marital and social life, women wrote letters, record-
ed stories, and composed autobiographies using the nǚshū script. When the times got diffi-
cult, they would write letters to their female friends to seek support and sometimes visit and
briefly stay with their sworn sisters for comfort and consolation. When traveling storytellers
came to the village, they were enthusiastic audience and would write down in nǚshū the tales
they heard from the performances.33 Autobiographic writing of the women was almost exclu-
sively focused on the tragic experiences of the authors. The writing was possibly presented
to the goddesses at the local temple, registering the authors’ wish to alleviate their pain and
suffering.
Despite the rich variety and significant volume of writing that has been done in nǚshū,
what remain for us to see today are a small number of samples. This had to do with the local
custom as well. At the time of a woman’s death, it was customary to burn her nǚshū documents
so that she would be able to read them in her afterlife. This practice resulted in an almost com-
plete loss of nǚshū writing and made it difficult to locate original documents created before
the early modern period.34 The materials did survive, however, revealed to us extraordinary
creativity and forbearance of the women writers of nǚshū.
Notes
1 It has been proposed that men, instead of women, were the creators of nǚshū before hànzì was in
use, and that women preserved the script when men adopted hànzì (see Chiang, 1995, p. 49). How-
ever, the idea remains a theory so far.
2 For example: ibid., p. 49; (Idema, 2009, p. 4).
3 This idea is mentioned in Idema and Grant (2004, p. 543).
4 Lee (2002) wrote: “[T]his group creates its own oral and written language that is transmitted over
a thousand years from one generation to another” and “scholars asserted that Nushu was widely
practiced in Shangjiangxu Township, Jiangyong County, southern Hunan Province of China, for
more than one thousand years” (p. 101), but no sources were cited for these claims. Lee also wrote:
“allegedly, the development of Nushu language was most likely to be initiated 1,000 years ago, and
just who were the creators of Nushu has been disputed and remained unknown” (p. 106). However,
in this second case, the sources Lee cited for the origin of Nushu were local legends (p. 115, Note 9).
There did not seem to be any archeological or historical evidence that could confirm the claim.
5 Idema and Grant (2004, pp. 543–544).
6 See Lee (2002, p. 106) for references cited regarding men’s attitude toward nǚshū.
7 Chiang (1995, p. 47).
8 Ibid., p. xvi. According to Zhōu Shuòyí’s Bǎ idù Bǎ ikē (the Chinese equivalent of Wikipedia) page
(https://baike.baidu.com/item/周硕沂, accessed August 3, 2017), he reported the script to the
Central Academy of Social Sciences two years later, in 1956, and received responses from ZhŌu
Yǒuguāng (周有光), who was in charge of the writing reform at the time.
9 Idema and Grant (2004, p. 544).
10 For example, Lisa See’s 2014 novel Snow Flower and the Secret Fan, which was made into a film of
the same name in 2011.
11 Idema (2009, p. 4).
178 Part V Identity and gender in writing Chinese
12 Corresponding in a visual sense, but not in a strict sense between the two scripts, because the two
scripts are different in nature.
13 Chiang (1995, p. 55).
14 Endo (1999).
15 Chiang (1995, p. 68).
16 Ibid., p. 49; Endo (1999).
17 Ibid., p. 20.
18 Rogers (2005) mentions “the 1359 occurring syllable shapes in modern Chinese” on p. 29.
19 Chiang (1995, p. 22).
20 This number is based on 24 nǚshū texts that have a total of 1,535 graphs with variants. Ibid., p. 50.
21 Idema and Grant (2004, p. 543).
22 Idema (2009, p. 5).
23 Chiang (1995, p. 24).
24 Ibid., p. 16.
25 Ibid., p. 32.
26 The term in Chinese for sworn sisters was lǎotóng (老同 lit. ‘old same’) or tóngnián (同年 lit. ‘same
year’) if the girls were born in the same year or jiébài zı̌mèi (结拜姊妹/結拜姊妹 ‘sworn sisters’) if
they were not born in the same year.
27 Idema and Grant (2004, p. 547).
28 See Chiang (1995, p. 64).
29 Endo (1999).
30 Chiang (1995, p. 18).
31 Another version of description for sānzhāoshū maintains that it was delivered to the bride’s marital
home – instead of her natal home – on the third day of her marriage. (Lee, 2002, p. 105).
32 Idema and Grant (2004, p. 548).
33 Chiang (1995, p. 34); Idema and Grant (2004, p. 542).
34 See Chiang (1995).
20 Script choice in writing Japanese
T
HE WRITING OF JAPANESE in the modern, public context is probably more com-
plex than that of most languages today. Japanese writing employs a mixture of four
scripts: the morphographic kanji, the two moraic kana systems, hiragana and kataka-
na, and the Latin alphabet referred to in Japanese as rōmaji. Theoretically, three of the four
scripts – with kanji being the exception – would be able to fully represent the Japanese lan-
guage each on their own. In other words, if the writer so wished, she would be able to write
only and entirely in any one of those three scripts without major hindrance. However, a mixed
script is generally preferred. It is therefore important for an average writer to understand
when to use which script. Indeed, a writer of Japanese must be able to do so correctly in order
to be considered genuinely proficient. In this chapter, we take the inquiry a step further to
examine and understand both the linguistic constraints and the socio-cultural expectations
associated with script choice by native Japanese speakers. We will see that these choices, con-
sciously or unconsciously, are often associated with the writers’ social identities and may be
used to manipulate or construct such identities.
Linguistic constraints
Japanese script variation manifests predominantly at the lexical level (or, strictly speaking, the
morphemic level). That is, choices of which script to use are often made word by word. Words
of Sinitic origin that have long been part of the Japanese lexicon are usually represented in
kanji, and so are many native Japanese words. Almost all kanji have both an on reading mod-
eled on the original Chinese morpheme and a kun reading coming from the indigenous Japa-
nese morpheme. For example, the character 木 representing ‘tree’ may be used to write either
the Chinese loanword moku (mù in Modern Standard Chinese) or the Japanese word ki, both
meaning ‘tree.’ If the words in question are verbs, then the stems are often (partially) written
in kanji and the inflections are represented in hiragana. For instance, 食べた (tabeta) con-
sists of the kanji and hiragana 食べ(る) ‘to eat’ and the inflectional ending た that indicates
the completion of the action. In addition to grammatical particles and inflectional endings,
hiragana is also used for native Japanese lexical items, such as もう (mou ‘already’). Recent
loanwords, especially those from the West, are of course not associated with kanji. They are
usually written in katakana, for example, ビジネス (bijinesu ‘business’) and ミルク (miruku
‘milk’). Katakana is also used for onomatopoeia, as in ドキドキ (dokidoki ‘thump thump’)
describing a throbbing heart, or for indicating emphasis. In some cases, loanwords, especially
acronyms from languages written with the Roman alphabet, including English, French, Ger-
man, and Portuguese, can be represented directly in Roman letters or rōmaji.1 One example
180 Part V Identity and gender in writing Chinese
is the acronym OL, used in Japanese to mean “office lady,” referring in the Japanese context to
female office workers who perform secretarial tasks. A few more examples are: NG for “not
good,” CM for “commercial,” PV for “promotional video,” and JR for “Japanese Railway.” In
speech, they are pronounced as individual letters in Japanese-style pronunciation. OL will be
pronounced as ōeru, for example.
As you can imagine, one Japanese sentence is often written in a mixture of scripts. The
example below is a sentence with all four scripts represented.
私は東京のJETプログラムで英語を勉強しました。
Watashi-wa Tōkyō-no JET puroguramu-de Eigo-o benkyōshi-mashita.
‘I studied English at the JET Program in Tokyo.’
The pronoun 私 (watashi ‘I’), the nouns 東京 (‘Tokyo’) and 英語 (eigo ‘English’), and
part of the verb stem 勉強 (し) (benkyō ‘study’) are represented in kanji. Hiragana is used to
write the particles attached to the pronoun and nouns indicating topic (は -wa), possession
(の -no), location (で -de), and direct object (を-o). It is also used for the verbal inflection (ま
した -mashita) attached to the verb stem. The word プログラム (puroguramu ‘program’), an
English loanword, is written in katakana. The English acronym JET for the Japan Exchange
and Teaching Programme2 is in rōmaji.
Furthermore, there is not always a consistent one-on-one relation between a word and
its usual script. Although in most cases, a given word is conventionally written in one of the
scripts regardless of context – and it thus often makes sense to say “this is a kanji word” or
“that is a hiragana word” – a large number of lexical items can be written in two or more
scripts. For example, Table 20.1 are some of the words found to occur in kanji, hiragana,
or katakana with almost equal probability.3 For our convenience, we will refer to these as
triple-script words.
What, then, determines for the writer which form to use? As we will see in the next
section, script choices cannot be explained solely on the basis of grammatical or lexical con-
straints. They are motivated by extra-linguistic factors as well.
Chapter 20 Script choice in writing Japanese 181
Extra-linguistic factors
As a matter of fact, kanji, hiragana, and katakana have distinctive socio-cultural associations
in Japan. In one study, when asked to provide key words that describe the qualities they as-
sociate with each of the scripts, Japanese native writers came up with adjectives that clearly
separate the three scripts from each other, as shown in Table 20.2:4
Table 20.2 Descriptors by Japanese native writers for qualities associated with
script choice based on Iwahara et al. (2003)
Script choice Qualities associated with script
Kanji hard, difficult, intellectual, vigorous, old, male, formal
Hiragana soft, round, tender, simple, lovely, feminine, childish
Katakana hard, angular, foreign, cold, new, sharp, inorganic
These socio-cultural associations matter a great deal in the writing of triple-script words
and have been found to predispose native writers to choosing one script over another based
on the qualities most congruent with the given context. For example, in an interesting experi-
ment, Japanese university students were asked to come up with lists of Japanese celebrities they
regarded as most compatible with qualities associated with kanji, hiragana, and katakana re-
spectively. They had little difficulty coming to an agreement regarding such lists: the ex-Prime
Minister Hashimoto Ryūtarō (橋本龍太郎) and the enka5 singer Kitajima Saburō (北島三郎)
were perceived as kanji-compatible celebrities. The television comedian Tamori (タモリ) and
baseball star Ichirō (イチロー) were considered most compatible with katakana. The actress
and fashion designer 篠原ともえ (Shinohara Tomoe) and TV comedian and actor Akashiya
Sanma (明石家 さんま) were thought to be most closely associated with hiragana. Note that
such associations between the celebrities and the script have nothing to do with the script in
which the persons’ names are written; it is about the prominent personal characteristics that
the celebrities exhibit as perceived by the students and how well these characteristics match up
with the distinctive qualities the various scripts are perceived to possess.
Then, the students were given a made-up scenario in which the celebrities opened up
shops that needed to be named. They were asked to choose among the triple-script words
names for the shops and to write them down in any script they considered appropriate. No
surprise – they most frequently used the type of script most strongly associated with the per-
sonal characteristics of the celebrities. For instance, for “Mr. Hashimoto Ryutarō opened a
shop named _____,” a significantly higher percentage of students wrote the name in kanji than
in the other two scripts.
The choice of script is one of the most important factors in marketing Japanese con-
sumer commodities, and much thought goes into how brand names are written. The Japa-
nese automaker Toyota and motorcycle manufacturer Suzuki both use katakana even though
the brand names were originally family names, and family names are usually represented in
kanji or hiragana. The choice of katakana creates an impression of being modern, high-tech,
and international. In fact, most of the tech companies in Japan preferred katakana over kanji
branding. Toyota and Mazda switched to katakana years ago (though Nissan still uses kanji
along with Roman letters). Toyota’s logo design, in particular, went through several changes
in script.6
182 Part V Identity and gender in writing Chinese
By contrast, Japanese companies that have a distinguishingly long history often choose
to stay with their original kanji names to emphasize their time-honored heritage. Mitsui (三
井) and Sumitomo (住友), both multi-industry international corporations, trace their roots
to merchants in the Edo period (1603–1867) and have maintained their branding in kanji in
honor of their founders.7
Another example to illustrate the point here is from Japan’s alcoholic beverage industry.
As you may have noticed, brand names for sōchū or sake, which are traditional Japanese bev-
erages, tend to have kanji brands. Beer and whisky, which are later Western imports, usually
have their brands represented in katakana.
Violation of script choice conventions could achieve emphatic effects, much like
using bold or italics in writing English, and it is a tactic sometimes used in advertise-
ment.8 For example, a catchphrase in Japanese is used to market the plush toys of the
brand Tarepanda, and it is written entirely in rōmaji to achieve this eye-catching effect:
sawaruto yawarakaku igaito sittori siteiru,9 meaning “when you touch it, (you’ll find) it
soft and unexpectedly but comfortably moist.” In conventional orthography it would have
been written as 触ると柔らかく意外としっとりしている in a mixture of kanji and
hiragana.
Extra-linguistic factors that influence script choice may also include the gender or age
of the author or the target audience as well as the stylistic features of the writing. In gener-
al, the use of kanji is perceived by the Japanese as indicative of erudition on the part of the
author or the readership, who tends to be expected to be male or of middle age or above.
Since the cursive-stroked hiragana is usually associated with softness or femininity, the au-
thor or the audience is more often expected to be young or female. It also conveys a sense of
tradition as compared with katakana. Katakana, with its straight or angular strokes, is often
associated with modernity and pop culture, a perception that may have to do with the fact
that new, Western loanwords are usually written in katakana. This script is generally per-
ceived as suggesting a young and masculine identity. The use of rōmaji is more dominantly
related to commerciality or consumerism and is often associated with young, female authors
or readers.10
Chapter 20 Script choice in writing Japanese 183
Companion Website
These script stereotypes are reflective of the social stereotypes rooted in cultural-historical
context. For instance, hiragana is perceived as suggesting femininity and familiarity not just
because of its soft curved shape in contrast to the sharp angular lines of katakana or because
of its relatively simple form compared to kanji. It is also because historically hiragana was the
script used by women to communicate their private thoughts and to create literary works,
when kanji, the official script at the time, was almost exclusively used by men or for official cor-
respondence. Because of this, hiragana was referred to as onnade (女手 ‘woman’s hand’) while
kanji was also called otokode (男手 ‘man’s hand’). Kanji has also been the script associated with
formal education and thus intellectual cultivation and erudition from its early use to this day.
Similarly, the association of katakana with modernity and pop culture and that of rōmaji
with commerciality are rooted in the more modern reality of the Japanese society. Katakana is
the script used to write Western loanwords, which occur more predominantly in the context
of urban pop culture than elsewhere. In particular, katakana has come to convey the modern,
metropolitan, and often international feel of popular youth culture. Rōmaji, the use of which is
a legacy of the Occupation period (1945–1952), is frequently used in advertising or public sig-
nage to catch consumers’ attention for products and services. It signifies non-indigenousness
and conveys a cosmopolitan appeal. Often the meaning of the words or expressions matters
less than its decorative value, and the Roman letters were used to generate a visual impact.12
Companion Website
Notes
7 Noguchi (2009).
8 Iwahara et al. (2003, p. 379).
9 Tranter (2008, p. 142).
10 Based on Table 1 from Smith and Schmidt (1996, p. 50); also see Frank (2001, p. 209).
11 Koichi (2011).
12 Tranter (2008, p. 149).
13 Based on study conducted by Smith and Schmidt (1996).
14 Frank (2001).
15 Ibid., p. 222.
16 O’Mochain (2012).
17 Frank (2001, p. 217).
PART VI
What makes calligraphy a thriving art form in East Asia? It has to do with, among other
factors, the characteristics of the Chinese writing system. First of all, as a morphographic
script, the Chinese writing system requires a large number of symbols. Recall that basic lit-
eracy in Chinese usually means being able to read and write 2,000 to 3,000 characters. Cal-
ligraphers may master significantly more, and this large number of symbols provides them
with a rich visual repertoire for creating their works. Furthermore, Chinese characters are
composed of a variety of strokes. Calligraphers have considerable freedom to vary the shapes
and thickness of the individual strokes as well as how they relate to each other in the overall
compositions of the characters. Such are the expressive resources at the disposal of the callig-
raphers when they create new styles.
Chinese calligraphy matured as an art form along with traditional Chinese painting. In
pre-modern China, the painter-calligrapher, who was also usually a scholar, created calligra-
phy and painting in the same studios using the same tools and materials. For this reason, it is
perhaps no wonder that traditional Chinese calligraphy and painting share the same aesthetic
principles: vitality (气韵生动/氣韻生動 qìyùn shēngdòng ‘life-like energy’), balance, and in-
dividuality are among the fundamental criteria in evaluating the merit of either painting or
calligraphy.
You may wonder whether, to appreciate and enjoy Chinese calligraphy, one needs to be
able to read the Chinese script and understand what is written. After all, calligraphy works
usually contain meaningful texts, whether they are thousands of characters or single charac-
ters. To answer this question, it is important to understand that Chinese calligraphy is primar-
ily an abstract art. It is much less about what is written than the rhythm, vitality, and balance
of the marks created by the successive spontaneous movements of the brush. Indeed, Chinese
calligraphy is valued “purely for the sake of the satisfactory nature of its lines and groups of
lines.”4 In this sense, it speaks a universal artistic language that transcends the linguistic infor-
mation encoded in Chinese characters.
To calligraphers, the historical variations in the form of Chinese characters offer rich
artistic resources. An accomplished calligrapher usually masters more than one style and is
capable of drawing inspiration from multiple varieties to create her or his own personal style.
Students of calligraphy typically start with one style – most commonly the standard script –
tracing, copying, and emulating masterpieces produced by renowned calligraphers in the an-
cient past, and they then move on to similar practice with other styles. In what follows, we will
take a look at each of the major script styles, in particular their historical background, visual
characteristics, and the stories behind their creation.
Oracle-bone script
The earliest recognized archeological evidence we have to date of Chinese writing is characters
inscribed on so-called oracle bones during the late Shāng (商) dynasty (1200–1050 bce). The
writing was done on ox shoulder blades (牛骨 niú gǔ) or turtle undershells (龟甲/龜甲 guıˉ jiǎ)
for divination purposes, hence the name jiǎgǔwén (甲骨文: 甲 jiǎ ‘shell,’ 骨 gǔ ‘bone,’ 文 wén
‘writing’) in Chinese and “oracle-bone script” in English. Ancient kings and their diviners used
oracle bones to predict the future: A piece of bone or shell would first be inscribed with a question
or an inquiry. It would then be subject to heat until it cracked. A diviner would look at the cracks
and come up with an answer to the inquiry by interpreting the patterns the cracks had formed.
The prediction made by the diviner would then be carved on the same piece of bone or shell.
Sometimes, what actually transpired would also be recorded following the actual event. In this
fascinating manner, a piece of oracle bone constitutes a historical record of events that took place
several thousand years ago. The oracle bones discovered to date contain 4,000–4,500 different
characters, only about 1,200 of which have been deciphered.6 It is likely that many of the char-
acters dropped out of use, and only about 20% still exist in a transformed form in today’s script.
What does the oracle-bone script look like? Figure 21.2 offers one example. As seen
in the image, the characters contain almost entirely straight lines: horizontals, verticals, or
slants. There are few curves, circles, or dots, stroke types that are more commonly used in later
scripts. The lines are thin and vary little in width. These characteristics probably have to do
with the hard surface and the sharp writing tools used in the inscription of the characters. We
may also observe that the size of the characters are not entirely uniform, some distinctively
larger than others. The characters are arranged in columns but are not yet aligned horizontally.
FIGURE 21.2 Oracle bone from the reign of King Wu Ding (late Shaˉng dynasty, ca. 1200 bce) National
Museum of China
Source: CC BY-SA 3.0 (https://creativecommons.org/licenses/by-sa/3.0)
192 Part VI Chinese characters in art and literature
Bronze script
The Shāng (商) and the Zhōu (周) dynasties (16th–3rd centuries bce) saw the use of the
bronze script (金文 jīnwén, 金 ‘metal’). This style is also known as zhōngdı̌ngwén (钟鼎文/鐘
鼎文: 钟/鐘 zhōng ‘bell,’ 鼎 dı̌ng ‘(a type of ceremonial vessel)’) because it is most often seen
on ritual bronze vessels cast during the Shāng and Zhōu dynasties. As shown in Figure 21.3,
characters may be inscribed on the inside of a bronze vessel, arranged in a way that comple-
ments the design of its exterior, achieving an overall aesthetic harmony.
The vessel in Figure 21.3, called Duke Máo’s Cauldron (毛公鼎 Máogōng Dı̌ng), in fact
bears a prime example of the bronze script. The inscribed essay records how the King Xuān of
the Zhōu (周宣王), in a time of unrest, charged his uncle Duke Máo with governing state af-
fairs and bestowed on the Duke a wealth of gifts. The Duke had the cauldron cast to record the
honor given by the king and intended for the vessel to be used with respect by his descendants
for generations to come. The nearly 500-character inscription inside the cauldron is not only
the longest text on a bronze vessel but also one of the most elegant. Figure 21.4 is a rubbing of
half of the inscription inside the vessel. As we can see, compared with the oracle-bone script,
the broze script appears more elaborate. The characters tend to have a greater number of
strokes, and their structure is more complex. They also make more use of curved lines and less
of angular lines, giving the characters a softer look. The sizes of the characters have become
more uniform by this stage, even though they may not all fit in equal-sized square cells. Like
the oracle-bone script, the characters are usually arranged in columns and do not yet form
clear rows.
FIGURE 21.3 Duke Máo’s Cauldron with bronze script inscription on the inside
Source: Collection of National Palace Museum, Táiwān
The great seal script synthesized the stylistic features of previous scripts and further
matured from the bronze script. Stone-drum characters appear even more sophisticated than
the characters on Duke Máo’s Cauldron. As we can see from Figure 21.5, the characters in
general have a greater number of strokes and contain more curved strokes. Each character is
well balanced in structure and is roughly the same overall shape and size. The characters are
Chapter 21 Chinese calligraphy 195
not only aligned in columns but are also neatly arranged in rows. Since their discovery in the
7th century, the stone drums, extolled for the distinctive beauty of their inscription, have been
a source of inspiration for many calligraphers.
A simplified version of the great seal script, the small seal script does not represent a
drastic stylistic departure from the great seal. The individual characters are more refined and
balanced, but the strokes are still curved and are still of even thickness. Figure 21.6 shows a
Qín dynasty brick with 12 characters cast on its surface. Like the great seal script, the small
seal cannot be written quickly. It was soon replaced in official use by the clerical script de-
signed for more expedient writing.
Clerical script
The development of the clerical script (隶书/隸書 lìshū) is a critical step in the calligraphic
evolution of the Chinese writing system. It marks the transition of the Chinese script from the
archaic style to the modern style. Chinese characters written in the clerical script and onward
are more easily recognizable to today’s readers educated in contemporary characters, because
they bear significantly more visual resemblance to the contemporary script than do any of
the previous styles. Figure 21.7 shows a rubbing of the Cáo Quán (曹全) stele10 of the Eastern
Hàn dynasty, a representative work in clerical script. The curved lines of the seal scripts are
replaced by straight or slanted strokes or sharp bends. Circles have entirely disappeared. The
width of the strokes now varies, adding another dimension of artistic expression. For example,
196 Part VI Chinese characters in art and literature
a phrase often used to highlight the stylistic feature of the horizontal stroke in the clerical
script is “silkworm head and goose tail (蚕头雁尾/蠶頭雁尾 cán tóu yàn wěi),” vividly cap-
turing the dynamic change in contour as the stroke proceeds from the beginning to the end.
FIGURE 21.7 Partial rubbing of the Cáo Quán (曹全) stele (185 ce)
FIGURE 21.8 “Silkworm head and goose tail” – horizontal stroke of the clerical script
Companion Website
Following the clerical script, the stylistic repertoire of Chinese calligraphy dramatically
expanded. Three major variants of the clerical script developed almost concurrently from the
2nd to the 4th century ce. These were cursive script (草书/草書 cǎ oshū), running script (行
书/行書 xíngshū), and standard script (楷书/楷書 kǎ ishū). The standard script is the style
Chapter 21 Chinese calligraphy 197
used commonly today in printing and careful handwriting and is also the style Chinese chil-
dren are first taught in school. More experienced writers may use the running script in hand-
writing. As previously mentioned, students of Chinese calligraphy usually start their practice
with the standard script and then move on to running and cursive scripts. We will follow this
order in the following sections as we take a closer look at each of these styles.
Standard script
The so-called standard (aka regular) script (楷书/楷書 kǎ ishū) has been the standard style of
writing in China for thousands of years, and it remains so today. You may recall that, in impe-
rial China, candidates taking the civil service exams (604–1904) were required to write their
exam papers in the standard script. Being able to write a good hand in the style was believed
to enhance one’s chance of passing the exams. In modern China, schoolchildren learn to read
and write in the standard script. It is also the style people use in handwriting for formal pur-
poses, such as in filling out official paperwork. Print media, including books, newspapers, and
magazines, generally use typefaces of the standard script as well.
The standard script is visually close to the clerical script. Perhaps the most distinctive
difference between them is in the proportion of the characters. In the standard script, the
characters are almost perfectly square, while those of the clerical script tend to be longer in
the horizontal dimension than the vertical, looking almost like a squished-down version of
the standard script. Another difference is in how certain strokes begin and end. Horizontal
strokes in the standard script do not usually have “silkworm heads” or “goose tails” but are
written in a more subdued manner compared with the clerical script.
That said, calligraphers of the standard script have considerable freedom in developing
their individual styles. Works of the “four grandmasters of the standard script” (楷书四大
家/楷書四大家 kǎishū sìdàjiā) – Yán Zhēnqīng (颜真卿/顏真卿 709–785 ce), Liǔ Gōngquán
(柳公权/柳公權 778–865), Ōuyáng Xún (欧阳询/歐陽詢 557–641), and Zhào Mèngfǔ
(赵孟頫/趙孟頫 1254–1322) – are the most well-known. Although the majority of these
works are thousands of years old, calligraphy students today, more likely than not, still begin
their practice modeling on one of the four masters. Printed models of these works are also
readily available in Chinese bookstores.
Running script
It is easy to recognize the running script (行书/行書 xíngshū) in comparison with the stan-
dard script. Characters written in the standard script have distinctly separate strokes, each
carefully executed and deliberately positioned within an imaginary square. By contrast, as its
name suggests, the running script uses brush strokes written in swift motion in a manner that
creates continuous energy from one stroke to the next. As a result, strokes are semi-connected
within the characters, although the overall integrity of individual strokes is largely preserved
and the strokes are clearly discernable. The size of the characters may vary, and they may not
always conform to a square shape.
The famed calligraphy sage (书圣/書聖 shū shèng) Wáng Xīzhī (王羲之 321–379 ce)
is credited for having created the running script. He also wrote in this script, in the year 353,
what is perhaps the most well-known piece of work throughout the Chinese calligraphic his-
tory, Preface to the Poems Composed at the Orchid Pavilion (兰亭集序/蘭亭集序 Lántíngjí
198 Part VI Chinese characters in art and literature
Xù). The script allows the artist considerable freedom in modifying the strokes or the overall
structure of the characters, encouraging an inventive use of brush and ink. For example, in
Preface, Wáng wrote the character 之 (zhī ‘(grammatical marker)’) 21 times, each in a different
form. Such spontaneous variation is highly prized. However, it is also believed to be inimita-
ble. It is said that Wáng wrote the Preface after having had a few drinks with his friends. He
liked it so much and tried to replicate it later on but could not do so to his own satisfaction.
Nonetheless, the Preface is widely regarded as a masterpiece – legendarily so, in fact – and it
has served as a model for numerous calligraphers and amateurs to this day.
Cursive script
The cursive script (aka grass script,11 草书/草書 cǎ oshū) was originally invented during the
Hàn dynasty as a quicker version of the clerical script. Characters are routinely simplified by
replacing components with single, continuous strokes. Often, several characters are written
in one unbroken stroke as the brush moves continuously downwards. As you can imagine,
the cursive script is the most expressive of spontaneous, free-flowing energy. Perhaps for this
reason, it is the preferred script for poetic inscription in paintings during the Táng dynasty
(唐 618–907 ce).
The Táng monk Huái Sù (怀素/懷素 737–799) was a grandmaster of the cursive script.
Only a few pieces of his works survived, among which his Autobiography (自敘帖 Zìxù Tiē)
represents the epitome of cursive-style calligraphy for its unparalleled artistic accomplish-
ment and influence (Figure 21.9). The surviving copy is now housed at the Palace Museum in
Taipei and is on view once every 16 years.
We have taken a brief survey of the major script styles in the history of Chinese calligra-
phy. To conclude, it is important to keep in mind the following two points. First, these styles
did not develop in a strict linear order. For example, there was significant overlap between the
great seal, small seal, and clerical scripts, as well as between the clerical, running, cursive, and
standard scripts in terms of their predominant time periods. The running, cursive, and stan-
dard scripts have coexisted from their inception to this day. Second, although we have learned
about the script styles as distinct categories, they are not strict, mutually exclusive categories.
In practice, calligraphers may choose to blend two or more styles and write in a manner that
falls somewhere between, for example, the typical running and the typical cursive scripts.
Therefore, not all calligraphy work ever created could be clearly labelled with one of the script
names we have reviewed above. There are also sub-styles within the major categories. In daily
writing, standard and running scripts are the most commonly used styles, but all the major
script styles find their place today in works created by practitioners of traditional and modern
calligraphy.
Notes
1 The Shāng dynasty divination inscription is dated to 1250–1050 bce (Wilkinson, 2015, p. 681).
2 See Sullivan (1999).
3 Practitioners of water calligraphy (aka ground calligraphy) often use extra-large brushes dipped in
water to write on cement surface of public areas, such as on sidewalks or in public parks, as a form
of physical exercise.
4 Chiang (1973, p. 110).
5 Chapter 7.
6 Wilkinson (2015, p. 684).
7 Ibid., p. 682.
8 The state of Qín defeated the other states and united China to establish the Qín dynasty.
9 Wilkinson (2015, p. 693).
10 A stele is a stone slab, usually taller than it is wide, that is erected as a memorial. It generally carries
inscription on at least one side (the side considered to be the front) of its surface.
11 The name “grass script” for cursive script may have come from the idea that the brush strokes look
like leaves of grass blowing in the wind.
22 A calligraphy workshop
I
F YOU HAVE NEVER had the experience of writing with brush and ink, in this chapter,
you will get a taste of what it is like. As you can imagine, developing highly refined skills
in East Asian calligraphy takes an enormous amount of training. Many accomplished cal-
ligraphers start their practice in childhood and persist over a lifetime. Without substantial
experience handling calligraphy tools and materials, it is nearly impossible to produce highly
skilled works. This practice, however, will be useful for you to gain direct and personal un-
derstanding of East Asian calligraphy through hands-on experience. The end-product may or
may not be exactly the way you want it, but it will be a remarkable piece of work nonetheless
as an earnest attempt at a challenging art form.
Getting ready
Tools and materials
The main instruments and materials you will need for calligraphy practice are commonly
referred to in Chinese as wénfáng sì bǎ o (文房四宝/文房四寶), “four treasures of a schol-
ar’s studio.” These are brush (笔/筆 bı̌), ink (墨 mò), paper (纸/紙 zhı̌), and inkwell (砚/硯
yàn). In pre-modern times, they were indispensable items for students and scholars in the
East Asian cultural sphere. The manufacturing of these items became highly developed both
technically and aesthetically, so that the finest examples, especially of brushes and inkwells,
became valuable collector items and find their places in museums and auction houses. With
the adoption of Western writing tools from the mid-19th to the early 20th centuries, brushes
and inkwells have long been replaced by fountain pens or ballpoint pens for daily writing pur-
poses. However, the sustained popularity of traditional art ensures that these items are readily
available today from stationery sellers or in art supply stores. In the West, they can usually be
purchased at Asian bookstores, large art supplies shops, or online sellers based in East Asia.
For this workshop, we will use student-grade tools and materials. They are reasonably
good in quality, quite affordable, and possibly the most readily available. The following guide-
lines aim to help you select the right types of “treasures” to use for this practice.
The brush you choose should be medium to large, with a bristle length of 1.5–2 inches.
Brushes of this size are ideal for writing characters that fit comfortably in square cells of 3–4
inches in width and height. Although this is much larger than the characters used in actu-
al writing for everyday purposes, big characters are considered a wise choice for beginners,
because writing big makes it easier to grasp the characteristics of individual strokes and to
appreciate the composition of whole characters. In fact, practicing calligraphy is colloquially
referred to as xiě dàzì (写大字/寫大字) “writing big characters” in Chinese.
Chapter 22 A calligraphy workshop 201
Calligraphy ink comes in two main forms: solid ink sticks and bottled liquid ink. Ink
sticks are the traditional form. They need to be ground and mixed with water right before
writing takes place. Compared with ready-mixed liquid ink, solid sticks allow the calligrapher
more control over the ink’s consistency. The action of ink grinding can also be a calming and
meditative process that prepares the artist for the writing. For our purpose, however, bottled
ink is a better choice. It is more convenient and affordable, and it also saves the worry about
achieving the right consistency.
Calligraphy paper, in Chinese, is commonly referred to as “Xuān paper (宣纸/宣紙
xuān zhı̌).” Although in English it has the common name “rice paper,” rice is not its main
ingredient. Xuān, in fact, comes from the ancient place name Xuānzhōu (宣州 ‘Xuān prefec-
ture,’ today’s 宣城 Xuānchéng ‘Xuān city’), where a highly prized kind of calligraphy paper has
been produced since the Táng dynasty (618–907 ce). Authentic Xuān paper uses the bark of
elm trees indigenous to the area as its main ingredient. Many calligraphy students start with
writing on gridded practice paper that is of a lesser quality than Xuān paper. The printed grids
are an excellent aid for beginners, so we will also use this kind. Such practice paper is gridded
either with a rectangle inside each cell like the character 回 (huí) or with crossing lines like
the character 米 (mı̌), as shown in Figure 22.1 Our calligraphy model uses the mı̌-grid, so it
is better to use the mı̌-gridded paper. Each sheet usually contains 12 or 15 cells. For today’s
practice, you will need 8–12 sheets.
FIGURE 22.1 Huí gridlines and mǐ gridlines on typical calligraphy practice paper
Since we will not be grinding ink, a dedicated inkwell or ink stone may not be necessary
and can be substituted by a small, shallow container for liquid, such as a ceramic saucer.
In addition, there are a few items you may consider using to make the writing process
smoother. You may use a piece of felt underneath the calligraphy paper to keep the ink from
seeping through onto the desk surface or use a couple of sheets of newspaper for the same
purpose. You may also want to use a paperweight to keep the paper in place while writing.
A few sheets of paper towel may come in handy when you need to wipe ink off the desk.
Before you start writing, take your time to arrange the tools and materials on your desk.
If you are right-handed, place the calligraphy model to the left of the practice paper, and
the brush, ink, and inkwell to the right, as is shown in the diagram (Figure 22.2). If you are
left-handed, then you may want to switch the items on the two sides.
the calligraphy brush is almost always held perpendicular – that is, at a 90-degree angle – to
the writing surface (see Figure 22.3), unlike a pen, which is usually held at a 30- to 45-degree
angle. To do this, the key is to bend your wrist upward, so that the palm is almost upright.
Place your fingers as shown in Figure 22.3. At the top, your index finger, middle finger, and
thumb work together to hold the brush. The index and middle fingers are on the same side of
the brush, while the thumb counterbalances them from the opposite side. The index and mid-
dle fingers may spread apart from each other to give a steadier hold. Below them, the brush
rests against the top knuckle of the ring finger. Then, the little finger rests against the inside
of the ring finger to provide additional support. The five fingers assume different angles in
relationship to the writing surface: the index finger is more or less horizontal, while the others
are almost vertical. The middle, ring, and little fingers point downwards, while the thumb
points upwards. Note that your fingers naturally curve around the brush without touching the
palm. They do not clench together to form a fist; rather, they should be quite relaxed, leaving
an empty space between them and the palm, and the size of this space is – in conventional
pedagogical lingo – large enough to fit in an egg.
Holding the brush in this manner, you should be able to move its tip forward and back-
ward relatively freely. When you do so, the thumb and the index finger are stationary, making
a pivot where they touch the brush. The other three fingers then move in coordination to bring
the brush away from you and then toward you. This is the basic gesture used when writing a
vertical stroke. In writing a horizontal stroke, most of the movement is accomplished by mov-
ing the entire hand left to right, with the wrist or the elbow as the pivot. To help keep a steady
Chapter 22 A calligraphy workshop 203
hand, beginners may wish to rest the wrist on the desk while writing. This will limit the size
of the characters, however, and is why sometimes more strict training may require a “hanging
wrist” with the wrist hanging a few inches above the writing surface instead of resting on it.
It is also important to sit in a correct posture when practicing calligraphy. Sit with a
straight back. Do not lean forward against the edge of the desk. There should be a two- to
three-inch distance between your body and the desk. Your forearms should naturally rest on
the desk. If you are right-handed, you can use your left hand to hold down the calligraphy
paper as you write, and vice versa. Do not cross your legs. Place both of your feet flat on the
ground without extending your legs forward or bending them backward. Your overall body
posture should be straight yet relaxed to avoid fatigue or overstraining your muscles. Do not
hold your breath. Keep breathing naturally while you write.
Warm-up
Writing using a soft brush may feel very different from using a pen. Before we start writing
Chinese characters, let’s do a few exercises to warm up.
We will use the center-tip technique (中锋/中鋒 zhōng fēng) in these exercises. Pay at-
tention to how the mark you make changes as you alter the pressure and direction of the brush
tip. One constant factor is that the tip of the brush should always remain in the center of the
mark you make. For example, when writing a vertical stroke, the brush tip always makes a
symmetrical contact on the paper along the direction of its movement without leaning either
to its left or to its right. This is what is called the center-tip technique. It is the most commonly
used brush technique.
First, make some straight and even lines. Start by making a few horizontal lines across
the width of your paper. Maintain steady pressure so that the width of the lines remains even.
Then make a few vertical lines that cross the horizontal lines at even intervals. Keep the width
of the vertical lines constant and consistent with the horizontal marks. Do this repeatedly
until you feel comfortable using the brush to create straight and even lines.
204 Part VI Chinese characters in art and literature
Then, vary the pressure on the brush to change the width of the lines. Use less pressure
for thinner lines and more pressure to make the lines thicker.
In the next exercise, use the brush to make half circles, still keeping the width of the lines
even. Do this in a variety of directions.
In the last exercise, make diagonal lines and vary the pressure of your brush as you write.
Start out relatively thin and then make the line thicker while you put more pressure on the
brush tip. Then, reduce the pressure again until eventually lifting the brush off of the paper.
Basic strokes
Chinese characters are composed of a variety of strokes. Although the number of characters
is enormous, the types of strokes are quite limited. The most basic ones include horizontal
(横/橫 héng), vertical (竖/豎 shù), down-left sweep (撇 piě), down-right sweep (捺 nà), bent
(折 zhé), hook (钩/鉤 gōu), and dot (点/點 diǎn). Let’s practice writing each one of them.
First, the horizontal stroke. Before we start, look at the model (Figure 22.5) carefully.
What characteristics do you see in the form of this stroke? Is it even in width? Is it perfectly
horizontal? Is it positioned right in the center of the square cell? You will notice that the two
ends of the stroke are slightly thicker and the middle portion is relatively thin. The stroke
slants subtly upward as it goes from left to right. The overall stroke is in the center of the cell.
Therefore, it begins somewhat below the middle line and ends a little bit above it. Written in
this manner, the horizontal stroke is shaped almost like a piece of bone, demonstrative of its
inner strength.
Now let’s write the horizontal stroke. Start off by pressing down the tip of the brush at
a 45-degree angle. Then drag the brush toward the right while slightly and gradually lifting it
as it goes toward the middle. As it passes the middle point, slightly press it down as it moves
forward. To end the stroke, lift the brush so that only the tip lightly touches the upper right
edge of the stroke, press down on the brush at a 45-degree angle, and then bring it back up to
round off the corner, resulting in a somewhat rounded end of the stroke.
There are two types of vertical strokes: one with a rounded ending called “hanging dew-
drop” (垂露/垂露 chuí lù; see Figure 22.6), and the other with a sharp tapered ending referred
to as “suspended needle” (悬针/懸針 xuán zhēn; Figure 22.7). The beginning of both of the
Chapter 22 A calligraphy workshop 205
vertical strokes is the same and similar to that of the horizontal stroke. The ending of the
“hanging dewdrop” is similar to the ending of the horizontal stroke. The “suspended needle”
ends somewhat differently: instead of lifting and pressing the brush to round off the end, it
simply lifts to form a sharp needlepoint at the end. This kind of vertical stroke is usually used
when it is the last stroke of a character.
The down-left (Figure 22.8) and down-right (Figure 22.9) sweeps are similar to the
suspended-needle vertical stroke in terms of brush technique. The one exception is the ending
of the down-right sweep. As you see in Figure 22.9, the stroke has a smooth curve on its upper
edge and an angular shape on its lower edge. This effect is achieved by changing the direction
of the brush while maintaining a smooth top edge.
206 Part VI Chinese characters in art and literature
The bent combines a horizontal stroke and a vertical stroke (Figure 22.10). Where
the two parts meet it forms a bent. At this juncture, lift the brush almost but not entirely
off the page to change its direction, either from horizontal to vertical, or from vertical to
horizontal. When you press the brush down again to go in the new direction, it should be
at a 45-degree angle, much like the technique used at the start of a horizontal or a vertical
stroke.
The hook might be one of the most difficult strokes to write for beginners. It is similar
to the bent in that the brush needs to change directions. However, it is also different because
it usually tapers off into a pointed ending at an angle greater than 90 degrees. The hook can be
attached to a horizontal or a vertical, as shown in Figures 22.11 and 22.12.
Chapter 22 A calligraphy workshop 207
The “dot” is really a cover term for a variety of short strokes. Unlike what the term may
imply in English, a “dot” in Chinese calligraphy is almost never round. Rather, it is sometimes
shaped like a raindrop, sometimes appears to have a hook coming out of it, and at other times
looks more like a shorter vertical stroke, all depending on which part of the character it com-
poses. What these various shapes have in common is their brevity. They are so much shorter
than a regular horizontal, vertical, or sweep that conceptually it is natural to think of them all
as dots. Figure 22.13 is what can be called a “left dot” and Figure 22.14 is a “right dot.”
A single stroke can also be a combination of several of the basic stroke types. For
example, the character 力 (lì ‘strength, power’) consists of two strokes, a simple piě (down-left
sweep) and a more complex héng-zhé-gōu, that is, a combination of horizontal, bent, and hook.
208 Part VI Chinese characters in art and literature
Stroke order
In writing a character, which stroke comes first, and which comes second? Does stroke se-
quence matter? Yes, it does. In fact, stroke order matters a great deal in calligraphy. One
reason is that strokes are often connected in running and cursive scripts. A group of strokes
may be done in one continuous motion of the brush. The sequence of the strokes thus con-
tributes to the spontaneity of the resulting characters and may determine whether the char-
acters “feel right” or not. Following commonly established stroke orders also ensures that
strokes are connected in a consistent manner so that the characters written by one callig-
rapher are recognizable to others. The role of stroke order in the standard script may be
less apparent but is just as important, since mastering the standard script, for perhaps the
majority of calligraphy students, serves as the foundation for progressing to the running and
cursive scripts.
There are a few major patterns or principles when it comes to stroke order:
It may not always be completely straightforward to determine the exact stroke order of a char-
acter. Even native writers sometimes vary among themselves in how they write certain char-
acters. That said, are you up for a few challenges? See the companion website for stroke-order
exercises.
Companion Website
Composition
Writing well does not only require beautiful execution of individual strokes. It also de-
mands close attention to the spatial relationship between the strokes. In general, calligra-
phers may consider it better to have imperfect strokes but a balanced arrangement of the
strokes than the other way around, which shows how important it is to master the compo-
sition of characters.
At this stage of your practice, learning to use the printed guidelines on your calligra-
phy paper will be particularly beneficial. Before you start to write, observe closely, and when
you write, imitate as much as possible. Take the following character 山 (shān, ‘mountain,
hill’) as an example (Figure 22.15). What do you notice? First of all, there is a great amount
of empty space around it within the area defined by the gridlines. None of the strokes are
touching or even approaching the outer frame of the square cell. Then, notice how the three
vertical strokes are arranged, and how they are spaced in relation to each other. The middle
stroke is right in the center, with almost equal length above and below the middle horizon-
tal gridline. The two shorter verticals are about equal distance from the middle one, and
along the horizontal dimension each is positioned at about the middle point of its half of
the grid. On the vertical dimension, the two shorter verticals sit at about the same level, yet
they are not the same height. The one on the left is somewhat shorter, not quite coming up
to the middle gridline, while the one on the right crosses slightly over the gridline. Finally,
notice how the horizontal stroke is positioned. Is it perfectly horizontal? It is easy to see that
it is not, if you compare the distance between this stroke and the middle gridline on the left
end versus the right end of the horizontal stroke. The horizontal stroke in fact slants upward
as it goes from left to right. This subtle curvature gives it a tension – and thus an unyielding
inner strength – like that in a piece of bone or in a bamboo stem that would only slightly
curve while under pressure. As you have seen, the gridlines are very useful if you take the
time to “read” the character. Reading the character in this manner will no doubt train your
eyes to see. Being able to see the defining characteristics of a character is the prerequisite
to producing the desired result, so do take the time to look closely before you start writing.
210 Part VI Chinese characters in art and literature
Writing exercise
Now you are ready to do some hands-on practice. Use Figures 22.16 to 22.19 as a model to
write the four characters we have already looked at in the stroke-order exercise: 日 (rì), 心
(xˉı n), 永 (yǒng), and 道 (dào).
If you are interested in practicing more, you can ask your instructor for additional mod-
els to make a complete piece.
Enjoy!
23 Modern calligraphy in China
B
Y THE MID-1980S, the grand tradition of Chinese calligraphy had been increasing-
ly seen as a rigid, limited, and outdated form of expression by some calligraphers in
China who we now refer to as the Modernists. Through theoretical debates, technical
exploration, and artistic innovation, these calligraphers launched a movement to modernize
the tenets and practice of calligraphy, challenging and redefining its boundaries. They also
created and exhibited new works that were drastically different from the classical style. In this
chapter, we will learn about the sociocultural background of the Modernists’ endeavors and
look at a few of their works.
Companion Website
Another approach of the Modernists was to play with the characters – taking them
apart, recombining them, and generating new or additional meaning in the process. The piece
by Sà Běnjiè consists of two characters, 覺 (jué ‘to be aware’) and 悟 (wù ‘to understand’), with
their components reconfigured into one coherent image. The parts in red compose the char-
acter 悟, which has 心 (xıˉ n ‘heart,’ written as 忄as a radical) on the left and 吾 (wú ‘I, my’) on
the right. The bottom-left component in black, 見 (jiàn, 见 in simplified script), meaning “to
see,” is the bottom element of 覺. The top-right black component is the top part of 覺 and does
not constitute a stand-alone character. By recombining the components of the two characters
into one image, this piece may convey synthesized meaning derived from the characters. As
the artist explains, it shows that the meaning of the word 覺悟, “consciousness,” consists of “to
see” (見), “my” (吾), and “heart” (心). Therefore, being conscious can be understood as being
able to see one’s heart.
Chapter 23 Modern calligraphy in China 215
Companion Website
Qiū Zhènzhōng’s Modernist calligraphy works sought to break the rules of traditional
calligraphy, but only one or two at a time. While retaining the general look of classical pieces –
his works appear to contain a number of characters, words, or phrases written in black ink on
white paper – they were undeniably modern. One example is his New Poetry Series – Promise
(新诗系列-保证/新詩系列-保證 X īnsh ī Xìliè – Bǎ ozhèng; Figure 23.1). Unlike classical cal-
ligraphy, in which a poem is written in columns arranged from right to left, in this piece the
characters appear from left to right in rows arranged from top to bottom, consistent with the
page layout of modern texts. Qiū also allowed the brush to be almost completely drained of
ink before recharging, creating a highly dynamic ink effect. Another example is his Characters
to be Deciphered Series – No. 9 (待考文字系列 – No. 9 Dài Kǎo Wénzì Xìliè; Figure 23.2). The
characters in this piece are unrecognizable to viewers because they are ancient characters that
have not been deciphered. The content of the piece would be an unlikely choice for traditional
calligraphy, which was used to record and present important textual information. In addition,
the way the characters are scattered on the writing surface instead of being arranged in col-
umns or rows also breaks the convention.
Wáng Dōnglíng, in some of his modern calligraphy works, took yet another approach
to innovation. These works contain very few – or perhaps no – characters. They seem to shift
the visual scale of calligraphy by having the viewer zoom in so close as to see only part of a
calligraphic stroke. One example, Being Open and Empty (守白 Shǒu Bái), is shown in Fig-
ure 23.3. In this type of work, there are no discernable characters, only abstract shapes and
lines that suggest some resemblance of characters. The visual quality of calligraphy clearly
exists – there are traces of the brush made with black ink on white paper – yet at the same
time it is no longer calligraphy in the traditional sense. To the artist, breaking away from the
traditional form allows for the freedom to fully express his feelings and his understanding of
the calligraphy art.4
Wáng is also a master of cursive calligraphy written on a grand scale. Figure 23.4 shows
him writing the four-character phrase 道法自然 (dào fǎzìrán ‘Dao Operates Naturally’) in
2014 at the The Metropolitan Museum of Art. Such works are often done with a large crowd
of spectators on site. The act of writing thus becomes a performance characterized by an ebul-
lient outpouring of energy, making the creative process as integral a part of the artwork as the
final product.
FIGURE 23.4 Wáng Dōnglíng creating 道法自然 Dao Operates Naturally at The Metropolitan Muse-
um of Art, January 11, 2014
Source: Courtesy of the artist
Accompanying the changes in calligraphy was an intense debate between the Modern-
ists and the traditionalists about its essence: what was calligraphy, and what was it for? One
of the central questions was whether or not calligraphy must purely be the writing of Chinese
characters. On this issue, the Modernists held a much more liberal view than the tradition-
alists. As you have seen, some of them created works that integrated non-textual elements
or rendered characters into illegible forms. These works bore much resemblance to Western
abstract painting and were thus more approachable to Western viewers than traditional Chi-
nese calligraphy. In this sense, it might be appropriate to say that such works were the result of
cross-fertilization between traditional Chinese art and Western modern art. As for the ques-
tion of whether the calligraphy Chinese artists create without using legible characters can still
be considered “calligraphy,” it may remain a matter of debate for the foreseeable future.
Notes
3 Some art historians, for example Liu (2010), do not distinguish between the Modernists and the
Avant-gardists and consider them both modern.
4 Wáng Dōnglíng talks about the tension between his classical training and his efforts to continu-
ously innovate in calligraphy in a video interview by the Creative Thinking Project (2017). In this
interview, he also discusses his “chaotic script” (luànshū 乱书/亂書), an outcome of calligraphic
innovation for the past three decades.
24 Chinese characters in Avant-garde art
I
n THE PREVIOUS CHAPTER, we learned about the Modern Calligraphy movement that
came to fore in the 1980s. Given the sociocultural conditions of that time, it was no co-
incidence that calligraphy practitioners were pushing for change and innovating for new
forms of expression. Of course, such artistic experimentation was not limited to calligraphers.
In fact, thousands of painters, sculptors, print makers, mixed-media artists, performers, and
other artists launched an unprecedented Avant-garde art movement that eventually turned
Chinese contemporary art into a global phenomenon. Some of the most interesting works
created during and since the ’85 New Wave make use of Chinese characters or calligraphy. In
this chapter, we will look at a few examples with a focus on the period during and not long
after the New Wave, that is, from 1985 to 2000. We will examine the fascinating new perspec-
tives these works produced and reflect on some of the questions they raised.
Background
Several factors contributed to the flourishing of new concepts, perspectives, and approaches
to making art during and after the New Wave. Besides the ending of the Cultural Revolution
and the implementation of an economic reform and open policy, more direct impact on artists
came from the increasing accessibility of Western intellectual works translated into Chinese
and made available through bookstores. Books on Western philosophy, literary criticism, and
cultural theory became increasingly popular in Chinese intellectual communities and provid-
ed fertile stimuli for artists eager to speak the language of modernity and internationalization.
Reading fever
The following is an excerpt from an interview with artist Huáng Yǒngpíng (黃永砯 1954–), who was
at the forefront of the ’85 New Wave.1 Huáng’s personal account illustrates the impact of Western
texts on contemporary artists working in the 1980s.
At the end of the 1970s and through the beginning of the 1980s, there was an up-
surge in reading. What had happened during the Cultural Revolution, as many of you
know, was that books were limited and mostly banned, and people were encouraged
to read only one, the little red book, consisting primarily of quotations by Chairman
220 Part VI Chinese characters in art and literature
Mao. At the end of the 1970s, policies changed and bookstores re-opened, offering
an increasing number of translations of Western texts, as well as some traditional Chi-
nese texts. This period has been called “Reading Fever.” At this time, the intellectual
community was gripped with a desire to read a wide range of texts.
I love reading and read quite a lot during that period. Bookstores would constantly
be changing the titles on their shelves, so one had to read really quickly before the
next set of books replaced the old books on the shelves. I could read really fast, but
the information wouldn’t always stick! [Laughter.] I loved reading about philosophy,
but philosophical texts took quite a bit of time to absorb so I would spend a lot of my
free time focusing on these texts. I especially enjoyed reading Wittgenstein. Most of
us [artists] would eagerly await new translations of books and when they were made
available, we would immediately devour them.
At the same time, Chinese philosophical and literary works also became available again.
Books on Daoism and Buddhism – in particular Chánzōng (禅宗/禪宗, ‘Zen sect’) Bud-
dhism – as well as Confucianism were especially sought after. Artists continued to draw ideas
from classical Chinese intellectual works but began integrating them with Western concepts
and approaches in their artworks.
Artworks
Companion Website
The title of this work, Tiānshū (天书/天書: 天 ‘sky, heaven,’ 书/書 ‘book’), literally
means “sky book,” or “book from the sky.” In Chinese, this term is used to refer to writing that
is incomprehensible to ordinary people due to either its divine nature, or, in a joking manner,
its poor execution. In the case of this artwork, of course, the writing is unreadable for yet
another reason – none of the seemingly Chinese characters is real. Invented by the artist by
reassembling the original strokes in new ways compatible with the compositional principles
of Chinese characters, they look authentic even to native eyes. But they are not associated
with any sounds or meanings in the Chinese language and are not actually used for functional
communication between any writers or readers of Chinese.
FIGURE 24.2 Book from the Sky (details), hand-printed book page from wooden blocks inscribed
with false Chinese characters
Source: Courtesy of Xú Bīng Studio
222 Part VI Chinese characters in art and literature
In fact, Xú B ī ng invented some 4,000 characters. Not only that: he hand-carved each
character on a wooden block, printed the text using the movable blocks, created books strictly
following the bookbinding conventions of the Sòng dynasty (960–1279), and made exquisite
wooden boxes to hold the books. The entire project took him more than four years to com-
plete.
You may have all kinds of questions to ask by this point: Why did Xú B īng create so
many unreadable characters? Why did he put in so much effort to create a text that is utterly
incomprehensible? What did he intend to say with this artwork? It is natural to want to know
the artist’s intention when the artwork is so provocative and puzzling at the same time. How-
ever, the artist did not provide us with responses to any of these questions. His job was done
as soon as the artwork was completed, and he has done his job well when you, the viewer, start
asking questions. As for answers, you will need to come up with your own.
Imagine yourself walking into the exhibition hall that houses this installation. How
will you react to Book from the Sky? If you are like one of the viewers at the first show in
Beijing, you may feel a variety of emotions, depending on whether you consider yourself
Chinese or non-Chinese, whether you read and write the Chinese script, or whether you
have learned Chinese as a native speaker or a second-language learner. Perry Link (2006)
analyzed these reactions in his essay “Whose assumptions does Xu Bing upset, and why?”2 He
found that viewers who could read Chinese naturally tried to comprehend the text but then
were disappointed, frustrated, and even angered upon realizing that the characters were not
real. To them, looking for meaning where no meaning could be found was a rather irritating
experience.3 To those who did not read much or any Chinese, however, it could be a relief.
This was because not knowing the characters no longer constituted a deficiency, and they
knew as much or as little as any other viewer.4 To yet another group of viewers with a rebel-
lious spirit, the work could be exhilarating in its subversive nature: characters were created
by an individual instead of sanctioned by the nation-state, the use of characters represented
a lack of meaning rather than the creation of meaning, and books filled with characters were
printed but not to be read.5 The work called into question the fundamental idea of commu-
nication through writing, demonstrating how something that looked official and seemed to
demand serious attention – the printed text – could easily be manipulated by its creators and
ultimately end up manipulating its readers.
typical of models made from rubbings of ancient stone steles. On the table is a practice book
with the same characters as in the model printed in red outlines, guidelines for you to trace
as you write, also a typical practice for beginners. For your reference is a calligraphy manual,
black characters on white paper, with drawings illustrating the shapes of certain brush strokes.
That is also typical of traditional calligraphy manuals. You pick up the brush and are ready to
write.
So, what is there to figure out? You look at the model, and this is when you realize – wait
a minute, what is it I am writing? What is this script? It looks Chinese, but a native reader or
someone with knowledge of Chinese characters can tell that it is not Chinese. The characters
are not readable in Chinese. Are they invented characters again like those in Book from the
Sky? You look again, and again, and perhaps you take a glance at the calligraphy manual. You
notice the English letters in the manual, and it dawns on you that the script is no other than
that of English. Each letter is made to look like a component in a Chinese character, and each
English word is written as though it is a Chinese character. At this moment, you discover that
you can in fact read what is in the model: “Little Bo Peep/Little Bo Peep/Has lost her sheep
and can . . .” Voilà, you figured it out. What appears mysterious now becomes completely com-
prehensible, and what seems strange turns out extremely familiar.
Companion Website
Square Calligraphy Classroom, like Book from the Sky, challenges our assumptions
about language, writing, and identity. To Western viewers uninitiated to the content, norm,
and styles of Chinese calligraphy, the classroom, at least at first glance, may represent an un-
fathomable world of a different culture. To Chinese viewers who have grown up practicing
calligraphy, the brush, ink, red-lined tracing book, white-on-black characters of the printed
model, and nature-inspired metaphors in the manual are all cues of a familiar cultural ex-
perience. Indeed, both groups of participants come with their own culturally acquired as-
sumptions; however, both are likely surprised in the end. The English-speaking participants
would realize that they can actually read what is written, while the Chinese viewers would
find themselves puzzle over the incomprehensible script. Despite the differences in reaction,
the message may be one and the same: we are only constrained by our own culture. As we
realize that we have fallen into another one of Xú Bıˉ ng’s “traps,” we may start to contem-
plate – what other cultural traps are we in that we have yet to recognize, and how do we free
ourselves from them?
In Chinese culture, seal stamping is a traditional practice by which the owner of the
seal claims authorship or ownership or an artwork. When looking at examples of calligraphy
works, you may have noticed the red marks of the artists’ personal seals. The ones bearing the
artists’ names indicate authorship. Besides the artists, collectors may also stamp their own
seals on a work of calligraphy or painting they own. The painting in Figure 24.3, for example,
is a highly prized masterpiece, as shown by the great number of collectors’ seals.
FIGURE 24.3 Night-Shining White by Hán Gàn (active ca. 742–756), ca. 750.
Source: Collection of the Metropolitan Museum of Art
Sòng Dōng’s action to repeatedly stamp the surface of the river with a seal may be un-
derstood as an attempt to impose a certain kind of human authority over nature. However, the
character 水 (shuı̌ ‘water’) is semantically identical to the receiving surface of the stamping,
making it a pointless action. Nature exists regardless of human recognition of its existence.
Furthermore, as one can imagine, the action in itself is only futile. Any imprints on the water’s
surface are only temporary and instantly disappear as the river flows on. Perhaps the very
inability of the character to maintain any lasting existence on the river surface suggests to us
nature’s resistance against linguistic categorization – or any kind of artificial classification.
This may be one of many interpretations of this work.6
FIGURE 24.4 Copying the “Orchid Pavilion Preface” a Thousand Times by Qiuˉ Zhìjié
Source: Courtesy of the artist
What is the significance of this work? The meaning is, of course, open to interpreta-
tion. In Qiū’s own words, once the characters became indistinguishable from each other and
merged into one block of blackness – that is, for about 95% of the copying process – the act
of writing became a meditative practice of Zen.7 If you know something about the practice of
copying in traditional calligraphy training, the act of copying itself might not seem strange
to you. After all, almost all calligraphy students start by copying the works of the masters and
usually do so for years before writing without a model. There was also nothing unusual about
the choice of the model. As one of the most celebrated masterpieces, the Preface had been used
by numerous students as a model. What was unusual, however, was that each iteration of the
copying was done on top of the previous ones. With sufficient repetitions, the written words
began to disappear into the blackness; that is, with sufficient repetitions, even the best master-
piece turned into nothingness. Could this be a critical analogy for blindly following cultural
conventions? Could this be a warning against mindlkess reproduction without innovation?
What else do you think could be the message?
Notes
Y
OU MAY WONDER: WHAT have Chinese characters got to do with Western Mod-
ernist poetry? They appear to be very different subjects rooted in vastly distant tradi-
tions. It is also true that, we have so far almost always examined the Chinese writing
system within the Asian context. Only occasionally have we made comparisons between the
Sinitic script and the Roman alphabet – and in that case, the relevance is obvious. What can
possibly be the connection here? In fact, the connection is an important and fascinating one,
and we will delve into the details in this chapter. To understand the role Chinese characters
have played in the development of Modernist poetry in the English language, however, we
first need to know something about Imagism, the movement that marked the beginning of
Modernist poetic aesthetics in the West.
A poem
In a Station of the Metro
The apparition of these faces in the crowd;
Petals on a wet, black bough.
—Ezra Pound, 1913
Chapter 25 Chinese characters and Western Modernist poetry 227
In this poem, Pound presents a vivid image – a glimpse of faces appearing amid the crowd at a
dimly lit subway station in Paris – through a metaphoric equation. Instead of describing what he
sees, he has the image speak for itself. He compares the faces to bright and delicate petals on a
dark, wet tree branch. The concrete and exact details of the observation evoke a crisp image in the
reader’s mind’s eye. The language that he uses is akin to common speech, free of ornate diction
or verbose phrasing. This poem exemplifies the precise observed details, concrete metaphors, and
concise language that the Imagists aimed to pursue.
Chinese characters came to serve as illustrative support for Imagist poetics because of
Ernest Fenollosa and Ezra Pound’s collaborative work on Chinese poetry. The collaboration
was a rather unusual one. Pound, who was at the front and center of the Imagist movement,
had been actively promoting the work of Imagist poets by anthologizing and publishing
their poetry since 1912.3 By that time, Ernest Fenollosa (1853–1908) had already passed
away. In 1913, Fenollosa’s widow contacted Pound and handed him her husband’s unpub-
lished notes on Chinese poetry. Fenollosa was an American art historian with a focus on
Japanese art. While in Japan, he wrote about poetic aesthetics based on his study of Classical
Chinese poetry. His widow found that Imagist ideas – their emphasis on clarity, precision,
and economy of language – resonated with those of Fenollosa’s, and she wished for Pound
to complete the manuscripts her husband had left behind. In 1915, despite the fact that he
had never studied the Chinese language, Pound published Cathay, a collection of Classical
Chinese poems translated into English based on Fenollosa’s word-for-word rendering in his
notes. This work received critical acclaim for its “supreme beauty.”4 T. S. Eliot even called
Pound “the inventor of Chinese poetry for our time.”5 In 1918, Pound edited and published
in London Fenollosa’s essay titled “The Chinese Written Character as a Medium for Poet-
ry.” This essay, as you probably have understood by now, was where the Chinese script and
Western Modernist poetry came together. Although Pound’s poetry translation and the es-
say were criticized by Sinologists for their inaccuracy or misunderstanding of the Chinese
language and script, they were tremendously influential on the development of Modernist
aesthetics for English-language poetry. In the remainder of this chapter, we will take a closer
look at Fenollosa’s essay.
In what sense can verse, written in terms of visible hieroglyphics, be reckoned true
poetry? It might seem that poetry, which like music is a time art, weaving its unities out
of successive impressions of sound, could with difficulty assimilate a verbal medium
consisting largely of semi-pictorial appeals to the eye.8
In other words, Fenollosa suggests that Chinese poetry is primarily made of written
characters – what he terms “semi-pictorial appeals to the eye” – rather than speech sounds.9
Poetry, on the other hand, is an art of sounds, consisting of “successive impressions of sounds”
strung together. Thus, he doubts that Chinese poetry written in characters can be considered
true poetry.
Fenollosa has in this instance confused Chinese writing with the Chinese language
(speech) and has taken the written symbols of the Chinese language as its essential form of
existence. He seems to have failed to see that Chinese characters represent speech sounds
as much as any other full writing systems do and failed to understand that the Chinese lan-
guage or speech is not essentially different from the English language or speech, regardless of
whether it is written in characters or in an alphabetical script. The choice of words in a line
of Chinese poetry, for example, is not in a significant way determined by their picturesque
qualities, but rather, as in English poetry, by their sound and semantic values. Despite being
represented by a writing system that appears more “pictorial” than the Roman alphabet, the
Chinese language is as much speech in essence as the English language, and Chinese poetry is
as much a “verbal medium” as Western poetry.
Drawing from the observation that the earliest Chinese characters are predominantly
pictographic, Fenollosa claims that many of the simple pictograms and radicals contain “a
verbal idea of action”10 and are “shorthand pictures of actions or processes.”11 He gives one
example to illustrate this idea: the character 言 (yán ‘to speak’), as he describes, consists of a
mouth (口) on the bottom with two words and a flame coming out of it.12 He suggests that,
with concepts pictorially coded in this manner, expressing one’s thoughts in Chinese would be
as easily accomplished by drawing pictures as by speaking, and reading Chinese would be like
“watching things work out their own fate.”13 As fascinating as these ideas may seem, they are
hardly corroborated by the actual experience of native Chinse writers and readers. Decom-
posing characters and taking into account the meaning of individual components are hardly
part of their natural writing or reading process. Rather, characters are perceived as symbols
representing speech sounds in much the same way that Roman letters are processed in the
minds of English writers and readers.
The verbal quality encoded in Chinese written symbols, Fenollosa argues, is even more
pronounced in compound characters. He believes that when two or more components are put
together, a relationship between them emerges that is both fundamental and poetic. “Things
in motion, motion in things, and so the Chinese conception tends to represent them.”14 He
supplies five examples to illustrate this point:
Chapter 25 Chinese characters and Western Modernist poetry 229
Fenollosa seems to suggest that the meaning of the compound characters is the sum of their
components. This view, however, does not entirely hold up. As we know, the great major-
ity of Chinese characters are semantic-phonetic compounds. At least two of the examples
Fenollosa gives, 伙 ‘messmate’ (huǒ) and 洀 ‘ripple’ (zhōu), belong to this category. From our
earlier discussion in Part II Chapter 3, we also know that the so-called semantic-semantic
compounds are most likely a retrospective category made up for pedagogical purposes and
may not represent the true etymology of the characters. The most plausible semantic-semantic
compounds may be characters with reduplicated components such as “森” (sēn ‘forest’) and
“林” (lín ‘woods’), and the others are more likely semantic-phonetic compounds for which
the phonological clues have become opaque due to historical sound changes. If this is true,
then春 (chūn ‘spring’) and男 (nán ‘male’) are more likely semantic-phonetic compounds as
well. The remaining example, 東 (dōng ‘east’), is generally considered a simple character, not
a compound. All in all, it is more likely than not that none of the examples above can serve as
valid evidence for Fenollosa’s argument.
For Fenollosa, the visual poetic quality of characters extends to larger segments, such as
phrases, and adds additional layers of meaning through a mechanism he calls the “pictorial
method.”17 For example, in 日昇東 (rì shēng dōng ‘sun rises [in the] east’), the 日 component
occurs in all three characters, creating a visual connection that may be difficult to achieve in
Roman alphabetical writing. In Fenollosa’s imagination, the composition in Chinese char-
acters is varied enough for the poet to choose “words in which a single dominant overtone
colors every plane of meaning,” and he even goes so far as to call this “the most conspicuous
quality of Chinese poetry.”18 Of course, given what we have discussed up to this point, it is not
difficult to see that the pictorial method arises from Fenollosa’s imagination and is not some-
thing known to have been used as a creative device in composing classical Chinese poetry.
Overall, Fenollosa’s misguided claims may remind us of the ideographic myth (see
Chapter 8) that is at the root of much of the misunderstanding about the Chinese writing
system. He has greatly exaggerated the pictographic origin of the writing system and mis-
takenly considered Chinese characters as pictures representing ideas, actions, or processes
directly without referencing speech sounds. As we have previously learned, it is crucial to
understand that the Chinese writing system, like any full writing systems, is essentially visual
representation of speech sounds. Without the language (speech) behind the writing, it would
be impossible to read the script, and writing would also be devoid of reliable meaning or
communicative function.
of inflections, such as Japanese and Spanish, do not have to follow this order), for example,
“farmer – pounds – rice.” He reasons that such universality of the “agent-act-object” form
must imply its naturalness: it “was forced upon primitive men by nature itself. It was not we
who made it; it was a reflection of the temporal order in causation.”19 If the sentence denotes a
fact, then the verb is “the very substance of the fact denoted” and “erects all speech into a kind
of dramatic poetry.” According to Fenollosa, the dominance of the verb in a Chinese sentence
gives us the model of a terse, fine style.
Companion Website
Fenollosa does not agree with professional grammarians’ definition of what constitutes
a sentence. First of all, he thinks it is problematic to define a sentence as something that “ex-
presses a complete thought,” because in nature there is no completeness. “Acts,” as Fenollosa
sees it, “are successive, even continuous; one causes or passes into another.”20 Therefore, he
argues, there can be no complete sentences, and recognizing this gives the poet the creative
license to employ phrases, rather than grammatically complete sentences, for their succinct-
ness and sharpness.
Fenollosa thinks it is also problematic to define a sentence as something that “unites a
subject and a predicate” as grammarians do. He thinks of this definition as an artificial con-
struct out of “pure subjectivity,” because the subject and the predicate are both decided by the
speaker. In that sense, “the sentence . . . is not an attribute of nature but an accident of man as a
conversational animal.” If so, “then there could be no possible test of the truth of a sentence.”21
The distrust Fenollosa has of grammarians applies similarly to logicians. In fact, he
considers the study of logic developed in the Middle Ages the source from which prescrip-
tive grammarians have derived their linguistic theories. Both logic and grammar deal with
abstraction, while Fenollosa believes that thought deals with concrete things and that the lan-
guage of poetry must therefore also strive for concreteness.
and conjunctions all derive from verbs, and even prepositions contain “secrets of verbal
metaphor.”25
As an example, Fenollosa explains that the Chinese character 明 (míng ‘bright’) serves
as a verb, a noun, and an adjective. The compound consists of two parts, 日 (rì ‘sun’) and 月
(yuè ‘moon’). Thus, when writing “the cup’s brightness,” one is literally writing “the sun and
moon of the cup.” In this case, 明 serves as a noun. It can also function as a verb to mean “the
cup shines.”26 It is unclear what expressions Fenollosa is referring to in this case. “The cup’s
brightness” could be 杯之明 (bēi zhī míng; 之 zhˉı indicates possession), but “the cup shines”
cannot be *杯明, as this is not a legitimate word. Fundamentally, however, this again reveals
Fenollosa’s confusion between language and writing. He seems to have failed to recognize that
the part of speech of a morpheme is determined by how it functions in the language and has
little to do with how the morpheme is visually represented in writing.
Fenollosa thinks that parts of speech are not natural and are invented by grammarians.
He also believes that grammar for Chinese is imposed by foreigners. The idea that certain
words are nouns and certain others are verbs or adjectives is indeed a recent Western import
for describing the Chinese language.27 Fenollosa’s observation about the flexibility in the parts
of speech of Chinese also has some merit. However, the concept of a morpheme functioning
in multiple parts of speech may be more applicable to Classical Chinese, when words are more
often monomorphemic. In modern Chinese, however, significantly more words are polymor-
phemic, so different parts of speech of related words do tend to be in different forms.
Fenollosa’s essay makes for an excellent lesson for us – students of the Chinese writing
system. Given that the purpose of Fenollosa’s essay is philosophical, it might seem beside
the point to scrutinize the linguistic details it uses as evidence, since the inaccuracy of the
evidence does not necessarily invalidate the principal arguments. Yet, there is a difference be-
tween knowing and not knowing just what makes the linguistic evidence questionable in this
essay. Only when we know what the linguistic problems are can we fully appreciate the merit
of the essay in a context beyond linguistics. Critiquing Fenollosa’s understanding of Chinese
writing requires us to be able to tease apart “truth” from “fiction” before drawing conclusions
that may be founded on more solid evidence. Finally, it is worthwhile to note that despite the
many linguistic errors in Fenollosa’s analyses, his vision of what counted as good poetry was
remarkable and invaluable in the development of Modernist poetic aesthetics. It is in this spir-
it that we have examined closely a few of the key arguments raised in his essay.
Notes
9 Fenollosa does seem to also recognize the representation of speech sounds by Chinese characters:
“Chinese poetry has the unique advantage of combining both elements [sound and image]. It speaks
at once with the vividness of painting, and with the mobility of sounds. It is . . . more dramatic. In
reading Chinese we do not seem to be juggling mental counters, but to be watching things work
out their own fate” (Fenollosa et al., 2011, p. 45). But his focus is so overwhelmingly on the written
characters that consideration on the sound component is nearly non-existent.
10 Ibid., p. 45.
11 Ibid., p. 46.
12 Ibid., p. 46.
13 Ibid., p. 45.
14 Ibid., p. 46.
15 Fenollosa’s essay obviously refers only to traditional characters. Because it references a large number
of characters and often focuses on their appearance, to avoid confusion, simplified forms of the
Chinese characters are not provided for this chapter as they are in the other chapters of the book.
16 Modified from examples on p. 46 of Fenollosa et al. (2011).
17 Ibid., p. 59.
18 Ibid., p. 60.
19 Ibid., p. 47.
20 Ibid., p. 47.
21 Ibid., p. 47.
22 Ibid., p. 51.
23 Ibid., p. 48.
24 Ibid., p. 51.
25 Ibid., p. 52.
26 Ibid., p. 51.
27 The first scholarly work written by a Chinese that systematically described the grammar of Chinese
was Mǎ JiànzhŌng’s (马建忠/馬建忠) book Mǎshì WéntŌng (马氏文通/馬氏文通, ‘Basic Principles
for Writing Clearly and Coherently by Mister Ma’) published in 1898. Prior to that there were several
works on Chinese grammar written by Westerners.
References
Academy of American Poets. (2004). A brief guide to imagism. Retrieved December 28, 2018, from www.
poets.org/poetsorg/text/brief-guide-imagism
Aldington, R., H.D., Flint, F. S., Pound, E., Lowell, A., & Williams, W. C. (1914). Des imagistes: An anthology.
New York, NY: Albert and Charles Boni.
Alves, M. J. (2001). What’s so Chinese about Vietnamese? Papers from the Ninth Annual Meeting of the South-
east Asian Linguistics Society. Arizona State University, Program for Southeast Asian Studies, pp. 221–
242.
Alves, M. J. (2009). Loanwords in Vietnamese. In M. Haspelmath & U. Tadmor (Eds.), Loanwords in the
world’s languages: A comparative handbook. Berlin: Mouton de Gruyter.
Asia Art Archive. (2010). Conversation with Huang Yongping. Retrieved December 28, 2018, from www.
aaa-a.org/programs/conversation-with-huang-yongping/
Attané, I. (2012). Being a woman in China today: A demography of gender. China Perspectives, 2012, 5–15.
Barrass, G. S. (2002). The art of calligraphy in modern China. Berkeley, CA: University of California Press.
Bergman, J. (2010). China’s TV dating shows: For love or money? Time, June 30, 2010.
Berwick, R., & Chomsky, N. (2016). Why only us: Language and evolution. Cambridge, MA: Massachusetts
Institute of Technology Press.
Bodde, D. (1991). Chinese thought, society, and science: The intellectual and social background of science and
technology in pre-modern China. Honolulu: University of Hawaii Press.
Boltz, W. G. (1993). “Shuo wen chieh tzu 說文解字”. In M. Loewe (Ed.), Early Chinese texts: A bibliographical
guide, Early China Special Monograph Series, 2 (pp. 429–442). Berkeley, CA: Society for the Study of
Early China, and the Institute of East Asian Studies, University of California.
Boltz, W. G. (1996). Early Chinese writing. In P. T. Daniels & W. Bright (Eds.), The world’s writing systems
(pp. 191–199). New York, NY: Oxford University Press.
Bouchot, J. (1925). Pétrus Truong-Vinh-Ky, érudit cochinchinois, 1837–1898. Saigon: Bouchot.
Brannigan, M. C. (2015). Japan’s March 2011 disaster and moral grit: Our inescapable in-between. Lanham,
MD: Lexington Books.
Brown, L., & Yeon, J. (2015). The handbook of Korean linguistics. Chichester: John Wiley & Sons.
Campbell, L., & Mixco, M. J. (2007). A glossary of historical linguistics. Salt Lake City, UT: University of Utah
Press.
Camus, Y. (2007). Jesuits’ journeys in Chinese studies. Paper presented at the World Conference on Sinology
2007, Renmin University of China, Beijing, March 2007.
Chen, P. (1999). Modern Chinese: History and sociolinguistics. Cambridge: Cambridge University Press.
Chiang, W. (1995). We two know the script; we have become good friends: Linguistic and social aspects of the
women’s script literacy in southern Hunan, China. New York, NY: University Press of America.
Chiang, Y. (1973). Chinese calligraphy: An introduction to its aesthetic and technique (3rd revised and enlarged
edition). Cambridge, MA: Harvard University Press.
Chinese Academy of Social Sciences. (2007). Contemporary Chinese dictionary [Xiàndài Hànyǔ cídiǎn] (5th
edition). Beijing: The Commercial Press.
Chung, Y. (2008). Use hangeul, not hangul. The Korea Times, January 2, 2008.
DeFrancis, J. (1950). Nationalism and language reform in China. Princeton, NJ: Princeton University Press.
DeFrancis, J. (1977). Colonialism and language policy in Viet Nam. The Hague: Mouton.
DeFrancis, J. (1984). The Chinese language: Fact and fantasy. Honolulu: University of Hawaii Press.
DeFrancis, J. (1989). Visible speech: The diverse oneness of writing systems. Honolulu: University of Hawaii
Press.
Eliot, T. S. (1928). Introduction. In Selected poems by Ezra Pound. London: Faber & Gwyer.
Endo, E. (1999). Endangered system of women’s writing from Hunan, China. Presentation given at the Associ-
ation of Asian Studies Annual Conference, March 1999.
Fan, C. C. (1996). Language, gender, and Chinese culture. International Journal of Politics, Culture, and Soci-
ety, 10(1), 95–114. doi:10.1007/BF02765570.
234 References
Fei, D. (2007). 85 New Wave: The birth of Chinese contemporary art. Shanghai: Shanghai People’s Publishing
House.
Fenollosa, E., Pound, E., Saussy, H., Stalling, J., & Klein, L. (2011). The Chinese written character as a medium
for poetry: A critical edition. Ashland, OH; London: Fordham University Press.
Fincher, L. H. (2013). China’s entrenched gender gap. New York Times, May 20, 2013.
Fisher, S. R. (2001). A history of writing. London: Reaktion Books.
Foust, M. A. (Ed.). (2016). Feminist encounters with Confucius. Leiden; Boston, MA: E. J. Brill.
Frank, H. (2001). Identity and script variation: Japanese lesbian and housewife letters to the editor. In K.
Campbell-Kibler, R. J. Podesva, S. J. Roberts, & A. Wong (Eds.), Language and sexuality: Contesting
meaning in theory and practice. Stanford, CA: CSLI Publications.
Goldin, P. R. (2000). The view of women in early Confucianism. In C. Li (Ed.), The sage and the second sex:
Confucianism, ethics, and gender (pp. 133–161). Chicago, IL: Open Court.
Gottlieb, N. (1995). Kanji politics: Language policy and Japanese script. London; New York, NY: Kegal Paul
International.
Gu, G. (1987). The construction of modern calligraphy. [Xiàndài shūfǎ gòuchéng]. Beijing: Běij ī ng Tı̌yù
Xuéyuàn Chūbǎnshè.
Gu, G. (1990). The three steps of modern calligraphy. [Xiàndài shūfǎ sān bù]. Beijing: Zhōngguó Shūjí Chūbǎn-
shè.
Gulik, V. (1974). Sexual life in ancient China. Leiden: E. J. Brill.
Hannas, W. C. (1997). Asia’s orthographic dilemma. Honolulu: University of Hawaii Press.
Hannas, W. C. (2003). The writing on the wall: How Asian orthography curbs creativity. Philadelphia, PA: Uni-
versity of Pennsylvania Press.
He, Z., & Jiao, H. (2014). Shàng shìjì wǔshí niándài Táiwān Zìyóu Zhōngguó zázhì duì hànzì jiǎnhuà de hūyù.
[Taiwan’s Free China periodical advocates for character simplification in the 1950s]. Fujian Forum
(Humanity and Social Sciences Edition), 2, 115–119.
Hoosain, R. (1991). Psycholinguistic implications for linguistic relativity: A case study of Chinese. Hillsdale, NJ:
Lawrence Erlbaum.
Huang, H. (1996). Portuguese settlement in Macao and cultural exchange between China and the outside
world. Review of Culture, Macau, S., 2(29), 67–84.
Hue, C., & Hu, Z. (2003). Number of characters a college student knows. Journal of Chinese Linguistics, 31(2),
300–339.
Idema, W. L. (2009). Heroines of Jiangyong: Chinese narrative ballads in women’s script. Seattle, WA: University
of Washington Press.
Idema, W. L., & Grant, B. (2004). Reality. In The red brush: Writing women of imperial China (pp. 542–566).
Cambridge, MA: Harvard University Asia Center.
Iwahara, A., Hatta, T., & Maehara, A. (2003). The effects of a sense of compatibility between type of script and
word in written Japanese. Reading and Writing: An Interdisciplinary Journal, 16, 377–397.
Jeon, J. (2015). Some pushing for more elementary school instruction of Chinese characters. Hankyoreh.
Retrieved September 16, 2018, from http://english.hani.co.kr/arti/english_edition/e_national/704331.
html
Joshi, R. M., & Aaron, P. G. (2006). Handbook of orthography and literacy. London: Taylor & Francis.
Kaiho, H., & Nomura, Y. (1983). Kanjijōho no shori no shinrigaku. [Psychology of kanji information processing].
Tokyo: Kyōiku.
Kim, B. (2017). Hanja in elementary textbooks is still controversial. Korea National University of Education
Bulletin, Vol. 77, May 31, 2017.
Kim, S.-J. (2015). Debate grows over teaching Chinese characters. Korea Times, September 14, 2015.
Kindaichi, H. (2017). The Japanese language. North Clarendon, VT: Tuttle Publishing.
King, R. (1996). Korean writing. In P. T. Daniels & W. Bright (Eds.), The world’s writing systems (pp. 218–227).
New York, NY: Oxford University Press.
Koichi. (2011). The ultimate Kanji test: Kanji Kentei. Tofugu. Retrieved December 10, 2018, from www.
tofugu.com/japanese/kanji-kentei/
Kornicki, P. F. (2018). Languages, scripts, and Chinese texts in East Asia. Oxford: Oxford University Press.
Kraus, R. C. (1991). Brushes with power: Modern politics and the Chinese art of calligraphy. Berkeley, CA:
University of California Press.
Kubler, C. C. (2011). Basic written Chinese: An introduction to reading and writing for beginners. North Clar-
endon, VT: Tuttle Publishing.
Le, M., & O’Harrow, S. (2007). Vietnam. In A. Simpson (Ed.), Language and national identity in Asia. Oxford;
New York, NY: Oxford University Press.
Lee, I., & Ramsey, S. R. (2000). The Korean language. New York, NY: State University of New York Press.
References 235
Lee, K., & Shim, J. (2015). Use of Chinese characters in primary school textbooks. Korea Focus. Retrieved
September 16, 2018, from www.koreafocus.or.kr/design2/layout/content_print.asp?group_id=105802
Lee, L. (2002). Creating a female language: Symbolic transformation embedded in nushu. In X. Lu, W. Jia, &
D. R. Heisey (Eds.), Chinese communication studies: Contexts and comparisons (pp. 101–118). Westport,
CT: Greenwood Publishing Group.
Lee, T.H.C. (2000). Education in traditional China: A history. Leiden: E. J. Brill.
Li, C. (Ed.). (2000). The sage and the second sex: Confucianism, ethics, and gender. Chicago, IL: Open Court.
Lin, J. Y. (1995). The Needham Puzzle: Why the industrial revolution did not originate in China. Economic
Development and Cultural Change, 43(2), 269–292.
Lin, Y. (1977). My country and my people. Heinenmann: Asia.
Link, P. (2006). Whose assumptions does Xu Bing upset, and why? In J. Silbergeld & D.C.Y. Ching (Eds.),
Persistence-transformation: Text as image in the art of Xu Bing. Princeton, NJ: Princeton University Press.
Liu, C. (2010). Zhōngguó xiàndài shūfǎ shǏ. [The history of modern Chinese calligraphy]. Nanjing: Nanjing
University Press.
Liu, M. (1957). Zhōngguó y ī nbiāo zì shū. [Chinese phonetic script]. Beijing: Script Reform Publishing [Wénzì
Gǎigé Chūbǎnshè].
Liu, Y. (1991). Difficulties in Chinese information processing and ways to their solution. In V. H. Mair & Y. Liu
(Eds.), Characters and computers (pp. 1–8). Amsterdam; Washington, DC: IOS Press.
Logan, R. K. (1986). The alphabet effect: The impact of the phonetic alphabet on the development of western
civilization. New York, NY: Morrow.
Mair, V. H. (1991). Preface: Building the future of information processing in East Asia demands facing lin-
guistic and technological reality. In V. H. Mair & Y. Liu (Eds.), Characters and computers (pp. 1–8).
Amsterdam; Washington, DC: IOS Press.
Mair, V. H. (1996). Modern Chinese writing. In P. T. Daniels & W. Bright (Eds.), The world’s writing systems
(pp. 200–208). New York, NY: Oxford University Press.
Mair, V. H. (2000). Advocates of script reform. In W. T. De Bary, I. Bloom, W. Chan, J. Adler, & R. J. Lufrano
(Eds.), Sources of Chinese tradition (Vol. 2, pp. 302–308). New York, NY: Columbia University Press.
Mair, V. H. (2002). Sound and meaning in the history of characters: Views of China’s earliest script reformers.
In M. S. Erbaugh (Ed.), Difficult characters: Interdisciplinary studies of Chinese and Japanese writing
(pp. 105–123). Columbus, OH: National East Asian Languages Resource Center, Ohio State University.
Man, B. (1990). Backhill/Peking/Beijing. Sino-Platonic Papers, p. 19.
Needham, J. (1956). Science and civilization in China (Vol. 2). Cambridge: Cambridge University Press.
Needham, J. (2013). Grand titration: Science and society in East and West. London: Routledge.
Nguyen, N. B. (1984). The state of Chũ’ Nôm studies: The demotic script of Vietnam. Fairfax, VA: Indochina
Institute, George Mason University.
Ni, H. (1948). Zhōngguó p ī ny ī n wénzì yùndòng shǏ. [History of Chinese phonetic script movement]. Shanghai:
Shídài Shūbào Chūbǎnshè [Times Publishing].
Ni, H. (1987). Lād ī nghuà X ī n Wénzì yùndòng shǏmò hé biānnián jìshì. [Chronical of Latin Xua Sinwenz move-
ment]. Shanghai: Zh ī shí Chūbǎnshè [Knowledge Press].
Noguchi, M. S. (2009). The crafty names of Japan’s cleverest companies. The Japan Times Online, June 17, 2009.
O’Mochain, R. (2012). Listserv communication by LGBT and heterosexual Japanese youth: A survey of three
features. US-China Foreign Language, 10(11), 1734–1744.
Otis, E. (2015). Inequality in China and the impact on women’s rights. The Conversation, March 19, 2015.
Peng, H. (2010). Xu Teli and the new characters movement of Latinization. Journal of Changsha Normal
College, 2010(2), 14–18.
Penney, M. (2012). The Fukushima anniversary: Japanese press reactions. The Asia-Pacific Journal. Retrieved
from http://apjjf.org/-Matthew-Penney/4710/article.html
Qian, X. (1999). Qian Xuantong anthology (Vol. 3). Beijing: Renmin University Press.
Qiu, Z. (1995). A one-thousand-time copy of Lantingxu. Retrieved July 26, 2017, from www.qiuzhijie.com/
worksleibie/calligraphy/e-lanting.htm
Renmin chuban she. (1956). Zhōngguó wénzì gǎigé de dìy ī bù. [First step in China’s script reform]. Hong Kong:
Zhōngguó Yǔwén Xuéshè.
Rogers, H. (2005). Writing systems: A linguistic approach. Malden, MO: Blackwell.
Sampson, G. (1985). Writing systems. Stanford, CA: Stanford University Press.
Seaton, P. (2016). Japanese war memories and commemoration after the Great East Japan Earthquake. In A. L.
Tota & T. Hagen (Eds.), Routledge international handbook of memory studies (pp. 345–356). London;
New York, NY: Routledge.
See, L. (2014). Snow flower and the secret fan. London: Bloomsbury Publishing.
Seeley, C. (1991). A history of writing in Japan. Leiden; New York, NY: E. J. Brill.
236 References
Seybolt, P. J., & Chiang, G. K.-K. (1979). Language reform in China: Documents and commentary. White
Plains, NY: M. E. Sharpe.
Shibatani, M. (1990). The languages of Japan. Cambridge [England]; New York, NY: Cambridge University
Press.
Smith, J. S. (1996). Japanese writing. In P. T. Daniels & W. Bright (Eds.), The world’s writing systems (pp. 209–
217). New York, NY: Oxford University Press.
Smith, J. S., & Schmidt, D. L. (1996). Variability in writing Japanese: Towards a sociolinguistics of script
choice. Visible Language, 30(1), 47–71.
Snow, D. (2004). Cantonese as written language: The growth of a written Chinese vernacular. Hong Kong: Hong
Kong University Press.
Sohn, H. (2001). The Korean language. Cambridge [England]; New York, NY: Cambridge University Press.
Sohn, H. (2011). Korean language in culture and society. Honolulu: University of Hawaii Press.
Song, J. J. (2005). The Korean language structure, use and context. London: Routledge.
State Council, The (2013). Guówùyuàn guānyú gōngbù “Tōngyòng gu ī fàn hànzì biǎo” de tōngzh ī . [The
State Council’s notice about the publication of the “List of standardized characters for general use”].
Retrieved from http://www.gov.cn/zwgk/2013-08/19/content_2469793.htm
Su, J. (1999). Sexism in Chinese characters. [Hànzì zhōng de xìngbié qíshì]. Journal of Chinese, 1999(4), 38–41.
Sugimoto, T., & Levin, J. A. (2000). Global literacies and the World-Wide Web. London: Routledge.
Sullivan, M. (1999). The three perfections: Chinese painting, poetry, and calligraphy. New York, NY: G. Braziller.
Tahara, S. (2012). ‘Fukushima’ o ‘Fukushima’ to yobu no wa, mō yameyou! [Let’s stop calling “Fukushima (福
島)” “Fukushima (フクシマ)!] Retrieved from www.taharasoichiro.com/cms/?p=789
Tang, D. (2013). 5,000-year-old primitive writing generates debate in China. NBC News, July 11, 2013.
Taylor, I., & Taylor, M. M. (1995, 2014). Writing and literacy in Chinese, Korean and Japanese. Amsterdam;
Philadelphia, PA: John Benjamins.
Taylor, K. W. (1983). The birth of Vietnam. Berkeley, CA: University of California Press.
Taylor, K. W. (2013). A history of the Vietnamese. Cambridge: Cambridge University Press.
Tranter, N. (2008). Nonconventional script choice in Japan. International Journal of the Sociology of Language,
2008 (192), 133–151. Retrieved from https://doi.org/10.1515/IJSL.2008.040
Tsu, J., & Ellman, B. A. (Eds.). (2014). Science and technology in modern China: 1880s-1940s. Leiden: E. J. Brill.
Tuminez, A. S. (2012). Rising to the top? A report on women’s leadership in Asia. Lee Kuan Yew School of Public
Policy (Singapore) and Asia Society (New York City).
Unger, J. M. (2001). Functional digraphia in Japan as revealed in consumer product preferences. International
Journal of the Sociology of Language, 150, 141–152.
Unger, J. M. (2004). Ideogram: Chinese characters and the myth of disembodied meaning. Honolulu: University
of Hawai’i Press.
Vinograd, R. (2011). Making natural languages in contemporary Chinese art. In H. Tsao & R. T. Ames (Eds.),
Xu Bing and contemporary Chinese art: Cultural and philosophical reflections (pp. 95–115). Albany, NY:
State University of New York Press.
Vovin, A. (2014). Out of Southern China? Philological and linguistic musings on the possible Urheimat of
Proto-Japonic. Journées de CRLAO, June 2014.
Walle, W. V. (2013). The encounter between Europe and Asia in pre-colonial times. In J. Cauquelin, P. Lim, &
B. Mayer-König (Eds.), Asian values: Encounter with diversity (pp. 164–200). New York: Routledge.
Wang, H.-C., Hsu, L.-C., Tien, Y.-M., & Pomplun, M. (2014). Predicting raters’ transparency judgments of
English and Chinese morphological constituents using latent semantic analysis. Behavior Research
Methods, 46(1), 284–306.
Wang, J. (2009). Dāngdài Zhōngguó de wénzì gǎigé. [Writing reform in contemporary China]. Beijing: Dāngdài
Zhōngguó Chūbǎnshè [Contemporary China Publishing House].
Wang, L. (2002). Hànyǔ p ī ny ī n yùndòng de huígù jiānjí tōngyòng p ī ny ī n wèntí. [On Chinese pinyin’s history
and Romanization issues]. Zhongguo Yuwen, 2, 165–173.
Wang, X., & Liu, Y. (2016). Hànzì wénhuà yǔ shūfǎ yìshù. [Chinese character culture and the art of calligraphy].
Beijing: Zhōngguó Wénlián Chūbǎnshè [China Federation of Literary and Art Circles Publishing].
Wilkinson, E. P. (2015). Chinese history: A new manual (4th edition). Cambridge, MA: Harvard University
Asia Center.
Xiang, Y., Wu, J., & Yang, J. (Eds.). (2016). Shūfǎ sānshí liù kè. [Thirty-six lessons on calligraphy]. Guilin:
Líjiāng Chūbǎnshè [Líjiāng Publishing House].
Xing, J. Z. (2006). Teaching and learning Chinese as a foreign language: A pedagogical grammar. Hong Kong:
Hong Kong University Press.
Xiong, V. C. (2006). Emperor Yang of the Sui dynasty: His life, times, and legacy. Albany, NY: State University
of New York Press.
References 237
Yamada, H. (1991). Is logography a better writing system? In V. H. Mair & Y. Liu (Eds.), Characters and com-
puters (pp. 141–145). Amsterdam; Washington, DC: IOS Press.
Yang, F. M. (2001). Luó Míngjiān hé Lì Mǎdòu de Pú-Hàn cídiǎn (lìshιˇ yǔyánxué dǎolùn). [Michele Ruggieri
and Matteo Ricci’s Portuguese-Chinese Dictionary (A historical-linguistic introduction)]. In J. W. Witek
(Ed.), Dicionário Português-Chinês [Portuguese-Chinese Dictionary]. Biblioteca Nacional Portugal. San
Francisco, CA: Ricci Institute for Chinese-Western Cultural History (University of San Francisco).
Yang, J. (2015). Jiǎng Jièshí wèihé fàngqì “jiǎntιˇ zì”? [Why did Chiang Kai-shek give up “simplified charac-
ters”?]. Téngxùn Wǎng Duǎn Jì [Tencent Brief History], March 9, 2015. Retrieved September 25, 2017,
from http://view.news.qq.com/original/legacyintouch/d313.html
Yen, Y. (2005). Calligraphy and power in contemporary Chinese society. London ; New York, NY: Routledge-
Curzon.
Yin, B. (1990). Chinese romanization: Pronunciation & orthography = Hanyu pinyin he zhengcifa. (M. Felley,
Ed. & Trans.). Beijing: Sinolingua.
Yin, B. (1991). Pinyin-to-Chinese character computer conversion systems and the realization of digraphia in
China. In V. H. Mair & Y. Liu (Eds.), Characters and computers (pp. 26–36). Amsterdam; Washington,
DC: IOS Press.
Yuan, L. (2005). Reconceiving women’s equality in China: A critical examination of models of sex equality. Lan-
ham, MD: Lexington Books.
Zhang, S. (2016). Chinese characters are futuristic and the alphabet is old news. The Atlantic, November 1,
2016.
Zhang, S., Wang, T., Li, Q., & An, N. (Eds.). (1997). Jiǎnhuà zì shù yuan. [The origins of simplified characters].
Beijing: Yǔwén Chūbǎnshè.
Zhang, T. (2005). Literacy education in China. Paper commissioned for the EFA Global Monitoring Report
2006, Literacy for Life, UNESCO.
Zhao, S., & Baldauf, R. B. (2008). Planning Chinese characters: Reaction, evolution or revolution? Dordrecht:
Springer.
Zhong, X. P. (2009). Four interpretations for the slogan “Women hold up half the sky.” Nankai Journal,
2009(4). doi:10.3969/j.issn.1001–4667.2009.04.007
Zhou, Y. (1961). Hànzì găigé gàilùn. [Script reform overview]. Beijing: Wénzì Gǎigé Chūbǎnshè [Script
Reform Publisher].
Zhou, Y. (1991). Intrinsic features of Chinese language as applied in word processing on computers. In V. H.
Mair & Y. Liu (Eds.), Characters and computers (pp. 20–25). Amsterdam; Washington, DC: IOS Press.
Zhou, Y. (2003). Zhōngguó yǔwén de shídài yǎnjìn. [The historical evolution of Chinese languages and scripts].
Columbus, OH: National East Asian Languages Resource Center, Ohio State University.
Zhu, W. (1956). Wǒ hé “Jiāngsū x īn zìmǔ” [“Jiāngsū x ī n zìmǔ” and I]. Yǔwén Jiànshè, 4, 3–5.
Zhuang, G. (2016). Shūfǎ jiànshǎng fāngfǎ lùn [On approaches to calligraphy appreciation]. China Fine Arts
Education, Visual Art, 8 [造型艺术卷(8)], 446–449.
Index
Note: Page numbers in italics refer to figures; page Chinese characters 75n19; in Asia 64–75; in
numbers in bold refer to tables. avant-garde art 219–225; Cantonese-specific
61; changes in appearance of 190; computer
agglutinative languages 73–74 processing of 145; coverage rate 145; debate
allographs 2, 4, 5 over reform 140; Fenellosa’s essay on 227–231,
allomorphs 2, 4–5 232n15; formation of 46–50; liberated 138;
allophones 2–5, 39 number of in major dictionaries 144; ordering
alphabetical order 149–150 of 51–53; representation of speech sounds by
Asia: borrowing of Chinese script in 64–75; map 65 232n9; sexism in 166–167; as visual shorthand
aspiration 4, 20, 113, 120, 122, 125n6 of motion 228–229; and Western modernist
avant-garde art 219–225 poetry 226–231; see also character simplification;
Chinese script; Chinese writing system; folk
báihuà (plain speech) 41–42, 44n3 characters
biángbiáng noodles 35 Chinese language and dialects 30–32, 73–74;
Book from the Sky (Xú Bīng) 220–222, 221 dialect groups 31, 32; map of dialect groups 32;
Bopomofo (Zhùyīn Fúhào) 122–123; compared to speech-and-writing divide 116; standard style
pīnyīn 123; origin of symbols 123 41; words in English 120; written 40–41; see also
borrowing: from Chinese writing system 68–69; dialects
general strategies 69–75, 69; sound-only 69–70; Chinese script: bronze script 51, 137, 139, 139, 166,
variations and parallels 72–75 189, 192, 193, 194; calligraphic evolution of 189–
bronze script 51, 137, 139, 139, 166, 189, 192, 193, 190, 199; clerical script 51, 139, 195–197, 196;
194 cursive script 51, 139, 143, 198, 198, 199n11;
brush talks 66 demythifying 62; great seal 51, 194–195, 194,
Buddhism 92, 103, 220 199; nature of 45–46; oracle bones 55n2, 139,
Buddhist sutras 65 166, 190–191, 191, 192; phonographic 12–13,
140; running script 51, 139, 197–198; seal 51, 82;
Cài Jīng 159 small seal 195, 195; standard script 197; women’s
Cài Xiāng 159 script 169–177; see also Chinese characters;
Cài Xīyǒng 116, 118 Chinese writing system; script simplification
Cài Yuánpéi 136 Chinese Writing Reform Association 140
calligraphy 160; ‘85 New Wave 212–213; avant- Chinese writing system: borrowing from 68–75,
garde 213; basic strokes 204–208, 204–208; 69; changes in script style 50–51, 51; character
center-tip technique 203–204, 204; Chinese inventory 50; compared to Sino-Korean and
188–199; Classical 213; composition 209, 210; Sino-Japanese 69; different types of characters
huí and mĭ gridlines 201, 201; modern 212–217; 49; hànzì 128, 152, 171, 172; impact on science
Modernist 213, 214–215, 215, 216–217; and and technology 151–152; layout conventions 54;
moral preferences 161–162; Neo-Classical 213; as logographic 13–14, 45, 46; as morphographic
and personhood 161; stroke order 208–209; 13–14, 45; as morphosyllabic 13, 45; ordering
water 188, 199n3; writing exercise characters 51–53; section headers 52; semantic
210–211 transparency of 46; sexism in 163–167; structure
calligraphy workshop: gesture and posture of 51–55; see also Chinese characters; Chinese
201–203; holding the brush 203; setting up the script
writing space 202; tools and materials 200–201; chũ’ nho 101–102, 105
warm-up 203–204 chũ’ nôm 102–105, 155n18
Cantonese 30–31; compared to Mandarin 59–60; Chuányīn Kuàizì 118
shorthand system for 118 citation forms 24
center tip technique 203–204, 204 civil service examination 41, 51, 66, 75n9, 102,
Chao Yuen Ren 124 158, 197
character index 150 Classical Chinese 40–42, 67, 116; in Asia 65–68;
character simplification 136, 138–139, 142 impact on Modern Standard Chinese 42–43
Index 239
phonemic writing 12, 14–16, 26n1, 61, 77, 83, 95, Ruggieri, Michele 112–114
123, 127; see also pīnyīn running script 51, 139, 197–198
phonetic extension 48
phonetic kun 71, 91–92 sandhi 25
phonetic on 70, 91 script 8; running 51, 139, 197–198; simplified
phonetic scripts: Bopomofo 122–124; GR 124; 59–60, 136, 137–138, 141, 146, 152, 214;
Gwoyeu Romatzyh 122–123; later proposals traditional 59, 59–60, 138–139, 141–142, 146,
of 115–119; Latinxua Sin Wenz 124–125; 171; see also Chinese script; Japanese script;
motivations for 115–118; Postal Spelling System script simplification; writing systems
121–122, 121; progressive views on 119; Wade- script simplification: early attempts 136–139;
Giles system 119–120, 120–121; Yale system 122 General Table of Simplified Characters 141;
phonetic writing, before pīnyīn 112–125 historical roots of simplified characters 142–143;
phoneticization 118–119 international impact 146–147; by the PRC 139–
phonetics 2, 3, 27n3 146; set reduction 143–146; in Taiwan 138–139
phonographic scripts 12–13, 14, 14, 43, 57, 115, seal scripts 51, 82; see also great seal script; small
118–119, 122, 126n27, 127–129, 140, 149, seal script
155n18 seal stamping 224
phonology 2–3 section headers 52
pictographic scripts 10n3, 57, 214, 228, 229 semantic extension 48
pictographs/pictograms 10n3, 46–47, 47, 57, 58, semantic kun 70–71, 72, 79, 80, 91–92
166, 190 semantic on 70, 71, 72, 78, 79, 80, 91, 103
pictography 7, 10n3, 46–47, 49 semantic-phonetic compounds 47–49, 57, 79, 103,
pīnyīn 17–18, 112; as alternative script 128; 115, 229
compared to Bopomofo 123; compared to Wade- semantic-semantic compounds 48–49, 103, 229
Giles system 120–121; current uses of 130–131; sentences, as transitive processes 229–230
defined 127–128; GR in 124; history of 129–130; sexism 163–167
and ISO 26n1; on road signs 133; spelling rules shorthand 118
131–134; Tongyong 130 simple finals 18–19, 18
Pitman, Isaac 118 simplification see character simplification; script
Postal Spelling System 121–122, 121 simplification
Pound, Ezra 226–227 simplified script 59–60, 136, 137–138, 141, 146,
primacy of speech 9, 10n5 152, 214; see also script simplification
Printing on Water (Sòng Dōng) 223–224 Singapore 136, 146
punctuation 118, 126n28; in pīnyīn 133–134 Sinitic languages 31–32
pŭtōnghuà (common speech) 33, 60, 127, 129, 146 Sino-Japanese 66, 69, 69
Sino-Korean 66, 69, 69, 79
Qián Xuántóng 136–137 Sino-Vietnamese 66–67
Qiēyīn Zìmŭ 118, 119 small seal script 51, 195, 195
Qín Huì 160 Sòng Dōng 223–224
Qiū Zhènzhōng 215, 216 Sòng Shù 118
Qiū Zhìjié 224–225 speech 7, 9–10; Chinese language and dialects
quốc ngũ’ 105–108 30–32; mutually intelligible 30; primacy of 9,
QWERTY keyboard 152–153 10n5
spelling rules 21–23; marking syllable boundaries
radical index 150 23; in pīnyīn 131–134; representation of er as
radicals 52, 141, 150, 166 suffix 23; for zero-initial syllables 22
rebus principle 48 Square Calligraphy Classroom (Xú Bīng) 222–223
register 42, 67, 131, 144 standard script 50–51, 197
retroflex 21, 27n5 Sūn Guòtíng 160
rhyme 34 sworn sisters 175–177, 178n26
Ricci, Matteo 113–114, 114 syllabaries (syllabic writing systems) 12, 95, 173
rōmaji 87, 94–96, 100n3, 179–180, 182–184 syllables 12, 15, 33–35; boundaries of 23; English
Romanization: based on Italian orthography 34; in hangeul 84, 84; Japanese 96; in Korean
113; based on Portuguese orthography 113; 84–85; Modern Standard Chinese (MSC) 34;
of Chinese 77–78, 112–115, 119, 124–125, nucleus of 33; single-syllable poem 37
129–130, 147; of Japanese 87; of Vietnamese
107–108; Yale system 122; see\ also Wade-Giles Taiwanese 75n25, 146
Romanization System third-day book 176–177, 178n31
Romanized writing 12; see also Romanization tone sandhi 25
242 Index
tones 23–26, 36, 37, 156n22; contour 24–25, 36, of 174–175; before marriage 174–175; in pre-
37, 39n6; marking with diacritics 24; neutral 25, modern China 163–164
25–26, 39n5 Women’s Script 169–177; see also nǚshū
Tongyong Pinyin 130 word boundaries 54–55, 132
Tonkin Free School 108 word order 73
topolects 39n2 word segmentation 23, 27n8
traditional script 59, 59–60, 138–139, 141–142, writing 6, 8; defined 163; as secondary to speech
146, 171 9–10
Trigault, Nicolas 113, 114 writing systems 8, 10; classification of 12–15, 14,
triglossia 43n1 14; consonantal 12, 15; linguistic universal of
15–16; logographic 13–14, 45, 46; moraic 13, 15,
Unicode 152 93, 97, 146, 179; morphemic 15, 84, 179, 231;
universality myth 59–62 morphographic 13–14, 14, 15, 43, 45, 49, 57, 61,
84, 179, 189; morphosyllabic 13, 14, 16, 45, 154,
Vietnam, Chinese writing system in 65–66 173; phonemic 12, 14–16, 26n1, 61, 77, 83, 95,
Vietnamese 64–65, 72–75, 76n29, 155n18; colonial 123, 127; phonographic 12–13, 14, 14, 43, 57,
policy 107; Confucian script (chũ’ nho) 101–102, 115, 118–119, 122, 126n27, 127–129, 140, 149,
105; literary writing 104; national script (quốc 155n18; pictographic 10n3, 57, 214, 228, 229;
ngũ’) 105–108; Romanized 107–108; southern syllabic 12; as systematic 8–9
script (chũ’ nôm) 102–105, 155n18; writing 101, written language 10, 40–41; see also
104, 107 script
vocabularies, common 64–65 Wú language 31
voiced elements 3, 27n3 Wŭbĭ method 153
voiceless elements 4, 19, 27n3 Wùxū Biànfă 118
vowels: basic 83–84, 86n19; beginning a syllable
22; compound 24, 83, 86n19; conjoining 23; Xú Bīng: Book from the Sky 220–222; Square
double 124; final 82; in the final 21; geminate 96; Calligraphy Classroom 222–223
hangeul 83; in Hebrew 7, 12; in Japanese 94–96;
long 87, 96; main 17; marked with diacritics 24; Yale Romanization 122
medial 82; as nucleus of syllable 33; secondary Yán Zhēnqĭng 138, 159
17; voiced 27n3; see also diphthongs Yí language 8, 10n4; symbols representing syllables
13
Wade, Thomas Francis 119–120 Yuè Fēi 160
Wade-Giles Romanization System 119–120, 121,
129–130; compared to pīnyīn 120–121 Zen Buddhism 92, 220
Wáng Dōnglíng 216–217, 216–217, 218n4 Zhāng Huáiguàn 158
Wáng Zhào 117, 118 Zhèng Qiáo 115
water calligraphy 188, 199n3 Zhōngguó Yīnbiāo Zìshū 119
wényán (Classical Chinese) 40, 41, 42, 116 Zhōu Yŏuguāng 128, 129
women 165–166; during and after marriage Zhōu Zuòrén 136
175–177; geo-economic reasons for low status Zhū Wénxióng 119