CANTONESE

as Written Language
The Growth of a Written Chinese Vernacular

Don Snow

• *
~ ~ tl:: ltB. if.±.
HONG KONG UNIVERSITY PRESS
Hong Kong University Press
14/F Hing Wai Centre
7 Tin Wan Praya Road
Aberdeen
Hong Kong

© Hong Kong University Press 2004

ISBN 962 209 709 X

All rights reserved. No portion of this publication may be reproduced

or transmitted in any form or by any means, electronic or mechanical,
including photocopy, recording, or any information storage or retrieval
system, without permission in writing from the publisher.

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available
from the British Library.

Secure On-line Ordering

http://www.hkupress.org

Printed and bound by Condor Production Co. Ltd., in Hong Kong, China
Contents

List of Illustrations and Tables VII

Preface IX

1. Why Study the Development of Written Cantonese? 1

2. From Spoken Vernacular to Written Language 15

3. Spoken and Written Cantonese 45
4. Written Cantonese in Pre-modern Guangdong 67

5. The Hong Kong Dialect Literature Movement 101

6. Written Cantonese in Modern Hong Kong 125

7. Why Has Use of Written Cantonese Increased? 175
8. Epilogue: The Future of Written Cantonese 213
Appendix 1: Cantonese Texts 219
Appendix 2: Interviews and Public Lectures 245
Appendix 3: Titles of Publications and Published Works 249
Appendix 4: Characters for Chinese Terms 253
Notes 259
References 285
Index 307
3
Spoken and Written Cantonese

Introduction

As we have seen in Chapter 2, the development of a spoken L vernacular into

the realm of written language is partially an issue of growth in the language's
range of social functions. It is also a growth of the purposes to which the
language is put to use by the community. However, there are more specifically
linguistic aspects of the story which also deserve attention.
One important line along which the development of a written vernacular
needs to be traced is an increase in vernacular literacy - in other words, in
the number of people who can read and write in the vernacular. To a large
extent, this is affected by how easy or difficult it is to learn the written form of
the vernacular. Obviously, a written vernacular that is relatively standardized
and based on easily understood principles is easier to learn than one that is
less standardized or based on obscurer mechanisms. This issue is of particular
importance in the Chinese context where students already need to invest an
unusually large amount of time and energy in order to become literate. As
Chinese is written using a character-based system, it takes longer to learn to
read and write than do languages that use alphabetic (i.e. relatively phonetic)
writing systems. Given that the written forms of Chinese dialects are generally
not taught in Chinese schools, people need to learn to read and write dialects
on their own, and under these circumstances whether or not it is easy to learn
to read and write a dialect becomes a critically important factor in determining
the likelihood of its growth.
Another important line along which the growth of a written vernacular
needs to be traced is that of its vernacular "authenticity", i.e. how faithfully it
adheres to the norms of the spoken language. As we have already seen, the
written and spoken forms of any language generally differ to some degree, so
we should not expect written Cantonese to be entirely faithful to its spoken
counterpart. However, a developed written vernacular should be fairly
46 Cantonese as Written Language

consistent in following the norms of the spoken language it is based on, and
one of the most important measures of its growing maturity as a written
language is the increased consistency with which written Cantonese follows
the norms of the corresponding registers of spoken Cantonese - and diverges
from the norms of written Standard Chinese.
In order to set the background for these two story lines in the growth of
written Cantonese, in this chapter we shall examine the relationship of spoken
Cantonese to Mandarin, and also how one goes about writing in Cantonese.
We shall also consider the questions of how faithfully various kinds of written
Cantonese adhere to the norms of spoken Cantonese, and whether or not
written Cantonese should be considered a language variety distinct from
Standard Chinese.

The Relationship of Cantonese to Mandarin

Perhaps a good place to start is with the question of whether Cantonese should
be considered a language or a dialect. One problem with the term "dialect"
is that it may convey the false impression that Cantonese is more closely related
to Mandarin than it actually is, and this is one reason why some have argued
that Cantonese should more accurately be described as a distinct language
within the Chinese language family.! For example, in a controversial 1990
article, LiJingzhong explicitly rejects the term "dialect" (jangyan 1J10, arguing
instead that Cantonese should be considered an "independent language" (duli
de yuyan 1ILlz:a9~1§) within the broader Chinese language family. The m;:yor
reason given by those who hold this view is that while spoken Cantonese and
Mandarin descend from the same ancestor, hence have many similarities in
grammar and vocabulary, they also differ enough that they are mutually
unintelligible; in fact they are no closer to each other than are the various
members of the Romance language family.2
The major argument for considering Cantonese a dialect of Chinese rather
than a separate language is based more on sociolinguistic grounds than purely
linguistic ones. As Trudgill (2000: 4-5) points out, the primary factors in
determining what is a "language" and what is a "dialect" are often more
political and cultural than linguistic. Varieties that are autonomous and
dominant are classed as "languages," while those varieties that are
"heteronomous," i.e. that look to a standard or dominant language for their
norms, are classed as "dialects." As Deng (1980: 84-85) points out, the major
reasons Cantonese is considered a dialect of Chinese have to do with China's
tradition of political unity, the traditional unity of its written language, and
the subordinate position that Cantonese plays in this scheme of things. It is
in this sense that the term "dialect" is accurate in describing the role of
Cantonese within the Chinese language family. For this reason, in this book I
 Spoken and Written Cantonese 47

shall follow the practice of many scholars in referring to Cantonese as a dialect

of Chinese rather than as a separate language.
However, I also wish to emphasize that the degree of difference between
Cantonese and Mandarin is quite significant. Given that the development of
written Cantonese could be characterized as a gradual shift from the norms
of Standard Chinese to those of spoken Cantonese, it is important to examine
in more detail the differences that distinguish these two varieties of Chinese.

Grammar differences

Because Mandarin and Cantonese are different varieties within the same
language family, they share a great many similarities in syntax. In fact, grammar
is the level at which the two varieties are most similar. 3 However, we should
not overlook the fact that even at the syntax level there are differences between
the two varieties. 4 Some examples of such differences would include the
following:
(1) Word order in comparative structures: In Cantonese, the normal word
order for a comparative phrase would be: (Subject) + (adjective) +
(comparative maker) + (object). So, for example, the sentence "I am taller
than him" would be (C) Ngo gou gwo keoi ~i\!;~1e: (I + tall + more than +
him). In contrast, in Mandarin the normal order would be (Subject) +
(comparative marker) + (object) + (adjective), and the sentence above
would be (M) Wo hi ta gao ~ tt11B i\!; (I + compared to + him + tall).
(2) Positions of direct and indirect objects after the verb "to give": In
Cantonese the indirect object in constructions with "to give" would
normally precede the direct object. So, for example, the sentence "I gave
him money" would normally be (C) Ngo bei cin keoi ~f.!1j!.il1e: (I + give +
money + he). In contrast, in Mandarin the direct object comes first, so
the sentence would be (M) Wo gei ta qian ~Mrf1Bil (I + give + he + money).
(3) Deletion of relative marker: In some patterns Cantonese allows deletion
of the relative marker in constructions where it would be required in
Mandarin. For example, in a Cantonese phrase like (C) Keoi coeng (ge) go
zek go 1e:p~ (1lIJ%) IifmfHIx (He + sang + possessive + that + measure word +
song), the relative marker (C) ge is optional. In contrast, in the parallel
phrase in Mandarin, (M) Ta chang de na shou ge f1BP~ 1¥J~~§:m:, the marker
de cannot be deleted.

Despite the fact that grammatical differences between Cantonese and

Mandarin do exist, relatively little attention will be devoted to them in the
remainder of this study. The main reason is that through much of the history
of written Cantonese, distinctly Cantonese syntax is not very evident in texts
which have Cantonese elements. As we shall see, early texts are often
48 Cantonese as Written Language

"Cantonese" only in the sense that they contain occasional Cantonese words
inserted into a framework that is basically Classical Chinese or Standard
Chinese. As later texts become more consistent in their adherence to the
norms of spoken Cantonese, distinctly Cantonese syntax patterns do begin to
emerge. However, the appearance of such structures tends to follow along after
the appearance of distinctly Cantonese vocabulary, with vocabulary leading
the way. So separate examination of the development of grammar structures
does not contribute much to our tracing of the pattern by which written
Cantonese breaks new ground.
A second reason I devote relatively little attention to syntax is that when
Chinese people who do not speak Cantonese attempt to read texts written in
Cantonese, the primary difficulty they experience is with unfamiliar Cantonese
vocabulary rather than grammar. Similarly, when Cantonese speakers attempt
to learn Standard Chinese, the main difficulty they face is with vocabulary,
not syntax. 5 Given that vocabulary is the main challenge Chinese face in
mastering the difference between the two varieties, it seems appropriate that
this be our focus as well.

Pronunciation differences

The most important difference between the spoken forms of Cantonese and
Mandarin lies in pronunciation. Even though Cantonese and Mandarin have
many words in common, the pronunciation of such words is often very
different, and this is the major reason why the two languages are mutually
unintelligible in spoken form. 6 For some words, the pronunciation difference
is minimal, lying in a single sound. The Cantonese word for "good," (C) hou
(tff), for example, is quite close to Mandarin (M) hao both in pronunciation
and tonal value. However, there are other Cantonese words that do not sound
like their Mandarin counterparts at all. For example, the Cantonese words (C)
hai (*) (department) and (C) hokzaap (¥W) (to study) are substantially
different phonetically from the corresponding Mandarin (M) xi and (M) xuexi,
despite the fact that they have evolved from a common phonetic ancestor and
are written with the same characters. 7
While pronunciation is the main area of difference between the spoken
forms of Cantonese and Mandarin, pronunciation differences generally do not
affect the written form of the two varieties. When Cantonese and Mandarin
share a vocabulary item in common, it is virtually always written with the same
Chinese character, so the effects of pronunciation differences tend to
disappear in the written form of the languages.
 Spoken and Written Cantonese 49

Vocabulary differences

The main difference between the written forms of Cantonese and Mandarin
lies in their vocabulary systems. 8 This difference is particularly evident in the
most common daily words of Cantonese, many of which differ considerably
from their daily Mandarin translation counterparts. Consider the following
examples:

Table 3.1 Examples of vocabulary differences between Cantonese and Mandarin

English Cantonese Mandarin English Cantonese Mandarin

(possessive marker) ge pgt dei¥J that gollflM na fj~
(perfective aspect zo~ leT he/she keoi fE: ta ill!.
marker)
(pluralizer for
pronouns)
deill1!!! men fr~ now jigaa rm* xianzai :m:tE
no, not (negator) mPH bu /F to look taz IIi(l kan'ff
is (copula) haii* shi~ to give beif* gez,,'Il
.~

in, on, at (locative) hai IIf* zaz:tE to like zungji1!t;@': xihuan :glli:
this ni PJt zheEf to seek wanJJl! zhao t.14

Note that the list in Table 3.1 includes a number of very high frequency words,
including quite a few basic function words.
It is difficult to quantify precisely how different the vocabularies of
Cantonese and Mandarin are, and general estimates range from 30 to as high
as 50 percent.9 A more precise corpus-based estimate given by Ouyang (1993:
23) concludes that about one-third of the lexical items used in regular
Cantonese speech are not found in Mandarin. Ouyang, however, also points
out that the degree of lexical difference is considerably lower in more formal
registers. In an analysis of the transcription of a radio news broadcast, for
example, he concluded that only about 10.6% of the lexical items were
distinctly Cantonese (1993: 80-82).
As Ouyang's analysis suggests, the first step toward a more precise estimate
lies in making a distinction between formal and informal registers of Cantonese
because, due to the influence of written Chinese on formal Cantonese, formal
spoken Cantonese is considerably more similar to formal Mandarin than
colloquial Cantonese is to colloquial Mandarin. 1O Following Ouyang's lead, let
us compare the language of two formal Cantonese news broadcasts with that
of a colloquial Cantonese radio talk show.
Transcript #1: Examination of a Hong Kong Commercial Radio news
bulletin demonstrates two points. ll First, as Ouyang found, the occurrence of
distinctly Cantonese words in formal Cantonese speech is relatively low - only
12 percent of the total number of characters are markedly Cantonese.
50 Cantonese as Written Language

Secondly, the great majority of these occurrences are accounted for by

repeated use of just a few Cantonese function words. Of all the marked
Cantonese characters I2 in the transcription of the broadcast, 42 percent (63
out of 120) are accounted for by just three Cantonese characters: 63
occurrences of (C) ge PIlt(possessive marker), 31 of (C) hai {,~(to be), and 26
of (C) hai Pf*(in, at).
Transcript #2: A transcript of a Hong Kong TVB news feature story provides
similar results, though the language is slightly more conversational. I3 Only 15
percent of the characters in the transcript are marked Cantonese (164 characters
out of 1,111), and the same three Cantonese words - (C) ge, (C) hai, and (C)
hai - tend to dominate, accounting for 43 percent of the occurrences of marked
Cantonese characters.
The percentages of marked Cantonese words above correspond quite
closely with Ouyang's results, and suggest that it is safe to generalize that in
formal spoken Cantonese the percentage of marked Cantonese items is
relatively low, hovering in the 10 to 15 percent range. It also appears that
formal spoken Cantonese is "Cantonese" primarily in its pronunciation and
use of Cantonese function words. Such markedly Cantonese vocabulary as
occurred in news broadcasts consists largely of a few high frequency function
words; much of the remaining vocabulary is vocabulary which Cantonese shares
in common with Mandarin. We shall later see similar patterns in relatively
formal written Cantonese texts. I4
Transcript #3: Rather different results appear when we examine the
transcript of an informal Cantonese radio talk show. I5 First, a much higher
percentage of the vocabulary used is markedly Cantonese. Out of all the
Chinese characters in the transcript, 42 percent (612 out of 1,448) are
markedly Cantonese. This percentage corresponds well with the estimates
suggested by Ouyang for informal Cantonese speech.
Second, in this transcript, the three function words (C) hai, (C) hai, and
(C) ge account for much less of the marked Cantonese that appears - only 18
percent. Comparing Transcript #1 and Transcript #3, we see that a much wider
range of Cantonese words appears in the informal speech of Transcript #3:

Table 3.2 Range of Cantonese vocabulary used in formal versus informal speech

Total Characters in Transcript Nwnber of Different Marked

Cantonese Characters
Transcript #1 2,353 43
Transcript #3 1,448 105

In fact, in comparison with Transcript #1, Transcript #3 has twice as many

different marked Cantonese characters in a transcript that is only half as long.
Clearly, the less formal language of the talk show not only has a considerably
 Spoken and Written Cantonese 51

higher total amount of markedly Cantonese vocabulary, but it also shows use
of a much broader range of marked Cantonese vocabulary. We shall later see
a similar pattern in the written Cantonese of colloquial novels. 16
These percentages are important for two reasons. First, they establish a
rough baseline from which we can judge how "authentic" a written Cantonese
text is; i.e. how faithfully it adheres to the norms of spoken Cantonese. Written
language is generally not an exact reproduction of spoken language, even
when that written language is intended to approximate speech. For example,
many of the features of natural speech - incomplete sentences, errors,
repetitions, and so forth - are filtered out when language is written down,
so in a written Cantonese text, we might expect to see less particle use than is
found in natural speech. However, on the basis of the evidence above, it would
seem that in Cantonese texts which attempt to faithfully and consistently
replicate spoken Cantonese, we should expect that (1) in formal written
Cantonese marked Cantonese characters should account for approximately 10
to 15 percent of the total characters in the text; and (2) in informal written
Cantonese marked Cantonese characters should account for between 25 to 40
percent of the characters in the text.
These percentages are also significant because they underscore the extent
to which Cantonese is lexically different from Mandarin Chinese. If a written
Cantonese text includes as many marked Cantonese vocabulary items as
colloquial Cantonese speech does, the text becomes very difficult for someone
who knows only Standard Chinese to read. When 30 percent or more of the
vocabulary in a text is unfamiliar, there simply is not enough comprehensible
text remaining to make guessing reliable. When a person literate in Standard
Chinese cannot read material written in Cantonese (or finds it so difficult that
the effort will rapidly be abandoned), it clearly becomes necessary to consider
written Cantonese a distinct writte~ variety of ChineseP

Writing Cantonese

How Cantonese words are written

As we have seen above, there is a considerable degree of lexical overlap

between Cantonese and Standard Chinese. Thus, for many Cantonese words,
there is a pre-existing written form that is known by anyone who is literate in
Standard Chinese. The key challenge for a writer of Cantonese is finding
written forms for Cantonese words which do not have a pre-existing Standard
Chinese written form. As we have seen in the case of written Taiwanese, this
problem of "having the sound but no character" (you yin wu zi ifii1!!f=¥) can
be a serious obstacle for those who attempt to write or read a Chinese dialect.
The issue of how Cantonese words are represented in written form has
52 Cantonese as Written Language

been dealt in some detail elsewhere, most notably in Bauer (1988), Luke
(1995), and Cheung and Bauer (2002).18 My more limited goal here is to
present a general overview of the most common mechanisms used, and
consider the implications that choice of mechanism has for the growth and
development of written Cantonese.
It may help to begin with a hypothetical example. Consider the following
Cantonese dialogue, in which a disappointed suitor speaks to the object of
his ardor:
Boy: 1.!::!.~, ~IH$!U!i1$~f>iJ~-Jl'f~tii 0

Daanhai, camjat nei zung waa nei jiu tung ngo jat-cai sik faan.
But yesterday you still say you would with me together eat (rice).

Girl: ~nDl:t!lli 1$Jj;.~~DM~ !

0

Ngo mou gam gong. Nei zan hai jau di toujim!

I didn't so say. You really are a bit annoying!

A reader who understands Standard Chinese will see that quite a few of the
Chinese characters in the sentences above - fifteen to be exact - are familiar
in both their form and usage. However, eleven other characters, the ones in
bold, are "marked Cantonese" which clearly adhere to the norms of spoken
Cantonese rather than to the norms of Standard Chinese. Within these eleven
Cantonese characters, we see representative examples of the following major
categories of ways in which writers represent Cantonese words that fall outside
the range of Standard Chinese.
(1) Standard Chinese characters used in uniquely Cantonese ways: When
writing Cantonese, in some cases there is no problem finding the appropriate
Chinese character because it already exists in Standard Chinese. This is true
not only for the majority of Cantonese words which are the same as Mandarin
words in every respect except pronunciation, but also for a considerable
number of words that exist in both Cantonese and Mandarin but are used
quite differently in the two varieties. For example, in the first sentence of the
dialogue above, we find the word (C) waa (~I'i) as a verb meaning "to say."
This word is also common in Mandarin as a noun meaning "speech," but in
Mandarin it cannot function as a verb. Similarly, the words (C) tung (iPJ "with")
and (C) sik (~ "eat") also exist in Mandarin, but their usage is not the same
as in Cantonese.
Many words in this category preserve usages that existed in ancient
Chinese but have died out in modern Mandarin. Strictly speaking, then, these
are not cases of character formation, because characters for these Cantonese
words already exist. However, the way these characters are used in Cantonese
texts would often cause difficulty for a reader who only knows Mandarin.
(2) New (previously non-existing) Cantonese dialect characters: Sometimes
when confronting the "sound but no character" problem, Cantonese speakers
 Spoken and Written Cantonese 53

have resorted to the strategy of creating a new character to represent a

Cantonese word. The best-known example of this, the Cantonese character
::ff, appears in the second sentence of the dialogue above representing the
Cantonese word (C) mou (did not, not have). Historically, however, this
particular method of character formation has not been very productive, and
it is rarely used. 19 To the extent that "new" Cantonese characters are created,
they are virtually always combinations of pre-existing and familiar elements.
Even the character::ff above is based closely on its Standard Chinese antonym
(M) you ~ (to have, to be), the only difference being that two strokes in the
center are missing.
The reasons new characters are rarely created from scratch are partly
cultural. For most of China's history there has been great reluctance to create
entirely new characters, or at least give formal recognition to them. Even in
ancient times, new Chinese characters were often combinations of existing
elements, and in modem times new Standard Chinese words are created with
new combinations of existing characters rather than through the creation of
new characters. These character formation traditions have affected Cantonese-
speaking Chinese as much as they have Chinese people elsewhere. 20 Another
reason why entirely new characters are rarely created is that they present more
difficulties to a reader than do characters with recognizable elements. Not only
would a totally new character be unfamiliar to a reader, but it would also not
be obvious how the reader should go about decoding it.
(3) Romanization: Another way the "sound but no character" problem is
sometimes solved is through use of romanization. This approach has often
been taken by Western missionaries for the writing of Chinese dialects, and
as we have seen in the case of written Taiwanese, it has also sometimes been
used by Chinese people for writing in dialect.
Historically, however, this approach has rarely been used by writers of
Cantonese, and until recent decades Cantonese was almost always written in
Chinese characters. Even today many Cantonese texts do not use romanization
at all. This may be due in part to the fact that there are relatively few common
Cantonese words that cannot be represented adequately in characters.21 There
are, however, several words that are frequently written using the Roman
alphabet, the best-known example being the use of "D" (seen in the dialogue
above) to represent the Cantonese word (C) di (nag a bit), which often appears
in newspapers and books. There are other examples of alphabet letters being
used to represent Cantonese words, but none that has attained the wide
acceptance of "D."22
While romanization has traditionally not been a very popular solution for
dealing with the problem of how to write down Cantonese words, this strategy
has become more popular in recent years. Younger Cantonese writers,
especially pocketbook writers, often spell Cantonese words out in Roman
letters when they do not know what Chinese character to use or feel a Chinese
54 Cantonese as Written Language

character is not widely enough known by readers. This most frequently

happens with newly coined slang expressions, but sometimes one will find
writers spelling out a word even when written Cantonese already has an
accepted character for the word. For example, in the Diary of The Little Man
series, the word (C) mit (to tear off) is often written as "Mit" despite the fact
that the word is also written with the character M. Romanization of Cantonese
words has also become common in computer-based interpersonal
communications among young people in Hong Kong (ICQ) , so may eventually
come to be more widely used in published Cantonese writing as wel1. 23
(4) Borrowing on the basis of pronunciation: When they do not know how
to write a word in Cantonese, the strategy most often adopted by Cantonese
people - both past and present - is phonetic borrowing, i.e. using a Chinese
character that has the same sound when pronounced in Cantonese as the word
to be written down. (When borrowing characters in this way, the original
meaning of the character is generally ignored.) This strategy works because
literate Cantonese speakers know the Cantonese pronunciation of Chinese
characters.
One example of the phonetic borrowing strategy appears in the first
sentence of the text above, use of the character !t to represent the Cantonese
word (C) zung (still). In both Standard Chinese and written Cantonese, !t is
a commonly-used character meaning "heavy." In Cantonese this character is
pronounced (C) cung, and because (C) cungsounds like the Cantonese word
for "still" - (C) zung- the character!t is borrowed to represent (C) zung
(still). (This particular borrowing took place several hundred years ago; for
example !t appears with the meaning "still" in The Flowery Scrol~ a text dating
from the late Ming dynasty.) 24
Sometimes phonetic borrowings are marked in various ways. Traditionally,
Cantonese phonetic borrowings were often marked by the addition of a
"mouth" radical (kou IJ) to the borrowed character. Many Chinese characters
are composed of two parts, a semantic element (usually on the left side of
character), and a phonetic component (usually on the right) .25 Phonetically
borrowed characters in Cantonese are often marked by the addition of a
"mouth" (kou IJ) radical at the left side of the character, with the borrowed
character appearing on the right as the phonetic element. 26 In the dialogue
above, (C) cam ~ in camjat ~ B (yesterday), and (C) gam Jtit (so) are examples
of characters borrowed for their sound and then marked with the "mouth"
radical. In more recent decades, phonetic borrowings are sometimes marked
by being placed in brackets, particularly if a writer is concerned that readers
might not immediately recognize a particular character as a phonetic
borrowing. For example, in the book Diary of The Little Man, the Cantonese
word (C) laai (to lick) is represented by the graph r~J .
 Spoken and Written Cantonese 55

This tendency to mark phonetic borrowings with a mouth radical or

brackets illustrates the predominance of phonetic borrowing as the strategy
favored by Cantonese speakers when attempting to write their language, and
is also no doubt part of the reason why literacy in the written from of
Cantonese has become popularized to a greater degree than the written form
of other dialects such as Taiwanese. Such marking not only alerts the reader
to the fact that a character is being used in a special way, but also suggests
the strategy the reader should use in making sense of the character. However,
it is important to note that in many cases phonetic borrowings are not marked
in any special way at all, even in relatively ancient texts that make use of
Cantonese (see Appendix 1, Sample Texts 1-4). The early development of a
widespread informal preference for the phonetic borrowing strategy means
that even when Cantonese readers are not alerted to a phonetic borrowing
by any explicit marking of an unfamiliar character, they still know which
strategy is most likely to result in the word's successful decoding.
(5) "Seeking original characters": Finally we should note one additional
approach to the "sound but no character" problem which is pointedly not
represented in the sample dialogue text above, and rarely used in written
Cantonese - ben zi kao (*'=¥~), i.e. "seeking original characters."
The historical fact that Chinese dialects often preserve words and uses
from ancient Chinese that have now disappeared in Standard Chinese has
given rise to a form of study called ben zi kao (*'=¥~) which involves searching
ancient texts to find the Chinese characters with which dialect words were
written in the past. This approach presumes that many dialect words once had
a written form that has since fallen out of use.
One modern example of this kind of study for Cantonese is a 1960 article
by Luo Zhengping in which he suggests possible ancient characters for 58
Cantonese words. This article is not only a good example of a "seeking original
characters" study, but also of the attitude that scholars who engage in such
study usually take toward the widely accepted phonetically borrowed forms of
Cantonese words. Luo considers these to be "mistaken characters" (bie zi }jIJ
'=¥), and he clearly feels that the ancient characters he presents are more
legitimate written representations of the Cantonese words in question (Luo
1960: 129).
The "seeking original characters" approach has had some influence
among those who advocate written Taiwanese, and some scholars of Cantonese
have also engaged in such efforts.2' However, this approach tends to appeal
to scholars more than to the general public, and the dissemination of such
scholarly efforts through articles and dictionaries has apparently had relatively
little impact on most people who frequently use written Cantonese.
56 Cantonese as Written Language

Written Cantonese and phonetic borrowing

Having described the various mechanisms by which Cantonese words are

represented in written form, it is important to stress that by far the most
important and productive of these mechanisms is phonetic borrowing. 28 One
indication of how dominant this tendency toward phonetic borrowing comes
from a study conducted by Huang Xiaoya in 2000 in which she categorizes
characters in Cantonese dictionaries according to the mechanisms by which
they were chosen/created. Of Huang's five categories, the one with the highest
percentage of characters (38 percent) is phonetic borrowing. Furthermore,
an additional category she calls "popular dialect characters" (jangyan su zi 1f
1rift'¥) contains many characters which are actually phonetic borrowings to
which a "mouth" radical has been added to create new characters.29
However, the predominance of phonetic borrowing becomes even clearer
if an actual Cantonese text, rather than a dictionary, is used as the corpus.
For example, when we examine the two texts in Appendix 1 which are most
consistent in following spoken Cantonese norms (Texts 11 and 12, a selection
from a Cantonese newspaper article and a popular Cantonese novel) we find
that every marked Cantonese character in each of these two texts carries
significant phonetic clues to its pronunciation. For a few of the marked
Cantonese characters in these two texts, such ~ (C) tai (to look), the
pronunciation of the phonetic element is not exactly the same as that of the
Cantonese word it represents (the Cantonese pronunciation of the phonetic
element of this character is dai); however, even in these cases the
pronunciation is quite similar. In other words, not only are most Cantonese
characters chosen on the basis of phonetic borrowing, but the most commonly
used Cantonese characters are virtually all chosen on this basis.
Here we might pause to note that this tendency to rely on phonetic
borrowing is not confined to the writing of Cantonese; in fact, most Chinese
characters used in Standard Chinese today have been created through a
process of phonetic borrowing. As noted above, many Chinese characters
consist of a combination of elements (radicals), and most Chinese characters
used today have a phonetic element - usually on the right hand side of the
character - which sounds similar to or even exactly the same as the new word
being represented. In traditional Chinese nomenclature, these are referred
to as pictophonetic (xingsheng *1l) characters, and have long made up the
majority of Chinese characters in use. Even as early as the Shang dynasty (1700-
1045 BeE), approximately a third of all Chinese characters included a phonetic
component, and by the eighteenth century over 90 percent of the characters
in the influential Kangxi Dictionary had a phonetic component. Traditional
Chinese discussion of character types also recognizes an additional category
 Spoken and Written Cantonese 57

of phonetic loan (jiajie 'ffi[1tf) characters, those which are borrowed in their
entirety because they have the same sound as the word to be represented. 30
This tendency to use Chinese characters phonetically is also found among
neighboring peoples who have used Chinese characters to represent their
languages. For example, the Zhuang people of southern China developed a
writing system that made use of Chinese characters to write the Zhuang
language, and borrowed Chinese characters for their sound value. 31 Also,
before colonization by the French, the Vietnamese language was often written
in a system called Nom which used Chinese characters for their sound value
to represent Vietnamese words. Sometimes these Nom Chinese characters were
borrowed in their original form for their sound value; in other cases, new
composite characters were created by borrowing two elements from different
Chinese characters, one for meaning and one for sound. 32
The fact that most Cantonese characters are created through phonetic
borrowing is very important to the spread of literacy in Cantonese. From the
reader's point of view, the predominance of this mechanism means that
readers know what strategy they should use in trying to figure out any
unfamiliar Cantonese characters they encounter. Furthermore, this strategy
is relatively easy to apply. As long as readers have a reading knowledge of
Standard Chinese and the ability to speak Cantonese, they can usually figure
out unfamiliar words quite easily merely by sounding them out in context. This
means that readers who are already literate in Standard Chinese can learn to
read in Cantonese without undergoing an education process like that needed
for literacy in Standard Chinese.
Widespread consensus in favor of the phonetic borrowing mechanism is
also helpful for writers of Cantonese. There are still Cantonese words for which
there is no widely-agreed-on written form, a fact which poses problems for
those who wish to write in Cantonese. However, the consensus in favor of
phonetic borrowing means that writers of Cantonese have a convenient and
widely accepted strategy for dealing with this problem, and this makes it much
easier for people to write Cantonese.

The Standardization of Written Cantonese

Written Cantonese is not very standardized in the sense of widespread

agreement as to which characters should be used to represent which words.
In fact, it is not unusual for two or even more different characters to be in
widespread use even for fairly common Cantonese words. The following list
(drawn from both published texts and dictionaries) provides a few examples
of such words:
58 Cantonese as Written Language

Table 3.3 Different graphs used to represent Cantonese words

Cantonese English Characters Cantonese English Characters

word gloss used word gloss used
bei to give 1* :W je thing 1Ilf!l!f
dez (plural suffix) Ilt!!.:it!! jigaa now m* 1t<*
di a little i¥J n~ D ngai to plead ~~~
gam this way IIftIlfOC tam to coax ~1@sZ4s
Tum
go that UflM1mI*1i zo (perfective) ~tcll!lll
ha to bully
.p~

still !I! 1Jjl

*
zung
hai to be t~ zungji to like i.l2: "P2:

For some of these words, such as (C) jeand gam, the first character listed above
is clearly in wider use than the alternative form. However, for other words such
as (C) bei, dei, di, go, zungji, zung, and jigaa, more than one form is in
widespread use.
One might suppose that the widespread use of divergent forms would
cause confusion among readers, but this confusion is considerably reduced
by two factors: (1) the Cantonese pronunciation of all or most of the variants
is either the same or very similar; and (2) in many cases the difference between
the variant forms is minimal, often the presence or absence of a "mouth"
radical.
One reason for this lack of uniformity in which characters are used to
represent which Cantonese words is that Cantonese dictionaries do not always
agree as to which written form should represent any given Cantonese word.
It is not uncommon for the same word to be represented with different
characters in different dictionaries. 33 Furthermore, no single dictionary has
yet been accepted as an authoritative standard around which others would
converge. The influence of dictionaries and reference tools in the process of
standardizing a language - particularly its written form - is often
considerable. For example, the spelling of English words varied widely from
writer to writer until the publication of Samuel Johnson'S dictionary in 1755
established a widely accepted standard for English spelling. The absence to
date of a widely accepted authoritative reference work therefore helps account
for the lack of standardization in written Cantonese.
Of the Cantonese dictionaries available in Hong Kong, the first to make
much impact on the writing of Cantonese was the 1981 Cantonese Dialect
Dictionary (Guangzhouhua Fangyan Cidian J1HN t!ijJ1§~JI!t.) compiled by Rao
Bingcai (~*7f), Ouyang Jueya (lWc~i-l;.5.l), and Zhou Wuji ()WJ~,8.).
Although the authors reside in mainland China, the dictionary was published
 Spoken and Written Cantonese 59

by the Commercial Press in Hong Kong, and a few Hong Kong authors and
publishers began to use it as their standard reference. 34 However, the impact
of this dictionary was limited by several factors. First, it has a relatively small
number of entries and does not include all the characters writers and
publi~hers need. 35 Second, while the entries in this dictionary generally provide
widely used forms of characters rather than more scholarly or obscure forms,
Hong Kong writers find that it does not always accurately reflect Hong Kong
usage. 36 Finally, some writers in Hong Kong have been hostile toward the
dictionary simply because it is the work of scholars from mainland China rather
than from Hong Kong. One example of this appeared from 27 February to 9
March 1990 in a series of articles in Hong Kong's Tin Tin Daily's "Chewing
over words" (Yao wen jue zi Il5e)cPi'=¥) column in which the author, (C) Lau
Sing Jtl~(JungJuk ~:E') not only criticized flaws in the dictionary, but even
questioned the qualifications of the authors to produce such a dictionary.37
The fact that no single Cantonese dictionary has yet established itself as
authoritative is no doubt one of the main reasons why most Hong Kong
authors do not rely much on formal reference tools when writing Cantonese.
Even when existing reference works are unanimous as to their choice of what
character should be used for a certain Cantonese word, it is not unusual to
find that writers have used another, often simpler, character which shares the
same pronunciation.38 However, lack of an authoritative reference work is only
part of the picture; there also seems to be an underlying sense that uniformity
of character use in written Cantonese is not terribly important to the average
writer. In fact, it is not unusual for an author to use different characters on
different occasions, apparently relying more on sound than precedent. 39
While one could not say that written Cantonese is standardized in the
conventional sense, it is important to recognize, as noted above, that there is
standardization of a sort in the widely held consensus in favor of using phonetic
principles in choosing Cantonese characters. This consensus is important
because wide agreement on this easy-to-use principle is sufficient to permit
the language to function and grow. Writers know how to go about putting
Cantonese words into print, and readers know how to go about decoding the
Cantonese words they see in print. While lack of complete standardization is
no doubt at times an annoyance for readers and writers of Cantonese, it seems
to be little more than that. As Li (2000: 201) notes, users of written Cantonese
are "somewhat inconvenienced by the absence of a standardized writing system.
Such a handicap, however, does little to prevent speakers of Cantonese in
Hong Kong from using the vernacular they know best to represent their
thoughts in writing." The fact that both writers and readers can readily go
about the task of using written Cantonese means that over time it is likely that
higher levels of standardization will eventually emerge.
60 Cantonese as Written Language

What Is "Written Cantonese"?

One might assume that "written Cantonese" is essentially Cantonese speech

written out in Chinese characters. This would be a fair description of written
Cantonese as it appears in contemporary works such as the Diary of The Little
Man novel series, which are actually radio program scripts later published in
book form, and are thus quite literally written representations of speech.
However, from a historical perspective written Cantonese has not always so
closely followed the norms of Cantonese speech; in fact, early Cantonese texts
are essentially written in Classical Chinese with only an occasional sprinkling
of Cantonese words. Furthermore, not all written Canontese texts found even
today in Hong Kong follow the same conventions in how closely or loosely
they adhere to a spoken Canontese norm. In other words, not all "written
Cantonese" is the same, and in order to trace the development of the language
we shall need to distinguish between at least several basic kinds of written
Cantonese: 40
(1) Occasional Cantonese terms in Standard Chinese texts: In some texts,
particularly early texts that contain Cantonese, occasional Cantonese words
or phrases appear in a text that clearly follows Classical Chinese or
Standard Chinese norms. In such texts, Cantonese words are relatively few
in number, and may either have been included to give a sense of local
color or perhaps slipped in more or less by mistake. Into this category
fall many of the verses and songs found in the "wooden fish books" (see
Chapter 4).
(2) Saam Imp dai: This is a clearly defined style of writing in which Classical
Chinese, Standard Chinese, and Cantonese are all used (see Chapter 6).
Cantonese is spread evenly through these texts, as are the other two
language varieties and in virtually every sentence there are elements of
all three language varieties. This writing style was most popular in Hong
Kong newspaper columns from the 1950s to 1980s.
(3) Random mixing of Cantonese and Standard Chinese: In some texts,
Cantonese terms are mixed into Standard Chinese in a way that is not
consistently based on any single spoken or written language norm. Into
this category falls much newspaper Cantonese and Cantonese fiction (see
Chapter 6). While it may be possible to find a rationale for the author's
decision to include Cantonese in these texts, it may not be possible to
explain why Cantonese words were used at any specific point.
(4) Patterned mixing of Cantonese and Standard Chinese: In some texts that
are dominantly Baihua or Standard Chinese, Cantonese is included at
specific points for a discernable purpose. The switch from Standard
Chinese to Cantonese, for example, might signal a change in the tone of
a dialogue. This approach to Cantonese use is found in Cantonese opera
texts (see Chapter 4).
 Spoken and Written Cantonese 61

(5) Standard Chinese narration/Cantonese dialogue: In some texts, Standard

Chinese is consistently used for the narrative portions of a text, and
Cantonese for dialogue. This approach has been taken by some of the
writers associated with City Magazine (see Chapter 6).
(6) Authentic Cantonese: In some texts, the language adheres consistently to
the appropriate register of spoken Cantonese. Writers during the Dialect
Literature Movement generally used this approach (see Chapter 5), as
have many contemporary newspaper and fiction writers, particularly those
who have also worked in film and television script-writing (see Chapter
6).

While even this six-category taxonomy inevitably involves

overgeneralization, it provides a starting point for more nuanced discussion
of the various ways in which written Cantonese has historically been used, and
also serves to make the point that in discussing "Cantonese" texts, some
attempt must be made to sort out the elements of Standard Chinese,
Cantonese, and even Classical Chinese. In particular, distinctions between
different kinds of written Cantonese texts need to be made if we are to see
the gradual development of written Cantonese toward more faithful adherence
to a spoken Cantonese vernacular norm.
The taxonomy above also serves to suggest that the development of written
Cantonese has not progressed simply by a gradual increase of the percentage
of Cantonese words found in texts. Rather, its development has progressed
in a more clearly patterned way, largely through the appearance of new genres
with new conventions that allow greater use of Cantonese. In this sense, the
development of written Cantonese shows close parallels to the development
of Baihua in China, which, rather than usurping the role of Classical Chinese
in established literary genres, tended to appear in new genres which performed
new functions (usually related to oral literature ). Once established in a genre,
Baihua tended to dominate that genreY This pattern also appeared in the
development of written Japanese.
Similarly, Cantonese has made much of its progress as a written language
through its use in new genres which, from their inception, made use of
Cantonese in a particular way. For example, while we cannot be certain that
genres such as "wooden fish songs" and "southern songs" (see Chapter 4)
always contained dialect, this seems probable because even the oldest extant
works in these genres make use of Cantonese. Similarly, other later genres
that we will encounter, from the Cantonese love songs of the 1800s to modern-
day diary-type novels, have all had clear and relatively stable conventional
norms for use of Cantonese. There have also been a few exceptions to this
pattern; for example, Cantonese opera originally contained no dialect and
experienced only a gradual increase in use of Cantonese. However, in general,
the pattern by which written Cantonese has grown is one of the appearance
62 Cantonese as Written Language

of new genres, each with conventions of language use that allow closer
adherence to the norms of spoken Cantonese than allowed by previous genres.

Is Written Cantonese a Distinct Language Variety?

In this book, I have assumed that written Cantonese can and should be
considered a written language variety in its own right, one that is (at least in
the later stages of its development) distinct from the written form of Standard
Chinese. However, this is not necessarily an assumption that the average person
in Hong Kong would always share. This point was driven home to me the first
time I went into a Hong Kong bookstore and asked (in Cantonese) whether
or not they had any books written in Cantonese. While the point of the
question seemed perfectly obvious to me, it was not clear at all to the staff at
the store, and it was only after considerable explaining that one clerk suddenly
said: "Oh, he means those colloquial books!" (The actual Cantonese term she
used for "colloquial" was (C) haujyufaa Cl~{r..) As this experience repeated
itself in one store after another, it slowly dawned on me that what I perceived
to be a difference between two separate language varieties - Standard Chinese
and written Cantonese - was sometimes perceived in Hong Kong mainly as
a difference in register, with texts written in Cantonese seen as simply being
more colloquial.
At the risk of overgeneralizing, I would suggest that the popular view most
people in Hong Kong have toward written Chinese - and the role of written
Cantonese within that written language - is essentially the following:
(1) The main language distinction people in Hong Kong make is between
"Chinese" (C: Zungman *::lC), and English. While they are aware of
differences between Mandarin and Cantonese, the Chinese/English
distinction tends to be much more salient in people's minds, and
differences in varieties of Chinese tend to be viewed as variations within
a language rather than as different languages. 42
(2) There is relatively little emphasis on the tie between written Chinese
(Standard Chinese) and spoken Mandarin. While it is of course generally
known that Standard Chinese is based on Mandarin, from a Hong Kong
perspective the tie between these two language varieties is somewhat less
salient than it would be in other parts of China. Instead, the tendency in
Hong Kong is to view written Chinese as relatively independent of any
spoken variety of Chinese. Instead, it is seen more as a supra-dialect written
language that stands in roughly equal relationship to all spoken varieties
of Chinese, much as Classical Chinese did in pre-modern times. This
tendency is due in part to the fact that the Baihua movement reached
Hong Kong relatively late, and that Classical Chinese played major role
in school curriculums longer in Hong Kong than elsewhere in China. 43
 Spoken and Written Cantonese 63

(3) Elements of Standard Chinese that are more distinctively Mandarin than
Cantonese (in other words, vocabulary which is common in Mandarin but
not in spoken Cantonese) are still considered to be part of "our Chinese
writing" in Hong Kong because they are identified primarily with a written
language rather than an (outsider) spoken language.
(4) The written form of distinctively Cantonese words is often viewed as use
of local colloquial language, a lower register rather than elements of a
distinct language variety.

It should also be noted that language users in Hong Kong frequently are
not sure precisely where the lines are between proper Standard Chinese and
written Cantonese. While an attempt is made in schools to teach students what
is and is not acceptable Standard Chinese, it has been repeatedly observed
that neither students nor teachers are always entirely sure which words are
acceptable in Standard Chinese and which are not. 44 In part as a result of this
confusion, Standard Chinese in Hong Kong has absorbed some Cantonese
words that have now largely or completely displaced their Standard Chinese
counterparts. For example, one will almost always see the characters for (C)
gaasi 1*{fA (furniture) on the signs of furniture stores rather than the Mandarin
term (M) jiaju *J'l" and it is not unusual to find (C) gaasi also appearing in
Hong Kong literary works ostensibly written in Standard Chinese. Other
Cantonese terms such as (C) sezilau~*. (office), (C) saanggwo1=.* (fruit),
and (C) syutgou ~~ (ice-cream) also appear from time to time in Hong Kong
Standard Chinese literary works in which there is no apparent evidence of
conscious intent to use Cantonese vocabulary. In Hong Kong, all of these terms
are becoming accepted as Standard Chinese, and people are not always aware
that these terms would be considered non-standard outside the Cantonese-
speaking regions of China. 45
However, it would be going too far to say that people in Hong Kong do
not make any distinction between Standard Chinese and written Cantonese,
or to say that written Cantonese is always considered just a lower register of
Standard Chinese. To the contrary, there is a general awareness that a
distinction should be made in writing between (acceptable) Standard Chinese
and (unacceptable) Cantonese dialect. As noted above, this issue is one that
is given much attention in schools, and attempts to teach what is and is not
acceptable in writing Standard Chinese absorb much of teachers' time and
attention. 46 So even if students and teachers are not always sure which terms
are acceptable when writing Standard Chinese, they are well aware that some
terms are acceptable and that others are not. They also know that the norm
they should follow in writing is that of a written language - Standard Chinese
- rather than that of their Cantonese speech.
Furthermore, as we will see in the following chapters, there are often fairly
clear patterns as to where and how written Cantonese is used, which suggests
64 Cantonese as Written Language

that its use is a conscious choice involving awareness of written Cantonese as

a distinctive language system rather than merely an occasional shift into a lower
register. Perhaps the clearest example of such a pattern is found in texts where
the narration is consistently in Standard Chinese but the dialogue is
consistently in written Cantonese (a pattern also seen in some Wu dialect
literature) .

Toward the Story of Written Cantonese (II)

In Chapter 2 we suggested that the story of written Cantonese could be viewed

as one of shifting diglossic balance, with written Cantonese (L) moving into
the range of functions previously held only by Standard Chinese (H). In this
chapter I have suggested several other lines along which we should trace the
development of written Cantonese, summarized as follows:

Increasing consistency in use of Cantonese in writing

One of the main lines of development we shall see in the history of written
Cantonese is an increasing tendency to adhere to the norms of spoken
Cantonese, even when they conflict with the norms of written Standard
Chinese. In large part we shall see this in a gradual increase in the percentage
of Cantonese lexical items used in texts, and we shall follow this development
by looking for the presence of "marked Cantonese" terms in texts. In early
texts containing Cantonese, th~ percentage will be very small, and is not
reflective at all of how Cantonese is normally spoken. However, over time, texts
will come closer to percentages of distinctly Cantonese lexical items that are
normal for spoken Cantonese, approximately 10 percent for more formal
registers of the spoken language (news reports and so forth), and 30 percent
or more for informal and conversational registers.
At later stages of the development of written Cantonese, we shall also look
for the decreasing use of "marked Standard Chinese" as an indicator of the
extent to which a Cantonese norm is established and maintained. "Marked
Standard Chinese" words are those commonly used in spoken and written
Standard Chinese, but rarely if ever used in spoken Cantonese. Use of these
terms, therefore, suggests adherence to a Standard Chinese rather than written
Cantonese norm, and their disappearance suggests that the norms in a given
text have become fully Cantonese. For two reasons, however, less weight can
be placed on this indicator than on the presence or absence of marked
Cantonese. First, in one sense, there is no such thing as truly "marked"
Standard Chinese because even words that are more typical of Mandarin than
of Cantonese are sometimes still used in higher registers of Cantonese, though
 Spoken and Written Cantonese 65

often in contexts that are closely associated with written texts. Secondly, the
use of marked Cantonese is a much more stigmatized choice in writing than
the use of marked Standard Chinese ever is; the presence of marked Cantonese
is therefore more significant as an indicator of a choice to move toward a
Cantonese norm.

Increasing Cantonese literacy

A second main line of development has to do with spreading literacy in written

Cantonese. In some cases, particularly early in the history of written
Cantonese's development, this may be an issue of primary literacy, of written
Cantonese being the first language that people (primarily women) learned
to read. However, in more recent years the issue is one more of how well
Cantonese speakers who have been educated in the use of Standard Chinese
are also able to read and even write in Cantonese. As suggested in this chapter,
given that schools do not generally teach written Cantonese, the ease with
which Cantonese speakers can attain literary in written Cantonese is a vital
factor in facilitating the increase of Cantonese literacy.

Increasing number of genres which allow use of Cantonese

A third important line of development has to do with the kinds of texts and
published materials within which written Cantonese is used. As noted above,
increase along this line has generally not resulted from Cantonese appearing
with increasing frequency in well-established genres of writing. In fact, the
evidence will show that once the language conventions of a genre are set, tliey
generally do not change much over time, and a genre in which people expect
Standard Chinese to be used will generally not begin to admit use of
Cantonese. Instead, increase in the use of written Cantonese normally occurs
through the creation of new genres within which written Cantonese has a
prominent role right from the start.
8
Epilogue: The Future of Written Cantonese

Will the Growth of Written Cantonese Continue?

Given that much of this book has been dedicated to tracing the growth of
written Cantonese, it seems appropriate to close by casting an eye to the future
and asking whether this pattern of growth is likely to continue.
The most obvious reason to suspect that the role of written Cantonese
might possibly not continue to grow has to do with the reintegration of Hong
Kong into China, hence into a nation where Mandarin (Putonghua) is the
overwhelmingly dominant language. Clearly the return of Hong Kong to
Chinese sovereignty brings the Hong Kong speech community into even closer
relationship with mainland China, and - returning to the analogy used in
Chapter 2 - this will inevitably have some impact on the marketplace in which
Hong Kong people make their language choices. Because of China's enormous
population, growing wealth, growing political and military power, and
expanding cultural influence, the appeal of Putonghua in the Hong Kong
market cannot help but increase. The question is: In what precise ways?
One safe prediction is that more people in Hong Kong are likely to learn
Putonghua than in the past. The Hong Kong government has made more of
an effort in recent years to promote study of Putonghua, both among
government employees and among the populace at large. l Perhaps more
important, the case for learning Putonghua is ever easier to make in Hong
Kong because the practical advantages of learning Putonghua are so obvious.
Not only is there increasing opportunity, and even need, for Hong Kong
people to do business on the mainland; it is even becoming increasingly
common for young Hong Kong people to seek jobs on the mainland with
mainland companies and institutions. Likewise, the flow of people from the
mainland into Hong Kong for tourism, business, or simply to visit friends and
relatives is also increasing, thus expanding the opportunities for people in
Hong Kong to practice whatever Putonghua skills they learn. Because of the
214 Cantonese as Written Language

similarities between Putonghua and Cantonese, Cantonese speakers can learn

to function in Putonghua quite rapidly if given study and practice
opportunities, and once they begin to actually use Putonghua, their Putonghua
skills are likely to keep improving. This generates the prospect that Putonghua
could become a living language in Hong Kong, one which most members of
the community can and do regularly speak, and this would cause Putonghua
to seem less like a foreign language.
It seems likely that the role of Putonghua in Hong Kong's schools will
also continue to increase, a development which could potentially have
significant implications for the future growth of written Cantonese. As we
have seen, Putonghua has already become a required subject in Hong Kong
schools. Furthermore, as of the time of this writing (2004) there are reports
that educational authorities in Hong Kong have begun seriously examining
the possibility of moving toward use of Putonghua as the medium of
instruction in Hong Kong schools (although no specific plan has been
presented to date).2 Were the role ofPutonghua in Hong Kong schools to
increase dramatically - in particular, if it were to become the medium of
instruction in Hong Kong schools - the conditions which have facilitated
the growth of written Cantonese would be affected in two important ways.
First, as Bauer (2000: 44) has pointed out, the ease with which Cantonese
speakers learn to read and write Cantonese is based on their ability to
pronounce Chinese characters in Cantonese. If the Hong Kong education
system were to adopt Putonghua as the medium of instruction, and if
students no longer learned to pronounce characters in Cantonese, it would
become much harder for students to learn to read and write Cantonese.
Secondly, if Putonghua came to dominate the education system to the degree
that Mandarin dominated schools in Taiwan for many decades, the result
might well be generations of students who were more comfortable expressing
themselves in Putonghua than in Cantonese, a development which would
not only make written Cantonese harder to learn to read and write, but also
undermine its effectiveness as a language of identity.
However, it seems that such a scenario is unlikely in Hong Kong, at least
within the near future. First, while knowledge of Putonghua will no doubt
continue to increase in Hong Kong, this does not in and of itself necessarily
mean a decrease in the ability or willingness of people to speak Cantonese; in
particular, it does not necessarily mean that Putonghua will replace Cantonese
as the language of identity in Hong Kong. Rather, what is happening so far is
that people in Hong Kong are adding the ability to function in Putonghua to
their toolkit of language skills, rather than replacing their Cantonese with
Putonghua. This addition of Putonghua language skills, rather than
eradication of Cantonese, is also the goal at which the Hong Kong government
aims in its "promote Putonghua" campaigns. Second, the level of Putonghua
instruction in Hong Kong schools is currently so low that it does not threaten
 Epilogue: The Future of Written Cantonese 215

the role of Cantonese, and does not even provide much support for students'
efforts to learn to read and write Standard Chinese. Putonghua would need
to eclipse Cantonese almost entirely within the education system if Putonghua
were to replace Cantonese as students' language of identity or undermine their
ability to pronounce Chinese characters in Cantonese. It does not seem likely
that the role of Putonghua in Hong Kong schools would increase to this
degree, at least not in the near future. In fact, if the Hong Kong government
continues its recent policy of requiring most schools to use "mother tongue"
(spoken Cantonese and written Standard Chinese) as the medium of
instruction, rather than using English, it is possible that more young people
in Hong Kong will be exposed to greater amounts of educated discourse in
Cantonese and develop an even higher degree of proficiency in that language.
Therefore, at least for the immediate future, I suspect that written
Cantonese is likely to retain its current role in Hong Kong society, and perhaps
even expand its role in Cantonese-speaking areas of mainland China. I make
this prediction with some hesitation, because predicting the future - and
putting those predictions in print - is always a risky business, but there are
several reasons why I feel reasonably confident in this assessment. The first is
simply the evidence of the past. As I have endeavored to demonstrate in this
book, the trend over the past several centuries has been for the role of written
Cantonese to gradually but steadily increase, following the rising prosperity
of first Guangzhou and the Pearl River Delta region, and then Hong Kong.
While the past is by no means always a reliable guide to the future, there is
good reason to suspect that existing trends will continue unless there are
significant changes in the factors that have given rise to them.
A second point also has to do with the continued prosperity of Cantonese-
speaking regions of mainland China. The growing wealth of Guangzhou and
the Pearl River Delta region has contributed substantially to the growing vitality
of the spoken Cantonese language, a vitality which is reflected in a variety of
interesting ways.
As noted in Chapter 7, the unusually high prestige of Cantonese within
its own speech community is suggested by the fact that Cantonese-speaking
areas are the only part of China where a significant amount of local television
and radio broadcasting is done in a Chinese dialect. A significant amount of
this Cantonese-language broadcasting originates in Hong Kong; in fact, since
the 1980s the Hong Kong broadcast media have had a very substantial
audience in Guangdong province. 3 However, Cantonese is also the language
used in a significant amount of broadcasting in Guangdong, and even parts
of Guangxi.
The geographic range in which Cantonese is spoken as first language is
actually expanding. In some parts of Guangdong province where other dialects
were originally spoken, Cantonese is moving in at the expense of the original
dialects. This is happening, for example, in Shanwei, in Min-dialect-speaking
areas in Zhongshan, and around Huizhou and Boluo.4
216 Cantonese as Written Language

The influence of Cantonese on Putonghua is growing as more and more

Cantonese words are coming to be accepted in Putonghua and used
throughout China, many even now appearing in standard Chinese dictionaries.
Tang and Tu, for example, count more than 160 Cantonese expressions which
appear in the 1996 edition of the widely used Modern Chinese Dictionary (Xiandai
Hanyu Cidian ~1-t~mf~.) (2000: 367).5
Even in places far from Guangdong, such as Beijing, Shanghai, Nanjing,
Tianjin and Wuhan, many Chinese study Cantonese as a second language. 6
This interest in Cantonese study is reflected in the surprisingly large number
of textbooks and tape sets available throughout China for Chinese people
interested in learning Cantonese. My personal collection of such materials
includes over twenty textbooks and tape sets. While many of these are
published in Guangzhou, others are published further afield in cities such as
Beijing, Shanghai, Shenzhen, Hainan, Changsha, and even Dalian.
This prominence of Cantonese in Guangdong and throughout China
creates a climate in which the relatively high prestige and influence of
Cantonese in Hong Kong is easier to maintain. In fact, it seems that the use
of written Cantonese is even experiencing modest and gradual expansion back
into its ancestral home. For example, Cantonese terms now appear more often
in Guangzhou newspapers, though such usage consists of isolated Cantonese
terms (only one or so per page, often marked with quotation marks and
sometimes supplied with a gloss).? Also, young Cantonese speakers on the
mainland have also begun to use Cantonese in informal personal written
communication, much as it is often used by young people in Hong Kong. 8
A final reason I predict a continued role for written Cantonese has to do
with the continuing gap between written and spoken language in Hong Kong.
Despite the introduction of some Putonghua instruction in Hong Kong
schools, for most Hong Kong students Standard Chinese remains a written
code which is largely divorced from their spoken language. As long as this
situation does not change, the basic conditions which make written Cantonese
effective as a language of identity remain in place. So barring a dramatic
change in Hong Kong language policy; a dramatic increase of the role of
Putonghua in Hong Kong schools; or a dramatic decline in the economic
fortunes of Hong Kong and Guangzhou, it seems probable that written
Cantonese will continue to playa role in the life of Hong Kong, and perhaps
even increasingly in Guangdong.

Where Is the Growth of Written Cantonese Going?

If we accept that the role of written Cantonese may continue to expand, the
final question is: How far? Is written Cantonese likely to eventually replace
Standard Chinese as the written language of Hong Kong or even Guangdong?
Despite my belief that written Cantonese may continue to grow and
 Epilogue: The Future of Written Cantonese 217

develop, I do not believe the final outcome of this shift in social roles is likely
to be a replacement of Standard Chinese by written Cantonese, in Hong Kong
or anywhere else. As an important language of wider use the practical value
of Standard Chinese for people in Hong Kong is too high for the idea of its
replacement by written Cantonese to generate much support. Furthermore,
the political implications of abandoning the national Chinese written language
in favor of a written dialect would also not be acceptable to the majority of
people either in Hong Kong or elsewhere in China.
However, the more important point to make is that there is little or no
evidence that the replacement of Standard Chinese by written Cantonese is
desired by anyone in Hong Kong, including those who regularly use written
Cantonese. There is certainly no organized body working for the replacement
of Standard Chinese by written Cantonese, and to the extent that expressions
of support for written Cantonese appear in Hong Kong's press and media,
these voices suggest toleration of its limited use, not its conquest of Standard
Chinese.
The final destination of this growth of written Cantonese seems not to be
replacement of Standard Chinese. Rather, it appears to be a new diglossic
balance which better reflects the identity of Hong Kong's Cantonese-speaking
community, and perhaps to some extent that of other Cantonese-speaking
areas in China. In this new balance, Standard Chinese would continue to be
the language of choice for most texts, especially those with serious or formal
purposes. It would also serve as the written language of wider communication
for interaction with the broader Chinese-speaking world. Finally, it would serve
as a vehicle for expressing the Chinese component of the Hong Kong identity.
Written Cantonese, in contrast, is consolidating a limited role in the Hong
Kong community in texts that simulate spoken interaction between Cantonese
speakers, and in texts for light entertainment purposes. Underlying this, it is
becoming the language of choice for informal written interaction between
Cantonese speakers, an in-group language that allows Cantonese speakers to
express the local Hong Kong component of their identity. Such a diglossic
balance of roles reflects the mixed culture and identity of Hong Kong in a
way that neither Standard Chinese nor written Cantonese alone are able to
do. It seems likely that this new diglossic balance will continue to characterize
the language situation in Hong Kong for the foreseeable future.
Notes

CHAPTER 1
1. In the Chinese context, the term "dialect" is used to refer to a variety of Chinese
that differs significantly from Mandarin in pronunciation, vocabulary, and to some
degree in grammar.
2. A complete listing of books mentioned in this book, including both English and
Chinese titles, can be found in Appendix 3.
3. The Chinese titles of this and other periodical publications mentioned in this book
are listed in Appendix 3.
4. This particular list of article types is drawn from examination of the 11 November
2001 issue ot Apple Daily.
5. See, for example, DeFrands (1950: 194); Yuan (1960: 177); Cheung (1985: 191);
Ramsey (1987: 99); and Norman (1988: 215).

CHAPTER 2
1. Ferguson defines diglossia as "... a relatively stable language situation in which, in
addition to the primary dialects of the language (which may include a standard
or regional standards), there is a very divergent, highly codified (often
grammatically more complex) superposed variety, the vehicle of a large and
respected body of written literature, either of an earlier period or in another
speech community, which is learned largely by formal education and is used for
most written and formal spoken purposes but is not used by any sector of the
community for ordinary conversation" (1959: 336).
2. For additional discussion of differences between Hand L, see Fasold (1984: 53)
and Trudgill (2000: 96).
3. Chejne (1969: 39-41).
4. Frangoudaki (1992: 366).
5. Alexiou (1982: 162).
6. Keller (1982: 82). Ferguson also mentions Chinese briefly in his 1959 article,
proposing the pre-modern situation in China as a further instance of diglossia,
with Classical Chinese as the H language and colloquial Mandarin as L; in fact he
suggests that it "probably represents diglossia on the largest scale of any attested
instance" (337-8).
260 Notes to pages 77-32

7. Fasold (1984: 48-50).

8. Hong Kong Government, Census and Statistics Department (2002).
9. Hong Kong Government, Census and Statistics Department (2002).
10. Even this three-level scheme oversimplifies the situation in Hong Kong to some
degree. A more detailed description can be found in Luke (1998: 148-9).
11. Situations where H is only used for writing and L only for speaking have been
referred to as "biglossia" (Britto 1986: 42).
12. Haas (1982: 25).
13. Cheung (1992: 210).
14. Alexiou (1982: 159); Frangoudaki (1992: 369-70).
15. Rickard (1989: 81).
16. Anderson (1991: 38).
17. Twine (1991: 8); Mair (1994: 731).
18. Gellner (1997: 52).
19. Miller (1995: 33).
20. Hudson (1980: 33).
21. Twine (1991: 37-38).
22. Throughout Asia, Buddhist writings were often among the first vernacular texts
to appear in many societies. For discussion of this phenomenon, see Mair (1994).
23. Takeuchi (1999: 9). .
24. Twine (1991: 201).
25. Twine (1991: 85, 98).
26. Takeuchi (1999: 36).
27. Baihua is essentially the same written language as Standard Chinese, but it seems
inappropriate to use the term "Standard Chinese" in discussing periods before
Baihua had become a national standard.
28. Mair (1994: 707-8) argues that Classical Chinese was a drastically abbreviated "sort
of demicryptography largely divorced from speech" which descended from oracle
shell and bone inscriptions. While granting that it must also have been affected
by spoken Chinese, Mair feels the disparity between Classical Chinese and spoken
Chinese was always of an unusual order of magnitude.
29. Norman (1988: Ill); Hanan (1981: 10-11).
30. Hanan (1981: 14).
31. Gao (1993: 8-10); Lee and Nathan (1985: 380-1).
32. Gunn (1991: 39); Gao (1993: 10).
33. Chen Ping (1999: 70).
34. While the vernacular was promoted in order to make literature a more effective
tool for reforming the nation, literature was also used as a vehicle for promoting
the vernacular. "In a retrospective essay on the origin of the vernacular language
movement, Hu Shi commented explicitly that he and his colleagues sought to
promote the vernacular not through appeal to the government and its education
system, but rather by creating major works of literature that would appeal to the
general reading public and recommend themselves as guides to writing as well as
to thinking" (Gunn 1991: 38).
35. Gao (1993: 58).
36. DeFrancis (1984: 249); Chen Ping (1999: 79).
37. For a full taxonomy, see Giles et al. (1977: 309); also Landry and Allard (1994:
22). The Chinese concept of the "competitive ability" (jingzhengli tt* jJ) of a
 Notes to pages 32-40 261

language variety, as used by Li Rulong (1997: 235), describes essentially the same
phenomenon.
38. Allard and Landry (1994: 117).
39. See Ye (1996) and Luo (2000) for discussion.
40. Approximately 7.5% of China's population speaks a Wu dialect as home language,
as opposed to approximately 4.5 for Cantonese (Summer Institute of Linguistics
2003).
41. Hu ([1926] 1974: 24-25).
42. Hu ([1926] 1974: 33-34); Lu Xun cited in Cheng (1978: 52).
43. Zhang (1983: 19); Duval (1980: 177). Duval notes that Nzne-tazled Turtles was the
most popular of the Wu dialect novels at its time of publication.
44. As in the case of Cantonese southern songs, the brothel setting was probably also
important in providing sanction for dialect use. See Chapter 4.
45. Link (1981: 60).
46. Dolezelova-Velingerova (1980: 9).
47. To date the most comprehensive English-language study of written Taiwanese is
Henning Kloter's Written Taiwanese (2004). Unfortunately, it appeared too late for
me to make use of its wealth of information in this book.
48. Yao (1989: 16); Hsiau (2000: 135). See also Van der Loon (1992).
49. Hsiau (2000: 37).
50. Gu (1989: 80); Hsiau (2000: 40-2).
51. Hsiau (2000: 38-41).
52. Hsiau (2000: 45).
53. Hsiau (2000: 44).
54. Hsiau (2000: 54-5).
55. Hsiau (2000: 137).
56. Chen Ping (1999: 122-3); Hsiau (2000: 139).
57. Hsiau (2000: 7; 105).
58. Lin (1988: 13); Song (1988: 194-5); Lin (1989: 142); Lin Qiyang, interview, 5 June
1990.
59. For analysis of the advantages of romanization, see Zheng (1988b: 32).
60. Interestingly, in response to Chen Ruoxi's complaint, advocate of written Taiwanese
Lin Yangmin effectively concedes the point, arguing that this just demonstrates
the need for promotion of Taiwanese in the education system (1999: 200).
61. In Song Zelai, A Weak Small People (Ruoxzao Mznzu ~~/J\~~) (Taipei: Qianwei
chubanshe, 1987), 173-263.
62. Hsiau (2000: 140-4).
63. Tsao (1999: 349); Hsiau (2000: 131).
64. Tsao (1999: 369).
65. Such textbooks as I have seen include texts in written Taiwanese, and sometimes
also in Taiwanese romanization.
66. There is controversy as to the extent to which the prompt books were used by
storytellers. For example, Jaroslav Prusek holds that these texts were used by
storytellers, though he concedes that they were also intended for a reading public
(1970: 427). Patrick Hanan, on the other hand, feels that these texts were primarily
intended for a reading public (1981: 9).
67. Hanan (1981: 6); Norman (1988: 111).
68. Hanan (1981: 20).
262 Notes to pages 41-50

69. This pattern is similar to that in which vernaculars first appeared in many
European nations. See Joseph (1987: 76).
70. Hanan (1981: 10-11).
71. Hanan (1981: 6); Norman (1988: 111).
72. On the idea that a form of diglossia is developing in Hong Kong's written Chinese,
see also Zhong (2000: 156).
73. Halliday (1989: 29-30).

CHAPTER 3
1. Another problem with the term "dialect" is that it can carry negative connotations
of inferiority; in fact, the sinologist Victor Mair has proposed that in discussion of
China the term "dialect" be replaced with the more neutral term "topolect," as
he does in Mair (1994) and elsewhere. While acknowledging the problem, I have
chosen to continue using the term "dialect" in part because of its familiarity, in
part because the social role of Cantonese and other Chinese dialects is more
limited than that of Mandarin.
2. DeFrancis (1950: 195; 1984: 39); Deng (1980: 84-5). Incidentally, there are, of
course, different sub-dialects of Cantonese, some of which differ considerably from
the varieties of Cantonese spoken in Guangzhou and Hong Kong. Historically,
however, Cantonese has been relatively homogeneous, and until recent years the
form of Cantonese spoken in Guangzhou was clearly the prestige dialect of the
region and has provided a regional standard (Norman 1988: 214-5). In recent
years, Hong Kong Cantonese has taken over much of this role.
3. Zeng (1982: 343); DeFrancis (1984: 63); Chan and Kwok (1986: 407).
4. Bruche-Schulz (1997: 3011); Matthews and Yip (2001: 266). See Zeng (1982: 358-
75) and Matthews and Yip (2001) for a more extensive discussion of grammatical
differences between Cantonese and Mandarin. For a comprehensive introduction
to Cantonese grammar, see Matthews and Yip (1994).
5. Zhong (1994: 224).
6. Zeng (1982: 343).
7. Obviously the difference apparent here to a reader of English results partly from
different romanization systems being used for Cantonese and Mandarin, but even
so these renderings give a good idea of how different the words sound.
8. As Hanan (1981: 14) notes, this was also the major difference between classical
and vernacular Chinese in traditional Chinese literature.
9. Zhan (1986: 47); Deng (1989: 3); Mai (1993: 66-7). LiJingzhong (1990: 42)
proposes an even higher estimate, suggesting that the level of lexical difference
between Cantonese and Mandarin may be in the range of 77 to 98 percent, but
his results are biased significantly by the fact that the sample corpus Li uses
(Cantonese dictionaries) generally only contain lexical items that are unique to
Cantonese.
10. Chan and Kwok (1986: 408); Bauer (1988: 249). See Luke (1998: 148-9) for more
detailed analysis of the different registers of speech in Hong Kong, many of which
include an admixture of English.
11. 5 a.m., 18 September 1989, Commercial Radio, taped and transcribed by the
author.
12. Here and throughout this book, the term "marked Cantonese" refers to use of
 Notes to pages 50-53 263

characters (and occasionally romanization) which is distinctly Cantonese in the

sense that: (1) these characters are unique to written Cantonese, and not found
in Standard Chinese; or (2) they are used in a way that represents a distinctly
Cantonese usage, and would not normally be acceptable in Standard Chinese. All
calculations of percentages of marked Cantonese in this book are based on
numbers of marked Cantonese characters rather than words (some Cantonese words
are multi-syllabic and are written with a combination of two or more characters).
When some of the characters in a multi-character Cantonese word are the same
as those in a Standard Chinese counterpart, but others are different, I have only
counted the latter as "marked Cantonese." See the introduction to Appendix 1
for further discussion.
13. 7 p.m., 10 March 1990. The transcript was supplied by the reporter, and then
checked against his actual broadcast.
14. See Appendix 1, Text 12, for discussion.
15. 5.10 p.m., 13 September 1989, Commercial Radio. The program consisted of an
interview with a retired police chief.
16. See Appendix 1, Text 11, for discussion.
17. I have only had limited opportunity to test how well readers in mainland China
can read texts written entirely in Cantonese. However, in the few informal tests I
did run (using an excerpt from Diary a/the Little Man), I found that readers who
could speak Cantonese but had little previous exposure to written Cantonese were
able to decode it slowly but accurately. Those who did not speak Cantonese were
generally able to figure out much of the main outline of the story, but were also
often inaccurate in their explanations of individual sentences. These latter readers
reported that they were relying heavily on guessing skills, and that the task was
burdensome enough that they would not attempt it unless they had to.
18. See Bauer (1988, especially pp. 257-76) and Luke (1995) for discussion of the
various mechanisms by which Cantonese words are represented in written form.
Cheung and Bauer (2002) provides the most comprehensive compendium to date
of the Chinese characters used to represent Cantonese words.
19. In Bauer's discussion of newly created Cantonese characters (1988: 260-2), he
concludes that this is both an important and productive method for the creation
of written forms of Cantonese words. Our difference of opinion here is more
apparent than real, however, because many characters Bauer classifies as new I
have considered to be cases of borrowing for pronunciation. The difference in
our classification of such characters is due to the fact that Bauer is concerned
primarily with the precise written form of Cantonese characters, and classifies them
on that basis. In contrast, here I focus on the underlying mechanisms by which
characters are created or chosen.
20. DeFrancis (1984: 84-85; 1989: 99-100). This reluctance to create new characters
was stated to me explicitly by Dialect Literature Movement author Hua Jia, who
said that when he encountered the problem of not knowing what character to
use for a Cantonese word, he avoided creating new characters. He preferred,
rather, to work within the framework of already existing characters. Interview, 18
June 1990.
21. As Li (2000: 219) points out, some Cantonese words are difficult to represent in
Chinese characters because there is no Standard Chinese character which has a
sound close enough to the Cantonese word in question, and romanization is often
used for such words. The number of these is, however, not very large.
264 Notes to pages 53-59

22. Among the other words that have been represented with single letters are some
obscene terms. It may be that the use of romanization in these cases emotionally
distances the author and audience somewhat from the impact of taboo terms, but
more likely the practice was borrowed from the rather handy English custom of
just putting down the first letter of an obscene word and then leaving the rest
blank, something that is not easy to do with Chinese characters. For example, see
Wu Gun-man's Growing up in Hong Kong, pp. 16-18.
23. As K K Luke points out, ICQ chats tend to be short and rapid, and romanization
is often used for Cantonese words in part because English letters are easier to type
rapidly than Chinese characters. Interview, 26 September 2002.
24. Xue (1985a: 22). For more on The Flowery Scroll, see Chapter 4.
25. See DeFrancis (1984), Chapters 6 and 7 for discussion.
26. Bauer (1988: 261). Chan and Kwok (1986: 423) note that the "mouth" radical is
also often added when Chinese characters are used in Hong Kong to transliterate
English words.
27. In Taiwan scholars have played a more active role in the creation and promotion
of written Taiwanese, and one result of this is that there is a greater variety of
principles being used for character selection/creation. Less educated readers and
writers are most likely to choose characters for Taiwanese words on the basis of
sound, but such choices are often considered ignorant by scholars and more
educated groups (Cheng 1978: 311). For Cantonese examples of such studies, see
Luo (1960), Bai (1980) and Tan (1994).
28. Bauer (1988: 261, 276); Li (2000: 206).
29. Huang Xiaoya (2000: 246).
30. Hayford (1987: 152); DeFrancis (1989: 99); Norman (1988: 68).
31. Ramsey (1987: 242-3).
32. DeFrancis (1977: 21-24).
33. Huang Xiaoya (2000: 247).
34. The publishing arm of TVB, Publications Holdings Limited, one of Hong Kong's
leading publishers of pocketbooks, took the Cantonese Dialect Dictionary as its
standard, as did a number of writers associated with Publications Holdings (Kwan
Wing-kei, interview, 25 September 1990; Joseph Yau, interview, 3 July 1990; Ng
Chun-bong, interview, 3 July 1990). No other publishers that I spoke with used
this dictionary, or any other, as their standard reference.
35. Ng Chun-bong, interview, 3 July 1990.
36. Kwan Wing-kei, interview, 25 May 1990; Ng Chun-bong, interview, 3 July 1990.
37. This was one of several series of articles criticizing Cantonese reference books
produced by mainland Chinese scholars. Lau's suspicions as to the credentials of
the authors of the Cantonese Dialect Dictionary typify his attitude toward the other
works as well.
38. For example, reference works are quite unanimous in assigning the character III
to the word (C) laan (to crawl). However, A Foon's Diary of The Little Man series
uses ~ instead. We find the same pattern with the word (C) mit (to tear off). The
reference works use ~, while Diary of The Little Man uses "Mit."
39. For example, in the Diary of The Little Man series, the following Cantonese
words have varying forms: (C) caan (to rush): ~ and rJl~J ; (C) daap (to taste):
"DEP" and ~ ; (C) duk (to poke): r~J and rtf J ; (C) wt (to squeeze): it
r
and 1t J ; (C) zit (to tickle): nfp and I!lt. The same phenomenon can be seen
 Notes to pages 60-72 265

in other books as well. In Gam Gwok-loeng's Vanishing from .the Human Race,
(C) di (a little) is sometimes written as I¥J and sometimes as nag.
40. For another taxonomy, see Liang (1987: 65-6).
41. Hanan (1981: 5; 10). For example, rather than replacing classical Chinese as the
language of the classical tale (chuanqi 1$*), use of Baihua appeared in a new kind
of vernacular short story (Baihua xiaoshuo B lm/J\Jm). A number of features which
distinguish the classical tale from the vernacular story have to do with the role of
the narrator. For example, in classical tales the narrator never addresses the
audience, and makes "editorial" comments only before or after the tale. In the
vernacular story, on the other hand, the narrator interjects comments and
questions to the audience throughout the story. For a complete discussion of these
and other differences, see Hanan (1981: 20-26).
42. Lin (1996: 80).
43. Chang (1981: 61-62); Chan (1987: 235).
44. Chan (1987: 243); Deng (1989: 3); Kwo (1992: 205); Zhong and Zhang (1993:
24-26); Zhong (1994: 229); Yu (2001: 2-3).
45. Another feature of the Standard Chinese of Hong Kong and Cantonese-speaking
areas in Guangdong province is the use of English vocabulary that was first
borrowed into Cantonese, and then worked its way into local Standard Chinese
usage. Even in Guangzhou, one frequently finds signs for (C) diksi I¥J± ( taxi)
and (C) baasi B± (bus) rather than for (M) chuzuche tl:lm* and (M) gonggong
qiche -0~7~*, and in Hong Kong the two Cantonese terms have won out
completely, becoming the normally used terms for "taxi" and "bus" in any writing,
no matter how formal or standard.
46. Zhou (1987); Tong (1989: 15).

CHAPTER 4
1. Siu (1993: 20); Lary (1996: 16).
2. Liang (1982: 88); Huang (1999: 215).
3. Wiens (1967: 132); Huang (1999: 34-35).
4. Huang (1999: 36); 209; Lary (1996: 9).
5. Huang (1999: 215; 219).
6. Wakeman (1966: 43);Jiang and Fang (1993: 121); Faure (1996: 38); Huang (1999:
44); Tsin (1999: 17).
7. Faure (1996: 38).
8. Schafer (1967: 45).
9. Wiens (1967: 143-4).
10. Faure (1996: 38).
11. Schafer (1967: 28); Wiens (1967: 143).
12. Huang (1999: 215); Siu and Faure (1995: 2); Jiang and Fang (1993: 158-66).
13. Wiens (1967: 182).
14. Huang (1999: 210).
15. Faure (1996: 38-9).
16. Siu (1993: 22-23); Faure (1996: 44); Siu and Faure (1995: 11-12).
17. Wakeman (1966: 57,121); Liang (1982: 78-79). An interesting manifestation of
this ability to defy national authority while claiming true patriotism may be found
in the Triads of the southeast Chinese coast. These secret societies, common in
266 Notes to pages 72-79

both Fujian and Guangdong, were ostensibly devoted to the overthrow of the Qing
and the restoration of the Ming, though in practice they were often primarily
organized crime groups (Wakeman 1966: 119-20).
18. Liang (1988: 28); Naquin and Rawski (1987: 181).
19. Naquin and Rawski (1987: 180). The most complete account of this warfare is to
be found in J.A.G. Roberts, "The Hakka-Punti War" (PhD dissertation, Oxford,
1968).
20. Wakeman (1966: 56-58). The role inter-group conflict played in creation of a
distinct Cantonese identity was by no means unique. Despite the fact that Hakka
people had been present in southeastern China for centuries, it was only when
the Hakka began entering Guangdong in the Ming dynasty and found themselves
in conflict with neighboring groups, first the She (~n people and later other Han
Chinese, that a common sense of Hakka identity began to grow. Significantly,
Hakka-speaking people in southern Jiangxi who did not experience such conflicts
were also slow to develop any sense of Hakka identity (Leong 1997: 35-39; 63;
Lutz and Lutz 1998: 11).
21. Vogel (1969: 32); Eastman (1974: 255).
22. Eastman (1974: 255-60). Goodman and Feng (1994: 179) point out that, unlike
the several independent dynasties that established themselves in Guangdong over
the centuries, which aspired to be national rather than regional governments,
Chen Jitang's government in the 1930s attempted to promote a localist political
identity. However, they also note that this had much to do with his competition
with Chiang Kai-shek. Lary (1996: 18) describes the recent "re-discovery" of Chen
by Chinese historians, and the tendency to portray him sympathetically as one who
was able to keep the Guangdong economy going during the Depression.
23. Attempts of the national government to assert its control over Guangdong are a
major theme of Ezra Vogel's 1969 Canton Under Communzsm.
24. Li (1990: 32): Huang (1999: 91).
25. See also Norman (1988: 210-11) and Bauer and Benedict (1997: xxxix).
26. Mai (1993: 69-70).
27. Liu (2000).
28. Naquin and Rawski (1987: 173).
29. Yang (2000).
30. In a recent research project, when Cantonese, Hakka and Chaozhou people in
Guangdong were asked how they determined who belonged in which group, the
first criterion cited by most people was language. Blood ties were second, and home
region third (Huang 1999: 478).
31. Huang (1999: 217).
32. Wakeman (1966: 57).
33. Li (1989: 1); Goodman and Feng (1994: 178); Huang (1999: 222).
34. Wakeman (1966: 57); Pan (1991: 15); Ching (1996: 54); Fitzgerald (1996: 159).
35. Similar arguments have been made for how it was that the vernacular came to be
used in the written languages of other Asian countries. For example, John
DeFrancis suggests that early vernacular writing in Vietnamese can also trace its
origins to Buddhist literature (1977: 21).
36. Liang (1988: 19).
37. Liang (1978: xiii); Xue (1985a: 108-9).
38. Tan and Tan (1982: 7).
 Notes to pages 79-83 267

39. Rawski (1979: 6, 116-7); Hegel (1998: 6). While there is no conclusive evidence
as to the period when wooden fish books were first published, Liang Peizhi believes
it was in the late Ming. In interviews with wooden fish book publishers, he found
that they had seen woodblocks carved as early as the late 1500s (1978: 245-47).
See also Tan and Tan (1982: 2). Gazetteers also preserve some evidence of
distinctive Cantonese writing. For example, New Words about Guangdong (Guangdong
Xinyu JJDIl:*'Tj'fg) by Qu Dajun OHljc±IJ) (1630-1697) has a section on "local words"
(tu yan ± 1r) which includes a number of Cantonese terms. However, the section
fails to list many of the common Cantonese terms which were already appearing
in wooden fish books, and is thus not a very good indicator of the state of the
written language at the time.
40. Rawski (1985a: 25-27).
41. Fu (1949: 42); Liang (1978: xiv, 226, 247-52).
42. Topley (1975: 75); Liang (1978: 232); Rawski (1979: 6); Xue (1985a: 2).
43. Tan and Tan (1982: 2, 16); Xue (1985a: 1); Hu ([1926] 1974: 24-25).
44. Liang (1978: xiii); Xue (1985a: 107).
45. Chen (1998: 289); Chan Sauyan, interview, 12 March 2003; Wong Yiu-kwan,
interview, 26 April 2003.
46. Xue (1985a: 108-9). There is not complete agreement as to how the term "wooden
fish song" (C: muk jyu go *:*Jf1q should best be used. Some scholars use it as a
general term covering a broad variety of genres (Tan and Tan 1982: 19-20), while
others use it in reference to a particular genre of long narrative song (Xue 1985a:
108-9; Liang 1988: 18; Yung 1989: 138). The confusion arises from the fact that
there is little difference between the verse structures of the wooden fish songs and
southern song genres (Yung 1989: 193).
47. Yung (1989: 138); Xue (1985a: 1-2).
48. Liang (1978: 256-7); Tan and Tan (1982: 17); Xue (1985a: 110); Yung (1989: 139).
49. The English title, The Flowery Scroll, is taken from the translation by Sir John
Bowring published in London by Wm. H. Allen and Co. in 1868.
50. Xue (1985a: 4); Liang (1988: 18);Ye (1996: 59).
51. Tan and Tan (1982: 17); Xue (1985a: 2). There may have been two authors; see
Xue (1985a: 4) and Ye (1996: 59) for discussion.
52. For explanation of what constitutes "marked Cantonese" see Chapter 3, and also
the introduction to Appendix 1.
53. Xue (1985a: 2). One interesting measure of the influence of the work is perhaps
the number of times it has been translated into Western languages. A prose
translation into English was published by Sir John Bowring as The Flowery Scroll
(London: Wm. H. Allen and Co., 1868). Earlier verse translations into English
( Chinese Courtship by P. P. Thomas) and Dutch (Geschiedenis van het gebloemde Brief
Papier, by "Mr Schlegel") had also been published (Bowring 1868: v-vi).
54. For discussion of different theories, see Chen (1998: 280).
55. Liang feels that local folk songs formed the base of the southern song genre, and
discusses the arguments of other scholars who feel that the southern song owed
more to other genres imported from outside Guangdong. In either case, the
consensus is that elements of both local and outside genres went into the formation
of southern songs. See Liang (1988: 31-33) for further discussion of this theory.
For alternative theories, see Chen Zhiqing (1999: 19).
56. Fu (1949: 44); Liang (1988: 24).
268 Notes to pages 83-89

57. For example, the Xiangjiang Wanbao, a Hong Kong newspaper which was published
between 1922-24, included a southern song, a dragon boat song, or a Cantonese
love song in its entertainment section almost every day.
58. Fu (1949: 44); Xu (1958: 30); Liang (1978: 256-7); Tan and Tan (1982: 17); Yung
(1989: 138).
59. This was the highest national imperial examination.
60. For discussion of this theory, see Liang (1988: 85,98-99) and Tan and Tan (1982:
23-25). Another possibility is that the author was a native of Hangzhou writing
under the name Liao Lianxian (Ye 1996: 53).
61. Here and following I use the term "Standard Chinese" to refer to a written
language which was somewhere between Classical Chinese and modern Mandarin.
As Hanan notes, early "Baihua" texts were actually written in a mix of Baihua and
Classical Chinese (1981: 6-7). My point is to distinguish between Cantonese dialect
and forms of written Chinese used widely throughout China.
62. Interview, 17 May 1990.
63. Wong Yiu-kwan, interview, 26 April 2003. See Yung (1989, Chapter 9), for
discussion of padding syllables as used in Cantonese opera.
64. Liang (1988: 19); Chen Zhiqing (1999: 20); Yung (1989: 138); Ye (1996: 49-50).
65. Fu (1949: 44). This is not to say that dragon boat songs were never published.
One collection from the late 1800s, Improved Lingnan Notes (C: Goiloeng Lingnaam
Ziksi rj( IHJtl¥i lW~) contains a number which are very colloquial. (Reprint; Macao:
Xinsheng chubanshe, pp. 24-28).
66. Liang (1988: 24-25); Chen Zhiqing (1999: 20).
67. Liang (1988: 149, 151).
68. This was the degree granted for passing the imperial examination at the provincial
level.
69. Jiang and Fang (1993: 498). For biographical information, see Tan and Tan (1982:
24); and Liang (1988: 156).
70. Chen Zhiqing (1999: 23). The translation of Jyut Au as "Cantonese Love Songs"
follows Sir Cecil Clementi's 1904 translation and also Peter Morris' 1992
translation. Liang (1988: 165) notes that there was at least one other writer of
Cantonese love songs who was a contemporary of Zhao Ziyong, a holder of the
jinshi degree named Feng Xun (l~mu). Unfortunately, however, Feng's works are
now all lost.
7l. See Liang (1988: 215) for discussion.
72. Liang (1988: 228-9).
73. Clementi (1904: 1); Chen (1986: 8-9). Over ten editions of the Cantonese Love Songs
have survived, and according to Chen (1986: 5-6) even non-Cantonese speakers
have read and enjoyed these verses. Hu Shi, for one, was familiar with them and
cited them as the leading example of Cantonese dialect literature ([1926] 1974:
27).
74. For text, see Liang (1988: 235-6).
75. This single case is the usage of zai as the preposition of place, and even this case
is not clear. While hai is the more common preposition of place in Cantonese,
even today sometimes zai (C: zot) will be heard, and it often appears in texts which
are otherwise quite purely Cantonese, a fact that would tend to indicate that it is
generally not considered marked Standard Chinese.
76. The edition of Cantonese Love Songs found in volume 56 of the Folklore and Folk-
 Notes to pages 89-94 269

literature Series edited by Lou Zukuang [Lou Tsu-k'uang] (Taipei: Orient Culture
Service, 1971) includes the dialect term glossary, which explains 74 Cantonese
words and expressions. Unfortunately, while More Cantonese Love Songs is included
in the same publication, its dialect glossary is not included.
77. Yet a third possibility is that the glossaries reflect the needs of Cantonese people
living in Shanghai, Beijing and other places who might have been interested in
such works but not gained literacy in Cantonese. Chan Sau-yan, interview, 12 March
2003.
78. A substantial literature in Cantonese was produced by missionaries in the late
1800s, written for the most part in Chinese characters, and this book may be an
example of such efforts. Ball (1894: vii-xvi) provides an annotated bibliography
of 64 texts either written in or translated into Cantonese by missionaries.
Unfortunately, Ball provides no information as to how widely these texts were
circulated.
79. By contrast, the romanization system for the Xiamen (Amoy) dialect of Fujian
devised in 1850 still plays a role in the church in Taiwan today, and outside the
church romanized Taiwanese is now advocated by a number of writers as the best
system for writing down Taiwanese words for which there are no agreed upon
characters (Zheng 1988b: 32).
80. This text is reproduced in volume 122, set 7, of the Folklore and Folk Literature
Series of Peking University and Association for Chinese Folklore edited by Lou
Zukuang [Lou Tsu-k'uang] (Taipei: Orient Cultural Service, 1974). While no
publication information is given, it was probably published in 1929, the same date
as other materials in the volume. It could not be much earlier as Nanjing is referred
to as China's capital in the text.
81. For biographical information on these two men, see Luk (1984).
82. This was the degree granted for passing the imperial examination at the county
level.
83. There is some doubt here, because the copies in my possession were rescued from
a library garbage heap some years ago by David Faure, and the two texts which I
discuss in this paragraph mayor may not have been published in one single bound
volume along with Chen Ronggun's Essentials for Women and Children. The string
binding had deteriorated, so it was no longer clear what exactly had been bound
to what; they seem to have originally been published separately, and then
republished together in 1919 by the Shanghai Geyan Congjishe. The similarity of
the titles of these books to Chen's other texts also suggests that Chen was the
author.
84. Ching (1996: 56).
85. Lai and Huang (1988: 12).
86. Liang (1982: 7, 174); Liang (1988: 259); Yung (1989: 9).
87. Liang (1982: 153-4). For the story ofLi Wenmao, see Lai and Huang (1988: 13-
17); also Lai (2001: 127-31).
88. Liang (1982: 55-56).
89. Liang (1988: 259-60).
90. Lai and Huang (1988: 2~27); Lai (2001: 206).
91. Liang (1982: 174, 179); Liang (1988: 20).
92. Liang (1982: 179); Liang (1988: 259-60).
93. Liang (1982: 177).
270 Notes to pages 94-/05

94. Huang (1971: iv, ix).

95. Yung (1989: 18).
96. The dialogue in this particular selection is entirely spoken rather than sung, but
the general pattern of dialect use shown is not unique to spoken dialogue sections
in Cantonese opera.
97. Hayes (1985: 91) describes this as an early twentieth-century collection, but it is
older. In 1904 Ball guessed that it was published "in the last century or two" (xxi).
Ogaeri dates it sometime after 1871 (1964: 172), and Ye (1996: 315) dates it to
1864. Huang (2002: 53) suggests this was the first novel written in a style called
saam kap dai that combines Classical Chinese, Standard Chinese, and Cantonese.
For more on saam kap dai, see Chapter 6.
98. Ye (1996: 315).
99. Ball (1904: xxii).
100. Ye (1996: 305). Hayes also mentions a collection of stories called Xun huan jian
which are written to some extent in Cantonese and, unlike Casual Chats in the
Colloquia~ actually set in Guangdong (1985: 91).

CHAPTER 5
1. As a very large percentage of the Chinese laborers working in the United States
were from the Pearl River delta region, it is not surprising that Guangzhou became
one of the main centers of the boycott movement. See Liang (1988: 266-74) for
a detailed account of the role of traditional verse forms in the boycott.
2. Liang (1988: 267).
3. Ye (1996: 52).
4. Xiangjiang Wanbao, 4June 1922, p. 4.
5. Xiangjiang Wanbao, 3 May 1922, p. 1.
6. This journal only lasted a few months and ceased publication in 1933 when Ouyang
Shan and other leftists had to flee Guangzhou (HuaJia, interview, 18June 1990).
The journal evidently had little influence (Jing 1949c: 6).
7. Mao Dun was in Hong Kong in 1938 and was the editor of one of Lih Pao's feature
sections (Lu 1987: 145).
8. See page four of the 5 July, 12 July, and 14 November 1938 issues of Lih Pao
respectively.
9. Lih Pao, 9 July 1938, p. 4.
10. Despite the apparent lack of success of the Lih Pao experiment, it may have had
some influence on the Dialect Literature Movement through Mao Dun. In his role
of editor on the Lih Pao staff, he would almost certainly have been aware of the
column, and in his postwar involvement with the Dialect Literature Movement it
is reasonable to assume that he would have passed this experience on to other
writers more deeply involved in the movement than he was.
11. It should be noted, however, that some early proponents of Baihua, including Qian
Xuantong (~1rIRJ) and Hu Shi, had a relatively positive attitude toward the
inclusion of dialect terms in Baihua (Gao 1993: 63-64).
12. Ramsey (1987: 12).
13. Gunn (1991: 46); Holm (1986: 7-8).
14. Gunn Piing (1991: 46); Chen Ping (1999: 80); Ramsey (1987: 13); Gao (1993: 119).
15. Gao (1993: 117-8, 135).
 Notes to pages 105-113 271

16. Holm (1986: 8-10, 13-15).

17. For a detailed summary of the history of the DLM, see Huang (1988: 158-72).
18. Jing (1949c: 6). It later became the Dialect Literature Research Society (Fangyan
wenxue yanjiu hui 1J1f:>C¥liffJ"Elt).
19. Liu (1997: 129).
20. HuaJia, interview, 18 June 1990; Lou Qi, interview, 19 June 1990.
21. Jing (1949c: 6) lists seventeen names of writers of either theoretical or literary
works - eighteen including himself. "DLM participants" HuaJia and Lou Qi stated
that even they did not know who many of these writers were, citing frequent use
of pen-names as one reason. Interviews, 18 June 1990; 19 June 1990.
22. Feng and Quan (1948: 48); Hua (1949: 40).
23. Cited in Huang (1949: 18-19); Mao (1948: 16).
24. Liu (1997: 127).
25. HuaJia, interview, 18June 1990; Lou Qi, interview, 19June 1990.
26. Liu (1997: 137).
27. Liu (1997: 140).
28. See the "People's Wind" (Min feng ~Ji.) section of the 17 March 1949 issue of
Sing Tao Jih Pao and the "Literary Weekly Supplement" (Wenshi zhoukan :>C.'1::J!J
til) section of the 17 January 1948 issue of Wah Kiu Yat Po for examples. While it
seems odd that papers which tended to lean to the right politically would publish
material written by leftist writers, the editors of the supplement sections tended
to be more liberal than the other editors (Hua Jia, interview, 18 June 1990).
29. Jing (1949c: 6).
30. Fung Gwong-Iit, interview, 11 July 1990.
31. Lin (1947: 7); Hua (1947: 22).
32. Lou Qi, interview, 19 June 1990.
33. Ta Kung Pao, 9 March 1949, p. 7; Fangyan Wenxue (1949: 146-50).
34. First installment, Zheng Bao, 29 November 1947, p. 27; Ta Kung Pao, 29 June 1949,
p. 7. Also in Hua (1949: 117-37).
35. Ta Kung Pao, 9 March 1949, p. 7; and Ta Kung Pao, 23 March 1949, p. 7. Also in
Hua (1949: 61-67).
36. An interesting parallel is Hu Shi's famous 1917 article "A Preliminary Discussion
of Literary Reform" (Wenxue gailiang chuyi :>C¥i!)( ~~~) which advocated that
China should replace Classical Chinese with Baihua. The article itself was written
in Classical Chinese.
37. Hua (1947: 5), Hua (1949: 22);Jing (1949a: 4).
38. Chen Canyun's "Third Aunt Yang" is one rare example.
39. Hua (1949: 40).
40. Hua (1947: 10); Hua (1949: 40); Jing (1949b: 7); Huang (1949: 26).
41. Feng and Quan (1948: 46). The term "mosquito press" is borrowed from Link
(1981: 118). For discussion ofthe mosquito press in Hong Kong, see Chapter 6.
42. Cited in Mao (1948: 16). One issue which the DLM writers did not discuss was
the extent to which these other "dialect" works were actually written in dialect. As
we shall see in Chapter 6, the dialect works published in the mosquito press were
written in a mix of Standard Chinese and dialect, and were thus quite different
in their language from the works of the DLM writers.
43. Frankly, I must confess some sympathy for Lan Ling as I am also not entirely sure
what the title means. In a later Zheng Bao article (IS November 1947) HuaJia
mocks Lan Ling for being unable to read the title, but Hua never explains it either.
272 Notes to pages 114-121

44. Zheng Bao, 8 November 1947. Ironically, however, the letter itself is written in rather
formal Standard Chinese.
45. Huang (1988: 161).
46. HuaJia (1949: 34-35).
47. HuaJia, interview, 18June 1990; Fung Gwong-lit, interview, 11 July 1990.
48. Holm (1986: 13).
49. Feng and Quan (1948: 53).
50. Fu (1947: 16); Feng and Quan (1948: 47); Mao (1948: 17).
51. Hua (1949: 22); Liu (1949: 6).
52. To the best of my knowledge, neither Lan Ling nor Lin Luo was able to publish
anything related to the DLM after 1947, and in Jing Wen's 1949 list of the
movement's major theorists they are not included despite the amount of debate
generated by their 1947 articles (1949c: 6). Even though they were silenced fairly
early in the movement, the fact that in 1949 HuaJia still felt the need to refute
their views suggests that other writers shared their feelings on this issue.
53. See HuaJia's "Counting Dead Grass" (C: Syun sei cou .jE1j!.) in Hua (1949: 100-
16); Chen Canyun's "Third Aunt Yang" (Ta Kung Pao, 9 March 1949, p. 7); and
Gan Niu's "The Strange Case of the Guangzhou Baby Murder" (C: Joengseng saat
jing kei on $~~lJUf~) (First installment, Zheng Bao, 3 April 1948, p. 27).
54. Huang (1988: 167-8).
55. Hua Jia, interview, 18 June 1990.
56. Hua (1987: 7).
57. Hua (1949: 63-64); Huang Sheng (1949: 24).
58. Hua (1949: 34-35).
59. Lou Qi, interview, 19 June 1990.
60. Hua (1949: 61).
61. Hua (1949: 62).
62. Lou Qi, interview, 19 June 1990.
63. Hua Jia said he did not recognize any of the names of the writers who wrote dialect
works in the Nanfang Ribao in 1950 and 1951. Interview, 18 June 1990.
64. Fung Gwong-lit, interview, 11 July 1990.
65. Of the writers and newspaper personnel I talked to in Hong Kong, most had never
even heard of the DLM, or remembered it only vaguely. The exceptions were all
people who had been involved in leftist politics, such as Cheung Chor, editor of
the Hong Kong Commercial Daily. As a youth he studied the DLM works while
working in a factory, and even wrote dragon boat songs in that style. His interest
in dialect writing, and the influence it has had on Hong Kong Commercial Daily's
policies toward dialect material, were the one concrete legacy I was able to find
of the DLM's influence on the growth of written Cantonese. Cheung Chor,
interview, 10 July 1990.
66. For the beer advertisement, see Narifang Ribao, 10 March 1951.
67. DeFrancis (1984: 226).
68. Peterson (1994: 934-5).
69. Yu et al. (1979: 462); Hayford (1987: 167).
70. For example, see Chen Canyun's 1960 "Duck-House Notes" (Yaliao ji shi ~~~
~). There are four dialect terms used in a 13-page story, all footnoted. In
Guangdong Zhong-duanpian Xiaoshuo Xuan (Jjt~ i=fmJt /J\~j1!), vol. 2 (Guangzhou:
Guangdong renmin chubanshe, 1979), pp. 33-46.
 Notes to pages 121-131 273

71. Yang Chonghua, interview, 18 June 1990.

72. Dai (1987: 24).
73. Tang and Tu (2000).

CHAPTER 6
1. Fonoroff (1988: 297).
2. LiJiayuan (1989: 31).
3. See Link (1981: 18-19) for discussion of the mosquito press in Shanghai.
4. Choi notes the predominance of Cantonese people, especially those from Hong
Kong, in the early Mandarin as well as Cantonese film industry, and attributes this
to the greater familiarity of the Cantonese with Western culture (1990: 63).
5. Various explanations have been offered for the origin of the term saam kap dai.
At a dinner with several of the "elder statesmen" of Hong Kong's newspaper world,
Hon Zung-syun (.'PiiJE), editor of Sing Pao Daily; Leung Tol (mIl), best known
as the columnist Lou Gam (.~); and the columnist Lau Sing (~tl~), the
following two explanations were offered to me. One possibility is that the term
may somehow be derived from the traditional Chinese examination system. In
Cantonese, the three top places in the provincial examinations are known
collectively as the (C) saam jyun kap dai (-='j(;'&~), or (C) saam kap dai (-=..&~)
for short. A second explanation is that the term comes from a kind of porridge
eaten in Guangdong which is made with three different kinds of pig innards and
is called (C) saam kap dai zuk (-=..&~~) or (C) kap dai zuk (.&~~). A final
possible explanation, suggested in Huang (2002: 3) is that the term originally
referred to badly cooked rice that had three levels: uncooked on top, cooked in
the middle, and burned on the bottom.

yi*,
6. The Classical Chinese elements are generally particles, such as (M) ye -tE and (M)
used at the ends of sentences, and (M) yue B (to say) which is normally
used to introduce quotations.
7. Such as (M) yue B, (M) yun z;; , (M) yi
8. LiJiayuan (1989: 29-30).
*, and ke fou PJ~.

9. Mao (1948: 6); Feng and Quan (1948: 46);Jing (1949a: 10).
10. Even in 1938, (C) Jam Wu-faa (fffii:ft:.), the same man who was later to start Hung
Lok, started a short-lived newspaper called Xian Dao (JIG.) which made
considerable use of saam kap dai. Huang (2002: 92).
11. The focus of the paper on the working class is suggested by one regular column
called "A Record of Workers' Grievances" (C: Daagungzai soujyun luk trlff~~
~). As the writer of the 10 August 1947 column asserts, Hung Lok is the only place
a poor worker can go to have his case heard.
12. One interesting point is that the word (C) saau Pl'l (to skim - in reading) is
marked in brackets, probably indicating that this is a term the author felt readers
might not be familiar with, and that they needed to be warned to pronounce this
word rather than read it for its normal meaning, "to whistle."
13. This last-mentioned feature was cited in an article by Mao Dun as an example of
the success of many urban writers (shimin zuojia mB.Ht~) (1948: 6).
14. To some extent, popular kungfu novels (wuxia xiaoshuo ~f!JvHm.) were also a part
of the history of published Cantonese in this period. While most novels in this
genre did not use Cantonese, the books of authors such as (C) Chan Ging (~!fYJ;
274 Notes to pages 732-138

pen-name Ngo si saanjan 1'Jt~ ill A, "The Mountain Man"), were written in a saam
kap dai style that included some Cantonese.
15. LiJiayuan (1989: 63).
16. Fung Gwong-lit, interview, 11 July 1990.
17. Fung Gwong-lit, interview, 11 July 1990.
18. This honor may belong to odd opinion columns written by (C) Gam Ngaaji (~
3f =) in the mosquito press newspaper Xian Dao in the late 1930s. Huang (2002:
92).
19. For a collection ofSaam So's articles, see Selected Odd Opinions of San Su (C: Saam
So gwaaileon syun -=:,*'~liifU~) (Hong Kong: Zuojia shuwu, 1975). For English
translation of one of Saam So's odd opinion articles, see Renditions 29 & 30: 68-
70.
20. For example, (C) Haa Gong (*0), one of Hong Kong's better-known odd
opinion columnists, more often wrote without using Cantonese than with it.
21. Interview, 3 July 1990.
22. Gou Dak-hung (Ginggei Laa), The Stockbroker's Diary (C: Ginggei jatgei ~*c. S ae)
(Hong Kong: Dagong shuju, 1953).
23. For a collection of Haa Gong's odd opinion articles, some of which contain
Cantonese, see Haa Gong, Haa Gong's Odd Opinions (Haa Gong gwaaileon *0'~
liifU) (Hong Kong: Xingji chuban youxian gongsi, 1986). For English translations
of Haa Gong columns, see Renditions 29 & 30: 320-28; and Seeds of Fire: Chinese
Voices of Conscience, edited by Geremie Barme and John Minford (New York: Hill
and Wang, 1988).
24. Among the mass-market newspapers, through the 1980s, Oriental Daily published
a column called "Third Uncle's Odd Opinions" (C: Saam Suk gwaaileon -=:*~'~liifU),
later replaced by "Brother Liao's Odd Opinions" (C: Liu Go gwaaileon T{If'~liifU).
Sing Pao Daily ran the column "Chats" (C: Taam tin ~7C), and Tin Tin Daily had
one called "Sound from Beyond the String" (C: Sin ngoi jam ~?T~{f). Among leftist
newspapers during the same period, Wen Wei Po ran a column called "Ms Lu's Odd
Opinions" (C: Luk neoisi gwaaileon 1lt:9:±'~liifU) which had some of the
characteristics of the odd opinion genre. A column which appeared in Ta Kung
Pao in the mid-1980s, "Go Naa's Notes" (C: Gonaa saugei ~JF¥ae), was also
arguably at least descended from the odd opinion tradition. While it used little
Classical Chinese, it did introduce direct speech with the classical yue B, and while
the author did not refer to herself by name, she continued the tradition of strong
narrator presence by writing in the first person.
25. Interview, 3 July 1990.
26. Interview, 10 July 1990.
27. For example, as of the early 2000s, Oriental Daily Olt:n S ¥l) had an odd opinion
column entitled "The Crow's Odd Opinions" (Wu Ngaa gwaaileon ,~~'~mfU), and
Apple Daily OijL* S ¥l) had a column entitled "GG's Tidbits" (GG sai jyu GG #.lIB
l'ia) with all the traits of an odd discourse column.
28. Ng (1989: 392); Rafferty (1989: 44).
29. Vogel (1989: 58); Rafferty (1989: 65).
30. Cheung (1985: 99).
31. Luk (1989: 55-56).
32. Chan and Kwok (1986: 407); Chen (1989: 33).
33. Zou (1989: 3).
 Notes to pages 138-147 275

34. Vogel (1989: 52-3, 61).

35. Siu (1996: 83).
36. Lau and Kuan (1988: 2, 181).
37. See, for example, Bolton and Kwok (1990: 65); Hong Kong Federation of Youth
Groups (1996: 7); Public Opinion Programme (1998); Brewer (1999: 94); Tong
et al. (1999).
38. Choi (1990: 70-1).
39. Chan and Choi (1989: 294).
40. Choi (1990: 72).
41. Fonoroff (1988: 306).
42. Choi (1990: 68-170).
43. Fonoroff (1988: 306); Choi (1990: 09, 73).
44. Liu (1984: 21). Kwan Wing-kei noted that it was only in the 1980s that songs in
Cantonese gained widespread acceptance. Interview, 25 May 1990.
45. Siu (1996: 84).
46. Ng Chun-bong, interview, 3 July 1990; Chan Hing-kai, interview, 10 July 1990.
47. Choi (1990: 73). For example, the (C) mou lei tau 1m1!UIi "nonsense humor" style
of Chow Sing-chi's (Zau Sing-ci )i!f]J:iL~I!l) films are rich in such word play.
48. J. Chan (1987: 2-13).
49. Chan and Choi (1989: 312).
50. Steven Yao, interview, 25 May 1990.
51. Cheung Chor, interview, 10 July 1990.
52. Fung Gwong-lit, interview, 11 July 1990. Fung noted that the newspaper considered
Cantonese acceptable in columns which are not considered very serious, mainly
those intended for entertainment.
53. Rafferty (1989: 31-32).
54. Leung Tol, interview, 3 July 1990. Leung noted that many of the early "horse
classic" writers were more familiar with horses than with the rules of proper written
Chinese, a judgment shared by Cheung Chor, interview, 10 July 1990. Leung also
argued that because both races and soccer matches were broadcast over the radio,
people were accustomed to hearing reports of these events in Cantonese and were
thus more accepting of Cantonese in written reports as well.
55. See Hong Kong Economic Journal 3 July 1980, p. 3, for example.
56. Hong Kong Commercial Daily, 5 January 1970, p. 2.
57. See Hong Kong Commercial Daily, 4January 1975, p. 1, for example.
58. In the earliest issue of Tin Tin Daily I was able to locate (1 April 1978), there was
also a Playboy advertisement with a Cantonese line in large print, and two large
advertisements (which had Cantonese in a speech bubble) for a restaurant called
the "Foreign Devil Street Stall" (C: Gwailou daai paai dong *1~*~:fi).
Unfortunately, Tin Tin Daily did not fulfill the promise of this colorful start, and
later issues show its Cantonese advertisements to be rather more standard and
tame.
59. Much of my information about comics in the 1970s comes from a 1974 report
entided "Sex and Violence in Comics" (Gongzaishu zhi baoli ji seqing0fHlZ~jJ
lH~:If).
60. Gongzaishu (1974: 3, 22-28).
61. Chan Hing-kai, interview, 10 July 1990.
62. Andy Wong, interview, 16 July 1990.
276 Notes to pages 148-159

63. Figures for number of columns containing Cantonese per day are approximate.
Features regularly appear and disappear in Hong Kong newspapers, and colUmnists
are not always consistent in their choice of style or language.
64. Chan and Choi (1989: 312).
65. To some extent, the relatively large number of Cantonese features in top-selling
mass-market newspapers is accounted for by the fact that the mass-market papers
tended to have many pages, hence more features than some other newspapers
such as Express or Jig Yip Ma Po. However, some newspapers with relatively few
Cantonese features, such as Mzng Pao Daily and Sing Tao Jih Pao were also large
newspapers with many pages, so in these cases the difference in total Cantonese
use was accounted for by policy rather than size of the newspaper.
66. Far Eastern Economic Review, 27 September 1990, p. 26.
67. Chan and Choi (1989: 312).
68. Leung Shu-chiu, interview, 9 July 1990. Hon Zung-syun, editor of Sing Pao Daily
also attributed the success of his paper (it had the second largest circulation in
Hong Kong through the 1980s) to the fact that it was responsive to the culture of
Hong Kong, a culture which he explicitly argued was not simply Chinese but rather
something unique and distinctive. Interview, 7 July 1990.
69. Steven Yao, interview, 25 May 1990.
70. Sing Pao Daily, 13 September 1989, p. 32.
71. Tin Tin Daily, 13 September 1989.
72. Like odd opinion articles, "Notes" articles are also generally written by upper-level
members of the newspaper staff. Cheung Chor, interview, 10 July 1990.
73. See Nan BeiJi, 16 October 1982, p. 2, and 16 March 1986, p. 8, for examples.
74. It is not unusual even for Hong Kong publications which are written entirely in
Chinese to have an English title along with a Chinese title. In fact, for some Chinese
publications like Movie and Fun Magazine, the English title is displayed far more
prominently on the cover than the Chinese title.
75. Cheung (1985: 97).
76. See City, October 1984, p. 8; December 1984, p. 8; and October 1989, p. 46.
77. See, for example, the January 1990 special issue on young Yuppies. Incidentally,
the English word ''Yuppie'' is generally used, even in Chinese text.
78. Cheung (1985: 91).
79. City, January 1980, p. 91.
80. City, December 1982, p. 30. For another article defending City's liberal approach
to use of dialect, see Ye (1981: 82).
81. In a 12 June 1990 interview with Joseph Yau, one of the founders of City and a
frequent contributor during its early years, he noted that the magazine never had
a formal language policy and that there was considerable diversity of viewpoints
within the staff. Support for the views he presented in his December 1982 article
was by no means unanimous.
82. Chan Hing-kai, interview, 10 July 1990.
83. Liang (1989: 5).
84. Among the better-known pocketbooks writers are numbered one college professor,
a CPA, and a middle school Chinese teacher. To the best of my knowledge, most
pocketbooks writers are graduates of tertiary institutions.
85. Luke (1995: 102).
86. Kwan Wing-kei, interview, 25 May 1990.
 Notes to pages 160-172 277

87. Kwan Wing-kei, interview, 25 May 1990.

88. Kwan Wing-kei, interview, 25 May 1990.
89. Chan Hing-kai, interview, 10 July 1990.
90. Kwan Wing-kei, interview, 25 May 1990.
91. Liang (1989: 5).
92. Kwan Wing-kei, interview, 25 May 1990.
93. Lo Wai-Iuen, interview, 13 June 1990; Luke (1995: 1).
94. Liang (1989: 6).
95. Interview, 10 July 1990.
96. Interview, 10 July 1990.
97. Ng Chun-bong, interview, 10 July 1990;Joseph Yau, interview, 12June 1990; Chan
Hing-kai, interview, 10 July 1990.
98. Tsui (2000: 3).
99. Lee (1992: 398).
100. Public acceptance of use of Cantonese in quotations now seems quite widespread.
For example, one school teacher I interviewed noted that while she is generally
opposed to use of written Cantonese in the media, she can accept its use in direct
quotations as long as no foul language is involved. Fok Lai-zan, interview, 24
September 2003.
101. Chen (1993: 216). Schools generally try to prohibit students from reading Yes! and
similar magazines, though this is generally more because of their content than
because they use so much written Cantonese. Kwok Wing-ki, interview, 14
September 2002. Chung Ling-sung, interview, 26 September 2002.
102. 21 July 1992 issue.
103.25 February 2000 issue.
104. Chin (1997: 87).
105. Shi (1999: 56).
106. Shi (1999: 2).
107. Chin (1997: 82).
108. K. K. Luke, interview, 26 September 2002; Stephen Matthews, interview, 26
September 2002; Leung Wai-mun, interview, 26 September 2002.
109. Bauer and Mo (2001: 1).
1l0. Huang (2000: 354) lists more than forty monolingual Cantonese dictionaries that
have been published since the 1930s, more than half since 1990. Many of these,
however, have been produced in mainland China, and have thus had relatively
little impact on the development of written Cantonese in Hong Kong.
Ill. K. K. Luke, however, notes that the Hong Kong Style Cantonese Dictionary has fewer
entries than the Rao et al.'s Cantonese Dialect Dictionary (1981), and is thus unlikely
to thoroughly replace it. Interview, 26 September 2002.
112. As Stephen Matthews points out, the fact that the grammar does not include
Chinese characters limits its appeal to Chinese readers, so most buyers have tended
to be English-speakers who are studying Cantonese. Interview, 26 September 2002.
113. Hu (1994: 25). Increasing interest in this romanization scheme from the Hong
Kong government may eventually lead to a larger role for it in Hong Kong, as
may its future adoption by software designers as an input method for computers.
K. K. Luke, interview, 26 September 2002; Stephen Matthews, interview, 26
September 2002.
114. As of 2004, the website is: www.info.gov.hk/digitaI21/ eng/hkscs/ download.html.
278 Notes to pages 172-182

115. See Yau (1992: 6) and Chin (1997: 80) for expression of such concerns.
116. I make this claim realizing full well that Cantonese has so far only "proven itself'
in light entertaining texts. However, in theory at least, formal texts written in
Cantonese should be easier to read than colloquial Cantonese texts because formal
Cantonese is closer to Standard Chinese than is colloquial Cantonese (Bauer 1988:
249).

CHAPTER 7
1. Joseph (1987: 38); Crystal (2000: 138).
2. Ho (1976: 551-4); Harrell (1993: 93-94); Lewis (1999: 339); Hansen (2000: 104).
3. Ramsey (1987: 3-4); Fitzgerald (1996: 13).
4. DeFrancis (1984: 225).
5. DeFrancis (1984: 226); Hayford (1987: 167).
6. Fitzgerald (1996: 14).
7. See Li Yunhan (1997: 190) and Pei (1997: 34) for views on the proper use of dialect
in Chinese literature.
8. Chen Ping (1999: 118).
9. Chan Chee-shing (1987: 242); Yu (1987: 229): Kwan-Terry and Luke (1997: 291-
2).
10. Fung Gwong-lit, interview, 11 July 1990; Leung Tol, interview, 26 June 1990;
Cheung Chor, interview, 10 July 1990. See also Chen (1985: 85) and Deng (1980:
86).
11. Kwok Wing-ki, interview, 14 September 2002; Fok Lai-zan, interview, 24 September
2002. For examples of such accusations, see Yin (1981: 38-40); Yang (1982: 36);
Lin (1986: 116); Liu (1986: 79): He (1987: 39).
12. Chan Man-hung, interview, 12 July 1990; Robert Lord, interview, 17 July 1990.
Kwan-Terry and Luke (1997: 292).
13. Fu (1975: 85-86; 1987: 28-29); Liang et al. (2001: 2).
14. T'sou (1985: 16); Zou [T'sou] (1989: 3).
15. Si (1981: 101); Tong (1989: 14).
16. Deng (1980: 87); Chan Chee-shing (1987: 243); Chen (1988: 18); Li (1988: 42).
17. Chang (1981: 61-62); Chan Chee-shing (1987: 235).
18. Deng (1980: 87); Li (1988: 41-42). While Deng does not stress the value ofa
knowledge of spoken Mandarin in the process of learning to write Standard
Chinese, he does note that knowledge of written Standard Chinese would facilitate
students' acquisition of spoken Mandarin.
19. Kang (2000: 7-9); Yu (2001: 2).
20. Zhong and Zhang (1993: 17); Kwo (1992: 205); Huang (1998: 3); Liang et al.
(2001: 4).
21. See also Yang (1982: 37); Fu (1987: 29); Zhou (1987: 71); Deng (1989: 3).
22. Deng (1989: 3).
23. Kang (2000: 10); Yu (2001: 2).
24. See also Kwo (1992: 209); Yu (2001: 3). As more teachers in Hong Kong have
studied Mandarin (Putonghua) in recent years, the ease with which they can
determine what is and is not acceptable as Standard Chinese is also increasing.
Chung Ling-sung, interview, 26 September 2002.
25. Yang (1987: 53).
 Notes to pages 182-186 279

26. Liang (1987: 66). Similarly, Zheng Liangwei points out that for more than twenty
years after 1945, Taiwan's noted authors were almost all native Mandarin speakers
from outside Taiwan, and feels that even now the ability of Taiwanese authors to
break into the creative writing scene is somewhat hampered by their non-native
Standard Chinese skills (1984: 5).
27. Interview, 3 July 1990.
28. Interview, 13 June 1990.
29. Interview, 18 June 1990. This view was echoed by Guangzhou publisher and author
Yang Chonghua (interview, 18June 1990), and also by the Dialect Literature
Movement writer Lou Qi, who commented on the unusual difficulty Cantonese
speakers seem to have in learning to write Standard Chinese (interview, 19 June
1990).
30. Choi Po-king, interview, 29 November 1989; Joseph Yau, interview, 12June 1990.
31. As LePage and Tabouret-Keller (1985: 188) note: "For some people in the world
it is true that one particular variety of linguistic behavior has a peculiar force and
intimacy from being powerfully associated with early childhood ... "
32. Zhong (1994: 227). As David Li points out, an additional reason why Hong Kong
people may prefer to use Cantonese terms to Standard Chinese equivalents is that
the Standard Chinese "equivalents," having arisen in a cultural context different
from that of Hong Kong, may actually not mean exactly the same thing as the
Cantonese term in question (personal communication).
33. Writers to whom I spoke generally felt that the only real difficulty involved in
learning to write in Cantonese was that of choosing characters for less common
dialect terms. Joseph Yau (pen-names Gu Sai-mung, Wu Gun-man), for example,
noted that the only composition problem he faced when working in Cantonese
was that of finding characters for obscure dialect words, a problem he can generally
solve by reference to a dictionary. Interview, 12June 1990. When faced with the
same problem, Ng Chun-bong (pen-name Ng Hau) said he often simply avoids
the obscure word, partly out of concern that if he does not know the character,
readers might not know it either. Interview, 3 June 1990. Chan Hing-kai (pen-
name A Foon) said he usually either spells such words out in romanization or
simply uses another character which has the same sound. Interview, 10 July 1990.
None of the above said they found learning to write Cantonese difficult.
34. This is the experience described to me by a number of people in Hong Kong.
Kwok Wing-ki, interview, 14 September 2002; Fok Lai-zan, interview, 24 September
2002; Leung Wai-mun, interview, 26 September 2002. Incidentally, it also describes
my own experience of learning to read written Cantonese.
35. Kwok Wing-ki, interview, 14 September 2002; Fok Lai-zan, interview, 24 September
2002; Leung Wai-mun, interview, 26 September 2002. Discussions I had with
university and middle school students elicited very similar responses.
36. As mentioned in Chapter 2, Hu Shi felt that heavy use of dialect was a m;yor reason
for the low readership of Shanghai Flowers; other Wu dialect novels which made
much more limited use of dialect sold much better. The sheer difficulty in reading
works written entirely in Taiwanese has also been a m;yor factor in their poor sales
to date.
37. Kwan Wing-kei, interview, 25 May 1990.
38. As of 1990, approximately 20,000 copies had to be sold for a book in mainland
China to break even. Yang Chonghua, interview, 18 June 1990. Yang stated that
280 Notes to pages 186-197

while a few books of the Guangdong publishing house are marketed mainly in
Guangdong, most are distributed nationally.
39. Luke (1995: 101).
40. Lin Qiyang, interview, 5 June 1990.
41. For example, the Diary of The Little Man radio scripts were later published as books
with little further editing. Chan Hing-kai, interview, 10 July 1990.
42. Kwan Wing-kei, interview, 25 May 1990.
43. Singer (1998: 56).
44. Giles and Coupland (1991: 96). See also Giles et al. (1977: 307).
45. Psychologist Marilyn Brewer argues that two of the most basic human motivations
are: (a) the need to be included by and assimilated with other people, a need
met within the in-group; and (b) the need to be differentiated from other people,
a need which is met by making distinctions between in-groups and out-groups
(1999: 188).
46. Tong et al. (1999).
47. Public Opinion Programme (1998).
48. For similar survey results, see also Brewer (1999: 194).
49. Hong Kong Federation of Youth Groups (1994, 1998,2000).
50. Rawski (1985b: 400).
51. Pierson (1994: 81).
52. Li and Thompson (1982: 87).
53. See Cheung (1985: 195-8); Bauer (1988: 282-3); Snow (1991: 260-82, 1993: 143);
Li (2000: 221-3); Wu (2000: 91).
54. Pierson (1992: 1295).
55. This belief that Cantonese has no grammar was one of the motivations behind
the writing of Cantonese: A Comprehensive Grammar. As Virginia Yip, one of the
authors, noted: "We want to debunk the myth that Cantonese has no grammar.
This is entertained by so many educated people - even our scientist friends and
students. They were so shocked and surprised to hear that we have produced a
whole book for Cantonese grammar." Quoted in Tsui (2000: 3).
56. Kalmer, Zhong, and Xiao (1987).
57. See Pierson (1994); Tong et al. (1999).
58. Report (1993: 3). See also Zhan (1997: 36).
59. Bauer (1988: 285).
60. Wu Changjiang, interview, 18 December 1991.
61. Kwan-Terry and Luke (1997: 289).
62. Bauer (1984: 298); Lo and Wong (1990: 29); Huang (1999: 171). For examples
of such attitudes, see He (1982: 45); Li (1982: 60); Yang (1982: 37); Li et al. (2000:
288); Tacey (2000: 9); and Tsui (2000: 3).
63. Liang (1987: 65); Li (2000: 222).
64. This quotation came from an informal survey I conducted in a university evening
school linguistics class. While many students in the class said they found the
language of a selection from Diary of The Little Man lively, all but three commented
that it was "sub-standard," "bad," or "unacceptable." Even those who, on the whole,
liked the passage felt compelled to note that the language was not appropriate.
See also Pierson (1992: 195).
65. See iI\l#l9='X (C: Gamsau Zungman, Splendid Chinese), Hong Kong: Education
Department, 2001.
 Notes to pages 191-202 281

66. Kwok Wing-ki, interview, 14 September 2002.

67. Rickard (1989: 19).
68. Barber (1993: 177).
69. Twine (1991: 28, 198-206).
70. DeFrancis (1977: 27, 37).
71. DeFrancis (1989: 198).
72. Rawski (1985b: 400).
73. DeFrancis (1984: 249).
74. Hu (1989: 72); Dai (1987: 27).
75. See Joseph (1987: 92) for further discussion of how European vernaculars were
considered rude or vulgar during the early phases of their rise.
76. Similarly, women's literature provided one of the earliest and most important
contexts in which vernacular Japanese made its appearance in written form (Twine
1991: 39).
77. Unfortunately, I have only been able to examine a few copies of mosquito press
newspapers from that period, and thus cannot conclusively say that I know what
Lan was referring to. I have been unable to discover material that is explicitly
pornographic, but my sample has also been smalL
78. For example, the pornographic newspaper Kam Yeh Pao (~~¥Ii) not only made
less use of Cantonese than the mass-market newspapers; in fact, it contained little
Cantonese at all, and the more explicitly pornographic articles were written
primarily in Standard Chinese. A similar case is provided by the now-defunct Blue
Cover Mystery Magazine (C: Laam pei syu ~Stl!), sometimes mentioned as a
magazine which contained both soft-core pornography and written Cantonese.
While such a magazine no doubt helped establish in people's minds a connection
between sex and written Cantonese, in issue number 431, the only issue I have
been able to examine, the two do not coincide. There are quite a few
pornographic, or at least titillating, articles, but none make use of Cantonese. The
only feature which does make use of Cantonese, a crime story written in saam kap
dai, portrays no sex.
79. I had this experience with the works of two Hong Kong writers, Xia Fei (~~)
and Yang Tiancheng (1'Gtx~). In both cases, I was assured by academics, writers,
and booksellers that these authors wrote pornography in Cantonese. However, for
Xia Fei, I was able to examine quite a number of books without finding any
Cantonese whatsoever, and in the case of Yang Tiancheng, even a series to which
I had been specifically directed proved to be written in Standard Chinese.
80. Deng (1980: 79); Song (1984: 23).
81. Wang (1978: 31-32); Yu (1987: 229); Chan Chee-shing (1987: 231, 233-5).
82. 29 October 1987, p. 34.
83. Lan (1947: 18); Feng and Quan (1948: 48); Hua (1949: 40); Huang Sheng (1949:
26) ; Jing (1949b: 7).
84. Ng Chun-bong, interview, 3 July 1990; Ng Honjyun, public lecture, 23 June 1990.
Joseph Yau justified use of dialect in his fiction because of its "forcefulness."
Interview, 12 June 1990.
85. Chen (1988: 18). For example, poet Tsai Yim-pui (~~f:ff) notes that he generally
avoids use of dialect in his work (a fact which may be due partly to the fact that
he also works for Ming Pao, a newspaper which discourages use of Cantonese),
but he does feel the need to use Cantonese on occasion in order to give a stronger
282 Notes to pages 202-210

Hong Kong flavor to his poems, and he believes that Hong Kong audiences
respond positively to written Cantonese. Interview, 25 May 1990. AB mentioned
earlier, essayist Lo Wai-Iuen (/J\,I[I" pen-name Siu Si Jal$~) both teaches and
writes in Standard Chinese, but has also made some use of written Cantonese in
her essays because she feels it gives her work more punch. Interview, 13 June 1990.
86. Wu (1992: 94).
87. Fok Lai-zan, interview 24 September 2002.
88. An interesting reflection of this is the strong tendency for Hong Kong young
people to spell out Cantonese words using the English alphabet when engaging
in ICQ chat on the computer. In this approach, students are writing down what
they would say - representing the sounds of their Cantonese speech - rather
than following a purely written norm. Cheng (2002: 26--27).
89. Liang (1989: 106); Ng Chun-bong, interview, 10 July 1990; Chan Hing-kai,
interview, 19 July 1990.
90. Joseph Yau, interview, 12June 1990. Chan Hing-kai said his target audience is the
14-30 age range, and noted that many of his readers are students. Interview, 10
July 1990.
91. Bi and Shu (1989: i). All of the works included in the reader were, however, in
Standard Chinese.
92. Interview, 12June 1990.
93. Kwan Wing-kei, interview, 25 May 1990.
94. Lau Sing, interview, 3 July 1990; Cheung Chor, interview, 10 July 1990; Fung
Gwong-Iit, interview, 11 July 1990.
95. Public panel discussion, 23 June 1990.
96. Interview, 12June 1990.
97. Kwan Wing-kei, 25 May 1990; Ng Chun-bong, interview, 10 July 1990. Two
pocketbooks which are readily comparable are Joseph Yau's (Gu Sai-mung) On a
Weekend Bed, published by Publications Holdings in 1986, and Chan Hing-kai's (A
Foon) Diary of The Little Man, published in 1988. Both share the same office setting
and content matter, and the same diary format. Both also made unusually heavy
use of Cantonese, and sold unusually well. However, Diary of The Little Man was
written entirely in Cantonese, was targeted at a younger audience, and sold many
times more copies. There are, of course, a variety of possible explanations for the
sales differences between these two books, but it is quite possible that the key lies
in the combination of colloquial Cantonese use and a young audience (Liang 1989:
106).
98. Public panel discussion, 23 June 1990.
99. Halliday (1989: 29-30).
100. In my own admittedly much more informal investigations of Nantong dialect
(Jiangsu province), I have also found that while people who speak the dialect do
have a sense of identity which is related in part to the dialect, they do not feel
their local Nantong identity - or the preservation of their distinct local dialect
- to be a terribly high priority.
101. An interesting case in point is provided by the role of the Chu Nom written
vernacular in Vietnam in the 1800s. While some were entirely against use of Chu
Nom, preferring Classical Chinese, and others wished to replace Classical Chinese
entirely with Chu Nom, many favored a middle ground in which Classical Chinese
was retained in the H role, but Chu Nom had a limited role in certain kinds of
literature, especially poetry. DeFrancis (1977: 44).
 Notes to pages 273-233 283

CHAPTER 8
1. Even as I write this chapter in Hong Kong, we are in the midst of a "promote
Putonghua month," with an abundance of billboards and television advertisements
encouraging people to study Putonghua. However, for perspective it should be
added that similar campaigns for promotion of English are even more common.
2. See the South China Morning Post, 9 November 2003, Agenda, p. 11, for articles
debating the relative merits of replacing Cantonese with Putonghua in Hong Kong
schools.
3. White and Li (1993: 170); Guldin (1995: 96-99); Huang (1999: 144-6).
4. Huang (1999: 101, 142); Chen Ping (1999: 51).
5. See also Chen (1994: 304); Qian (1995: 38); Zeng (2000: 28).
6. Zhan (1993: 52).
7. Tang and Tu (2000: 357-8; 370). See also Huang (1999: 141); Chen (2002: 6).
8. Through friends teaching in Guangdong I have seen at least a few examples of
personal notes passed between secondary school students which are written in
Cantonese.

APPENDIX 1
1. Use of a measure word without being preceded by a number or demonstrative is
distinctively Cantonese, so it is in bold as marked Cantonese. In later texts this
distinctly Cantonese grammar structure will appear repeatedly.
2. This text is taken from Xue (1985a: 14).
3. Soeng Ngo (~~) is a famous character from Chinese legend who stole a potion
of immortality and floated off to the moon.
4. Text from Liang (1988: 63-64).
5. SungJuk (SongYu *33.) was a poet of the Warring States period (circa 300 BeE).
6. This is an allusion to a line in the poem "Song of the South" (Jiang nan ge tt1¥i
WZ) by Tang dynasty poet Li Yi (*~) (748-827?).
7. Text from Fu (1949: 45).
8. Bags from Siu Hing (Zhaoqing), a city near Guangzhou, were made of a local
variety of reed, and did not stand up to wear and tear as well as bags made of
leather or cloth. The tendency of these bags to leak coins caused them to become
a symbol of bad luck (Guan 1990: 220).
9. Text from Folklore and Folk Literature Series #56 (Taipei: Oriental Cultural Service,
1972), pp. 37-38. This text is also in Morris (1992).
10. This is an allusion to a line in a poem composed by an early Ming dynasty poet
from Guangdong, Sun Fen (I*ji), as he was on his way to execution (Chen 1986:
99).
11. Text from Jiu ben Yueju congkan Cif*~"J~fiJ) #9 (Hong Kong: Shenzhou tushu
gongsi, 1980), no pagination. Originally published in Guangzhou by Da xin shuju,
probably in the 1920s.
12. There is apparently a misprint here of the Cantonese character 1E: (he).
13. This is also apparently a misprint; the character should probably be 1\I!l.
14. Text from Zheng Baa, 4 October 1947, p. 28.
15. Text from Hua (1949: 61). Originally published in Ta Kung Pao, 9 March 1949,
p.7.
16. This character is rather puzzling; it is neither Cantonese nor Standard Chinese.
284 Notes to pages 233-243

It probably represents an odd pronunciation of the Mandarin demonstrative na

(~~).
17. The brackets around this character are a good example of the way a writer warns
the reader to sound out a character rather than read it for its original meaning.
The original meaning of this character is "sentry post" or "whistle," but in
Cantonese its pronunciation is the same as that of the word for "to scan."
18. Text from Hung Lok, 10 August 1947, p. 3.
19. This last phrase has a very strong Cantonese flavor, but is quite intelligible to a
Mandarin speaker. This is a good example of a case where it is difficult to decide
whether or not to mark a phrase as Cantonese.
20. Text from San (1975: 118-9). Unfortunately, this collection does not give the dates
or locations of the original publication of these articles.
21. Text from Sing Pao Daily, 13 September 1989, p. 32.
22. A Hong Kong pop group which was very popular in the 1970s.
23. Of course, a Hong Kong reader will not normally read these lines aloud, and if
one did so, he or she might well translate the markedly Standard Chinese terms
into Cantonese. Many people in Hong Kong are quite proficient at "translating"
texts written in Standard Chinese into Cantonese when reading them aloud, and
it is not unusual for newscaster and other people who need to read scripts aloud
to write the script entirely in Standard Chinese but then translate it in to Cantonese
as it is read aloud. (Others just write the script in Cantonese right from the start.)
24. Text from the column "The Reporter's Notes" (i1f2.1!f-¥i1f2.), Tin Tin Daily, 13
September 1989, p. 14.
25. US$1.00 converts to approximately HK$7.80.
26. Text from Diary of The Little Man #2 (JJ\~ A)1IfJi1f['. #2) (Hong Kong: Tomakazu
Production House, 1988), pp. 5-6.
27. Text from YES! Magazine, 2 May 2003, p. 24.
28. Name of a film.
29. This text is taken from Apple Daily, 3 May 2003, p. A4.
Index

A Foon (see Chan Hing-kai) Bauer, Robert 6, 171

Ace Detective Maajyulung (Santaam Beijing 142, 158, 166, 190, 197,216,269
Majyuulung) 169 bian wen 40
advertising 2, 4-5, 126, 142, 146, 149, Bible 30
151-3,157,163-4,167,174, 196, Big Woman of the Little City (Siu seng daai
203, 275 (see also written Cantonese neoijan) 161, 205
- genres) biglossia 260
Ah Giu 242 Blue Cover Mystery Magazine (Laam pei syu)
Ah Sa 242 281
Alsace 189-90 Bohemia 26
Alsatian 190 Boluo 215
Punericansl02,190 bookstores 3, 131
Punoy 269 Bonds of Love, The (Seong si so) 89
Anderson, Benedict 19, 25 Bowie, David 157
Apple Daily 5, 151, 163, 165-8, 186-7, 243, Bowring, Sir John 267
259,274; sample text 167, 243 Breakthrough 166
Arabia 18, 70 Britain 73, 125, 178
Arabic 16-18, 20, 23 Brother Liao's Odd Opinions (Liu Go
Asia 23,27, 137, 198,260,266 gwaaileon) 274
Asia TV 237-8 Buddha 226
Austria 17 Buddhism 27, 30, 41, 77-79, 81,198,260,
266
BaJin 31
Baihua 9, 11-12, 29-31, 33-35, 40-42, Cai Yuanpei 31
60-61,67-68,99,104-5,111-2,118, Cantonese: A Comprehensive Grammar 171,
122, 127-8, 158, 198, 268, 270-1; 280
definition of 12,260; movement 7, Cantonese (dialect) 1,8-10,12-13,17,
12,30-31,62, lO4, 176, 188,260 (see 50,52,55,57,60-63,86,103,106,
also Standard Chinese; written 116, 118, 120, 137, 159, 180-1, 183,
vernacular - Chinese) 187, 192, 194, 216, 220, 222, 237,
Ball Bull Notes (Bo Ngau sauget) 144 261-2, 268, 270, 273, 275, 283-4;
Bat Waa-Iau (see Ng Honjyun) attitudes toward 141, 192, 194-7,
308 Index

210,215-6; in Cantonese operas 93- Cantonese romanization scheme 12

95,97; in films 125-6, 139-41, 182- Cantonese Talk about Events of the
3; grammar 47-48, 226, 262, 280, Times (jyutjyu sisi taam) 103
283; history of 73-76; influence on Casual Chats in the Colloquial (Zukwaa
Putonghua 216; norms of 10,46--48, kingtaam) 97-98, 109,270
50,52,60,62,64-65,67,86,95,98- cartoons 5,121,146-7,149,160,164,169
99, 115, 121, 153-4, 157, 159, 170, (see also written Cantonese -
172,203,219,224,231; in pop songs publication type)
140-1,182; prestige 195-6, 215-6; Cat Come Cat Go (Maau lai maau heot) 169
relationship to Mandarin 2, 11, 46- Categorized Cantonese (Fanloei tunghaang
51,62,171, 180,214,262; Gwongzauwaa) 91
romanization of 12, 172; role in Catholic 24
Hong Kong 1, 18, 139, 171-2, 175, Cau Hei 88
188, 215, 279, 283; pronunciation Chan Ging (Ngo si saanjan) 273-4
48, 56, 58, 83-84; in saam kap dai Chan Hing-kai (A Foon) 3, 141, 160-2,
127-35, 144, 234, 236; vocabulary 169, 196, 240, 264, 282
48-51,56,63-64,85,89,262-3,268, Chan Mung-gat 131
279,282 (see also marked Cantonese; Chan Shan 91
written Cantonese) Changsha 216
Cantonese Dialect Dictionary ( Guangzhouhua Chaozhou 83, 106, 266; dialect 17, 33,
Fangyan Cidian) 58, 264, 277 106, 121
Cantonese dictionaries 56-9, 170-2, 262, Chats (Taam tin) 274
277,279 Chen Canyun 109, 117, 119,271-2
Cantonese grammars 171-2 Chen Duxiu 30
Cantonese love songs (jyut au) 80-81, 87- ChenJitang 73, 125,266
90,94-95,97,102,108,129-30,199, Chen Mugong 102
209,226-7,268; sample text 88,225 Chen Ronggun 91-92, 269
(see also Cantonese Love Songs; Chen Ruoxi 38, 261
written Cantonese - genres) Cheng Ming 5, 157
Cantonese Love Songs (jyut au) 87,89,98, Cheng, Robert (see Zheng Liangwei)
226-7, 268; sample text 88,225 (see Cheng Yan-chee 243
also Cantonese love songs; written Cheung Chor 136, 272
Cantonese - genres) Cheung Kwan-hin 171
Cantonese opera 6, 60-61, 87, 93-97, Chiang Kai-shek 107, 266
101-2, 125, 237, 270; scripts 6, 83, China 1, 8-9, 11, 15, 19,24, 27, 29, 31,
93-95,97-98, 109, 131, 146, 148, 35-36,41-42,61-62,69-75,77,87-
169, 177, 187, 202, 268; sample text 88,90,98,101-5,108, Ill, 117-20,
95, 228-30 (see also written 123, 131, 136, 138, 175-8, 180-1,
Cantonese - genres; written 185,192,195,198-9,205,213,216,
Cantonese - publication type) 222, 234, 259, 261, 268-9, 271;
Cantonese (speaking) people 6, 8, 72-73, mainland China 1-2, 12, 42, 58-59,
77,85,89-90,98,106,112,118,120, 137-8, 140, 186, 190, 194, 215, 235,
125, 135, 138, 180-3, 187, 189, 190, 239, 263, 277 (see also People's
192, 194, 196-7, 201-2, 209, 214-7, Republic of China)
265, 268, 273, 279; culture 75-77, China Star, The 128
114; identity 69,73,75-77,191,266; Chinese (language) 1, 17, 27, 46, 62, 128,
sense of cultural marginality 69, 71, 137, 171, 178, 192-3, 216-7, 276;
76; pride 70-3, 76-7, 85, 92, 191 written vernacular 10, 30, 260, 262;
 Index 309

vernacularization of written Chinese Counting Dead Grass (Syun sei cou) 272
6-7 (see also Baih ua; Classical Crow's Odd Opinions, The (Wu Ngaa
Chinese; Mandarin; Putonghua; gwaaileon) 274
Standard Chinese; written Chinese) Cuba 87
Chinese characters 2, 29, 35, 38-39, 45, Cultural Revolution 137-8
52-57,60,171,207,277
Chinese civilization 27-29 Daai Loeng Aa Dau Gun 131
Chinese culture 6, 29, 31, 35, 38, 53, 73, Daan Ding (see Gou Dak-hung)
75-77, 111,126-7,139,176,187-8, Dante 197
192, 200, 205-7 Dalian 216
Chinese dialects 1-2, 33, 35, 51, 55, 73- Demotic 17, 21
74,83,91-92,104-5,128,175,180, dialect 46, 121, 259, 262; dialect
192, 195,200,215, 259, 262; written superiority 195 (see also Chinese
forms of 6-7,42, 45, 53, 77,91, 101, dialects)
104-6,176-7,184,208,217,232,278 Dialect Literature (Fangyan wenxue) 107
(see also dialects) Dialect Literature column (Fangyan
Chinese people 1, 190,270 (see also Han wenxue) 107,117,232
Chinese) Dialect Literature Movement (DLM) 11,
Chinese speech communities 8, 18 61,101-23,129-30,132,177-8,182,
Chinese vernacular fiction 40, 265 ' 199,201,203,231-4,263,270,272,
Chow Sing-chi 275 280; history of 106-7, 271; sample
ChuYun 109 texts 109, 1l0, 230-1 (seealsowritten
City Magazine 61,127, 157-62, 168-9, Cantonese - genres)
205-6, 276 Dialect Literature Research Society
Classical Arabic 18, 23-24, 29 (Fangyan wenxue yanjiu huz) 271
Classical Chinese 6, 8, 12, 19, 23-5, 35- Dianshi Zhoubao 136
36,48,60-62,67-68,75,78,82,84, Diao Yu Tai movement 138
92,99,106,112,156,176,180,188, Diary of a Hong Kong Kid (Hoenggongzai
192-3, 199, 203, 232, 238, 259-60, jatgez) 169
262,265,268,270-1,282; in Baihua Diary of a Madman (Kuangren rijz) 31
movement 29-31, 42, 104, 158; in Diary of The Little Man (Siu naamjan
Japan 27-29, 198; in saam kap dai 60, zaugei) 3, 54, 60, 131, 141, 161-2,
127,133-6,144,146,234,236,273- 169, 196, 199-200, 204-6, 240-1,
4 263-4,280; sample text 160, 240
classical tale (chuanqi) 265 diglossia 10, 15-40, 200, 210; definition
Clementi, Sir Cecil 268 16,259; diglossic shift 15-16, 21-40,
comic books 3,142,146-8,170,199,275 43,64; in Hong Kong 17-18, 41, 217,
(see also written Can tonese - 262; in Japan 27-8; in pre-modem
publication type) China 29
Commercial Press 3, 59 Dilhert 169
Communist Party (CCP) 73,101, 104-8, Dirge for Cau Hei (Diu Cau Hez) 88-89,
111-2,114-5,117-9,121,123,126, 226; sample text 88, 225
136,142,177-8 Divine Comedy 197
competitive ahility (jingzhenglt) 260 dog racing 144, 148 (see also written
Conggin Manfu 161 Cantonese - genres)
Corner of Guangzhou, A (Gwongzau jat Don't Think I Will be So Good to You
gok) 109 (Mai gu waa tung nei gam hou) 89
CouJat 169 dragon boat songs (lung zau) 80-81,86-
310 Index

87, 94, 102, 108-9, 121, 130, 224, Foshan 71, 79

231-2,268,272; sample text 86,223 France 24, 57, 103, 189
(see also written Cantonese - genres) French 15,24,26, 190, 197
Dream Come True 204 Fu Gongwang 108-9, Ill, 1I3, lI5, 224,
Dream of the Red Chamber (Hong Lou Meng) 230
226 Fujian 35, 71, 74, 210, 223, 266, 269
Drifting Clouds (Ukigomo) 28 Fun Magazine 157, 276
Duck House Notes (Yaliao ji sh~) 272 Futabatei Shimei 28

East River guerillas 107 Gam Gwok-Ioeng 141,265

Emigration - Again (sample text) 155, GamJung 182
239 Gam Ngaaji 196
England 28, 188, 198 Gam Sing-taan (Cing Zingjyu) 135, 185
English 1,9, 17-8, 26, 58, 62, 120, 125, Gan Niu 117, 272
135, 137, 141, 157-8, 161-2, 171, Garfield 3, 160, 169
178, 180, 188, 204, 215, 219, 241, genre 27-28, 30, 34, 38, 61, 82-83, 87-
261-5,267,274,276-7,282-3 88, 101-2, 108-10, 116, 118-9, 122-
Essentials for Women and Children (Fujyu 3, 126-7, 133, 135-6, 142, 147, 164,
seoi z~) 91, 269 173,209,223,226-7,265
ethnolinguistic vitality 16, 32-33, 43, 178, German 2,15,24,26,32,190
188-9, 210-1 Germany 17, 20, 26,189
Europe 15, 19-20, 23-26, 28-30, 210 Getting in Trouble (Re sh~) 36
Events in Lingnan (Lingnaam ziks~) 98 GG's Tidbits (GG sai jyu) 274
Express 276 Gingei Laa (see Gou Dak-hung)
Ginny Says the Five Tigers were Lousy
Faulkner 158 Dancers (sample text) 154, 137
Faure, David 269 Go Naa's Notes (Gonaa saugei) 274
FengJicai 121 Gong Shang Bao 117
Feng Naichao 112 Gong Zi-naam (see Wong Ling)
FengXun 268 Gongyang Commentary on the Chunqiu 92
Ferguson, Charles 16, 18,21,23 Gospels 198
films 125, 139-41, 145-6, 152-3, 159, 169, Gou Dak-hung (Dan Ding, Gingei Laa,
179, 182-3, 187, 209, 273 (see also Saam So) 131-6, 143,235-6,274
written Cantonese - factors in Great Leap Forward 121
growth of) Great Yu Pass 70-71
First International Conference on the Greece 17, 21
Written Representation of Chinese Growing Up in Hong Kong (Zoi Hoenggong
Dialects 171 zoeng daaz) 160,204,264
Five Admonitions to a Shrewish Wife Gu Liu 109
(sample text) 86, 223 Gu Sai-mung (seeYau,Joseph)
Flowery Scroll, The (Faa zin g~) 54,81,84, Guangdong 33, 67-68,74-76,80-81,89-
98, 222, 226, 264, 267; sample text 90,93-99,101-2,105-8,111-2,114,
82, 221 117-22,125,130-1,177,189,192,
Foreign Devil Street Stall (Gwailou daai 195,216,222,224,226,230-2,234,
paai dong) 275 266-7, 270, 273, 280, 283;
Forty Chapters in Prose (Saanjyu seisap zoeng) broadcasting in dialect 196, 215;
91 history 68-73 (see also Guangzhou)
 Index 311

Guangdong Dialect Arts Research Group 144,146,151,155,170,183,188,

( Guangdong fangyan wenyi yanjiuzu) 190,201-2,204,210,213,234,238,
106 241; attitudes toward language 62-
Guangxi 68, 93; broadcasting in dialect 63, 180, 188, 193-5, 197, 200, 216,
196, 215 262; audiences 126, 143-4, 150-1,
Guangzhou 11,67-68,70-74,79-80,83, 155, 174, 201, 221, 234; culture 6-7,
87,91,93,98,103,106-8,125,131, 10, 119, 126-7, 137, 139-40, 157,
185, 195, 210, 215, 223, 226, 231, 174,187-8, 197,200-1,204,206,
262,265,270,279,283;newspapers 276,279; diglossia 17-18; education
102, 120-1, 216; prosperity 72, 74, 1-2,91,137,172,178-80,183,192-
77, 80, 85, 87,93, 191,210, 215 (see 3, 196, 207, 214-6, 244, 277;
also Guangdong) government 1,18,126,137,172,
Guangzhou Wenyi 103 178, 180,207,213-5,277; identity 7-
guanhua 12, 192 8,41,127,136-41,189,190-4,204-
Guide to Cantonese, A (Gwongzauwaa 6,210,217; language education 10,
zinaam) 91 18,63,137,171-2,181,184,196-7,
Guilin 93 200, 204-5, 278, 283; language
Guo Moruo 105-6 situation 1, 17-18, 137, 141, 158,
Guo Qiusheng 35 172, 175, 179, 192, 217, 260, 262,
Guoyu 1, 12, 176 283; mass media 1, 126-7, 131, 139-
Gurani 18 42,148,157,159,174,188,215,273;
Guzi 128 as multilingual society 17; people 1,
103, 108, 118-9, 138-40, 143, 172-
Haa Gong (see Heoi Zi-ban) 3, 186, 213-4, 217, 239, 280;
Haa Gong's Odd Opinions 136, 274 prosperity 137-8, 142, 186-7, 210;
Hainan 216; Hainanese 121 publishing 8, 125-7, 131-2, 147-8,
Hakka dialect 17, 33, 36, 106, 118, 121, 159, 174, 178-9, 186-8, 264, 276;
171,266; people 39, 73, 106,266 role of English 1, 18, 158; society 2,
Half a Life as a Beast of Burden (Bun saang 137, 142, 158; speech community 2,
nagau maa) 133 7-8, 17, 189, 213, 217; students 2,
Han Bangqing 33 178-82, 282; writers 59, 105-7, 122-
Han Chinese 69-73, 93, 266 (see also 3,136,141,147,182-3,276,279,281
Chinese people) (see also comics, films, magazines,
Han dynasty 70, 73 newspapers)
Hangul 198 Hong Kong Commercial Daily 136, 143-6,
Hangzhou 268 272
He Huiqun 83 Hong Kong EconomicJoumall45, 167
Heoi Zi-ban (Haa Gong) 136, 143,274 Hong Kong Education Department 197
high (H) languages 16-25, 29, 32-33, 39- Hong Kong Federation of Youth Groups
42,64,183,210,259-60,280 191
Hindi 17 Hong Kong Polytechnic University 171
hiragana 27 Hong Kong Style Cantonese Dictionary (Gong
Holland 28 Sik Gwongzauwaa Cidin) 170, 277
Hon Zung-syun 273 horse racing 5, 126, 142, 144, 148, 166,
Hong Kong 1-2, 9-12, 41-42, 58-60, 62, 275 (see also written Cantonese -
67-68,91,99,101,107-9,117,119, genres)
123, 125, 130-1, 133, 135-40, 142, Hot Wind (Be feng) 107
312 Index

How can we write well? (Dimjoeng se zeng Journal of Travels in Heaven (Tintong
se dak hou?) 109 jaugei) 132
Hu Shi 31, 34, 260, 268, 270-1, 279 Joyce,James 157
HuaJia 106-11, 113, 115-9, 182, 232, 272 juren degree 87, 91
Hua Shang Bao 106-7,114,117 Jyut Ping 12, 172
Huang Chao 70
Huang Guliu 107 Kaifeng 71
Huang Luyi (see Xiang Mizi) Kaleidoscope (Maanfaa tung) 132
Huang Ningying 108 Kam Yeh Pao 281
Huang Sheng 111, 113, 117 Kang Youwei 73, 91
Huang Shihui 35 Kangxi dictionary 56
Huang Zunxian 30-31, 37 katakana 27
Huaqiao Ribao (see Wah Kiu Yat Po) Katharevousa 17, 21
Hui, Sam (Heoi Gun-git) 140 King James English 127
Huizhou 215 Koran 16
Hung Lok 129-30, 132, 234, 273 Korea 25, 198
Hungary 26 kun qu 33
kungfu novels (wuxia xiaoshuo) 130, 182,
ICQ 6, 54, 264, 282 273
I Have a Date with Spring (Ngo wo ceontin
jau go joekwut) 3, 169 Lai He 36
identity 189-91; and language 176, 189, Lai,Jimmy (Lai Chi-ying) 164, 166
193, 204-6, 210, 266 Lam Dik (see Lam Gaa-hung)
Imagined Communities 25 Lam Gaa-hung (Mung Zung Jan, Lam
Improved Lingnan Notes (Gailoeng Lingnaam Dik) 132, 144
Ziksz) 268 Lan Ling 111,113-5,199,271-2,281
In the Underworld (Jan zoi gongwu) 132 language death 7, 37, 175, 211
India 70 Lao She 31
Indonesia 70 Latin 15, 18-20,23-26,29, 197-8, 210
Inside and Outside the Goldfish Bowl Lau Sing (JungJuk, Dau Lung) 59, 134,
( Gamjyu gong noi ngoz) 145 136,264,273
Islam 16,24 Lau Waa-dung 131
Italy/Italian 15, 18,24,26, 197 Leftist Alliance of Writers (Zuoyi zuojia
lianmeng) 105
Jam Wu-faa 273 Legislative Council 137
Japan 19,25,27-28,30-31,35,41,94, Leon Man-zeoi 98
103,130,137,177,198 Letter Writing Reference Materials for Women
Japanese 19, 27, 35-36, 40, 198 (see also and Children (Fujyu seonzaat coiliu) 92,
written Japanese) 128
Jia Pingwa 121 Leung Shu-chiu 150
Jiangsu 282 Leung Tol (Lou Gam) 136, 273
Jiangxi 71, 266 Li, Aurthur 243
Jig Yip Ma Po 276 Li Jingzhong 46
Jing Wen 112 Li Poetry Magazine (Li Shi Kan) 37
jinshi degree 83, 268 Li Wenmao 93, 269
John and Mary 161 Li Yi 83,283
Johnson, Samuel 58 Liang Qichao 30-31, 73
 Index 313

Liao Fengshu 87, 89 182; grammar 47; prestige of 97,

Liao Lianxian 268 192; relationship to Cantonese 2,11,
Lih Pao lO3-4, 270 46-51,62; relationship to Standard
Lin Luo 108, 111-3, 115-6, 199, 272 Chinese 2, 9, 12, 18, 179-80, 193,
Lin Qiyang (Xiang Yang) 36 278; role in Hong Kong 1, 18, 179-
Lin Shuangbu 38 80, 192, 196; in Taiwan 36-37, 39,
Lin Yangmin 38 214,279
Lin Zongyuan 36-37 manyogana 27
Linguistic Society of Hong Kong (LSHK) Mao Dun 31, lO5-6, 270, 273
12-3,172 marked Cantonese 49-52, 56, 64-65, 82,
literacy 25,31,45,51,55,79-80, 121, 180, 84,86,88,97,109-10,114,129-30,
208 (see also written Cantonese - 135,154-6,161,165,172,219,221-
literacy) 4,226,229-32,234,236-7,239,241-
Little Hoodlum (Siu laumaang) 147 3, 267, 283; definition 219-20, 262
Lo Wai-Iuen (Siu Si) 182 (see also Cantonese language)
Lou Gam (see Leung Tol) marked Standard Chinese 64-65, 86, 97,
Lou Qi 106-8, 118, 120 109-10,155-6,172,224,226,231-
Love Diary of the Big Man (Daai naamjan 2, 239, 242, 263, 268 (see also
lyun'oi saugez) 161 Standard Chinese)
Lover's Day Is Really No-Lover's Day massification (dazhonghua) lO4-6, 113
( Cingjanzit sat hai mou cingjan zi zit Matthews, Stephen 171
leon) 134-5; sample text 134, 235 May Fourth Movement 31, 112,205
Loving Couple, The (Yuanyangzi) 107 Meeting of Monks (Heshang huz) 107
low (L) languages 16-25, 29, 32-33, 39- Meiji era 28, 31
42,45,64,210,259-60 Metropolitan Weekly 146
Lu Xun 31, 34, 105 Mid-Autumn festival 223
Lu Zijun 91 Middle Ages 20, 24
Luo Zhengping 55 Ming Coeng 159
Lynn 158 Ming dynasty 6,11,30,33,54,72,79,81,
83,93,185,222,266-7,283
Maan Hoeng 233 Ming Pao Daily 5, 136, 142-3, 145, 149-
Macao 68, 91, 115, 125 52,201,276,282
magazines 83, 94, 126, 146, 157-9, 164- Ming Pao Group 159
6,173-4,183,196,199,203,237,242 Ming Paa Monthly 5,157
(see also written Can tonese - Ming Pao Weekly 146
publication type) Ministry of Education 92
Magnificent Dreams in Shanghai (Hai shang minority languages 7, 19,32,39-40,42,
fanhua meng) 34 175-6, 207, 211
Mainlanders 36-37 missionaries 41, 53, 78, 91, 198,269
Malaysia 17; Malay 17 Modern Chinese Dictionary (Xiandai Hanyu
Manchu 72-73, 93 Cidian) 216
Mandarin 1,9-12,17,33,35,52,63-64, modernization 15, 24-25, 27-28, 30-34,
73,75-76,99,103-4,106,120,125, 42-43,94,126,176
140-1,176-7,181-2,187,197,200, Mongol 72,198
213, 236, 259, 262, 268, 273, 278, Monopoly 157
284; attitudes toward 192-5, 199; More Cantonese Love Songs (Zoi jyut au) 87,
definition ofl2; in films 140-1,179, 89, 269
314 Index

Morris, Peter 268 Nom 57, 198, 282

mosquito press 10, 113-4, 119, 123, 125- "nonsense humor" (mou lei tau) 275
6,128-32,199,203,233-4,271,273- Northern Expedition 73
4, 281 "Notes" (saugez) genre 156, 276; sample
"mouth" (kou) radical 54-56, 58, 264 text 155, 239
Movie 157, 276 Nuzhen 71
Mr Square (Loutou sinsaang) 141, 160
Ms Lu's Odd Opinions (Luk neoisi "odd opinion" (gwaai leon) columns 133-
gwaaileon) 274 6, 145, 147, 155-6, 236, 274, 276;
Mung ZungJan (see Lam Gaa-hung) sample text 134, 235
Office Almanac, The (Baangungsat tungsing)
Nan BeiJi 157 169
Nan Yue dynasty 69-70 Ohio 190
Nanfang Ribao 121, 272 On a Weekend Bed (Zaujat congsoeng) 158,
Nanhai 87, 106 160, 169, 205, 282
Nanjing 73, 216, 269 On the Art of Dialect Writing (Lun fangyan
Nantong dialect 282 wenyz) 107
nation building 16, 24, 32, 42,176 Opium War 30, 73
national language 19, 26, 32, 39, 42,104- Oriental Daily 5, 142-4, 150-2, 156, 166,
5,111,113,123,175-7,182,194, 168, 186, 274
197,217,260 Orthodox Pronunciation Academies 75
National Phonetic Alphabet 114 Ou Wen 107
nationalism 7, 15, 34-35, 43, 104, 138, Ouyang Jueya 49-50, 58
175 Ouyang Shan 103, 106,270
Nationalists (KMT) 36,73,107,142,176-
7 Paang Gin 237
New Cantonese Love Songs (San jyut au gaai paperback books 3, 8, 51, 126-7, 131,
sam) 87 151, 159-62, 169, 174, 183, 187-8,
New Moon (Xin Yue) 94 199 (see also pocketbooks; written
New Time Magazine 157 Cantonese - publication type)
New Words about Guangdong ( Guangdong Paraguay 18
Xinyu) 267 Passion (fyun joeng) 89
New Youth (Xin Qingnian) 30 Peach Blossoms Floating on Spring
newspapers 5, 8, 10, 28, 83, 102-3, 106, Waters (sample text) 95, 228
113-5, 120-1, 126, 128-9, 130-3, Pearl River 67-68,71-73,88,93,97,190,
136,142-6,148-57,163,166-9,174, 210,215,226,270
183,186,196,199-200,202-3,233- People's Liberation Army 107
4,237,239,244, 268, 274-6 (see also People's Republic of China 107, 120-2,
written Cantonese - publication 138-9,172,178,186,195,198,207,
type) 219 (see also China)
Newsweek 127 Persia 70
Next Magazine 6, 164-8 phonetic borrowing 54-57, 59, 68, 82, 90,
Ng Chun-bong (Ng Hau) 141, 160, 182 115,122,173,184,207-8
Ng Hau (see Ng Chun-bong) Pinyin 12-13, 121
Ng Honjyun (Bat Waa-lau) 160, 206 pocketbooks 8-9,159-62,165,169,173,
Nine-tailed Turtles (Jiu wei gui) 34, 261 186-8,196,202-5,240,264,276,280
Ningbo 33 (see also paperback books; written
No Help (Mou ho noi) 89 Cantonese - publication type)
 Index 315

pop songs 139-41,179,182,275 Rome 24

Preliminary Discussion of Literary Rude Collection, The (Bat man zaap) 3,159
Reform, A (Wenxue gailiang chuyi) Russian 104
271
prompt books (hua ben) 40, 261 Saam Coek 157
Protestant 24 saam kap dai60, 126-36, 142-7, 153, 156,
Publications Holdings Limited 159-62, 160, 164, 167, 185, 270, 273; in
205-6,264,282 fiction 129-36, 146, 159, 234, 274,
Putonghua 1, 12, 120-1, 172, 179, 192- 281; in odd opinion columns 133-6,
3,195,200,213-6,278,283 236-7, 274; sample texts 129, 134,
233, 235, (see also written Cantonese
Qi dialect 92 - genres)
Qian Xuantong 270 Saam So (see Gou Dak-hung)
Qin dynasty 69-70,73-74,176 Saam So's Odd Opinions (Saam So
Qing dynasty 40, 72, 74-75, 79, 81, 90- gwaaileon) 133
91,93,97,158,185,222,266 sacred language 19, 22-23,25-27,29,42,
Qing Ming festival 225 175,193
Qiu Tingliang 30 San'yutei Encho 28, 40
Qu Dajun 267 SARS 243-4
Qu Qiubai 105,112,177 Scraps (Ling ling seng seng) 145
Quan Lin 112 Second World War 6, 11, 36, 41, 99, 129-
Quanzhou 35 30, 144, 153
Qujiang 70 "seeking original characters" (ben zi kao)
Quraysh dialect of Arabic 16 38,55,208
Shakespeare, William 157
radio 49-50, 60,159-62,187-8,196,209, Shandong 87
215, 241, 275, 280 (see also written Shang dynasty 56
Cantonese - development and Shanghai 9, 33-34, 92, 98, 103-4, 125-6,
growth of) 216,234,269,273
Rao Bingcai 58 Shanghai Flowers (Hai shang hua liezhuan)
Readings in Cantonese Colloquial 97 33-34,279
Record of Workers' Grievances, A Shanwei 215
(Daagungzai sou jyun luk) 273 Shao Binru 97
Renaissance 24 She 266
Reporter's Notes, The (Geize sauget) 155- Shenzhen 216
6, 276; sample text 155, 239 Shi Shan 129
Representation of Cantonese with Chinese Shimuzu Usaburo 28
Characters, The (ji Hon zi se Jyutjyu) Shrimp Ball Chronicles (Haa kau gei) 107
171 Sighs Before Dawn (Taan ng gaang) 83
Republican 73, 91, 176 Sing Pao Daily 142-3, 150, 160, 166, 168,
Rhyming Book for Women and Children 186,274,276
(Fujyu wanjyu) 91 Sing Tao Evening Post 143
Roman Empire 18-19,24,29 Sing TaoJih Pao 107,143,276
Romance languages 46 Singapore 1
romanization 12,36,38, 114, 121, 261- Sino-Japanese war 28, 35
2, 269; of Cantonese 12, 53-54, 91, Sisters 6
104,115,122,171-2,263-4,277,279 Siu Hing (Zhaoqing) 224, 283
(see also written Cantonese - how Slow Steps on the Sports Stage (Tai taan
written) maan bou) 144
316 Index

Smart Ass 158 208

So eng Ngo 221, 283 stock market 5, 126, 142, 145, 148, 166-
Song dynasty 70-74 7 (see also written Cantonese -
Song of the South (Jiang nan gel 283 genres)
Song Zelai 37-38 Stockbroker's Diary, The (Ginggei jatgez)
songbooks (changben) 105 131-2, 135
"sound but no character" (you yin wu zi) Strange Case of the Guangzhou Baby
51-3,55 Murder, The (Joengseng saat jing kei
Sound from Beyond the String (Sin ngoi on) 272
jam) 274 Stray Notes at the Fishbowl's Edge (Gong
South China Morning Post 164 bin zaapgei) 145
Southern Han dynasty 71,74 Sub-Culture Magazine 157
Southern Min dialect 9, 35-36, 39, 210, subjective ethnolinguistic vitality 32
215 (see also Taiwanese) Sun 168
southern songs (naamjam) 80-81,83-87, Sun Fen 226, 283
94, 102, 108, 199, 223-4, 226, 261, Sun Jiazhen 34
267-8; sample text 83, 222 (see also Sun Vat-sen 73
written Cantonese - genres) SungJuk 222,283
Soviet Union 104 Suzhou 9,33,81,98,222
Spanish 18 Swahili 17
speech community 10, 32-33, 259 (see also Swiss-German 17, 20, 22, 32
Chinese speech communities) Switzerland 17, 20, 22, 32
sports 142, 144, 166,275 Syu Zing-cyun (see Yau, Joseph)
St Paul's College 91
Standard Chinese 1-3, 8--9, 12-13,36,40, TaKungPao107, 114, 117, 120, 132, 143-
48,55-57,60-65,84-85,88-90,92, 4,232,274
95-96, 103-4, 106, 110, 113, 118-21, Taipei 142
128, 137, 140-1, 145, 152-9, 161, Taiping Rebellion 93
164-6,168,174,187,192-4,199- Taiwan 1-2,9,11-12,35-39,72,140,177,
206, 219-20, 231-2, 237, 260, 263, 186-7,210,214,235,264,269,279;
268, 270, 272, 278-9, 281-3; in aboriginal people 36, 39; Hakka
comics 5, 147, 170; definition 11; people 36, 39; identity 35-39;
glosses 103-4, 122; in Hong Kong Taiwanese cultural nationalism 36-
17-18,41-2,63,172,175,215,217, 37
265; in Hong Kong newspapers 3, 5, Taiwan Literature (Taiwan Wenyi) 36
178, 243; of Hong Kong students Taiwan Youth (Taiwan Seinen) 35
178-80; learning to write 63, 176, Taiwanese (Taiwanhua) 35,39, 177,269
179-81, 183-4, 197,215,278-9; (see also Southern Min dialect;
relationship to Cantonese 46-7, 51- written Taiwanese)
4, 62-4, 89, 179-81, 203, 278; Taiwanese (people) 9,36-38
relationship to Mandarin 2, 8, 18, Tam, Alan 238
62-3,99, 193; in saam kap dai 60, tan ci 33,80
127, 130, 134-5,234,236; in Taiwan Tan Liqiang 91
36-7,279 (see also marked Standard Tang dynasty 27, 30, 70-71, 74, 283
Chinese; written Cantonese - styles Tanzania 17
of) Tears of Parting (Fan bit lem) 89
Standard German 17, 20, 22, 32 Teenage 157
standardization 45, 89, 172-3, 177, 195, Teens Weekly! 6, 166,206
 Index 317

television 139-41, 145-6, 153, 159, 182- Vietnam 19, 282

3, 187, 196, 209, 215, 237 (see also Vietnamese 57, 198,266
written Cantonese - factors in Violent Protest of Da Mao City (Khong-po
growth of) e phah niau chhz) 38
Texas 190
textbooks 90-93 (see also written Wah Kiu Yat Po 107,117, 143, 151
Cantonese - publication type) Wang Wei 83
Third Aunt Yang (joeng Saam Sam) 109, Wasted Youth (Fong waa hoei dozz) 129-30,
271-2 135; sample text 129, 233
Third Uncle's Odd Opinions (Saam Suk We All Grew Up This Way 204
gwaaileon) 274 Wen Wei Po 144, 274
This is the Real People's Army (Ni di zing Wenyi Shenghuo 107
hai janman ge guandeoz) l09 What should we write? (Se matje hou ne?)
Tianjin 216 109-10; sample text 110, 232
Tibetans 198 Widow Speaks at Night, The (Gwaafu je
Tin Tin Daily 10, 59,142,144-6,149-53, waa) 109
156,166-7,186,274-5 Willow Hand, The (Go lau sau) 109, 111,
To, Raymond 3,169 113-4; sample text 109, 230
Today Is Not Like the Past (Gamsi m tung Winter Leaves and Spring Comes (Dong qu
wongjat) 107 chun laz) 119
Tolstoy 158 Witness to Strange Things in Hong Kong, A
Tomakazu Production House 159-62, (Hoenggong mukdouze gwaai jinzong)
206 132
Toothless's Dog Talk (Ngaa Bang gong Wong Ling (Gong Zi-naam) 132
gau) 144 Wong Zim 3, 159, 169
topolect 262 wooden fish (muk jyu) 79
traditional stories 97-98, 109, 131, 147 wooden fish books (muk jyu syu) 6, 79-
(see also written Cantonese - genres) 80,84,86,90,92,94,97,105,116,
Translation of Words for Women and 119, 125, 130-1, 148, 198,202,222,
Children, A (Fujyu jikcz) 92, 128 224, 226, 267 (see also written
Traveler's Autumn Regrets, The (Haak Cantonese - publication type)
tou cau han) 83-84, 223; sample text wooden fish songs (muk jyu go) 79-82,
83, 222 147,221-4,232,267; sample text 82,
Triads 265 221 (see also written Cantonese -
Tsuk Wa Song Sam (Entertaining Tales in genres)
the Colloquial) 98 Worries of a Citizen, The (Gwokman
Twins Effect, The 242 jauhan) 102
TVB 50, 139-40, 159, 240, 264 written Cantonese (general) 2-3,6,8-12,
35,54,67-68,76,99,104,119,126,
United States 102, 270 128, 148, 155, 207, 221; academic
Upper, Middle and Lower Class Society interest 170-2; as distinct language
(Soeng, zung, haalau sewuz) 132 variety 10, 62-64; definition of 12,
Urban Council 137 60; in PRC 121-2, 283; relationship
to Standard Chinese 10, 49, 62-65,
Vanishing from the Human Race 141,265 203; significance of 6-7;
vernaculars 1, 15, 19, 23-27, 29, 33, 41, standardization of 57-59
45, 104-5, 262; (see also written written Cantonese - appeal and
vernaculars) popularity of9, 83, 85-86, 88, 90, 94,
318 Index

98, 101-2, 106, 113, 116, 118, 122- 61,87-89,98,226-7; in Cantonese

3, 129, 136, 146-7, 151, 159-60, 162, operas 95-97, 229-30; in Dialect
165-6, 173-4, 185, 187,201-3,209, Literature Movement 101, 109-10,
282; intimacy and vividness 113, 116, 114, 231-2; in dog racing articles
118,182,201-3,281-2 144, 148; in dragon boat songs 86,
written Cantonese - attitudes toward 11, 224,231; in entertainment features
41-42, 81-82, 86-87, 101, 122-3, 5-6,142, 145-6, 148, 164, 167; in
136,145,150,153,162,164,171, horse racing articles 5, 144, 148; in
188, 196-204, 277, 280; vulgarity 81, news items 164, 173, 188, 209, 243-
101,115-6,147,170,175,196-8, 4, 277; in Notes genre 156; in
200; sex 196, 199-200 personal correspondence 6, 9, 54,
written Cantonese - audiences 80, 91, 92, 201-2, 204, 216-7, 283; in
103-4,108,114,117-9,127,129-32, political writing 102-3; in
139-40, 143-4, 146, 148, 157, 160, pornography 5, 113, 115, 199-200,
162, 165, 167, 169-70, 173-4, 185, 281; in saam kap dai style 114, 120,
196, 198-200, 209, 221-2, 227, 232, 129, 133-5, 143,234-5; in southern
236,238,242,282 songs 61, 84-85, 88-89, 223-4, 226;
written Cantonese - characters 50, 52-54, in sports articles 142, 144; in stock
56-59,82,90,98,115,170-3,208, market articles 5, 142, 145, 148; in
241, 263-4, 279 traditional stories 97-98, 109, 131; in
written Cantonese - development and wooden fish songs 61, 81-82, 84,
growth of 6-11, 15, 29,40-41,43,45, 222-3
47,52,61,64,111,115,120,122-3, written Cantonese - how written 11, 43,
126,131,137,162-3,168,175,177, 46, 51-57, 92, 115, 122, 171, 184,
215, 219, 272; challenges faced by 203, 263
33, 42-43, 123; in modern Hong written Cantonese - literacy 8-9, 42, 45,
Kong 125-74, 215, 277; origins of 51,55,57,59,65,67,80,90,92-93,
77-79, 198-9; in pre-modern 97-98, 111, 113-5, 122, 147, 166,
Guangdong 67, 77-99; future of 173,184-5,203,207,263,269,279,
213-7 284
written Cantonese - factors in growth of written Cantonese - publication type 2-
8, 11,77, 176-206; difficulty learning 6,9-10,79; on billboards 2, 4-5,163;
Standard Chinese 176, 178-83; ease in Cantonese opera scripts 61, 93,
of learning written Cantonese 176, 98,202; in cartoons 3, 5, 121, 164,
184-5, 214; ethnolinguistic vitality 169; in comic books 5, 142, 146-8,
43, 188-9, 210; government policy 170; in gazetteers 267; in magazines
176-8, 207; identity 8, 41, 188-206, 3, 5-6, 87, 94, 146, 157-9, 164-6,
210,214,216; media 8,176,187-8; 172, 179, 203, 237, 239; in
middle class 127, 132, 137-8, 164, newspapers 3-6, 60, 87, 114, 120,
166-7, 169, 173-4, 204-6; regional 128-33,142-6, 148-57, 166-9, 172,
publishing industry 176, 185-7; 179, 186, 203, 233, 237, 239, 243,
student/youth culture 6, 8, 41, 127, 271,275-6; in paperback books 3, 8-
136, 139, 147-8, 162, 164-6, 168, 9,51,60-61,159-62,169,172,186,
172-4,179-81,204-6,216,242,282 188, 203, 241, 282; in textbooks 90-
written Cantonese - genres 8, 61-62, 65, 93, 98-99; in wooden fish books 60,
67, 79-81, 102; in advertising 102, 98,202,226
121, 123, 146, 151-3, 163-4, 167, written Cantonese - relation to spoken
203, 275; in Cantonese love songs Cantonese 43, 45, 51, 60-61, 67,81,
 Index 319

86,88-89,97-99,101,115,121-2, from spoken vernaculars 15, 33, 40;

130, 142, 145 ease of learning 34, 36, 38, 207-8;
written Cantonese - sanction for 101; ethnolinguistic vitality 16, 210-1;
association with speech 41, 78,90, genres 61, 209; government policy
99, 1l0, 116--7, 131, 135, 148, 153, 39, 207; identity 38, 210; light
156,158,168,173,197-8,217,220, content 41; low prestige 34, 37, 40-
222-4, 227, 230, 233, 236, 238-9, 42, 77; market factors 34, 38, 208-9;
241,244; light content 5,41,78,90, media 209; positive affective
95,99,128,133,135,145,147,159, response 28, 37, 101, 185;
165,168,174,187,217,220,222-3, relationship to spoken languages 7,
227, 230, 234, 236, 238, 241-2, 275; 16,20-21,28-29,31,40,45,51,78,
low prestige 40-41, 67, 78, 80-83, 180-1, 197; standardization 37, 208;
87-88,90,92,95,97,99,103,108, vulgarity 28,31,36,101, 116, 197-8,
148,168,174,196--200,224,261 208, 281 (see also Chinese dialects;
written Cantonese - styles of 60-61; vernaculars)
Cantonese only in dialogue 153, written Wu dialect 9, 33-35, 39-40, 42,
158-60, 169-70; Cantonese only in 64,98,171,261,279 (see alsoWu
direct quotes 6, 164-8,243-4,277; dialect)
colloquial (consistent) Cantonese Wronged Ghost Yu (JyungwaiJyu) 133
61,86--89,96--97,101,109-10,115, Wu dialect 33-34, 195, 261 (see also
127, 142, 148, 153-7, 159-61, 165, written Wu dialect)
172,196,203,219,226,230-2,239- Wu Gun-man (seeYau,Joseph)
42, 278; mixing of Cantonese and Wu Shan 103
Standard Chinese 8, 60-61, 84-85, Wu Yang Beer 121
89,95,98-99,107,109,111,114-5, Wuhan 216
120-2, 127-8, 130, 134, 144, 146--7; Wuxi Baihua Bao 30
153-6, 160, 165-6, 185, 208, 237-8, Wynners 237-8
271 (see also saam kap dai)
written Cantonese - vocabulary 49, 50- Xia Fei 281
52,54,56,63-64,82,84-86,88-89, Xiamen 269
95,97,103,109,114,129-30,135, Xian Dao 273-4
154-6, 161, 173, 185, 220, 222-4, Xiangjiang Wanbao 102-3, 268
226,230-2,234,236,238-41,243 Xiang Mizi 87, 89
written Cantonese - writers of 53, 55, 57, Xinsheng Wanbao 131-3, 136
59, 115, 117, 119-23, 136, 147, 151, xiucai degree 91
158-9,162, 171, 179, 182-5, 201-2, XuJidun 37
204, 206, 276, 279; film/television Xue Shan 106--7
script writers 61, 141, 159, 182-3, Xun huan jian 270
187
Written Cantonese in Hong Kong 6 Yang Tiancheng 281
written Chinese 29-30, 62, 75, 111-2, Yangzhou 83
188, 201 (see also Chinese) Yao 72, 74
written Japanese 27-28, 34, 41-42, 61, 90, Yao Li 116--9
281 Yau, Joseph (Gu Sai-mung, Syu Zing-
written Taiwanese 33, 35-40, 42, 51, 53, cyun, Wu Gun-man) 158, 160, 169,
55,171,198,210,261,264,279 205-6,264,282
written vernaculars 7, 9,27,33,45,77,82, Ye Tingrui 83, 223
175-6, 184, 207, 260; developing Yes! 6, 165-6, 187, 204, 206, 242, 277;
sample text 165, 242
320 Index

Yip, Virginia 171 Zhejiang 71

You Suo Wei Bao 102 Zheng Bao 106-7, 113-4, 117
Young Hooligans (Gu waak zai) 170 Zheng Liangwei (Robert Cheng) 37
Yuan dynasty 72 Zhongshan 215
Yue 69-71, 74 Zhongshan University 103
Zhou Wuji 58
Zhang Chunfan 34 Zhoumo Huabao 121
Zhang Jiuling 70-71 Zhu Ziqing 31
Zhangzhou 35 Zhuang 57, 74
Zhao Tuo 70 Zhuhai87
Zhao Ziyong 87-89, 98, 225-7, 268 ZijauJan 161