HTTP/1.1 200 OK
Date: Thu, 29 Mar 2012 06:53:53 GMT
Server: Apache
Connection: close
Content-Type: text/html

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
    "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">

<html xmlns="http://www.w3.org/1999/xhtml">
<head>
   <title>Taiwanese, Mandarin, and Taiwan's language situation</title>
   <meta http-equiv="Content-Type"
         content="text/css; charset=utf-8" />
   <style type="text/css"
          media="all">
   /*<![CDATA[*/
   @import "/main.css";
   /*]]>*/
   </style>
         <link rel="stylesheet"
         type="text/css"
         media="print"
         href="http://www.pinyin.info/print.css" />
   <meta name="ROBOTS"
         content="ALL" />
   <meta name="Copyright"
         content="Copyright M. Swofford" />
   <meta http-equiv="imagetoolbar"
         content="no" />
   <meta name="MSSmartTagsPreventParsing"
         content="true" />
   <meta name="author"
         content="Mark Swofford" />


<script type="text/javascript">
var gaJsHost = (("https:" == document.location.protocol) ? "https://ssl." : "http://www.");
document.write(unescape("%3Cscript src='" + gaJsHost + "google-analytics.com/ga.js' type='text/javascript'%3E%3C/script%3E"));
</script>
<script type="text/javascript">
var pageTracker = _gat._getTracker("UA-4582585-1");
pageTracker._initData();
pageTracker._trackPageview();
</script> 
   <meta name="keywords"
         content="taiwanese, taiwan, chinese characters, hoklo, hokkien, sinitic, written taiwanese" />

   <style type="text/css">
   /*<![CDATA[*/
   <!--
   p       {
        line-height: 1.5em;
        }
   #quotes {
        text-align: right;
        }
   .notenumber, .raised     {
        position: relative;
        bottom: .4em;
        }
   .rant   {
        padding: 1em 3em;
        }
   .rant p   {
        text-align: justify;
        }
   .rant h3        {
        text-align: center;
        }
   i .py   {
        font-style: normal;
        }
   .py     {
        line-height: 1em;
        }
   .poem pre       {
        font-family: georgia, serif;
        padding-left: 2em;
        font-size: 1em;
        line-height: 1.2em;
        }
   #bibliography p {
        padding-left: 3em;
        text-indent: -3em;
        }
   td      {
        text-align: center;
        padding: .5em;
        }
   #vernacular, #literary  {
        background-color: rgb(218,248,175);
        }
   -->
   /*]]>*/
   </style>
</head>

<body>
   <div id="container">
                  <div id="header">
         <h6>Pinyin</h6>
      </div>

      <div id="main">
         <h1>How to Forget Your Mother Tongue and Remember Your National
         Language</h1>

         <h2>by Victor H. Mair</h2>

         <p>University of Pennsylvania</p>

         <div id="quotes">
            <p>
            &#20116;&#26041;&#20043;&#27665;&#65292;&#35328;&#35486;&#19981;&#36890;<br />

             <span class="py">Wu f&#257;ng zh&#299; m&iacute;n,
            y&aacute;ny&#468; b&ugrave; t&#333;ng.</span><br />
             "The languages of the people in all directions are [mutually]
            unintelligible."<br />
             <i>Li ji</i> [<i>Record of Ritual</i>], "Wang zhi [Royal
            Institutions]"</p>

            <p>&#22235;&#26041;&#35527;&#30064;.<br />
             <span class="py">S&igrave; f&#257;ng t&aacute;n
            y&igrave;.</span><br />
             "Speech is different in the four directions."<br />
             Wang Chong, <i>Lunheng</i> [<i>Balanced Disquisitions</i>],
            "Ziji [Autobiography]"</p>

            <p>&#35328;&#35486;&#19981;&#36948;<br />
             <span class="py">Y&aacute;ny&#468; b&ugrave;
            d&aacute;.</span><br />
             "We cannot understand each other's languages."<br />
             <i>Zuo zhuan</i> [<i>Commentary of Zuo Qiuming</i>], Xiang
            14</p>

            <p>&#25105;&#25163;&#23531;&#25105;&#21475;<br />
             <span class="py">W&#466; sh&#466;u xi&#283; w&#466;
            k&#466;u.</span><br />
             "My hand writes [what] my mouth [speaks]."<br />
             Huang Zunxian (1848-1905)</p>

            <p>
            &#27597;&#35486;&#26159;&#25105;&#20497;&#30340;&#26681;<br />
             <span class="py">M&#468;y&#468; sh&igrave; w&#466;men de
            g&#275;n.</span><br />
             "Our mother tongue is our root."<br />
             <i>Taiyu wenzhai</i>, 2 [September 15, 1989], 148</p>
         </div>

         <h2>Abstract</h2>

         <p>The concept of <i>guoyu</i> ("national language") is deeply
         embedded in the consciousness of everyone who has grown up in
         Taiwan during the past half century. Lately, however, people have
         begun to speak of their <i>muyu</i> ("mother tongue") as being
         worthy of inculcation. <i>Guoyu</i>, of course, refers to Modern
         Standard Mandarin (MSM), which in China is called <i>putonghua</i>
         ("common speech"). Mandarin is not native to Taiwan, yet it is the
         national language of Taiwan's citizens and is the sole official
         written language. In contrast, the citizens of Taiwan are
         discouraged from writing their native languages (viz., Taiwanese,
         Hakka, and various aboriginal languages) and it is only recently
         that it has been possible to teach them in the schools. This paper
         will examine the complicated processes whereby the citizens of
         Taiwan are transformed from speakers of their mother tongues to
         speakers and writers of the national language. This transformation
         does not rely purely on educational activities carried out in the
         schools, but involves political, social, and cultural factors as
         well. The transformation of Cantonese and Shanghainese speakers
         into Mandarin speakers and writers will also be examined for
         comparative purposes.</p>

         <h2>Contents</h2>

         <ol>
            <li><a href="#part1">Preliminaries</a></li>

            <li><a href="#part2">The Fundamental Unwritability of the
            Nonstandard Sinitic Languages</a></li>

            <li>
               <a href="#part3">Mechanisms Whereby One's Mother Tongue Is
               Displaced by the National Language</a> 

               <ol>
                  <li><i><a href="#part3a">Basse-Vulgarisation</a></i></li>

                  <li><a href="#part3b">The Pursuit of Prestige</a></li>

                  <li><a href="#part3c">The Myth of Monolingualism</a></li>

                  <li><a href="#part3d">The Sinographic Snare</a></li>
               </ol>
            </li>

            <li><a href="#part4">The Prognosis for Written
            Taiwanese</a></li>

            <li><a href="#part5">Reflections</a></li>
         </ol>

         <h2 id="part1">I. Preliminaries</h2>

         <p>We may take it for granted that a child born in Taiwan of
         Taiwanese-speaking parents will begin his or her life speaking
         Taiwanese. <span class="notenumber"><a href="#n1">1</a></span> We
         may also take it for granted that, when it comes time for this
         child to acquire literacy, he or she will necessarily have to
         learn a second language, namely Mandarin. Not only is there no
         expectation that this child will learn to read and write
         Taiwanese, at the present moment in history it is virtually
         impossible for this child to read and write Taiwanese because
         there are no accepted norms for writing in that language. In other
         words, Taiwanese is fundamentally an unwritten (and arguably at
         this point in time an unwritable) language.</p>

         <div id="copyright">
            This article was written in 2003. Its appearance here on Pinyin
            Info marks its first publication.
         </div>

         <p>How did it happen that a society which values education highly
         would opt to seek literacy in a language other than its own? The
         purpose of this paper is to investigate this conundrum from a
         number of angles (political, social, cultural, historical, and
         linguistic).</p>

         <p>Perhaps it would be better to rephrase the question. It may not
         at all be the case that the Taiwanese have opted to seek literacy
         in a language other than their own. Rather, they may have been
         conditioned by circumstances to do so. Our task, then, will be to
         delineate these circumstances and describe how they have had such
         an enormous impact upon the Taiwanese people. We shall examine the
         processes whereby Taiwanese children are weaned away from their
         mother tongue and acquire in its stead a second language for the
         purpose of becoming literate. It is remarkable that this shift of
         emphasis from mother tongue to acquired language takes place
         within an environment where one's mother tongue is the common
         spoken language of society. It is not due to migration into
         another speech community where it is necessary to acquire
         competence in the local language, which is the usual environment
         for the displacement of an individual's mother tongue.</p>

         <h2 id="part2">II. The Fundamental Unwritability of the
         Nonstandard Sinitic Languages</h2>

         <p>Let us assume that there is an ardent Taiwan nationalist who is
         desperate to write literature in his or her own language (viz.,
         the mother tongue of the Taiwanese people). Certainly there have
         been no lack of such individuals during the past century and,
         indeed, up to the present moment. Unfortunately, such persons will
         be frustrated at every turn when they try to put their dearest and
         nearest thoughts and emotions in Taiwanese down on paper: there
         simply exist no established conventions for writing out Taiwanese
         language in its unadulterated form. Of course, a determined
         individual may devise various idiosyncratic, <i>ad hoc</i> methods
         for writing Taiwanese in Chinese
         characters,<span class="notenumber"><a href="#n2">2</a></span> in
         Japanese <i>kana</i>, in Mandarin phonetic symbols (<i>bopomofo;
         zhuyinfuhao</i>), in roman letters, etc., or some combination
         thereof. The problem is that whatever ingenious solutions this
         individual may come up with to write down Taiwanese to his or her
         own satisfaction or to the satisfaction of his closest associates,
         there is no guarantee that other Taiwanese speakers will
         understand what he or she may have written because there is no
         consensus among them about how their language should be written.
         Quite the contrary, as time passes, there is greater and greater
         confusion about how Taiwanese should be written. The number of
         competing schemes grows daily, and each scheme has its ardent
         supporters, leading to dissension and despair. Even the government
         cannot institute a workable scheme because its own advisers have
         different views on how to cope with the dilemma. What is more, the
         government lacks an unmistakable mandate from the people to
         institute a unified writing system for Taiwanese.</p>

         <p>Why is it so hard to write Taiwanese? Despite the fact that
         many people have a strong desire to write in their mother tongue,
         there are numerous factors militating against an easy realization
         of a workable script for Taiwanese. The first and foremost is that
         the characters are perfectly suited for writing Literary Sinitic
         (LS; Classical Chinese) but ill adapted for writing practically
         anything else. Thus, for nearly two millennia (from around 1200
         <span class="smcaps">BCE</span> till about the eighth century
         <span class="smcaps">CE</span>), the characters were restricted
         almost entirely to writing solely in LS. Still today, after a
         century of promotion of the national vernacular by social and
         educational reformers, there is a strong tendency to backslide
         into semi-literary (<span class="py">b&agrave;nw&eacute;n
         b&agrave;nb&aacute;i</span> &#21322;&#25991;&#21322;&#30333;)
         styles because they comport so readily with the genius of the
         script.</p>

         <p>The second factor is the conspicuous absence of a tradition of
         writing in the regional vernaculars. Indeed, until the mid- Tang
         period, there was not even a precedent for writing a national
         vernacular. Hence, the initial attempts to write extended
         vernacular texts were fraught with orthographic and other errors,
         and the basic matrix of writing remained LS, with only a
         relatively small amount of vernacular added in. During the
         medieval period, and largely under the aegis of popular Buddhism,
         a written <i>koine</i> haltingly developed. (Mair 1994a) In the
         following centuries, the <i>koine</i> gradually continued to
         mature until it became Early Mandarin under the Mongols,
         <span class="py">b&aacute;ihu&agrave;w&eacute;n</span>
         &#30333;&#35441;&#25991; under the Manchus, and then Modern
         Standard Mandarin (MSM) after the May Fourth Movement and with the
         support of the governments of the Republic of China and the
         People's Republic of China.</p>

         <p>Throughout this long gestation and maturation of the national
         written <i>koine</i>, there never was any question of writing the
         nonstandard regional vernaculars. It is worth pondering that the
         earliest extant texts containing a significant admixture of a
         regional vernacular are four Southern Min dramas from the late
         Ming and early Qing periods. (Wu and Lin 1976; Mair 1994b:
         1295-1298) Written Southern Min did not prosper, however, because
         it was insufficiently distinguished from LS and Vernacular Sinitic
         (VS). That is to say, the writing style of these Southern Min
         plays was greatly alloyed with LS and the national vernacular; it
         was certainly far removed from a pure representation of Southern
         Min spoken language.</p>

         <p>The third factor inhibiting the growth of regional vernaculars
         is a strong scholarly bias against such writing as crude and
         vulgar. This theme, which is a constant refrain in literati
         comments on popular culture, may be epitomized in the maxim
         <span class="py">b&ugrave; d&#275;ng d&agrave;y&#462; zh&#299;
         t&aacute;ng</span>
         &#19981;&#30331;&#22823;&#38597;&#20043;&#22530; ("cannot ascend
         the hall of great elegance").</p>

         <p>The fourth factor is outright political prohibition against
         writing the regional vernaculars. Until the lifting of martial law
         on July 14, 1987, not only was it considered subversive to write
         in Taiwanese, one could be punished for compiling Taiwanese
         dictionaries and instructional materials, or even for speaking
         Taiwanese in public arenas such as schools, governmental bodies,
         the media, and so forth. This tight control of the nonstandard
         vernaculars is still operative in mainland China. As MSM
         (<i>putonghua</i>) extends its reach across the land through
         education, mass media, commerce, and other channels, the noose
         around the neck of the regional vernaculars is inexorably
         tightening. Now, with the third successive generation after the
         founding of the People's Republic of China, youths are beginning
         to feel more comfortable speaking MSM than their mother
         tongue.</p>

         <p>As a consequence of these four factors, writing in the regional
         vernaculars is at best atrophied, if not entirely stillborn. The
         sole exception is that of Hongkong Cantonese, which itself can
         hardly be said to be flourishing, but at least there are those in
         the former crown colony who are actively seeking ways to write the
         full panoply of colloquial speech (Snow 1991; Gunn forthcoming).
         The main reason why written Cantonese could thrive in Hongkong to
         the extent that it has is due to the unique colonial experience
         under the British. Freed from the political controls and cultural
         dominance of the North China mandarinate for a century, the people
         of Hongkong were able to experiment with true vernacular writing.
         Since 1997, however, the restrictions on the use of the vernacular
         are once again beginning to be enforced, and rule from Peking is
         inevitably leading to greater use of MSM in the courts, in
         schools, in the media, and other public arenas.</p>
         <!-- ==================== PAGE 5 ========================== -->

         <p>The Japanese colonial experience in Taiwan was partially
         liberating for those who wished to write in Taiwanese, but the
         focus was very much on education and writing in Japanese. With the
         resumption of control by mainlanders from the middle of the
         twentieth century, <i>guoyu</i> was harshly imposed on the
         populace and there was no open talk of <i>muyu</i> (except by
         those who equated it with <i>guoyu</i>). Under such circumstances,
         there was little prospect for the development of writing in
         Taiwanese.</p>

         <p>Now that the people of Taiwan have taken political control into
         their own hands, there is a tremendous amount of energy directed
         toward the establishment of written Taiwanese, but no functional
         models have been created to actualize such hopes. To put it
         succinctly, until a little over a decade ago, the Taiwanese people
         have been deprived of the opportunity to publicly practice their
         written mother tongue, so it remained manifestly moribund.</p>

         <p>One of the greatest obstacles to writing Taiwanese is the
         undeniable reality that many of the most frequent morphemes,
         especially very common grammatical particles, cannot confidently
         be written with Chinese characters. Most estimates of the
         Taiwanese morphemes that lack an appropriate sinographic written
         form are about 20-25% for typical running
         texts<span class="notenumber"><a href="#n3">3</a></span> (Cheng
         1978: 307-308; Chiung 1999). Supposedly, the remaining 75-80% of
         Taiwanese morphemes are adequately covered by recognized Mandarin
         or LS cognates, or special graphs have already been invented and
         are fairly widely accepted by the Taiwanese-speaking populace.
         While these estimates may be more or less correct for highly
         Mandarinized Taiwanese styles, if one were to attempt to write in
         a more purely colloquial Taiwanese (i.e., closer to the way people
         speak in day-to-day circumstances), the percentage of morphemes
         that cannot be matched with Chinese characters would rise sharply.
         In learned studies of Taiwanese vocabulary, such as that of Hong
         Weiren in his <i>Taiwan lisu yudian</i> [<i>A Lexicon of Taiwanese
         Etiquette and Customs</i>], where profound scholarship is brought
         to bear on the proper sinographic way of writing colloquial terms,
         it is immediately obvious that there is much controversy over how
         to write basic Taiwanese words in characters. Even when there are
         fairly well-established ways of writing Taiwanese words, it is
         easy to demonstrate that the characters chosen for them are often
         "wrong" in the sense that they are arbitrary homophones or near
         homophones, and that their meanings are completely irrelevant.</p>

         <p>The very name "Taiwan" is perhaps the best example to begin
         with. Superficially (according to the surface signification of the
         two characters with which the name is customarily written),
         "Taiwan" means "Terrace Bay." That sounds nice, even poetic, but
         it is an inauthentic etymology and has nothing whatsoever to do
         with the actual origins of the name. (This is a typical instance
         of the common fallacy of
         <i><span class="py">w&agrave;ngw&eacute;nsh&#275;ngy&igrave;</span></i>
         &#26395;&#25991;&#29983;&#32681;, whereby the semantic qualities
         of Chinese characters interfere with the real meanings of the
         terms that they are being used to transcribe phonetically.) The
         true derivation of the name "Taiwan" is actually from the ethnonym
         of a tribe in the southwest part of the island in the area around
         Ping'an.<span class="notenumber"><a href="#n4">4</a></span> As
         early as 1636, a Dutch missionary referred to this group as
         Taiouwang. From the name of the tribe, the Portuguese called the
         area around Ping'an as Tayowan, Taiyowan, Tyovon, Teijoan,
         Toyouan, and so forth. Indeed, already in his ship's log of 1622,
         the Dutchman Comelis Reijersen referred to the area as Teijoan and
         Taiyowan. Ming and later visitors to the island employed a
         plethora of sinographic transcriptions to refer to the area
         (superficially meaning "Terrace Nest Bay" [Taiwowan
         &#33274;&#31389;&#28771;], "Big Bay" [Dawan &#22823;&#28771;],
         "Terrace Officer" [Taiyuan &#33274;&#21729;], "Big Officer"
         [Dayuan &#22823;&#21729;], "Big Circle" [Dayuan &#22823;&#22291;],
         "Ladder Nest Bay" [Tiwowan &#26799;&#31389;&#28771;], and so
         forth). Some of these transcriptions are clever, others are
         fantastic, but none of them should be taken seriously for their
         meanings.</p><!--          *** page 6 ***  -->

         <p>One of the most lively districts in Taipei is Wanhua
         &#33836;&#33775; ("Multitudinously Floriate"), which sounds like a
         lovely name for an urban area. When we look a little closer at the
         derivation of this place name, however, we find that the two
         characters chosen to write it are entirely specious with regard to
         the original meaning and are phonologically remote from the
         correct pronunciation. The old sinographic form of this name is
         &#33355;&#33338; , which would be pronounced
         <i><span class="py">b&#593;&#769;ng-kah</span></i> in Taiwanese.
         Here the radicals are useful in helping us to understand that this
         was an old word of the non-Sinitic indigenes for "boat." It was
         only during the Japanese occupation that the sinographic form of
         the name was changed to the two graphs meaning "Multitudinously
         Floriate." Phonologically this makes some sense in Japanese where
         they are pronounced <i>banka</i>, but not in Mandarin where they
         come out as Wanhua.</p>

         <p>If a goodly portion of the old place names in Taiwan are of
         this confused nature, there are even more common words for which
         the sinographic form is problematic and conflicted. One that has
         long intrigued me is <i>chhit-th&ocirc;</i> or
         <i>thit-th&ocirc;</i> ("play [around]"). This word has the fairly
         widely accepted sinographic form of <img src="chhit_tho.gif"
              width="32"
              height="15"
              alt="Sinographs for chhit-tho" /> (although there are those
              who vigorously dissent and state that it really should be
              written as &#12564;&#38518; or with some other totally
              different characters -- the trick is to find two graphs that
              sound more or less like what people say) (Hong 1986: 21;
              Zheng and Zheng 1977: 14). However, when we examine the two
              extremely obscure graphs used to write this Taiwanese word
              meaning "play," we find that they have nothing to do with it
              (the first means "near" and the second means "cunning,
              deceitful"). I suppose that people who stick with this false
              writing of <i>chhit-th&ocirc;</i> do so because radical 162
              (<i><span class="py">chu&ograve;</span></i> ["move fast and
              stop abruptly"] in both of the graphs reminds them of the
              same graph in the Mandarin word
              <i><span class="py">y&oacute;u</span></i> &#28216;, which can
              mean "play," and because playing is something that one would
              like to day and night (the phonetic parts of these two
              characters are "sun" and "moon"). One might also mention in
              this context the title of the first chapter of the Zhuang Zi
              [Master Zhuang] which modifies
              <span class="py">y&oacute;u</span> ("play / wander [around]")
              with <span class="py">xi&#257;oy&aacute;o</span>
              &#36877;&#36953; ("carefree[ly]"), both syllables of which
              are also written with radical 162. Such interpretations are
              forced, of course, and cannot persuade a skeptic that these
              are the best two characters for writing the Taiwanese word.</p>
              <!-- ==================== PAGE 7 ========================== -->

         <p>Many very common usages in sinographic Taiwanese writing are
         counter-intuitive to the reader of MSM texts. When I first started
         trying to read written Taiwanese, I could not understand why the
         graph for <i><span class="py">sh&agrave;ng</span></i> &#20663;
         ("injure; harm") occurred so often in circumstances that made no
         sense to me. Only later did I realize that in written Taiwanese
         this graph stands for the degree expression
         <i>siu<span class="raised">n</span></i> ("too"). (Cheng 1981: 6)
         Still later I found out that almost all characters have at least
         two often quite distinct pronunciations in written Taiwanese, and
         that these pronunciations may signify different meanings.</p>

         <table>
            <thead>
               <tr>
                  <th>&nbsp;</th>

                  <th><span class="py">d&uacute;yin</span>
                  &#35712;&#38899;</th>

                  <th><span class="py">y&#468;y&#299;n</span>
                  &#35486;&#38899; / <span class="py">jieshu&#333;</span>
                  &#35299;&#35498;</th>

                  <th>&nbsp;</th>
               </tr>

               <tr>
                  <th>&nbsp;</th>

                  <th>reading pronunciations</th>

                  <th>&nbsp;spoken pronunciations / explications</th>

                  <th>&nbsp;</th>
               </tr>
            </thead>

            <tbody>
               <tr>
                  <td>1.&nbsp;&#30333;</td>

                  <td>pek</td>

                  <td>p&egrave;h</td>

                  <td>white</td>
               </tr>

               <tr>
                  <td>2.&nbsp;&#38754;</td>

                  <td>bi&#257;n</td>

                  <td>b&#299;n</td>

                  <td>face</td>
               </tr>

               <tr>
                  <td>3.&nbsp;&#26360;</td>

                  <td>su</td>

                  <td>chu</td>

                  <td>book</td>
               </tr>

               <tr>
                  <td>4.&nbsp;&#29983;</td>

                  <td>se&#331;</td>

                  <td>se<span class="raised">n</span> /
                  si<span class="raised">n</span></td>

                  <td>student</td>
               </tr>

               <tr>
                  <td>5.&nbsp;&#19981;</td>

                  <td>put</td>

                  <td><span class="py" style="font-family: arial unicode ms;">m&#772;</span></td>

                  <td>not</td>
               </tr>

               <tr>
                  <td>6.&nbsp;&#35201;</td>

                  <td>i&agrave;u</td>

                  <td>beh / &agrave;i</td>

                  <td>want</td>
               </tr>

               <tr>
                  <td>7.&nbsp;&#36820;</td>

                  <td>ho&aacute;n</td>

                  <td>tn&#769;g</td>

                  <td>return</td>
               </tr>
            </tbody>
         </table>

         <p>The romanizations in the second column are called
         <i><span class="py">d&uacute;yin</span></i> &#35712;&#38899;
         ("reading pronunciations"), while those in the third column are
         known as <span class="py">y&#468;y&#299;n</span> &#35486;&#38899;
         ("spoken pronunciations") or, as with the last three items,
         <span class="py">jieshu&#333;</span> &#35299;&#35498;
         ("explications"). (Hong 1988: 344-345) The elaborate phonological
         regimen pertaining to the sinographs as practiced by
         precontemporary Southern Min scholars is exemplified in their
         careful reading of classical verse. (Branner 2002)</p>

         <p>Another instance of how such bewildering sinographic usages
         keep proliferating in written Taiwanese may be found in the very
         recent character conversion of texts that were formerly written in
         romanization. There is a well-known novella entitled
         <i>Kho<span class="raised">2</span>-ai<span class="raised">3</span>
         e<span class="raised">5</span>
         siu<span class="raised">5</span>-jin<span class="raised">5</span></i>
         (<i>Beloved Enemy</i>) that was originally composed in Church
         romanization by Lai Rensheng (Rev. Lai Jinsheng) and first
         published in 1960 by the Taiwan Presbyterian Press. In 1991, Zheng
         Liangwei (Robert Cheng) published a character version of this
         novella (a rough estimate is that about one eighth of the
         syllables in the text are still written in romanization). Even
         after conversion to characters, it is still impossible for a
         person who is sinographically literate but does not speak
         Taiwanese to follow the story. What is more, the transcriber has
         often chosen characters that are disturbing to native Taiwanese.
         For instance, the word for "gangster" appears on p. 20. In the
         original romanized text this was written
         <i>lo<span class="py">&#729;</span><span class="raised">5</span>-moa<span class="raised">
         5</span></i>, which surely must be cognate with MSM
         <i><span class="py">li&uacute;m&aacute;ng</span></i>
         &#27969;&#27667;, yet the transcription given is &#40056;&#39995;.
         Supposedly the transcriber chose these two particular graphs to
         make sure that his reader pronounces the syllables in question
         exactly the right way (remember that virtually all Chinese
         characters have multiple pronunciations in Taiwanese) to yield the
         meaning "gangster." Nonetheless, it is difficult to imagine why
         the transcriber would not write &#27969;&#27667; and expect his
         reader to make the necessary adjustments for Taiwanese
         pronunciation, rather than using two unusual characters with fish
         radicals. If this proposed sinographic usage is accepted for
         Taiwanese, not only will poor policemen and detective fiction
         writers have to write a total of 49 strokes every time they want
         to mention the word for "gangster" (or if they are typing in a
         computer there is a good chance that they will not be able to find
         these two odd characters expeditiously, if at all), the hapless
         Taiwanese reader will be struck with fishy cognitive dissonance
         ("perch-eel") each time he or she stumbles upon these two ungainly
         graphs. The character version of
         <i>Kho<span class="raised">2</span>-ai<span class="raised">3</span>
         e<span class="raised">5</span>
         siu<span class="raised">5</span>-jin<span class="raised">5</span></i>
         is replete with such perplexing assignments of sinographs to
         Taiwanese morphemes. With this sort of stumbling block being
         thrown in the way of readers and writers alike at every step, it
         is hard to imagine that Taiwanese writing in characters will ever
         become the working script for daily use of the island's
         people.</p>

         <p>This poorness of fit between Taiwanese words and Chinese
         characters is a far more pervasive and complex linguistic
         phenomenon than that of characters having multiple pronunciations
         (<span class="py">du&#333;y&#299;nz&igrave;</span>
         &#22810;&#38899;&#23383;) or "separate" pronunciations
         (<i><span class="py">p&ograve;y&#299;nz&igrave;</span></i>
         &#30772;&#38899;&#23383;) regularly encounters in reading LS or
         MSM texts. In the latter cases, the variant pronunciations are
         usually phonologically and semantically linked, and can be
         explained by historical evolution and grammatical roles (e.g.,
         <i><span class="py">w&ugrave;</span></i> ["loathe, hate"],
         <i><span class="py">&egrave;</span></i> ["evil, wicked"],
         <i><span class="py">&#283;</span></i> ["feel nauseous; feel that
         somebody / something is disgusting"],
         <i><span class="py">w&#363;</span></i> ["oh!"] for &#24801;, where
         the first three readings are etymologically closely related and
         the last is being used to express an interjection). The situation
         concerning the use of characters in written Taiwanese is similar
         to, but even more complicated than, that of Japanese where many
         common <i>kanji</i> have several Sinitic-style pronunciations
         (<i>ondoku</i> or <i>onyomi</i>) and several Japanese-style
         readings (<i>kundoku</i> or <i>kunyomi</i>). For example, &#30452;
         has the Sinitic-style readings <i>choku</i> and <i>jiki</i>
         ("straight, immediate, direct, correct") and the Japanese-style
         readings <i>nao</i>(<i>su</i>) ("fix, correct; revise; convert
         into; [as suffix] "re-, do over"), <i>nao</i>(<i>ru</i>) ("return
         to normal, be fixed / corrected, recover"), <i>tada</i>(<i>chi
         ni</i>) ("immediately"), <i>su</i>(<i>gu</i>) ("immediately;
         readily, easily; right [near]"), and <i>jika</i> (<i>ni</i>)
         ("directly, in person"). Despite all of the widely different
         pronunciations for this single <i>kanji</i>, all of the
         definitions are semantically
         related.<span class="notenumber"><a href="#n5">5</a></span></p>

         <p>Many old-fashioned scholars are of the opinion that they can
         solve the problem of sinographless vernacular morphemes by
         engaging in a search for what are known as
         <i><span class="py">b&#283;nz&igrave;</span></i> &#26412;&#23383;
         ("original characters"). Their belief is premised on the notion
         that words lacking characters are actually old usages that have
         survived in the spoken language and that all one has to do to
         remedy the deficiency is assiduously comb early lexicons and rhyme
         books for characters that sound more or less right and mean
         approximately the same thing. Undoubtedly, such searches sometimes
         result in valid identifications, but in far too many cases the
         rare characters culled from such sources as</p>

         <p><i>Shuo wen jie zi</i> [<i>Explanations of Simple and Compound
         Graphs;</i> 100 <span class="smcaps">CE</span>] and
         <i>Guangyun</i> [<i>Expanded Rhymes;</i> 1008] are self-fulfilling
         prophecies of the obsession with authentication from the national
         past. No matter how diligently one searches, it will be impossible
         to find <i>benzi</i> for such essential Cantonese words as
         <i>nei1</i> / <i>ne1</i> ("this") and <i>m4</i> ("not") because
         they are not based on common Sinitic roots. A similar situation
         obtains for all of the other nonstandard regional vernacular and
         colloquial languages including Pekingese, Sichuanese, and
         Taiwanese.</p>

         <p>If we were to set out to write pure, unadulterated (with as
         little unnecessary Mandarin admixture as possible) spoken
         vernacular Taiwanese in characters, well over 25% of the morphemes
         in a running text would be lacking characters, approximately
         another 25% would be written with arbitrarily chosen (but more or
         less conventionally accepted) homophones or near-homophones and
         concocted special characters, perhaps another 10% would be written
         with extremely rare but correctly identified <i>benzi</i>, leaving
         roughly 40% of the morphemes being written with the "correct"
         characters. In reality, more colloquial styles of Taiwanese would
         undoubtedly have fewer than 40% of their morphemes written with
         characters that everyone could agree were the right
         ones.<span class="notenumber"><a href="#n6">6</a></span> Given
         that so few morphemes in the nonstandard regional vernaculars are
         writable with undisputedly correct <i>hanzi</i>, it is no wonder
         that their literatures have been subject to arrested development,
         to put it mildly.</p>

         <h2 id="part3">III. Mechanisms Whereby One's Mother Tongue is
         Displaced by the National Language</h2>

         <h3 id="part3a">1. <i>Basse-Vulgarisation</i></h3>

         <p>The simplest way to make speakers of a language feel that their
         tongue is inadequate is to stigmatize it as <i>s&ugrave;</i>
         &#20439; ("vulgar"), <i><span class="py">l&#464;y&#468;</span></i>
         &#20442;&#35486; ("slang"),
         <i><span class="py">t&#468;hu&agrave;</span></i> &#22303;&#35441;
         (<i>"earthy</i> speech"), and so on. Once such evaluations are
         accepted by the speakers of a given language themselves, the
         psychological impact is tremendous. Subjected to such
         indoctrination, they lose confidence in their mother tongue and
         may become acutely embarrassed when outsiders hear them speaking
         it. In my travels around China, I constantly encounter the
         following types of scenes:</p>

         <div class="rant">
            <p>An apple seller in Chengdu, "I'm so ashamed of my vulgar
            native tongue."</p>

            <p>A father in a Shanghai bookstore to his pre-school child,
            "Speak <i>putonghua</i> or people will think you're
            stupid."<span class="notenumber"><a href="#n7">7</a></span></p>
            <!-- ==================== PAGE 10 ========================== -->

            <p>A taxi driver in Changsha, "Please forgive me; our local
            language sounds so horrible."</p>

            <p>A female postal worker in &Uuml;r&uuml;mchi, "I'm sorry. My
            Mandarin sounds too much like Uyghur."</p>

            <p>A scholar from Manchuria (Dongbei [the Northeast]) upon
            hearing Peking teenagers conversing in their local tongue,
            "These boys are uneducated. They don't know how to talk
            properly."</p>
         </div>

         <p>Speakers of local languages throughout China are customarily
         complicit in these characterizations. They willingly accept the
         inferior status and deficient nature of their native forms of
         speech in comparison with MSM. In such an environment, a sensitive
         person eventually wishes that he / she could forget his / her
         mother tongue and acquire a new and more respectable language
         (<i>guoyu, putonghua</i>).</p>

         <p>The debasement of local languages and cultures in China
         (whether they are Sinitic or non-Sinitic) is so ubiquitous that
         people become inured to it. They internalize the negative
         stereotypes associated with peripherality and sheer difference
         (from the orthodox language and culture of the center). This
         subtle (but sometimes also brutal) psychological conditioning
         extends even to the names people call themselves and the totemic
         myths with which they identify. For instance, the people of Fujian
         and Taiwan are proud to identify themselves as being from Min, but
         seldom do they consider that the character adopted to write this
         name over two millennia ago (it did not yet exist among the oracle
         bone and bronze inscriptions) includes the infamous
         <i>ch&oacute;ng</i> ("insect; serpent") radical. There it is
         staring you right in the face every time you look at the
         character: a bug inside of a door, but people do not see the
         insect / snake, perhaps because they do not want to see it or
         cannot bear to see it. Here is how Xu Shen explained the character
         used to write <i>m&iacute;n</i> around the year 100
         <span class="smcaps">CE</span>: "Southeastern Yue [i.e., Viet];
         snake race. [The character is formed] from [the] insect / serpent
         [radical and takes its pronunciation from] <i>m&eacute;n</i>."
         &#26481;&#21335;&#36234;&#34503;&#31324;&#20174;&#34411;&#38272;&#32882;
         (Xu 100: 282b)</p>

         <p>Southern Min speakers refer to themselves as
         <i>b&acirc;n-l&acirc;m-l&acirc;ng</i>, which is usually written
         with sinographs meaning "Southern Min person"
         &#38313;&#21335;&#20154;, but should actually be written with
         sinographs meaning "Southern barbarian fellow"
         &#34875;&#21335;&#20738;. (Hong 1988: 343) The graph pronounced
         <i>l&acirc;m</i> in Taiwanese is the notorious
         <i><span class="py">m&aacute;n</span></i> ("barbarians [of the
         south]") as pronounced in MSM. Here is how Xu Shen explains the
         graph used to write <i>l&acirc;m</i> /
         <i><span class="py">m&aacute;n</span></i>: "Southern barbarians
         [who are a] snake race. [The character is formed] from [the]
         insect / serpent [radical and takes its pronunciation from]
         <span class="py">lu&agrave;n</span>
         &#21335;&#34875;&#34503;&#31278;&#20174;&#34411;<img src="minus_shan.gif" width="24" height="15" alt="looks the the &#24018; character, minus the 'shan' at the bottom" />&#32882;."<span class="notenumber">
         <a href="#n8">8</a></span> (Xu 100: 282b) The
         <span class="py">M&aacute;n</span> inhabitants of
         <span class="py">M&#464;n</span> are thus doubly southern, doubly
         barbarian, and doubly serpentine. Since these explanations have
         been enshrined in the most authoritative, foundational dictionary
         of the sinographs, a dictionary which is still invoked with
         reverence today, there is no denying them. The impact that such
         designations have had on the consciousness of those who are on
         both the receiving end and the giving-end is enormous.</p>

         <p>Although their ancient ancestors may not have been Sinitic
         speakers, the Min people of today at least speak one of the
         languages among the so-called Southern and Northern Min "big
         topolects" (or branches) of
         Sinitic.<span class="notenumber"><a href="#n9">9</a></span> Other
         southern peoples who still do not speak a Sinitic language, of
         whom there are many tens of millions, have been subjected to more
         subtle forms of psychological manipulation. Just as the
         <span class="py">M&aacute;n</span> and the
         <span class="py">M&#464;n</span> are said to belong to the
         "serpent race," early Chinese commentators declared that many
         southern peoples were the descendants of dogs. As these accounts
         of ethnogenesis have come down to us, the southerners themselves
         believed that their ancestor was a dog. Now, this would not be a
         problem if one were a speaker of an Indo-European language, since
         most Indo-European speakers hold canines in great esteem. This is
         especially true among Iranian speakers where the dog is considered
         to be possessed of supernatural powers akin to those of gods.
         (White 1991; Brewer <i>et al.</i> 2001) But Sinitic speakers in
         the past have tended to despise the dog as a filthy creature
         worthy only of being eaten. Hence, for non-Indo-European speakers
         within the Sinitic world to state that they are descended from
         dogs puts them in an extraordinarily self-compromising position.
         As to the convoluted process by means of which Tibeto-Burman
         speakers and speakers of other non-Sinitic and non-Indo-European
         languages in the southern portions of what is now China acquired
         myths of canine origins, one may consult the author's
         "<a href="http://www.sino-platonic.org/index.html#n87">Canine
         Conundrums: Dog Ancestor Myths of Origin in Ethnic
         Perspective</a>." (Mair 1998)</p>

         <h3 id="part3b">2. The Pursuit of Prestige</h3>

         <p>In contrast to the negative images attributed to the regional
         vernaculars, Mandarin is constantly portrayed as
         <span class="py">y&#462;zhi</span> &#38597;&#33268; ("refined"),
         <i><span class="py">h&#462;ot&#299;ng</span></i> &#22909;&#32893;
         ("pleasant to the ear"),
         <i><span class="py">li&uacute;ch&agrave;ng</span></i>
         &#27969;&#26274; ("expansive"), and the like. Never mind that
         Mandarin is permeated with Tungusic, Mongolic, Turkic, and other
         "barbarian"
         features,<span class="notenumber"><a href="#n10">10</a></span>
         making it the furthest removed of all the vernaculars from earlier
         forms of Sinitic. Never mind that characterizations of the
         regional vernaculars as "vulgar," etc. and of Mandarin as
         "refined," etc. are entirely subjective. And never mind that there
         are plentiful resources for vulgarity in Mandarin, whereas a
         regional vernacular such as that of Suzhou may actually be
         hyperrefined. The fact remains that Mandarin <i>is</i> the
         prestige language of the People's Republic of China, the Republic
         of China on Taiwan, the Republic of Singapore, and the Hongkong
         Special Administrative Region.</p>
         <!-- ==================== PAGE 12 ========================== -->

         <p>The commanding position of Mandarin is premised upon the
         following factors which do not require amplification in a paper of
         this length: 1. its association with military might, political
         power, and economic clout, 2. its designation as the official
         national language (<i>putonghua, guoyu, huayu</i>), 3. its
         background in the <i>koine</i> of a thousand and more years ago,
         4. a long history of written literature in a wide variety of
         styles, 5. large amounts of scholarly research and reference
         materials devoted to it. In comparison with Mandarin which
         possesses all of these overwhelmingly favorable conditions, the
         regional vernaculars have a difficult time competing for
         attention. The result is that the dominant position of Mandarin
         grows ever stronger, while the place of the regional vernaculars
         grows ever weaker. The dynamics of the relationship between the
         regional vernaculars and MSM may be summed up in the following
         slogan seen on a wall in Shanghai during the summer of 2002:
         <i><span class="py">Z&#363;nzh&ograve;ng w&eacute;nhu&agrave;;
         xu&eacute; p&#468;t&#333;nghu&agrave;</span></i>
         &#23562;&#37325;&#25991;&#21270; &#23416;&#26222;&#36890;&#35441;
         ("Respect culture; learn Mandarin"). The obverse is obvious.</p>

         <h3 id="part3c">3. The Myth of Monolingualism</h3>

         <p>The myth that there is only a single Chinese (Han) language and
         that it is spoken by more than a billion people (Chen 1999: 1) is
         as widespread, relatively recent, persistent, and obnoxiously
         misleading as the myth that the Great Wall was the only man-made
         object on earth that could be seen from the moon. Furthermore,
         both of these myths were perpetrated by Westerners and foisted
         upon the Chinese. Fortunately, the Ripleyesque "Believe It or
         Not!" nonsense about the Great Wall (Waldron 1990: 214, 253n596)
         has at last been disproven by astronauts who went to the moon and
         could not see any man-made structures on earth with the naked eye.
         The myth about a nation of Chinese speaking a single language made
         up of countless "dialects" with only negligible differences of
         "accent" continues to root itself ever more deeply in the global
         consciousness, plaguing common sense and scientific observation
         all the while.</p>

         <p>The Chinese originally knew better. Right through the Ming
         Dynasty, the inhabitants of the Central Kingdom realized that
         there were innumerable varieties of mutually unintelligible speech
         in the provinces and districts of their vast empire, and they knew
         equally well that if one wanted to convey one's sentiments to
         persons fifty or a hundred miles away, one had better do one of
         the following: a. hire an interpreter, b. hire a scribe, c. learn
         the dead written language (LS) which no one had used as a flexible
         spoken medium for at least the previous two millennia, or, if one
         did not have the considerable resources to do a., b., or c., try
         d. learn a few hundred or a couple thousand characters so that one
         could pick one's way through texts written in the national
         vernacular (VS). It was only with the advent of sizable numbers of
         Westerners in China that people began to get the foolish notion of
         there being but a single Chinese (Han) language. This was the
         result of foreigners' own incomprehension of the multitude of
         local tongues as well as their utter confusion over the nature of
         the exotic Chinese script and its complex relationship to spoken
         languages.</p>

         <p>The mischief surrounding the myth of monolingualism was further
         exacerbated during the middle of the first half of the twentieth
         century with the mistranslation of the word <i>fangyan</i> as
         "dialect."<span class="notenumber"><a href="#n11">11</a></span> In
         linguistic classification, a dialect is normally defined as a
         variety of a certain language, the language being the larger unit
         and the dialect the smaller unit. Furthermore, although different
         dialects of the same language may have slightly different
         phonological, lexical, and grammatical properties, they are
         usually considered to be mutually intelligible. Naturally,
         subdialects -- which are smaller, lower level divisions of
         dialects -- have fewer differences among each other than do
         dialects. By uniformly translating <i>fangyan</i> as "dialect,"
         this gives the misleading impression that all Sinitic languages
         are mutually intelligible, but this is patently not the case.</p>

         <p>As for units of classification above the level of subdialects,
         dialects, and languages, there are -- at the top -- families, then
         groups, and then branches. Taking the Indo-European language
         family as a well-known example, it is composed of the following
         groups: Balto-Slavic, Germanic, Celtic, Italic (or Romance),
         Albanian, Hellenic, Armenian, Anatolian, Indo-Iranian, and
         Tocharian. The Germanic group, in turn, is composed of the
         following branches: North Germanic, East Germanic, and West
         Germanic. English is one of the languages of the West Germanic
         branch, which also includes Frisian, Dutch, German, and Yiddish.
         The dialects of English, which are all mutually intelligible (!),
         include those of Australia, Boston, Texas, Newcastle, Lancashire,
         and so forth. The subdialects of the Boston area include Back Bay,
         Beacon Hill, Cambridge, Quincy, Roxbury, and so forth.</p>

         <p>Since the inauguration of modern linguistic science in China
         from the twenties and thirties of the twentieth century, this
         entire system of classification has been imported to China --
         lock, stock, and barrel: family
         (<i><span class="py">x&igrave;</span></i> &#31995;), group
         (<i><span class="py">z&uacute;</span></i> &#26063;), branch
         (<i><span class="py">zh&#299;</span></i> &#25903;), language
         (<i><span class="py">y&#468;y&aacute;n</span></i>
         &#35486;&#35328;), dialect
         (<i><span class="py">f&#257;ngy&aacute;n</span></i>
         &#26041;&#35328;)<i>, sub-dialect</i>
         (<i><span class="py">c&igrave;f&#257;ngy&aacute;n</span></i>
         &#27425;&#26041;&#35328;), which I shall refer to below as the
         FGBLDS system. This system of classification is now adopted by
         Chinese linguists for all the other languages of the world,
         including the many non-Sinitic languages of China, but it has
         strangely never been applied comprehensively to the Sinitic group
         of languages. In the ensuing paragraphs, I shall examine the
         reasons for this failure to include Sinitic among the language
         groups of the world that are eligible for unrestricted
         classification.</p>

         <p>The importation of the FGBLDS system would appear to be all
         well and good by itself. A serious problem arises, however, from
         the fact that this modern system of linguistic classification has
         been superimposed upon a preexisting indigenous tradition of
         language studies. Traditional language studies in China, called
         <i><span class="py">xi&#462;oxu&eacute;</span></i>
         &#23567;&#23416;, ("minor learning") in distinction to
         <i><span class="py">d&agrave;xu&eacute;</span></i>
         &#22823;&#23416; ("major learning") which deals with questions of
         morality and values, emphasized investigations of the sounds,
         structure, and meaning of characters. It did not include grammar,
         etymology,<span class="notenumber"><a href="#n12">12</a></span>
         morphology, taxonomy, cladistics, or dialectology.</p>
         <!-- ==================== PAGE 14 ========================== -->

         <p>At the turn of the Western Han to the Eastern Han (around the
         beginning of the Common Era), the multitalented literatus, Yang
         Xiong (53 <span class="smcaps">BCE</span>-18
         <span class="smcaps">CE</span>) is said to have compiled a
         synonimicon known by the short title <i>Fangyan</i> (it also has a
         much longer title which shall not concern us here). By the term
         <i><span class="py">f&#257;ngy&aacute;n</span></i>
         &#26041;&#35328;, Yang Xiong (or whoever the compiler of this work
         actually was) meant simply "[examples of synonomous terms [from
         various] places." He did not mean what we mean by "dialect,"
         namely "a variety of language that is used by one group of persons
         and has features of vocabulary, grammar, or pronunciation
         distinguishing it from other varieties <i>of the same language</i>
         that are used by other groups." (<i>Encyclopaedia Britannica</i>,
         15th ed. [1988],4: 63a [emphasis added]) After the appearance of
         Yang Xiong's great work, a tradition of <i>fangyan</i> studies
         (now referred to as
         <i><span class="py">f&#257;ngy&aacute;nxu&eacute;</span></i>)
         developed. Its aims and methods remained essentially those of Yang
         Xiong: to catalog synonymous terms from various localities.</p>

         <p>When traditional <i>fangyan</i> studies encountered modern
         dialectology during the first half of the twentieth century, chaos
         ensued. A full accounting of the stark confusion resulting from
         the clash between <i>fangyanxue</i> and dialectology requires
         separate treatment. Suffice it here to say that the confrontation
         of the traditional and modern disciplines resulted in the creation
         of such monstrosities as the notion of
         <i><span class="py">d&agrave;f&#257;ngy&aacute;n</span></i> ("big
         <i>fangyan"</i>), which is comparable to a branch in the FGBLDS
         system. It also led to the disastrous identification of
         <i>fangyan</i> and "dialect" as used in the nomenclature of the
         FGBLDS system.</p>

         <p>Now, some may object that no harm has been done by coining
         euphemistic neologisms such as <i>dafangyan</i> for "branch" and
         equating <i>fangyan</i> with "dialect" because, after all, these
         terms mean approximately the same things. Such, however, is not
         the case. Regardless of what Yang Xiong may have originally meant
         by <i>fangyan</i>, by the late imperial period of Chinese history,
         <i>fangyan</i> had come to mean roughly "variety of speech typical
         of a certain place." The concept of "place" in this definition
         could be near or far, large or small. It is absolutely essential
         to bear in mind that the concept of <i>fangyan</i> in traditional
         Chinese language studies had no bearing on the issue of
         relatedness, whereas "dialect" in the FGBLDS system necessarily
         implies linguistic consanguinity. (The issues concerning
         linguistic classification being discussed here must not be clouded
         by non-linguistic and non-classificatory usages of the word
         "dialect.") In an earlier paper (Mair 1991: 4-5), I demonstrated
         that <i>fangyan</i> during the late Qing period could just as
         easily refer to a foreign language as it did to one of the Sinitic
         languages. Here I shall reinforce that demonstration with
         additional proof by pointing out that, in 1863, the renowned
         statesman Li Hongzhang established a Guang Fangyan Guan [Broad
         <i>fangyan</i> Office] where <i>fangyan</i> clearly refers to
         foreign languages. In emulation of Li Hongzhang's innovation,
         similar translation and interpreting bureaus were set up in the
         provinces. These were called
         <i><span class="py">f&#257;ngy&aacute;n
         xu&eacute;t&aacute;ng</span></i> &#26041;&#35328;&#23416;&#22530;
         [<i>fangyan</i> academies]. (Brunnert and Hagelstrom 1911: 254,
         263) It would be an egregious error to translate <i>fangyan</i> in
         such situations as "dialect," since the <i>fangyan xuetang</i>
         were in reality colleges where interpreters learned how to
         translate to and from English, German, French, Italian, Japanese,
         and other languages.</p>

         <p>In an attempt to clear up the massive confusion resulting from
         the mistranslation of <i>fangyan</i> as "dialect," I long ago
         proposed that <i>fangyan</i> be rendered in English as "topolect,"
         which conveys precisely the meaning of the Chinese term. I am
         pleased to report that this rendering has now been accepted by
         such major, authoritative dictionaries as <i>The American Heritage
         Dictionary of English</i> (2000: 1822a) and the <i>ABC
         Chinese-English Dictionary</i> (1996: 161b). A <i>fangyan</i> can
         be huge (Cantonese, Mandarin) or it can be tiny (Dunhuang,
         Meixian); it can be near (Pekingese, Shanghainese), or it can be
         distant (German, Hindi). The term <i>fangyan</i> simply designates
         a variety of speech characteristic of a certain place, no matter
         how small, large, near, or far it may be, and no matter whether it
         be related to other Sinitic languages or not</p>

         <p>If "topolect" solves the problem of how to translate
         <i>fangyan</i> into English, how then do we translate "dialect"
         into Sinitic? The best way to approach the problem is
         etymologically, just as was done with the word <i>fangyan</i>. The
         prefix and root (both Greek derived) of the word dialect
         respectively mean "through" and "[form of] speech." According to
         this analysis, the most straightforward and exact translation of
         "dialect" into Mandarin would be
         <i><span class="py">t&#333;ngy&aacute;n</span></i>
         &#36890;&#35328;. Aside from being far more accurate than
         <i>fangyan</i>, this rendering highlights the fact that dialects
         (especially in the FGBLDS system) are normally mutually
         intelligible. (The supreme irony of the old translation of
         "dialect" as <i>fangyan</i> is that most Sinitic <i>fangyan</i>
         are mutually unintelligible.) For those who are hesitant to accept
         this radically new Mandarin translation of "dialect," it may be
         pointed out that "dialect" is ultimately derived from the same
         Greek roots (<i>dia-</i> ["through"] + <i>legein</i> ["to speak"])
         as English "dialog(ue)" which means "a conversation between two or
         more people." Engaging in a dialog obviously requires a high
         degree of mutual intelligibility. It should be further noted that
         the usual translation of English "dialog" as Mandarin
         <i><span class="py">du&igrave;hu&agrave;</span></i>
         &#23565;&#35441; is correct in that it conveys the sense of
         talking across to or back and forth with someone.</p>

         <p>Once we clarify the distinction between traditional
         <i>fangyanxue</i>, on the one hand, and dialectology and
         linguistic cladistics on the other, the fallacy of the monolingual
         myth is exposed. Up to the present moment, however, a vigorous
         pretense is maintained that there has only been a single Hanyu
         (Sinitic) from its very beginning and wherever its countless
         <i>fangyan</i> are spoken. According to this phantasmagoric
         scenario, Hanyu is uniquely uniform throughout all time and space.
         Diachronically, it has no counterpart to Latin or Sanskrit;
         synchronically, there are no significant differences between
         Southern Min and Mandarin. It is all just one stupendous
         Hanyu.</p>

         <p>When one looks at language trees for Sino-Tibetan published in
         China, virtually all of them show an incredibly elaborate
         branching structure for Tibeto-Burman languages, but none
         whatsoever for Sinitic. Whereas there is a proliferation of scores
         of Tibeto-Burman branches and languages, many of which have only a
         few thousand to several ten thousand or a hundred thousand
         speakers, there is but one unbranched line for Hanyu with all of
         its billion and more speakers. The portrayal of Hanyu as utterly
         monolithic is a politico-cultural fiction. In secret, no honest
         linguist who has studied the huge lexical, phonological, and
         grammatical differences between LS and VS, and among the numerous
         varieties of VS, could possibly accept the diachronic and
         synchronic uniformity of Hanyu.</p>

         <p>If the truth be told, there have been coded attempts to break
         out of the ideological straitjacket that demands acquiescence in
         the monolingual myth. Both in the <i>Nationalities</i> volume
         (1986: 554b) and in the <i>Languages and Scripts</i> volume (1988:
         523b) of the authoritative <i>Zhongguo da baike quanshu</i>, we
         find the following rather weird sentence:
         <i><span class="py">H&agrave;ny&#468; z&agrave;i y&#468;y&aacute;n
         x&igrave;sh&#468; f&#275;nl&egrave;i zh&#333;ng
         xi&#257;ngd&#257;ngy&uacute; y&#299; ge y&#468;z&uacute; de
         d&igrave;w&egrave;i</span></i>
         (&#28450;&#35486;&#22312;&#35486;&#35328;&#20418;&#23660;&#20998;&#39006;&#20013;&#30456;&#30070;&#26044;&#19968;&#20491;&#35486;&#26063;&#30340;&#22320;&#20301;
         "In linguistic classification, Hanyu [Sinitic] occupies a place
         equivalent to a language group"). Apart from its grammatical and
         logical defectiveness (Hanyu... is equivalent to... a position),
         this sentence is peculiar in many other respects. First of all, in
         both instances it occurs in an article on "languages of Chinese
         national minorities," not in an article on Sinitic or even
         Sino-Tibetan. Secondly, it was written not by a specialist on
         Sinitic, but by a well-known expert on non-Sinitic languages, Fu
         Maoji. Thirdly, it is very careful not to come right out and say
         that Hanyu <i>is</i> (<i>sh&igrave;</i> &#26159;) a language
         group, but only that it occupies a position <i>equivalent to</i>
         (<i><span class="py">xi&#257;ngd&#257;ng y&uacute;</span></i>) a
         language group. Fu Maoji is bound by the fiction of a monolithic
         Hanyu not to admit that Hanyu really is a language group, for --
         if he were to do so -- it would automatically mean that it is
         composed of more than a single language. At the same time, common
         sense and linguistic reality demand recognition of the
         multiplicity of constituents within Hanyu. Hence the blatantly
         evasive wording.</p>

         <p>What does all of this discussion about the mistranslation of
         <i>fangyan</i> as "dialect" and the monolingual myth concerning
         Sinitic have to do with the main topic of this paper? Simply this:
         if there is only a single Hanyu, and if all of its constituent
         members are mutually intelligible dialects, then there is no need
         for education in the regional vernaculars, nor is there any reason
         to write them down. The only thing that matters is MSM /
         <i>guoyu</i> / <i>putonghua</i> since it miraculously subsumes and
         exemplifies all other varieties of Hanyu, which supposedly differ
         only in insignificant ways.</p>

         <p>This sort of elaborate masquerade might have succeeded decades
         ago when naive Westerners labored under the misapprehension that,
         because China only has one script and literate people from
         different parts of the country can read what is written in it,
         then surely there must only be one language in China. More recent
         studies, however, have revealed the following pertinent facts:</p>
         <!-- ==================== PAGE 17 ========================== -->

         <ol>
            <li>Those who wish to become literate must learn MSM, which has
            its own grammar, syntax, phonology, lexicon, and idioms that --
            for most people -- vary substantially from those of their own
            mother tongue. (In the past, they would have had to learn LS,
            which was even much more difficult to acquire than MSM.)</li>

            <li>The regional vernaculars are almost never written down. In
            the rare instances when heroic attempts are made to write them
            in an integral fashion, speakers of other Sinitic languages --
            including Mandarin -- cannot read them at all, or read them
            only partially and with great
            effort.<span class="notenumber"><a href="#n13">13</a></span></li>

            <li>Recent investigations of the grammar, morphology, syntax,
            vocabulary, and phonology of the regional Sinitic vernaculars
            show that they differ among themselves as greatly as do the
            Indo-European languages of Europe. (Yue Hashimoto 1972;
            Killingley 1993; Matthews and Yip 1994; Chappell 2001)</li>

            <li>Personal observations by fluent native and foreign speakers
            of Mandarin, Cantonese, and other Sinitic languages affirm that
            there are hundreds of varieties of mutually unintelligible
            speech classifiable as Sinitic.</li>
         </ol>

         <p>Although the monolingual myth is starting to crumble under the
         weight of empirical evidence that there truly exist many different
         Sinitic languages belonging to a number of branches, the
         politico-cultural conceit of a single Hanyu for all time and all
         space dies hard.</p>

         <h3 id="part3d">4. The Sinographic Snare</h3>

         <p>Still more seductive than the myth of monolingualism is the
         allure of the sinographic script. Even illiterate Chinese and
         foreigners who cannot recognize the simplest character are
         convinced not only that the script is a thing of great beauty, but
         that it is the cement that binds Chinese society and culture
         together. The corollary of this belief is that, sans the script,
         Chinese society and culture would dissipate or even disappear. But
         -- assuming that one accepts this mode of thought -- the
         characters present speakers of the nonstandard regional
         vernaculars with a troublesome dilemma: if they wish to remain
         patriotically loyal to the Chinese polity and culture, then they
         naturally must embrace the sinographic script wholeheartedly, yet,
         by doing so they make it well-nigh impossible to write their
         mother tongue.</p>

         <p>Zheng Liangwei ( 1990) has summarized some of the deficiencies
         of writing restricted solely to Chinese characters:</p>

         <ol>
            <li>low capability of auditory symbolism</li>

            <li>easy to confuse literary and vernacular styles, making it
            difficult to bring writing more in line with speech</li>

            <li>temporal imprecision and social division lead to linguistic
            confusion</li>

            <li>encourages cultural self-centeredness</li>

            <li>causes intellectuals to adopt a prejudicial view toward
            other scripts and cultures</li>

            <li>inconvenient for cultural exchange</li>

            <li>requires an educational system that overlooks, is biased
            against, or even prohibits indigenous, local cultures</li>
         </ol>

         <p>While I am not confident that I fully understand or agree with
         each of Zheng's points, there is no doubt that the sinographic
         script is inimical to writing the nonstandard vernaculars (i.e.,
         anything other than MSM, which is a very special type of
         deracinated national vernacular), whether it be Taiwanese,
         Cantonese, Shanghainese, Sichuanese, or even Pekingese.</p>

         <p>Wang Yude, who has authored an informative series of lectures
         on Taiwanese language, puts it this way: "When one uses Chinese
         characters to write Taiwanese, the first problem one confronts is
         that it is impossible to find correct characters for the more
         frequently appearing vocabulary items...." (Wang 1993: 62-63) Wang
         states that, according to his investigations, it is impossible to
         find correct characters for approximately one out of every four
         Taiwanese vocabulary items. He gives the following list of pairs
         of Taiwanese words where the items in the left column are
         vernacular terms, together with their customary sinographic form,
         and the items in the right column are the literary readings of the
         relevant Chinese characters. The translations in the center
         column, which apply to the items in both the left and right
         columns, have been added by me.</p>

         <table>
            <colgroup>
               <col span="2"
                    id="vernacular" />
               <col id="trans" />
               <col span="2"
                    id="literary" />
            </colgroup>

            <thead>
               <tr>
                  <th colspan="2">vernacular term</th>

                  <th>translation</th>

                  <th colspan="2">literary reading</th>
               </tr>
            </thead>

            <tbody>
               <tr>
                  <td>s&uacute;i</td>

                  <td>&#28418;&#20142;</td>

                  <td>beautiful</td>

                  <td>&#32654;</td>

                  <td>b&iacute;</td>
               </tr>

               <tr>
                  <td>si&aacute;u</td>

                  <td>&#30219;</td>

                  <td>wild, crazy</td>

                  <td>&#29378;</td>

                  <td>k&ocirc;ng</td>
               </tr>

               <tr>
                  <td>s&aacute;n</td>

                  <td>&#30246;</td>

                  <td>thin</td>

                  <td>&#30246;</td>

                  <td>s&oacute;<span class="py">&#729;</span></td>
               </tr>

               <tr>
                  <td>ko&acirc;n</td>

                  <td>&#39640;</td>

                  <td>tall</td>

                  <td>&#39640;</td>

                  <td>ko</td>
               </tr>

               <tr>
                  <td>t&eacute;</td>

                  <td>&#30701;</td>

                  <td>short</td>

                  <td>&#30701;</td>

                  <td>to&aacute;n</td>
               </tr>

               <tr>
                  <td>khi&#257;</td>

                  <td>&#31449;</td>

                  <td>stand</td>

                  <td>&#35918;</td>

                  <td>s&#363;</td>
               </tr>

               <tr>
                  <td>si&#363;<span class="raised">n</span></td>

                  <td>&#24819;</td>

                  <td>think</td>

                  <td>&#24819;</td>

                  <td>si&oacute;ng</td>
               </tr>

               <tr>
                  <td>h&ecirc;ng</td>

                  <td>&#36996;</td>

                  <td>return</td>

                  <td>&#36996;</td>

                  <td>ho&acirc;n</td>
               </tr>

               <tr>
                  <td>h&#333;<span class="py">&#729;</span></td>

                  <td>&#32102;
                  &#38632;<span class="notenumber"><a href="#n14">14</a></span></td>

                  <td>give</td>

                  <td>&#32102;</td>

                  <td>kip</td>
               </tr>

               <tr>
                  <td>ko&#257;i</td>

                  <td>&#40799;&#40812; , &#25654;&#22890;</td>

                  <td>go against</td>

                  <td>&#20054;</td>

                  <td>koai</td>
               </tr>

               <tr>
                  <td>l&acirc;ng</td>

                  <td>&#20154;</td>

                  <td>person</td>

                  <td>&#20154;</td>

                  <td>j&icirc;n</td>
               </tr>

               <tr>
                  <td>bah</td>

                  <td>&#32905;</td>

                  <td>flesh</td>

                  <td>&#32905;</td>

                  <td>ji&oacute;k</td>
               </tr>

               <tr>
                  <td>thang</td>

                  <td>&#31383;&#25143;</td>

                  <td>window</td>

                  <td>&#31383;</td>

                  <td>chhong</td>
               </tr>

               <tr>
                  <td>l&#257;u-p&#275;</td>

                  <td>&#29238;&#35242;</td>

                  <td>father</td>

                  <td>&#32769;&#29238;</td>

                  <td>l&oacute;<span class="py">&#729;</span>-h&#363;</td>
               </tr>

               <tr>
                  <td>tsa-po<span class="py">&#729;</span></td>

                  <td>&#30007;&#20154;</td>

                  <td>man</td>

                  <td>&#26619;&#21754;</td>

                  <td>tsa-po<span class="py">&#729;</span></td>
               </tr>
            </tbody>
         </table>

         <p>It is obvious that some of the characters in the left column
         have been chosen because they represent precisely or roughly the
         meaning of the Taiwanese terms in question, while other characters
         have been selected because they are either exact or near
         homophones of the given Taiwanese terms, but in no case do both
         the meanings and the sounds of the characters "borrowed" (Wang's
         word [<span class="py">j&#462;iji&egrave;</span>
         &#20551;&#20511;]) to represent the Taiwanese words coincide. The
         same clumsy techniques, with all of their concomitant difficulties
         and ambiguities, apply to sinographic writing of the other
         vernaculars.</p>

         <p>Already in the 1870s, Edward Parker realized that there were
         many words in the Sinitic vernaculars that he called
         "characterless." He compiled a list of a hundred words from the
         vernacular languages of Peking, Hangzhou, Canton, Fuzhou, and the
         Hakka that were considered to have no sinographic written form.
         With effort, he was able to assign characters to thirteen of these
         words, leaving 87 words without characters to match them.</p>

         <h2 id="part4">IV. The Prognosis for Written Taiwanese</h2>

         <p>Technically, writing Taiwanese in romanization would be a very
         easy thing to do. Indeed, during the twentieth century tens of
         thousands of Taiwanese-speaking individuals quickly became
         literate in what is popularly known as church romanization.
         (Christine Lin 1999; Albert Lin 1999) Although, due to the
         presence of tone marks, hyphens used to connect the syllables of
         words, superscript nasals, and a few other special symbols, church
         romanization appears somewhat ungainly, it is actually a fully
         functional script. It is used mainly within the Presbyterian
         Church for religious purposes, but many people (including my
         brother's own mother-in-law) have used it for private
         correspondence and to write literary works. But church
         romanization has not spread widely in secular society, probably
         because of its close identification with Christianity. There are
         at least half a dozen competing romanization schemes for Taiwanese
         on the scene, with new ones popping up all the time. (Chiung
         2003)</p>

         <p>The problem is not that workable schemes have not been proposed
         for the romanization of Taiwanese, but that the people of Taiwan
         do not have the collective will to ensure that a superior scheme
         is selected as an official script. The main obstacle to the
         creation of an effective romanization for the whole of society is
         undoubtedly the affection for Chinese characters felt by a
         majority of the population. Despite the manifest inadequacy of the
         characters for writing Taiwanese, people keep striving to force
         their mother tongue into the sinographic mold. For all of the many
         reasons outlined above, I believe that such efforts are doomed to
         failure. Sinographically written Taiwanese will never become a
         widespread phenomenon. If, on the other hand, Taiwan continues to
         maintain its current quasi-independence for a considerable period
         of time, it is quite possible that a romanized script for
         Taiwanese will gradually and naturally develop as the result of
         electronic information processing, the inducements of
         international commerce and cultural exchange, the increasingly
         frequent insertion of English words in sinographic texts and of
         romanized Taiwanese terms in English texts, and similar
         eventualities.</p>

         <p>We must remember that it has only been about fifteen years
         since the Taiwanese people have been sufficiently free to discuss
         publicly the desirability of writing their mother tongue. When the
         first attempts were made to raise the status of Taiwanese, they
         caused a violent reaction among proponents of <i>guoyu</i> and all
         that it entailed. The following newspaper article is a good
         example of the moral indignation and political outrage that is
         capable of being directed against the regional vernaculars:</p>

         <div class="rant">
            <h3>"Cultural Taiwan Independence" Is More Terrifying Than
            "Political Taiwan Independence"</h3>
            <!-- ==================== PAGE 21 ========================== -->

            <h3>"The Transformation of Taiwanese into a Written Language"
            Exposes Separatism</h3>

            <p>"Taiwanese" is a <i>fangyan</i> of Chinese
            (<i><span class="py">Zh&#333;nggu&oacute;hu&agrave;</span></i>),
            similar to Cantonese and Shanghainese. Although its sounds are
            different, the shapes of the characters are the same. There is
            truly no reason for it to set itself apart from Chinese
            characters.</p>

            <p>A person harboring strong desires for "Taiwan Independence"
            (<span class="py"><i>T&aacute;i</i>[<i>w&#257;n</i>]
            <i>d&uacute;</i>[<i>l&igrave;</i>]</span>) and employing
            twisted theories has falsely claimed that Taiwanese can be
            transformed into a written language. What is more, this
            individual has made a commitment to this doctrine by personally
            compiling a dictionary [of Taiwanese] to prove the feasibility
            of his theories. Not only is such blather ridiculous, it makes
            you clench your fists in anger. Everybody knows that Taiwanese
            is a branch (<i><span class="py">y&#299; zh&#299;</span></i>)
            of Hanic (<i><span class="py">h&agrave;ny&#468;</span></i>),
            not a language (<i>yuyan</i>) with an independent (<i>duli</i>)
            system. All the more, it is not the language of a people
            (<i><span class="py">m&iacute;nz&uacute;</span></i>
            &#27665;&#26063;), but rather should be considered as a kind of
            <i>fangyan</i> within Chinese (<i>Zhongguohua</i>). Any of the
            <i>fangyan</i> already has a script (i.e., the Chinese
            characters), so there is absolutely no need to "transform it
            into a written language"
            (<i><span class="py">w&eacute;nz&igrave;hua</span></i>
            &#25991;&#23383;&#21270;). Using Taiwanese to read the Four
            Books and the Five Classics (<i><span class="py">s&igrave;
            sh&#363; w&#468; j&#299;ng</span></i>
            &#22235;&#26360;&#20116;&#32147;) is like using Hakka,
            Cantonese, or Shanghainese to read them. Although all of these
            <i>fangyan</i> are mutually incommunicable
            (<i><span class="py">w&uacute; f&#462; g&#333;u
            t&#333;ng</span></i> &#28961;&#27861;&#28317;&#36890;), they
            all belong to the "Hanic Family"
            (<i><span class="py">H&agrave;ny&#468; x&igrave;</span></i>
            &#28450;&#35486;&#31995;), so they all can use written Chinese
            (<i><span class="py">Zh&#333;ngw&eacute;n</span></i>) to
            communicate, and there is no question of being "transformed
            into a written language."</p>

            <p>Within the territory of our country, the different languages
            having different scripts are Manchu, Mongolian, and Tibetan
            because these are different races
            (<i><span class="py">zh&#466;ngz&uacute;</span></i>
            &#31278;&#26063;) belonging to different language families
            (<i><span class="py">y&#468;x&igrave;</span></i>
            &#35486;&#31995;) and having different cultures. But Taiwanese
            is an authentic <i>fangyan</i> that uses written Chinese
            (<i>Zhongwen</i>) to communicate, so there is really no
            necessity for it to be "transformed into a written language."
            What is more, it definitely cannot be divided from Hanic and
            set up a separate language family.</p>

            <p>After the Second World War, when the backward peoples of
            Africa were liberated from the colonial oppression of the
            European imperialist countries, because they only had languages
            but no scripts, they had to promote the "transformation into
            written language" of their native tongues to strive for
            survival in modern, civilized societies. In contrast, Taiwanese
            (the correct name should be Southern Min), in terms of its
            standing with regard to culture and civilization, is not
            inferior to any of the <i>fangyan</i> on the mainland, but is
            actually more ancient and elegant. Why, then, would the
            Taiwanese want to reduce themselves to the level of the
            backward peoples of Africa? This is incomprehensible and shows
            the self-destructiveness of "Taiwan independence"
            (<i>Taidu</i>).</p>

            <p>"Taiwanese" has the same script
            (<i><span class="py">w&eacute;nz&igrave;</span></i>
            &#25991;&#23383;) as the national language (<i>guoyu</i>), and
            Taiwanese compatriots belong to the same nation as the
            compatriots of the other provinces [of China]. Willfully
            wanting to set up another script and another nation
            (<i><span class="py">m&iacute;nz&uacute;</span></i>) are
            obvious indications of "separatist consciousness"
            (<i><span class="py">f&#275;nl&iacute; y&igrave;shi;</span></i>
            &#20998;&#38626;&#24847;&#35672;) This kind of "Cultural Taiwan
            Independence" is more frightening than "Political Taiwan
            Independence" because political questions can be solved through
            compromise, but once culture takes root, it is hard to
            eradicate. I recognize that Taiwanese is a precious heritage of
            our [country], and that it is deserving of protection and
            respect. I also personally approve of using discretion to
            increase the amount of <i>fangyan</i> programs in broadcasting,
            and all the more would encourage our brothers from other
            provinces to learn Southern Minnan [speech]. But I am strongly
            opposed to viewing [Southern Minnan] as an "independent
            language" (<i><span class="py">d&uacute;l&igrave;
            y&#468;y&aacute;n</span></i>). I believe that "the
            transformation of Taiwanese into a written language" is a plot
            to destroy the country. Everyone should rise up to denounce it
            orally and in writing. (Taipei City, Li
            Shengwei)<span class="notenumber"><a href="#n15">15</a></span></p>
         </div>

         <p>The violence of the rhetoric in this article is palpable. The
         author feels threatened by the prospect of written Taiwanese and
         proposes extreme measures (<i><span class="py">k&#466;uzh&#363;
         b&#464;f&aacute;</span></i> &#21475;&#35461;&#31558;&#20240;
         ["execute with the mouth and cut down with the pen"]) to prevent
         its happening. The irrational fury and uncomprehending linguistic
         fallacies that pervade this indignant denunciation of the mere
         suggestion of written Taiwanese are clear proof of the kinds of
         intimidation and pressures that have been brought to bear against
         writing in the regional vernaculars in China. Yet, no matter how
         much ranting and raving Li Shengwei and his ilk might indulge in,
         the speakers of Taiwanese themselves feel a genuine and legitimate
         need to transform it into a written language, Furthermore, not
         only are they keenly aware of the tremendous differences between
         Taiwanese and Mandarin, they are also deeply frustrated by the
         intractability of the Chinese characters when it comes to writing
         out their mother tongue. With the lifting of martial law, they
         would no longer be intimidated by the fierce, dogmatic rhetoric of
         Great Han chauvinists like Li Shengwei. Hence, no sooner had a
         modicum of democracy been instituted on the island than a surge of
         publications concerning Taiwanese language ensued. Among these
         were <i>Taiyu wenzhai</i> (<i>Tai5-gi2 bun5-tiah4; Taiwanese
         Digest;</i> vols. 1-4, late eighties) and <i>T&acirc;i-B&ucirc;n
         Thong-S&igrave;n</i> (<i>Newsletter for Written Taiwanese;</i>
         late nineties). And now there are numerous easy-to-find sites
         dealing with Taiwanese on the World Wide Web.</p>

         <p>Among the more intriguing pedagogical offerings of recent years
         is a large format, thin textbook entitled <i>Kejia Taiyu
         jiaocai</i> [<i>Teaching Material for Hakka Taiwan Language</i>]
         that was published in September, 1998 by the Taipei Municipal
         Government under Mayor Chen Shuibian and with an energetic preface
         by him. The first volume in a series entitled <i>Taibei shi muyu
         jiaocai</i> [<i>Mother Language Teaching Materials for Taipei
         City</i>], this is a colorful, luxuriously produced textbook. In
         his Preface, Mayor Chen begins and ends by invoking pluralism
         (<i><span class="py">du&#333;yu&#462;n</span></i>
         &#22810;&#20803;), and in the middle stresses pedagogy and
         teaching materials for the mother tongue (<i>muyu</i>). It is
         curious that the Mayor, whose Preface is written in perfect
         Mandarin (<i>guoyu</i>), stresses the need for providing
         instruction in Hakka and aboriginal languages, which had been
         suppressed during the previous forty to fifty years as a result of
         the distorted views of the preceding government, but he does not
         mention a word about Taiwanese (Hoklo). The Mayor's Preface is
         full of subtexts that speak to his political opponents in Taipei,
         to Taiwanese nationalists farther south on the island, to the
         Communist authorities and the people of mainland China, and to his
         future aspirations for the presidency. Although it is only about
         550 characters long, this is a fascinating document that deserves
         full analysis elsewhere. Suffice it to say here only that the
         Mayor is an astute analyst of social and cultural dynamics who
         recognized early on the centrality of language in the political
         equation that will determine the fate of Taiwan.</p>

         <p>The Taiwanese author Zhong Zhaozheng has likened writing in
         Mandarin (i.e.,
         <i><span class="py">zh&#333;ngw&eacute;n</span></i>) to a type of
         translation. In a thoughtful article first published in <i>Lianhe
         bao</i> (<i>United Daily</i>), Zhong shares his reflections upon
         the rise of the <i><span class="py">m&#468;y&#468;
         y&ugrave;nd&ograve;ng</span></i> &#27597;&#35486;&#36939;&#21205;
         (Mother Tongue Movement) to replace the old
         <i><span class="py">gu&oacute;y&#468;
         y&ugrave;nd&ograve;ng</span></i> &#22283;&#35486;&#36939;&#21205;
         (National Language Movement) that had been in effect for the
         previous forty years:</p>

         <div class="rant">
            <p>I am a native of Taiwan, born and bred. When I was growing
            up, especially when I was seven years old and entered public
            school (during the Japanese occupation, the schools that were
            set up for local children were called "public" schools), I was
            forced to learn Japanese. Before that time, I had only used
            Hoklo and Hakka. This was because my father was of Hakka
            descent and my mother was of Hoklo descent. My relatives were
            also half Hakka and half Hoklo, so I grew up hearing both
            languages. After I went to school and gradually got older, my
            Japanese ability also advanced. By the time I entered middle
            school, while we were in school we used only Japanese. During
            those middle school years, I even thought only in Japanese. Now
            I've abandoned Japanese and switched to Chinese
            (<i>zhongwen</i>, i.e., Mandarin) when I write. After getting a
            bit used to it, I've also started to think in Chinese
            (<i>zhongwen</i>).</p>

            <p>But then a problem came along. Normally when I'm writing, I
            think in Chinese (<i>zhongwen</i>) and write my thoughts down
            in Chinese (<i>zhongwen</i>). This is as it should be, and I
            find nothing objectionable about it. But when I come to dialog,
            then there's a big difference. When a character in one of my
            stories says something, clearly it's one kind [of language],
            but when I write it down it's another kind [of language]. It
            goes without saying that, between these [two kinds of language,
            my writing has} to undergo a process of translation.</p>

            <p>The subjection of literary works to translation is something
            that has long been done throughout the world, so of course it
            can stand, and there is no need to be suspicious of it.
            Nevertheless, a given place has its own special language and
            mode of expression. Sometimes a short oral utterance can bring
            a character's status and personality vividly to life. However,
            after undergoing this process of translation, that special
            flavor is completely lost and that kind of freshness no longer
            exists. Consequently, the character's personality is also
            distorted. (Zhong 1992: 21-22)</p>
         </div>

         <p>It is interesting that I felt exactly the same type of
         frustration and unnaturalness when I was translating Pu Songling's
         <i>Liaozhai zhi yi</i> (<i>Strange Tales from Make-do Studio</i>).
         (Mair and Mair 1989) I had no particular problem or difficulty
         dealing with narration or description, but when it came to dialog,
         it was almost painful to contemplate the fact that Pu's characters
         could not possibly have spoken the LS that he has coming out of
         their mouths in his stories. In my long career as a translator,
         these were the most existentially unnerving challenges I had ever
         coped with. Often when I was agonizing over the inappropriately
         literary register and rarefied diction of Pu's dialogs, I would
         think of the strong local color and speech of William Faulkner,
         the Hispanic tinge and parlance of Ernest Hemingway's novels, and
         the uncanny ability of Tom Wolfe to recreate the speech and
         mannerisms of different social and ethnic groups. I suspect that
         much of such great writers' success in these respects relies on
         the phonological versatility of the alphabet.</p>

         <p>Will there ever come a day when Taiwanese authors can routinely
         write genuine Taiwanese language dialogs, not just formulaic prose
         translated through a Mandarin filter? According to Lin Zongyuan
         (b. 1955), an award-winning songwriter and poet known for his
         Taiwanese translations of Shakespeare and William Butler Yeats,
         the future of Taiwanese literature is closely bound up with
         political democratization.</p>
         <!-- ==================== PAGE 25 ========================== -->

         <div class="rant">
            <p>"I believe that if we ever get bilingual education,
            Taiwanese will become mainstream. If you look at the number of
            people speaking the southern Fukien dialect in Taiwan, mainland
            China, and Southeast Asia, it's quite significant. I think the
            future looks quite good. But right now we are still
            establishing the written language." (Balcom 1992: 73)</p>
         </div>

         <p>More than a decade later, the situation has still not changed
         much. If anything, the Taiwanese are further from achieving
         unanimity on how to write their mother tongue than they were
         twenty years
         ago.<span class="notenumber"><a href="#n16">16</a></span></p>

         <p>Nonetheless, it is now possible to put together a college-level
         textbook of poems, stories, and essays written in Taiwanese with
         introductions and language notes for each selection. (Cheng <i>et
         al.</i> 2000) There seems to be a growing market for works written
         in Taiwanese, and publishers have responded by offering an
         increasing number ofa wide variety of Taiwanese language materials
         (novels, short story collections, anthologies, dictionaries,
         vocabulary lists, primers, and so forth). Many such recent
         publications, however, display a strange phenomenon that does not
         bode well for the future of writing in Taiwanese. The editors of
         such publications, which are intended for broad consumption among
         the population of Taiwan, have adopted the practice of taking
         works which were originally written wholly in romanization and
         converting them into primarily <i>hanzi</i> texts with a slight
         admixture of romanization, or taking works that were originally
         written with a mixture of <i>hanzi</i> and romanization and
         drastically reducing the amount of romanized words in them.
         Superficially this might appear to be a reasonable strategy, since
         it comports with the general affection and familiarity felt by the
         people of Taiwan toward <i>hanzi</i> and recognizes that there are
         still many high frequency Taiwanese morphemes for which no
         suitable <i>hanzi</i> have been discovered. The problem, however,
         is that -- in a disturbingly large number of cases <i>-- hanzi</i>
         are assigned to Taiwanese words in a completely arbitrary and
         often rather amusing fashion, as we have seen above.</p>

         <p>When Carstairs Douglas published his monumental dictionary of
         Amoy vernacular in 1873, there was not a single character in it.
         In 1923, Thomas Barclay published from the Commercial Press in
         Shanghai a <i>Supplement</i> to Douglas's dictionary. Although
         Barclay added characters for many of the entries, he still left
         many entries without any characters assigned to them. For the
         completely new entries added by Barclay, most lacked characters.
         This is in sharp contrast to a bizarre dictionary compiled by the
         Department of Sinitic Topolects in the Institute for Chinese
         Languages and Script of Amoy University and published to great
         fanfare in 1982. All of the entries have characters and MSM
         pronunciations, by which they are ordered under head characters.
         All definitions are given in MSM and, indeed, the MSM elements of
         the dictionary are openly based on the <i>well-known Xiandai Hanyu
         cidian</i> [<i>Dictionary of Modern Sinitic</i> {<i>i.e.,
         Mandarin</i>}]. A sizable portion of the entries in this
         dictionary from Amoy University are not really authentic Southern
         Min terms at all, but are simply Mandarin words with Southern Min
         pronunciations added to them.</p>

         <h2 id="part5">V. Reflections</h2>

         <p>Just as the speakers of the regional vernaculars have been
         perennially sloughing off their mother tongues and acquiring their
         national language, for the last generation and the current one,
         they have started to forget their national language and remember
         the global language. Code switching involving English is very
         widespread in Taiwan, Hongkong, and Singapore, and it is now being
         heard ever more frequently in China. An impressive amount of
         traffic on the internet in these areas is being carried out in
         English, and parents are eager to have their children start
         learning English at younger and younger ages. It should be
         observed, however, that -- just as with the regional vernaculars
         -- when foreign languages are reduced to sinographic forms,
         surpassingly strange transcriptions are liable to be put forth.
         For example, in Taiwan and Hongkong, a very well- established
         sinographic transcription of "cool" (meaning "wonderful, neat,
         great" in youth speak) is <i><span class="py">k&ugrave;</span></i>
         &#37239; (<i>"cruel"</i>), and my Shandong father-in-law writes
         <i><span class="py">g&#466;ut&oacute;u
         m&#257;on&iacute;ng</span></i> &#29399;&#38957;&#35987;&#22144;
         ("dog's-head cat's
         meow").<span class="notenumber"><a href="#n17">17</a></span> The
         commonly accepted writing of "cigar" with characters that mean
         "snow" and "lotus stem" at least sounds plausible when pronounced
         in Shanghainese (where it was first coined), but it is totally
         baffling in both sound and sense when pronounced
         <i><span class="py">xu&#283;ji&#257;</span></i> &#38634;&#33540;
         <span class="notenumber"><a href="#n18">18</a></span> in Mandarin.
         Even more unsettling is the demonstrated ability of the
         sinographic script to incorporate letters of the alphabet as
         characters, e.g., BB for "baby" in Cantonese, BP or BB for
         "beeper" in Shanghainese, and so forth. (Hansel 11994)</p>

         <p>Li Yang, using his <i>Fengkuang Yingwen</i> ("Crazy English")
         methods, claims that he has taught English to 20,000,000 of
         China's citizens. Li says that he has a mission to teach everyone
         in China how to speak English, not because he loves the language
         but because, as he puts it, "Microsoft and Coca-cola own the
         world." All together, there are well over 250,000,000 Chinese
         actively engaged in learning English right now, and that number is
         sure to grow in the coming years. (McArthur 2002: 357) In 1990,
         bilingual education (<i><span class="py">shu&#257;ngy&#468;
         ji&agrave;oy&ugrave;</span></i> &#38617;&#35486;&#25945;&#32946;)
         in Taiwan signified instruction in Mandarin and Taiwanese or
         Hakka. (Lin 1990) In 2003, bilingual education in Taiwan is more
         likely to signify instruction in English and Mandarin or
         Taiwanese. The inclusion of English in the compulsory curricula of
         Taiwan, China, and Singapore (where it is still the most important
         language of education) may seem like a heavy burden, but the
         government of Hongkong has gone several steps further and
         instituted a policy of <i><span class="py">s&#257;n y&#468;
         li&#462;ng w&eacute;n</span></i> &#19977;&#35486;&#20841;&#25991;
         ("three [spoken] languages -- Cantonese, English, Mandarin -- and
         two [written] languages -- Chinese and English"). The government
         of Taiwan will soon make the momentous switch from vertical
         alignment (<i><span class="py">sh&ugrave;p&aacute;i</span></i>
         &#35918;&#25490;) to horizontal alignment
         (<i><span class="py">h&eacute;ngp&aacute;i</span></i>
         &#27243;&#25490;) of texts for ease of incorporating English in
         official documents. In the words of Premier Yu Shyi-kun, "As it's
         our goal is to make English the quasi official language within the
         next six to 10 years, the use of English is bound to increase in
         official papers and private publications " (Ko 2003) To paraphrase
         the title of a noteworthy book that came out earlier this year, "
         <a href="http://www.pinyin.info/readings/writing_on_the_wall.html">
         the writing is on the wall</a>." (Hannas 2003)</p>

         <p>Despite the deep inroads of Mandarin and the steady incursions
         of English, the people of Taiwan cling tenaciously to their native
         language. They clearly sense that forgetting one's mother tongue
         is the severest form of cultural amnesia. To show the depth of
         feeling that many Taiwanese have for their native language, I
         would like to close with a poem by Chen Lei (b. 1939):</p>

         <div class="poem">
            <pre>
Love of the Mother Tongue

You can beat my skin,
You can eat my flesh,
But you cannot take away
The right to my language.

The language my parents taught me
Is the feeling in my heart;
The voice my native soil gave me
Is the thought in my head.

Why? Why do you say I cannot express myself!
It's because you have your megalomanic ideology of a culture
     lasting five thousand years,
And half a century of despotic control on this island;
Thus you have become a chicken-livered, bird-brained person
With a baleful face and barbarous hand.

Just take a look: come to America
And speak Shanghainese, Cantonese,
Speak Shandongese, Pekingese;
Would anyone dare to beat you for it?
Just take a look: Why is it
Only you behave this way?

You can beat my skin,
You can eat my flesh,
But you cannot take away
The right to my language;
My voice is my feeling,
Language is my thought.

I refuse to be wrapped up in your "national language,"
And become a Taiwanese
Who has lost his own soul.
</pre>
         </div>

         <p>Chen Lei desperately wishes to assert his cultural identity,
         and he realizes that his native language is at the heart of that
         identity. At the same time, he decisively rejects the hegemonistic
         national language imposed upon him and his countrymen by an alien
         political force. Nonetheless, the irony of Chen's predicament
         escapes him: every Taiwanese syllable that he writes with a
         Chinese character merely reinforces the control of the central
         culture. If Taiwanese language is the heart and soul of Taiwanese
         identity, as Chen asserts, Chinese characters are the heart and
         soul of Chinese identity. To write in Chinese characters is to
         evoke the deep cultural memories that he trenchantly opposes and
         to consign to ultimate oblivion the sounds and words with which he
         articulates his innermost thoughts and feelings.</p>

         <h2 id="partnotes">Notes</h2>

         <p>I wish to express my gratitude to the following individuals:
         Grace Wu for lending me books relating to Taiwanese language and
         helping me to read Taiwanese texts; Linda Chance for checking
         Japanese terms; Mark Swofford for keeping me abreast of current
         events concerning the evolution of writing in Taiwan; Anthony C.
         Yu for telling me of his personal experiences with language in
         Hong Kong and Shanghai; Wu Shouli for introducing me to the
         intricacies of old Minnan writing; and Robert Cheng (Zheng
         Liangwei) for energetically endeavoring to comprehend the reasons
         for the non-development of Taiwanese writing, for striving against
         all odds to create a workable means that would enable the people
         of Taiwan to write their mother tongue, and for patiently
         explaining it all to me.</p>

         <ol>
            <li id="n1">Taiwanese (the English equivalent of Daiwanway [MSM
            Taiwanhua]) is a dialect of the Southern Min big topolect or
            branch of Sinitic (Hanyu). A century and more ago, it was still
            very close to the language of Amoy, but has now diverged
            sharply under the following influences: Malayo-Polynesian
            substrate borrowings, Dutch usages, Japanese elements,
            Mandarinisms, English loans, and so forth. Thus, by "Taiwanese"
            I mean the highly evolved form of Southern Min spoken on the
            island of Taiwan that can no longer be equated with any
            language spoken on the mainland of China.</li>

            <li id="n2">During the thirties of the last century, there was
            a movement to use Chinese characters strictly phonetically and
            without any regard for their meanings to transcribe the sounds
            of Taiwanese in what was known as <i>kua-a ts'eh</i> or
            <i>koa-&aacute;-chheh</i> &#27468;&#20180;&#20876; (Xu 1993: 4;
            Wang 1993: 45) From the name of the genre, it is apparent that
            this method was used primarily to write down playscripts. N.B.:
            1. Since there is no standard form of romanization for
            Taiwanese, I follow the various sources that I cite. 2. I refer
            variously to "Chinese characters," <i>H&agrave;nzi</i>
            &#28450;&#23376;, and "sinographs," all of which are synonymous
            for the purposes of this paper and have only slight differences
            of nuance.</li>

            <li id="n3">The figures are comparable for Hakka, which is
            supposedly much closer to Mandarin than is Taiwanese. In a
            glossary of 900 or so frequently used words (Luo 1985:
            306-325), 20% have no known sinographic form for one or more of
            their syllables. For the very common trisyllabic words
            <i>iatappe</i> ("here"), <i>ketappe</i> ("there"), and
            <i>naitappe</i> ("where"), for example, none of the syllables
            are written with Chinese characters. Because of greater
            substrate influences from Austroasiatic languages (Zhuang, Tai,
            etc.) and massive assimilation of English words and
            expressions, the percentage of sinographless morphemes would
            certainly be much higher in Hongkong Cantonese, were it not for
            a much more proactive invention of special graphs, which now
            amount to well over a thousand. (Cheung and Bauer 2002)</li>

            <li id="n4">A few scholars (e.g., Xu 1998: 36) mistakenly
            contend that the name Taiwan derives from the Siraya word
            <i>taian</i> or <i>tayan</i> ("outsider"). Although this
            ignores the evidence from Dutch and Portuguese records, it does
            not negate the fact that the origin of the name Taiwan has
            nothing whatsoever to do with terraces or bays.</li>

            <li id="n5">The difficulties with Japanese place names,
            however, are comparable to the problems one encounters with
            sinographic Taiwanese writing in all subject areas. For
            example, there is a town in Hokkaido named Otaru which is
            written with characters that mean "little barrel." In
            actuality, the name is derived from an Ainu word referring to
            the sandy riverbed nearby.</li>

            <li id="n6">These are, of course, only very rough estimates,
            and there is probably no way to arrive at a precise
            quantification of the various categories of <i>hanzi</i> usage
            in written Taiwanese because of the wide range of more or less
            Mandarinized styles versus more or less colloquial styles,
            subjective attitudes concerning correctness of fit between
            <i>hanzi</i> and Taiwanese morphemes, and so forth. One thing,
            however, is certain: an alarmingly high proportion of Taiwanese
            morphemes either have no established sinographs assigned to
            them or have sharply contested sinographic assignments.</li>

            <li id="n7">I am always astonished when I hear people from
            Shanghai (China's greatest city) tell me that their language is
            "crude" (<i>c&#363;</i> &#31895;) and that they are embarrassed
            by it. It should be noted that service personnel (of department
            stores) in Shanghai are required by the government to greet
            customers in Mandarin, not Shanghainese.</li>

            <li id="n8">In pre-Han times, the phonophore stipulated by Xu
            Shen was associated with graphs that might have been pronounced
            roughly as <i>**b/m/pl</i>[<i>j]wan</i>. (Karlgren 1957: 66-67
            [no. 178a-q]) The slashes indicate that the labials would
            alternate in different cognates, and the "j" in brackets
            indicates a "y" sound that occurred in certain derivates but
            not all.</li>

            <li id="n9">See below for a discussion of "big topolects" and
            branches in linguistic classification.</li>

            <li id="n10">The influence of northern non-Sinitic languages on
            Mandarin has been broached by historical linguists such as
            Mantaro Hashimoto, David Prager Branner, Charles N. Li, Jerry
            Norman, and Jiang Lansheng.</li>

            <li id="n11">I have not been able to determine the exact date
            when <i>fangyan</i> was first used as a neologism for
            "dialect," but it seems to crop up around 1927 or 1928. My
            suspicion is that this is one more in a long line of
            "round-trip words" that began with one meaning in China (in
            this case Yang Xiong's notion of "topolect[al synonymous
            expression]"), acquired a new, Western-inspired meaning in
            Japan (in this case "dialect"), and was then sent back to
            China. (Mair 1994c) After all, the Japanese (aside from Yang
            Xiong's <i>fangyan</i> [Jap. <i>hogen</i>]) had their own
            traditional usages that were somewhat akin to "dialect," viz.,
            <i>ben</i> &#36783; ("dialect, brogue, accent" -- as in
            <i>Osaka-ben</i>) and <i>namari</i> &#35355; ("an accent,
            dialect, patois"). Indeed, China itself had similar indigenous
            terms, e.g., <span class="py">xi&#257;ngt&aacute;n
            &#37129;&#35527;</span>, <span class="py">t&#468;y&#468;
            &#22303;&#35486;</span>, and so forth. It was only with the
            importation of Western linguistic science that <i>fangyan</i> /
            <i>hogen</i> ("topolect") acquired the additional, technical
            meaning of "dialect" as part of the FGBLDS system. (There is no
            space here to discuss the experimental rendering of "idiom" and
            other folkloristic English words with <i>fangyan</i> during the
            teens and twenties of the last century, apparently slightly
            before Yang Xiong's term became unhappily wed to
            "dialect.")</li>

            <li id="n12">Etymology is the study of the derivation of words
            with particular attention to their roots. This is a different
            matter than the study of the structure, origin, and evolution
            of characters. In the China field, the latter types of
            investigation are often incorrectly referred to as
            "etymology."</li>

            <li id="n13">My own informal tests of students, colleagues, and
            friends indicate that Mandarin speakers can make sense of only
            about 50% of Taiwanese texts written in sinographs and can make
            next to no sense of Taiwanese texts written in romanization.
            Mandarin speakers can make even less sense (about 30-40%) of
            pure Cantonese texts written in sinographs and almost no sense
            at all of Cantonese texts written in romanization. (The
            percentages vary with the degree to which the texts in question
            have more or less Mandarin and Literary elements in them.) The
            mutual readability of the written Sinitic vernaculars is every
            bit as limited as that which exists among the languages of
            Europe or the northern half of the Indian subcontinent.</li>

            <li id="n14">I have added the character for "rain" (MSM
            <i>yu3</i>, Taiwanese <i>ho</i>) because it is used by some
            writers to represent the homophonous Taiwanese word
            ("give").</li>

            <li id="n15">A photocopied facsimile of this article from
            <i>Qingnian bao</i> [<i>Youth Daily</i>] has been reprinted in
            <i>Taiyu wenzhai</i> (<i>Tai5-gi2 bun5-tiah4; Taiwanese
            Digest</i>), 12 (July 15, 1990), 168, so the original must have
            appeared during the first half of 1990 or slightly before, On
            the same page of <i>Taiyu wenzhai</i> is also reprinted an
            article from <i>Zhanghua bao</i> [<i>China Daily</i>] with the
            headline "The "Chinese People (<i>Zhangguaren</i>) Should
            Consider It Glorious to Speak the National Language
            (<i>guoyu</i>)," Although less inflammatory than the diatribe
            from <i>Qingnian bao</i>, this article insists that it is wrong
            for legislators in Taiwan to speak Taiwanese,</li>

            <li id="n16">In <i>Taiyu wenxue yu Taiyu wenzi</i>
            [<i>Taiwanese Literature and Taiwanese Script</i>], Hong Weiren
            (1992) addresses the perplexing issue of the fervor surrounding
            the Taiyu wenxue yundong [Taiwanese Literature Movement] and
            the failure to create a workable script with which to write
            it.</li>

            <li id="n17">I have collected several English phrasebooks
            written entirely in Chinese characters that date from the late
            Qing period to the late twentieth century. Almost every entry
            in them is hilarious if one takes into account the meanings of
            the characters used to write down the English.</li>

            <li id="n18">The second graph should not be read in MSM with
            its alternative pronunciation <i>qie2</i>, which means
            "eggplant, aubergine," for that would yield the wrong
            equivalent (<i><span class="py">d&#656;ia23-22</span></i>) in
            Shanghainese, where the graph, in any event, is usually
            pronounced <i>ga2-3</i>.</li>
         </ol>

         <div id="bibliography">
            <h2>Bibliography</h2>

            <p><i>The American Heritage Dictionary of the English
            Language</i>. 4th ed. 2000.</p>

            <p>Balcom, John. 1992. "Homegrown Poetry." <i>Free China
            Review</i>, 42.9 (September): 66-73.</p>

            <p>Branner, David Prager. 2002. "'Red Cliffs' in Taiwanese
            H&agrave;nb&ucirc;n." <i>Chinoperl Papers</i>, 24: 67-100.</p>

            <p>Brewer, Douglas, Terence Clark, and Adrian Phillips. 2001.
            <i>Dogs in Antiquity. Anubis to Cerberus: The Origins of the
            Domestic Dog</i>. Warminster: Aris &amp; Phillips.</p>

            <p>Brunnert, H. S. and V. V. Hagelstrom. 1911(?). <i>Present
            Day Political Organization of China</i>. Rev. by N. Th.
            Kolessoff, tr. from the Russian by A. Beltchenko and E. E.
            Moran. Peking: Kelly and Walsh (?); rpt. Taipei: Ch'eng Wen,
            1971.</p>

            <p>Chappell. Hilary. 2001.
            <i>Sinitic Grammar Synchronic and Diachronic Perspectives</i>.
            Oxford: Oxford University Press.</p>

            <p>Chen Guoqiang. 1990. "'Taiwan' mingcheng de youlai [The
            Origins of the Name 'Taiwan']." <i>Wen shi zhishi</i>
            [<i>Knowledge about Culture and History</i>], 4 (cumulative
            <i>l06</i>): 88-91.</p>

            <p>Chen Lei. See Huang Jinlian.</p>

            <p>Chen, Ping. 1999. <i>Modern Chinese: History and
            Sociolinguistics</i>. Cambridge. Cambridge University
            Press.</p>

            <p>Chen Zhangtai and Li Rulong. 1991. <i>Minyu yanjiu</i>
            [<i>Studies on Min LanguagesJ</i>. Peking: Yuwen Chubanshe.</p>

            <p>Cheng, Robert L. (Zheng Liangwei) 1978. "Taiwanese Morphemes
            in Search of Chinese Characters" <i>Journal of Chinese
            Linguistics</i>, 6.2 (June): 306-314.</p>

            <p>------, Chin-Chin Tseng, Ing Li, and Guangcheng Lu, eds.
            2000. <i>A Taiwanese Reader: A Collection of Writings for
            College Students</i>. Taipei: Yuanliu.</p>

            <p>Cheng, Susie S. 1981. <i>A Study of Taiwanese
            Adjectives</i>. Monographs on Modern Linguistics. Taipei:
            Taiwan xuesheng shuju.</p>

            <p>Cheung Kwan-hin and Robert S. Bauer. 2002. <i>The
            Representation of Cantonese with Chinese Characters</i>.
            Journal of Chinese Linguistics Monograph Series 18.</p>

            <p>Chiung, Wi-vun Taiffalo. 2003 (? -- visited August 8, 2003).
            "<a href="http://www.de-han.org/pehoeji/lomaji/index.htm">Romanization
            and Language Planning in Taiwan</a>." On the Web at:
            http://home.kimo.com.tw/de-han/pehoeji/lomaji/Shtm</p>

            <p>------(?). 1999 (visited August 8, 2003). "Writing
            Taiwanese: Scripts vs. Sound Notations." On the web at:
            http://daiwanway.dynip.com/tw/writing.shtml</p>

            <p>Douglas, Carstairs. 1873. <i>Chinese-English Dictionary of
            the Vernacular or Spoken Language of Amoy, with the Principal
            Variations of the Chang-chew and Chin-chew Dialects</i>.
            London: Tr&uuml;bner; new ed. with corrections by the author,
            1899.</p>

            <p>Hannas, William C. 2003.
            <i><a href="/readings/writing_on_the_wall.html">The Writing on
            the Wall: How Asian Orthography Curbs Creativity</a></i>.
            Philadelphia: University of Pennsylvania Press.</p>

            <p>Hansell, Mark. 1994. "
            <a href="http://sino-platonic.org/complete/spp045_chinese_alphabet.pdf">
            The Sino-Alphabet: The Assimilation of Roman Letters into the
            Chinese Writing System</a>."
            <i><a href="http://www.sino-platonic.org/">Sino-Platonic
            Papers</a></i>, 45 (May): 1-28.</p>

            <p>Hong Weiren. 1992. <i>Taiyu wenxue yu Taiyu wenzi</i>
            [<i>Taiwanese Literature and Taiwanese Script</i>]. Taipei:
            Qianwei chubanshe, 2nd printing, 1995.</p>

            <p>------.1988. "Tan Helaoyu de zhengzi yu yuyuan [A Discussion
            of the Correct Characters for 'Hoklo' Language and Its Origin."
            In Zheng Liangwei and Huang Xuanfan (Robert L. Cheng and
            Shuanfan Huang), ed., <i>Xiandai Taiwanhua yanjiu lunwenji</i>
            (<i>The Structure of Taiwanese: A Modern Synthesis</i>).
            Taipei: Wenhe chuban youxian gongsi.</p>

            <p>------.1986. <i>Taiwan lisu yudian</i> [<i>A Lexicon of
            Taiwanese Etiquette and Customs</i>]. Taipei: Zili wanbao she,
            1987, 2nd printing.</p>

            <p>Huang Jinlian, ed. 1995. <i>Chen Lei Taiyu wenxue xuan</i>
            [<i>A Selection of Literature in Taiwanese by Chen Lei</i>].
            Nanying Taiyu wenxue congshu [South Seas Taiwanese Literature
            Series] 2. Tainan: Tainan xianli wenhua zhongxin.</p>

            <p>Karlgren, Bernhard. 1957. <i>Grammata Serica Recensa</i>.
            Reprinted from the <i>Bulletin of the Museum of Far Eastern
            Antiquities</i> (Stockholm), 29.</p>

            <p>Killingley, Siew-Yue. 1993 <i>Cantonese</i>. Languages of
            the World / Materials 06. M&uuml;nchen - Newcastle. Lincom
            Europa.</p>

            <p>Ko Shu-ling. <i>2003</i>. "
            <a href="http://www.taipeitimes.com/News/taiwan/archives/2003/08/12/2003063299/print">
            Western Writing Style to Debut</a>." <i>Taipei Times</i>
            (Tuesday, August 12): 3. Downloaded from
            http://www.taipeitimes.com/News/taiwan/<br />
            archives/2003/08/12/2003063299/print</p>

            <p>Lai Rensheng / Renchuan. 1991
            <i>Ko<span class="raised">2</span>-ai<span class="raised">3</span>
            e<span class="raised">5</span>
            siu<span class="raised">5</span>-jin<span class="raised">5</span></i>
            (<i>Beloved Enemy</i>) Chinese character transcription by Zheng
            Liangwei Taipei Zili Wanban She Wenhua Chubanbu.</p>

            <p>Lin, Alvin. 1999 "
            <a href="http://www.sino-platonic.org/complete/spp089_taiwanese.pdf">
            Writing Taiwanese: The Development of Modern Written
            Taiwanese</a>,"
            <i><a href="http://www.sino-platonic.org/">Sino-Platonic
            Papers</a></i>, 89 (January): 1-4, 1-41, 1-4</p>

            <p>Lin, Christine Louise, 1999. "
            <a href="http://www.sino-platonic.org/index.html#n92">
            The Presbyterian Church in Taiwan and the Advocacy of Local
            Autonomy</a>."
            <i><a href="http://www.sino-platonic.org/">Sino-Platonic
            Papers</a></i>, 92 (January, 1999), i-xiii, 1-136.</p>

            <p>Lin Jixiong. 1990. <i>Taiyu jiaoxuefa</i> [<i>Pedagogical
            Methods for Taiwanese</i>]. Tainan: Daxia chubanshe.</p>

            <p>Luo Zhaojin. 1985. <i>Keyu yufa</i> [<i>Hakka Grammar</i>].
            Taipei: Taiwan Xuesheng Shuju, 1988, 2nd printing.</p>

            <p>Mair, Victor H. 1998 "
            <a href="http://www.sino-platonic.org/index.html#n87">
            Canine Conundrums: Dog Ancestor Myths of Origin in Ethnic
            Perspective</a>"
            <i><a href="http://www.sino-platonic.org/">Sino-Platonic
            Papers</a></i>, 87 (November): 1-74.</p>

            <p>------.1994a. "Buddhism and the Rise of the Written
            Vernacular in East Asia: The Making of National Languages."
            <i>Journal of Asian Studies</i>, 53.3 (August): 707-751.</p>

            <p>------, ed. 1994b, <i>The Columbia Anthology of Traditional
            Chinese Literature</i>. New York: Columbia University
            Press.</p>

            <p>------. 1994c
            "<a href="http://www.sino-platonic.org/complete/spp034_sinolinguistics_fanqie.pdf">East Asian
            Round-Trip Words</a>."
            <i><a href="http://www.sino-platonic.org/">Sino-Platonic
            Papers</a></i>, 34 (October), 5.</p>

            <p>------.1991. "
            <a href="http://www.sino-platonic.org/complete/spp029_chinese_dialect.html">
            What Is a Chinese 'Dialect/Topolect'? Reflections on Some Key
            Sino-English Linguistic Terms</a>."
            <i><a href="http://www.sino-platonic.org/">Sino-Platonic
            Papers</a></i>, 29 (September): 1-31.</p>

            <p>Mair, Denis C. and Victor H. Mair. 1989. <i>Strange Tales
            from Make-do Studio</i> by Pu Songling. Beijing: Foreign
            Languages Press.</p>

            <p>Matthews, Stephen and Virginia Yip. 1994. <i>Cantonese: a
            comprehensive grammar</i>. London: Routledge.</p>

            <p>McArthur, Tom. 2002. <i>The Oxford Guide to World
            English</i>. Oxford: Oxford University Press</p>

            <p>Moser, Leo J. 1985. <i>The Chinese Mosaic: The Peoples and
            Provinces of China</i>. Boulder and London: Westview Press.</p>

            <p>Parker, E[dward] H[arper]. 1879. "Characterless Chinese
            Words." <i>China Review</i>, 9: 85-88.</p>

            <p>Snow, Donald B[ruce]. 1991. "Written Cantonese and the
            Culture of Hong Kong: The Growth of a Dialect Literature."
            Indiana University Ph.D. dissertation.</p>

            <p>Waldron, Arthur. 1990. <i>The Great Wall of China: From
            History to Myth</i>. Cambridge: Cambridge University Press.</p>

            <p>Wang Huanan, ed. 1989. <i>Guyi angran hua Taiyu</i> [<i>On
            Taiwanese, Full of Ancient Qualities</i>]. Taiyu hanzi xilie
            [Taiwanese Sinographic Series], 1. Taipei: Diteng chuban tushu
            youxian gongsi.</p>

            <p>Wang Yude. 1993. <i>Taiwanhua jiangzuo</i> [<i>Lectures on
            Taiwanese</i>]. Tr. from the Japanese by Huang Guoyan. Taiwan
            bentu xilie [Taiwan Local Culture Series] 3, Number 14. Taipei:
            Zili wanbao she wenhua chuban bu.</p>

            <p>White, David Gordon. 1991. <i>Myths of the Dog-Man</i>.
            Chicago and London: The University of Chicago Press.</p>

            <p>Xu Jidun (Kno Kek-tun). 1998. <i>Taiwanyu gailun</i> [<i>A
            Survey of Taiwanese Language</i>]. Taipei: Qianwei
            Chubanshe.</p>

            <p><i>Shuo wen jie zi</i> [<i>Explanations of Simple and
            Compound Graphs</i> <i>J</i>.</p>

            <p>Wu Shouli and Lin Zongyi, ed. 1976. <i>Ming-Qing Minnan xiqu
            si zhong</i> [<i>Four Minnan Plays of the Ming-Qing
            Period</i>]. Tokyo: Dingjing Tang.</p>

            <p>Xiamen Daxue Zhongguo Yuyan Wenxue Yanjiusuo Hanyu Fangyan
            Yanjiushi [Department of Sinitic Topolects in the Institute for
            Chinese Languages and Script of Amoy University], comp. 1982.
            <i>Putonghua Minnan fangyan cidian</i> [<i>Mandarin-Southern
            Min Topolect Dictionary</i>]. Hongkong: Sanlian shudian.</p>

            <p>------ (Khor Kek-tuun). 1993. "Yong hanzi xie Taiyu de lilun
            yu shiji [The Theory and Practice of Using Sinographs to Write
            Taiwanese] (Original English title: Theory and Practice in the
            Writing Language of Taiwanese)." <i>Taiwan xueshu yanjiuhui
            zhi</i> (<i>Journal of Academic Research for Taiwan</i>). Li
            Zongfan boshi shishi jinian ji (In memory of Dr. Li
            Cong-phua<span class="raised">n</span>). Tokyo: Taiwan Xueshu
            Yanjiuhui (The Association of Academic Research for Taiwan).
            pp. 3-51.</p>

            <p>Xu Shen. 100 <span class="smcaps">CE</span>; 1966. Hongkong:
            Taiping shuju.</p>

            <p>Yue Hashimoto, Oi-kan. 1972. <i>Phonology of Cantonese</i>.
            Studies in Yue dialects 1. Cambridge: At the University
            Press.</p>

            <p>Zheng Liangwei (Robert L. Cheng). 1990. "Jiandan shiyong de
            jiaoyu wenzi [An Easy-to-Use Script for Education]." <i>Taiyu
            wenzhai</i> (<i>Taiwanese Digest</i>), 12 (July 15), 66-96.</p>

            <p>------. 1989. <i>Zou xiang biaozhunhua de Taiwanhuawen</i>
            [<i>On the Road to Standardized Written Taiwanese</i>]
            (Original English title: <i>Essays on Written Taiwanese</i>).
            Taipei: Zili wanbao she wenhua chuban bu.</p>

            <p>------ and Zheng Xie Shujuan. 1977. <i>Taiwan Fujianhua de
            yuyin jiegou ji biaoyin fa</i> (<i>Phonological Structure and
            Romanization of Taiwanese Hokkian</i>). Xiandai Yuyanxue
            Luncong (Monographs on Modern Linguistics). Taipei: Taiwan
            xuesheng shuju.</p>

            <p><i>Zhongguo da baikequanshu</i> [<i>The Encyclopedia of
            China</i>]<i>. Yuyan wenzi</i> [<i>Language and Script</i>]
            (1988). <i>Minzu</i> [<i>Nationalities</i>](1986).</p>

            <p>Zhong Zhaozheng. 1992. "Xiezuo ji fanyi? Muyu yundong xingqi
            hou de yi ge xingsi [Is Composition Tantamount to Translation?
            Reflections after the Rise of the Mother Tongue Movement]."
            <i>Tai<span class="raised">5</span>gi<span class="raised">2</span>
            bun<span class="raised">5</span>tiah<span class="raised">4</span></i>
            (<i>Taiwanese Digest</i>), new series 1 (cumulative 25), 5.1
            (January 15): 21-22.</p>
         </div>
      </div><!-- +++++  END OF MAIN DIV  +++++ -->
            <div id="footer">
         <p>Copyright &copy; 2010 <a href="/">&#25340;&#38899;/Pinyin.info</a> |
         <a href="http://validator.w3.org/check?uri=referer">XHTML</a> |
         <a href="http://jigsaw.w3.org/css-validator/validator?uri=http://www.pinyin.info/main.css">
         CSS</a></p>
      </div>
      <!-- +++++  END OF FOOTER DIV  +++++ -->
            <div id="navbar">
         <a href="/index.html"
              style="border-top: none;">home</a> 
              <a href="/readings/index.html"
              title="extensive selections from books related to romanization">readings</a>	
              <a href="/rules/index.html"
              title="how to use pinyin correctly">rules</a> 
              <a href="/tools/index.html"
              title="online tools to help you use pinyin and Chinese characters, esp. on Web sites">tools</a>											
              <a href="/romanization/index.html"
              title="romanization systems for Chinese">systems</a>
              <a href="/news/"
              title="romanization-related news stories">Pinyin News blog</a>
              <a href="/links.html"
              title="some other sites of interest">links</a> 
              <a href="/contact.html"
              title="e-mail the webmaster">contact</a> 
              <img src="/py.gif"
              alt="&#25340;&#38899;" />
      </div><!-- ***** END OF NAVBAR DIV ********  -->      </div><!-- ***** END OF CONTAINER DIV ********  -->
   </body>
</html>
