https://angeci.github.io/blog/tags/cantonese-%E7%B2%B5%E8%AA%9E/Recent content in Cantonese 粵語 on AngeCI’s BlogHugo -- gohugo.ioenFri, 24 Apr 2026 11:38:10 +0000https://angeci.github.io/blog/introduction-to-cantonese-phonology/Fri, 24 Apr 2026 11:38:10 +0000https://angeci.github.io/blog/introduction-to-cantonese-phonology/<p>The romanization system used in this article is <a class="link" href="https://en.wikipedia.org/wiki/Jyutping" target="_blank" rel="noopener" >Jyutping</a>.</p> <h1 id="initials"><a href="#initials" class="header-anchor"></a>Initials </h1><p>Standard Cantonese has 20 phonemic initial consonants. Although many people do mix up several pairs like /n/ and /l/, /ŋ/ and /ʔ/ ~ ∅.</p> <table> <thead> <tr> <th></th> <th>Labial</th> <th>Dental/Alveolar</th> <th>Palatal</th> <th>Velar</th> <th>Labiovelar</th> <th>Glottal</th> </tr> </thead> <tbody> <tr> <td>Unaspirated Plosive</td> <td>b /p/</td> <td>d /t/</td> <td></td> <td>g /k/</td> <td>gw /kʷ/</td> <td>∅ /ʔ/ ~ ∅</td> </tr> <tr> <td>Aspirated Plosive</td> <td>p /pʰ/</td> <td>t /tʰ/</td> <td></td> <td>k /kʰ/</td> <td>kw /kʰʷ/</td> <td></td> </tr> <tr> <td>Nasal</td> <td>m /m/</td> <td>n /n/</td> <td></td> <td>ng /ŋ/</td> <td></td> <td></td> </tr> <tr> <td>Unaspirated Affricate</td> <td></td> <td>z /ts/ ~ /tʃ/</td> <td></td> <td></td> <td></td> <td></td> </tr> <tr> <td>Aspirated Affricate</td> <td></td> <td>c /tsʰ/ ~ /tʃʰ/</td> <td></td> <td></td> <td></td> <td></td> </tr> <tr> <td>Fricative</td> <td>f /f/</td> <td>s /s/ ~ /ʃ/</td> <td></td> <td></td> <td></td> <td>h /h/</td> </tr> <tr> <td>Lateral</td> <td></td> <td>l /l/</td> <td>j /j/</td> <td></td> <td>w /w/</td> <td></td> </tr> </tbody> </table> <p>Alveolar and postalveolar affricates seems to be either allophones or free variation. Some people pronounce /ts/, /tsʰ/, /s/ before unrounded vowels, and /tʃ/, /tʃʰ/, /ʃ/ before rounded vowels, while some people only pronounce the alveolar variant in all positions. Linguistic materials suggest that these two sets of initial consonants are actually distinguished in pre-1950s Cantonese, however. See my other post “<a class="link" href="https://angeci.github.io/blog/chinese-dialectal-phenomena/#%e7%b2%b5%e8%aa%9e%e7%9a%84%e5%b9%b3%e7%bf%b9%e9%9f%b3" >A Brief Discussion on Several Chinese Dialectal Phenomena</a>” for further information.</p> <h1 id="rhymes"><a href="#rhymes" class="header-anchor"></a>Rhymes </h1><p>There are usually considered 8 main vowel phonemes in standard Cantonese, although there are about 11 realizations of these phonemes. Items marked with an asterisk * are rare rhymes.</p> <table> <thead> <tr> <th></th> <th>aː</th> <th>ɐ</th> <th>ɛː</th> <th>iː</th> <th>ɔː</th> <th>œː</th> <th>uː</th> <th>yː</th> <th>∅</th> </tr> </thead> <tbody> <tr> <td>-∅</td> <td>aa /aː/</td> <td>a* /ɐ/</td> <td>e /ɛː/</td> <td>i /iː/</td> <td>o /ɔː/</td> <td>oe /œː/</td> <td>u /uː/</td> <td>yu /yː/</td> <td></td> </tr> <tr> <td>-i</td> <td>aai /aːi/</td> <td>ai /ɐi/</td> <td>ei /ei/</td> <td></td> <td>oi /ɔːi/</td> <td>eoi /ɵy/</td> <td>ui /uːi/</td> <td></td> <td></td> </tr> <tr> <td>-u</td> <td>aau /aːu/</td> <td>au /ɐu/</td> <td>eu /ɛːu/</td> <td>iu /iːu/</td> <td>ou /ou/</td> <td></td> <td></td> <td></td> <td></td> </tr> <tr> <td>-m</td> <td>aam /aːm/</td> <td>am /ɐm/</td> <td>em /ɛːm/</td> <td>im /iːm/</td> <td>om* /om/</td> <td></td> <td></td> <td></td> <td>m /m̩/</td> </tr> <tr> <td>-p</td> <td>aap /aːp/</td> <td>ap /ɐp/</td> <td>ep /ɛːp/</td> <td>ip /iːp/</td> <td>op* /op/</td> <td></td> <td></td> <td></td> <td></td> </tr> <tr> <td>-n</td> <td>aan /aːn/</td> <td>an /ɐn/</td> <td>en* /ɛːn/</td> <td>in /iːn/</td> <td>on /ɔːn/</td> <td>eon /ɵn/</td> <td>un /uːn/</td> <td>yun /yːn/</td> <td></td> </tr> <tr> <td>-t</td> <td>aat /aːt/</td> <td>at /ɐt/</td> <td>et* /ɛːt/</td> <td>it /iːt/</td> <td>ot /ɔːt/</td> <td>eot /ɵt/</td> <td>ut /uːt/</td> <td>yut /yːt/</td> <td></td> </tr> <tr> <td>-ŋ</td> <td>aang /aːŋ/</td> <td>ang /ɐŋ/</td> <td>eng /ɛːŋ/</td> <td>ing /ɪŋ/</td> <td>ong /ɔːŋ/</td> <td>oeng /œːŋ/</td> <td>ung /ʊŋ/</td> <td></td> <td>ng /ŋ̩/</td> </tr> <tr> <td>-k</td> <td>aak /aːk/</td> <td>ak /ɐk/</td> <td>ek /ɛːk/</td> <td>ik /ɪk/</td> <td>ok /ɔːk/</td> <td>oek /œːk/</td> <td>uk /ʊk/</td> <td></td> <td></td> </tr> </tbody> </table> <p>Older Cantonese also has /om/ and /op/, though they are usually merged into /ɐm/ and /ɐp/ today.</p> <p>/aː/, /ɐ/, /ɛː/ and /iː/ are considered as unrounded vowels, while /ɔː/, /œː/, /uː/ and /yː/ are considered as rounded vowels. The roundness of the vowel may determine whether the /ts/, /tsʰ/, /s/ or /tʃ/, /tʃʰ/, /ʃ/ set of initial is used.</p> <p>All stop codas (i.e. -p, -t and -k) are unreleased, unlike most European languages.</p> <h1 id="tones"><a href="#tones" class="header-anchor"></a>Tones </h1><table> <thead> <tr> <th>Tone number</th> <th>Traditional name</th> <th>Description</th> <th>Value</th> <th>IPA</th> </tr> </thead> <tbody> <tr> <td>1/7</td> <td>陰平／短陰入</td> <td>high level/high falling</td> <td>55/53</td> <td>˥˥/˥˧</td> </tr> <tr> <td>2</td> <td>陰上</td> <td>medium rising</td> <td>35</td> <td>˧˥</td> </tr> <tr> <td>3/8</td> <td>陰去／長陰入</td> <td>medium level</td> <td>33</td> <td>˧˧</td> </tr> <tr> <td>4</td> <td>陽平</td> <td>low falling/very low level</td> <td>11/21</td> <td>˩˩/˨˩</td> </tr> <tr> <td>5</td> <td>陽上</td> <td>low rising</td> <td>23</td> <td>˨˧</td> </tr> <tr> <td>6/9</td> <td>陽去／陽入</td> <td>low level</td> <td>22</td> <td>˨˨</td> </tr> </tbody> </table> <p>短陰入 and 長陰入 are also referred as 高陰入 and 低陰入 respectively.</p> <blockquote class="alert alert-tip"> <div class="alert-header"> <span class="alert-icon">💡</span> <span class="alert-title">Fun fact</span> </div> <div class="alert-body"> <p>Cantonese is one of the handful Chinese variant that preserves the 陽上 tone. Most modern Chinese variant just merge it in 陽去 or 陰上.</p> </div> </blockquote>