I'm a newbie Chinese student and I have a question about the pronunciation of the word shān (山), meaning mountain. When I went to Taiwan recently, friends there pronounced it as sān, not shān. For example, they referred to the National Park Alishan (阿里山) as Ali SAN.

Is this particular to Taiwan, to my friends, or some other reason? 谢谢 for your help!

  • 1
    ‘Xièxiè’ may also come out a bit like ‘se-se’ in Taiwan too, have you noticed? :)
    – neubau
    May 28, 2014 at 16:35

4 Answers 4


Still should be shan1 but probably they have some 翘舌音 problems, i.e. cant pronounce ch, sh, zh etc. A lot of Chinese variants don't have 翹舌音 so these pronunciations become problematic.

Count it as an accent.

  • Thanks for your answer. I'll leave it open for a bit to see if anyone else has a different opinion. BTW: Does 翘舌音 mean something like tone-tongue twister?
    – Maarten
    May 28, 2014 at 15:50
  • 2
    翘舌音:retroflexes (the zh, ch, sh, r of pinyin)
    – Mou某
    May 28, 2014 at 15:54
  • 4
    I would like to add that most Taiwanese people do make a difference between s/sh, z/zh and c/ch, but it's not as prominent and the difference might be lost on people used to a Northern dialect.
    – Olle Linge
    May 29, 2014 at 0:47
  • @ollelinge how exactly do they make that difference?
    – MickG
    Dec 27, 2014 at 13:25

Many Taiwanese do not discriminate the phoneme pairs like z/zh, c/ch, s/sh. This is because their mother language, Taiwanese Minnan, does not compose of those retroflective sounds like zh, ch, sh. This is not a problem if those syllables carrying these phonemes are embedded in a multisyllabic words, like 阿里山。 Both a-li-shan or a-li-san are OK for local people to understand.


San is the Japanese pronunciation for 山.

Taiwan (or "Formosa") was part of Japan for nearly a century before 1945. So the "Taiwanese" pronunciation of 山 may be a relic of the Japanese occupation.

  • 5
    Since most retroflex initials are affected, I really don't think Japanese is relevant.
    – Olle Linge
    May 29, 2014 at 0:46
  • @OlleLinge: An "occupation," particularly if prolonged, might have affected "most retroflex initials." Because a different occupier might well have produced a different result. Look that the way that a number of Norwegian words became more like "Swedish" than "Danish" versions after Norway was transferred from Denmark to Sweden in 1814.
    – Tom Au
    May 29, 2014 at 13:29
  • Although you might be right in that a long occupation affects a language, I simply think this is completely unrelated to what we're talking about here. Are you suggesting that because 山 is pronounced "san" in Japanese, all retroflex sounds are affect in Taiwanese Mandarin, including zh/ch? Since the same phenomenon is prominent across southern China, I find it very hard to believe that this is related to Japanese.
    – Olle Linge
    May 30, 2014 at 2:06
  • Instead, this is the result of influences from other Chinese dialects, mostly 閩南, which (as far as I know) doesn't have these sounds.
    – Olle Linge
    May 30, 2014 at 2:14
  • @ollelinge I totally agree with you and no, Min Nan doesn't have retroflexes. I have heard many Hokkien songs and have never heard the retroflexes, and the alveolar-alveolopalatal distinction is merely, to my ears at least, a consequence of following vowels palatalizing alveolars. The fact that neither Peh-oeh-ji nor TLPA nor this page distinguish the retroflexes seem to support my claim.
    – MickG
    Dec 27, 2014 at 13:34

Rather than Japanese, Minnan (san/suan, see http://twblg.dict.edu.tw/holodict_new/index.html search for 阿里山 A-lí-san) and Hakka (san1) are probably more of an influence.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.