Which languages other than some Chinese languages have apical vowels? The "apical vowels" are the i in zi, ci, si (in IPA: z̩ (also seen as ɿ)) and ʐ̩ (also seen as ʅ). They are basically buzzed sounds after certain consonants.


4 Answers 4


The Mandarin apical vowels are examples of syllabic fricatives — that is, fricative sounds that form the nucleus of a syllable, as a vowel generally would. (Sometimes syllabic fricatives are also called "fricative vowels," though that's rarer.)

Here's a discussion from one of Ladefoged's books on "fricative vowels" a.k.a. syllabic fricatives in other parts of the world. He mentions examples from Czech and Bantu. Berber is also often described as having syllabic fricatives. Blackfoot has recently been claimed to have syllabic /s/.

(It's maybe worth mentioning too that you'll hear syllabic fricatives in English, if only as a fast speech phenomenon. So for instance, in my dialect, the word horses is [hɔɹsɨz] if I'm speaking carefully and [hɔɹsz̩], with a syllabic [z̩], in fast speech. If I'm in a hurry, the last syllable of horses sounds an awful lot like Pinyin si.)

  • 2
    Among the Slavic languages, at least Serbocroatian and Slovenian also have an "r" which can act like a vowel, including carrying stress I believe. In fact learning to think of the "r" as a vowel is the key to pronouncing some words which seem to lack vowels in those languages. Commented Sep 27, 2011 at 18:04
  • 1
    Right. In Czech, though, the "r" has a (rather unusual) pronunciation with quite a bit of friction in it. Some other Slavic languages have a syllabic "r," but it doesn't have the same unusual friction-y pronunciation, so it doesn't count as a syllabic fricative. Commented Oct 3, 2011 at 3:55
  • Well Czech has both "r" and "ř", the latter does have an especially unusual pronunciation, so is it that you're talking about or do you really mean "r"? Commented Oct 3, 2011 at 8:37
  • 2
    Yes, it's the sound spelled as R with a hacek that I'm talking about. Commented Oct 3, 2011 at 13:31

According to Wikipedia, there are other Sino-Tibetan languages with apical vowels. But I don't think they're very common at all: out of the major Chinese varieties, only Mandarin has it.

Also in East Asia, the Miyako language has an apical vowel, at least according to this page:


(例)ピㇲトゥ pɿtu(人)  カビㇲ kabɿ(紙)

The name of the language itself contains an apical vowel: [mjaːkufutsz̩]‎.

  • Apparently that symbol "ɿ" means "high back unrounded vowel, with frication from the preceding consonant" and has beenn replaced by ɯ in modern IPA. Can you shed any light on the pronunciation of those Miyako words? Commented Mar 12, 2018 at 12:56

I had the opportunity to talk about this issue with a Chinese linguist who is teaching a class I'm auditing at Fudan University.

Chinese-linguists have traditionally used the term "apical vowel" (and the otherwise unseen characters ɿ and ʅ) to describe the nucleus of the "zhi chi shi ri" and "zi ci si" syllables. The textbook we used was no exception.

As others have pointed out, these are often analyzed (or at least written as) syllabic fricatives by the linguistics community at large. I wasn't sure if my professor knew this, so I brought it up with him after class.

He was actually quite insistent that there is no frication in nuclei of these syllables (and was able to pronounce the syllables in a way consistent with that hypothesis). We eventually agreed that there was a fair amount of variation among Chinese people as to the realization of these phonemes.

So, if and when they're produced without frication, what are they? The retroflex one can be analyzed as a syllabic-retroflex-approximant. The apical one is harder, as both the front and the middle of the tongue are raised quite close to the roof of the mouth. A bit like a [ ɨ ] with the front of the tongue raised slightly.

  • If there is no frication, then it's just vowels; [sɯ] is just a plain consonant + vowel combination
    – iopq
    Commented Dec 24, 2022 at 1:38
  • @iopq are you unfamiliar with approximants? Or are you just making a normative statement about syllabic consonants? Commented Dec 24, 2022 at 15:55
  • If it's [sɰ] it still makes no difference. There's no such thing as an apical vowel, and no such thing will be added to IPA since there's no need for a syllable to necessarily have a vowel
    – iopq
    Commented Dec 28, 2022 at 14:42
  • @iopq I mean, I think this is just a question of notation. In some people's speech / in some syllables, there's frication (which non-sinologists would just write as syllabic fricatives). In other people's speech / for other syllables, there isn't, and non-sinologists would probably write with syllabic approximants. Sinologists often use their own notation for this, I think partly to emphasize the alveoleopalatal nature of the zi/ci/si set. I mean, at heart, the tongue just is where it is. Nothing but atoms and the void; the rest is just a human choice about notation. Commented Dec 29, 2022 at 4:47
  • there's no reason to invent new symbols. For example, /r/ can be an approximant, and it can be a fricative. We don't need to write / ɻ ~ ʐ / as a single symbol, you just pick one like /ʐ/ and everyone knows you mean pinyin r. Similarly, Chinese has a lot of variance of how 鱼 would be pronounced [ɥy] or [ʔy] - you just need to decide on a phonemic representation like /y/ and call it a day. You don't need a special symbol of a sound in between a vowel and an approximant to indicate the variation
    – iopq
    Commented Dec 29, 2022 at 5:22

Languages besides Modern Standard Mandarin that are thought to have apical vowels include Nuosu, Lisu, Ersu, Shixing, Achang, Ahi.

It is argued in this paper that the appropriate analysis of Modern Standard Mandarin "apical vowels" is as syllabic approximants:

  • the vowel in pinyin zhi chi shi ri (traditionally ʅ) is argued to be ɻ̩ (ɻ with syllabic diacritic)
  • the vowel in pinyin zi ci si (traditionally ʅ) is argued to be be ɹ̪̩ (ɹ with dental and syllabic diacritics)

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.