On the Wikipedia page for Hindustani phonology, it lists Hindustani as having ten vowels, three short and seven long. More importantly, it claims that there is a distinction between /o/ and /ɔ/, and between /e/ and /ɛ/. It says that /ɔ/ and /ɛ/ occur when there is a schwa in proximity of /ɦ/ for /ɛ/ and a schwa and a rounded vowel for /ɔ/, while /ɛ/ is also phonemic.

So firstly, would that not make both of them allophones for the schwa and any rounded vowel respectively? What makes them separate vowels?

And secondly, as a native speaker, I do not hear the distinction between the two pairs, even in the specific examples given. For /ɛ/, it does mention that it does not occur in all of the places listed, but even in the places where it says the /ɛ/ is phonemic, I do not distinguish it from /e/. And I feel like I, and other folk I hear speaking daily, pronounce both /e/ and /o/ as true mid vowels, not close mid. Is it just in my dialect or is it something else?

  • As far as I can tell, the Wikipedia page makes no such claim. It lists ten vowel phonemes (marked with /slashes/) and adduces a few extra (allo)phones (marked with [brackets]). I haven’t read the whole page, but the section about the vowels in question here writes only [ɛ] and [ɔ], meaning that these vowels are allophones, not phonemes. It says that optionally /ə/ can be realised as [ɛ] or [ɔ] in the contexts given. It also says that /e/ is varyingly transcribed as [ɛː] or [æː], so it’s not surprising if you don’t distinguish qualitatively between phonemic /ɛː/ and allophonic [ɛ]. Jul 2, 2021 at 15:24
  • @JanusBahsJacquet To quote directly: "Hindustani natively possesses a symmetrical ten-vowel system. The vowels [ə], [ɪ], [ʊ] are always short in length, while the vowels [aː], [iː], [uː], [eː], [oː], [ɛː], [ɔː] are usually considered long, in addition to an eleventh vowel /æː/ which is found in English loanwords. The distinction between short and long vowels is often described as tenseness, with short vowels being lax, and long vowels being tense." it does not specify them as allophones or distinguish them at all, they are all listed between brackets. I may not have understood your comment. Jul 2, 2021 at 15:44
  • Wikipedia: “[ɛ] occurs as a conditioned allophone of /ə/ (schwa) in proximity to /ɦ/, if and only if the /ɦ/ is surrounded on both sides by two underlying, orthographic schwas. This change is part of the prestige dialect of Delhi, but may not occur for every speaker.” @QuintusCaesius-RM — Are you a speaker of the dialect mentioned?
    – Yellow Sky
    Jul 2, 2021 at 16:43
  • No, but it also mentions a phonemic /ɛ/ just before that paragraph which I do not distinguish. Jul 2, 2021 at 17:17
  • 1
    right, only according to Ohala Jul 2, 2021 at 18:06

1 Answer 1


As a native Urdu speaker, /e:/, /ɛ:/, /o:/ and /ɔ:/ are always long and phonemic.

I will put forward some examples of minimal pairs where the contrast is essential for differentiation.

/bɛ:l/ (बैल) (bull/ox) /be:l/ (बेल) (vine)

/o:r/ (ओर) (direction/face) /ɔ:r/ (और) (and/more)

Not pronouncing these vowels correctly is considered non-standard and might be attributed to the influence of other regional languages. Even in words that do not form minimal pairs such as /dɔ:ɽ/ (दौड़) (race), saying /do:ɽ/ instead would be non-standard.

This Wikipedia article correctly lists all the eleven vowels that are phonemic in Hindustani. Nasalization of all these vowels is also phonemic but the vowel length does not matter in that case i.e. /i/ and /i:/, /e/ and /e:/ etc. are allophones when nasalized.

The schwa is indeed allophonic with /ɛ/ in words derived from Arabic and Persian such as Aham (اہم) and Shah(شہ) where the original schwa before an /ɦ/ (originally /h/) generally changes into /ɛ/ but the schwa can also be preserved. However, in Sanskrit-derived words I have never heard a schwa before an /ɦ/ e.g. the word रहना (to stay) is always pronounced /rɛɦna/ and never /rəɦna/, the latter would weird me out.

Your Answer

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

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