In the IPA.It has a diacritic, voiced.This mark always be used in AmE,japanese... such as water /ˈwɔːt̬ ər/,party /ˈpɑːrt̬i/,私 /ɰat̬aɕi/. So,if a voiceless consonant plus this voiced mark, how to distinguish them? For example t̬ and d.

  • Any example(s) of such words? – CinCout Oct 9 '17 at 6:14
  • 1
    Party AmE /'pɑːrt̬i/ BrE /ˈpɑːti/,I just wanna know,t or another voiceless consonant plus this voiced diacritic, how to read them,and what's the difference between a real voice consonant? – Berkeley Jones Oct 9 '17 at 9:31
  • I think you're getting confused between phonemic and phonetic notation (and perhaps concepts). The English phoneme /t/ has various realisations: usually [tʰ] at the start of a syllable, [t] elsewhere, but between vowels, it is often [d] in some dialects (especially American ones), and [ʔ] (glottal stop) in others. – Colin Fine Oct 10 '17 at 16:45

There is no general rule about how to phonetically differentiate a voiceless consonant symbol (like /t/) with a voicing diacritic from the equivalent voiced consonant symbol (like /d/) in IPA.

The "IPA", despite standing for "International Phonetic Alphabet", is actually not just (or even mainly) used for phonetic transcriptions, but also for phonemic and phonological transcriptions. For this reason, the same phone might be written different ways in different contexts, depending on its analysis.

The use of /t̬/ instead of /d/ in any IPA transcription is probably motivated by phonological, not phonetic criteria: we know that the sound somehow corresponds to /t/ historically, or at some level of the synchronic phonology. E.g. /ɰat̬aɕi/ seems to be a way of representing a pronunciation used in certain Japanese topolects (like Tōhoku?) of a word that is pronounced /ɰataɕi/, with a voiceless /t/, in standard/mainstream Japanese. Phonetically, the pronunciation in question may just be [ɰadaɕi]. But the consonant might be analyzed as being phonemically /t/, with a realization [d] derived by a rule of allophonic voicing.

American English /t̬/ in particular

Many native speakers of American English cannot hear any clear distinction between /t̬/ and /d/. I am one of them: to me, the pronunciation of the word water sounds like the pronunciation of the word wadder. From what I have read about English phonetics, I assume the phone I pronounce them with is something like [ɾ], a voiced alveolar "flap"/"tap", but I can't really hear much difference between this sound and a regular /d/ sound (as in the word "predict"). As far as I know, there is never any contrast between [ɾ] (used as a lenited /t/ or /d/ sound) and [d] in English. (Some British accents use [ɾ] in some contexts for the rhotic consonant /r/: in this context, [ɾ] may contrast with /d/.) So many people will impressionistically write the pronunciation of "water" with a voiced intervocalic consonant as something like "waw-der".

Other speakers have some kind of lenition, but feel like they hear a kind of distinction between lenited /t/ and lenited /d/, at least for some words. It's hard in most cases to describe what exactly the distinction is: it may be most salient in its effect on the preceding vowel sound. See the Language Log post Metal v. medal, by Mark Liberman.

One of the more stable contexts for a perceptual distinction is after the diphthong /aɪ/: what is called "Canadian raising" may apply before [ɾ] from /t/ but not (in most cases) before [ɾ] from /d/. In "Canadian raising" contexts, /aɪ/ is realized as something like [əɪ] or [ɐɪ].

If you are striving to acquire completely native-sounding American English, you will need to learn how to pronounce the English sound [ɾ] (My understanding is that it is similar, but not necessarily identical to the sounds transcribed with the IPA letter /ɾ/ in other languages, such as the Spanish "r" sound in pero).

But if you are not aiming that high and just want an understandable and reasonably natural-sounding accent, it is fine to maintain a distinction and use /t/ and /d/ in your own pronunciation. Liberman says that not all Americans voice the middle consonant in "water" (although I have never personally encountered one who doesn't), and in any case, [t] is not too surprising in this context because it corresponds to the spelling, and is commonly used in some native English accents like formal British pronunciation (less formal British pronunciation may have a glottal stop instead).

  • I'm really thanks for you're willing help, but I used water,even t,just For example.I mean voiceless consonants... – Berkeley Jones Oct 11 '17 at 18:29
  • @BerkeleyJones: Hmm, other voiceless consonants have different patterns of pronunciation. There is no general principle about how to interpret an IPA voiceless letter with a voiced diacritic vs. an IPA voiced letter. – sumelic Oct 11 '17 at 18:32
  • So, exactly what about this voiced diacritic?Just in the /t/,/t/ plus this mark equal /d/? – Berkeley Jones Oct 11 '17 at 18:37
  • @BerkeleyJones: Yes, I would say that you can generally assume that "t̬" is pronounced like [d]. To know more, you have to be familar with the particular language that is being transcribed. IPA is not very phonetically exact in most cases – sumelic Oct 11 '17 at 18:40
  • So, why created this diacritic...And voiceless diacritic... It just a grumble,good night. – Berkeley Jones Oct 11 '17 at 18:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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