IPA contains diacritics for indicating voiceless (/x̥/) and voiced (/x̬/) sounds. There are also different symbols for many voiced/voiceless pairs, e.g. /d/ and /t/ or /g/ and /k/.

Is there a difference between the sounds /d/ and /t̬/?

  • The chart I'm looking at also suggests "stiff voice" for /x̬/. Perhaps in situations like this, it is assumed to be more fine-grained? I also wouldn't be surprised if there were a lot of potential redundancies in IPA, considering the size of its symbol inventory. Where the phoneticians at, to resolve this crisis?
    – tdhsmith
    Sep 14, 2011 at 7:00

1 Answer 1


I usually only ever see the voiceless diacritic used on segments for which voicelessness is highly marked, like the devoicing that occurs in /Cr/ clusters in English. [tr̥æp]. Usually, a liquid like r is voiced, but in this particular instance, it is remarkably voiceless.

As for whether there's a difference between d and , I think the answer is that IPA transcription is frequently more of an interpretive exercise than a science.

The real question to ask is if you were presented with some audio from a language that contains what sounds like a voiced, alveolar stop, which symbol would you use? I think many times it depends on what else you know about the language, or word you're transcribing. What if you also knew that this language had no voiced stops, and that in this particular piece of audio, it sounds voiced because it's between two vowels? In that case, you might be tempted to use [t̬ ], to reflect your beliefs about the phonemic system. Or, to take the example from Theta30's comment, you might know that a word had an underlying /t/, but it gets voiced or flapped in some contexts, so you add a voicing diacritic to reflect that.

On the other hand, if you also knew the language in question does have voiced stops, or that this particular word does have underlying /d/, you might choose [d] to reflect those beliefs.

  • 1
    So your answer to the original questions, "Is there a difference between the sounds /d/ and /t̬/?", is "no", I understand. Sep 14, 2011 at 19:40
  • 5
    Well, if we want to get nitpicky, /d/ and /t̬/ aren't sounds, they're IPA symbols. The use of one and not the other in a transcription does convey some important difference, maybe even reflecting a language hearer's experience, but it's not necessarily one that could be confirmed by acoustic measures, say.
    – JoFrhwld
    Sep 14, 2011 at 19:51

Your Answer

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