According to The sounds of Japanese (Vance 2008):

When /ɾ/ is the first phoneme in an utterance, the tip of the tongue is already resting lightly on the alveolar ridge, and /ɾ/ is produced by rapidly releasing this contact. Strictly speaking this utterance-initial allophone isn't a tap.
  1. Can anyone elaborate on this utterance initial [ɾ] and its allophonic derivation, perhaps giving a reference for further reading? If it isn't a tap, then what is it? It sounds like it might be a voiced plosive, but it doesn't feel similar to either alveolar plosives [t, d]. It certainly sounds like there is an r-quality to it, but I wouldn't know how a plosive might acquire an r-quality.

  2. Are you actually realizing, and furthermore aware of, an allophonic variant of [ɾ] (specifically the apparent utterance-initial allophone)?

  • You should provide an example (a minimal pair).
    – user458
    Jul 15 '12 at 12:20
  • @sawa I know, but this is all the information I have! lolz. But this phone does not have a phonemic status, so a minimal pair in phonemes just doesn't exist. For a minimal pair in allophony, I think literally any other non-utterance-initial [ɾ] will serve as a phonetic contrast. Example: らちのらちです。the first [ɾ] must be the non-tap allophone where the second is a tap. This is really the best I can do.
    – taylor
    Jul 15 '12 at 12:30
  • I know this is answered, but is there a way you can clarify for future readers what you mean in layman's terms. For example "alveolar plosives" are not well known terms. Do you mean when is the R slightly rolled like a Scottish R and when is it more like an American L? I'm not a linguist and got lost pretty quick here... haha
    – BillyNair
    Jul 16 '12 at 20:25
  • @BillyNair okay I'll write an explanation, but it will take a bit of time
    – taylor
    Jul 17 '12 at 0:19

Wikipedia seems to say it is alveolar lateral flap, as opposed to the otherwise alveolar (central) flap. If you are serious, maybe you might want to try the article cited there. Trying this with myself, I seem to get the difference. I seem to be able to freely alternate an initial alveolar lateral flap with a centered one, but not a non-initial alveolar flap with a lateral one.

  • Stingy curmudgeons won't give it to me through my uni. And its a scant 3 pages. For $30US. For an international authority on phonetic transcription, their guidance is a little overpriced. Anyways, ty for helpful answer. I guess the features are [∓lateral] or [∓central] then.
    – taylor
    Jul 15 '12 at 14:35
  • @taylor Maybe it is ATR.
    – user458
    Jul 15 '12 at 16:07
  • @taylor: You can buy the entire book for around 30 USD. The article itself is available here. Sep 12 '12 at 1:38

According to "Handbook of the International Phonetic Association: Japanese" by Hideo Okada, /ɽ/ is pronounced

postalveolar in place rather than retroflex[...]. Initially and after /ɴ/, it is typically an affricate with short friction, [ɖɻ̝̆].

In a more Unicode-friendly notation, this could be transcribed as [ɖɻ̝̆] (where again, these aren't really retroflex but only somewhat retracted from the alveolar position). The retraction would impart the r-coloring you perceive.

Coincidentally, this sound is almost identical to the American English (John), except with the fricative portion extra-short, for those speakers that don't palatalize it.

The Handbook continues:

A postalveolar [l̠] is not unusual in all positions.

In summary, it sounds like [ɖɻ̝̆] and [l̠] are in free variation, with the former more common.

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.