Talk:Tap and flap consonants

flap or tap?

i think it would be good to discuss the distinction between flap and tap that some phonetictians/phonologists make. peace — ishwar  (SPEAK) 05:09, 2005 May 24 (UTC)

Japanese retroflex or alveolar lateral?

Previously the ramen in Japanese was assigned the retroflex flap. Now it has been changed to the alveolar lateral flap. Is there a source for this, because the phonology in the IPA handbook assigns the retroflex flap. From my studies of Japanese I've gotten the impression that it sounds exactly like the retroflex flap that is a common allophone of /l/ in many dialects of Swedish.

Peter ^Isotalo 04:05, Jun 9, 2005 (UTC)

Unfortunately I don't have Ladefoged's SOWL with me, and I can't recall offhand what he said. The sound is definitely lateral, at least in Kansai dialect. Impressionistically, it sounds rather central and retroflex before an /i/, but strongly lateral before an /o/. Perhaps it's a retroflex lateral flap?? I can't substantiate the change right now, so revert if you like. kwami 04:39, 2005 Jun 9 (UTC)

I just took a look at the IPA Handbook, and it says that r "is postalveolar in place rather than retroflex" (under 'Conventions'). They also say "A postalveolar [l] is not unusual in all positions." (There's a 'back' diacritic under the <l>.) Whenever a phonetic description says that an /r/ is often [l], or an /l/ is often [r], we should be suspicious that it's actually a lateral flap. That's the tack that Peter Ladefoged takes, and in my admittedly very limited experience it seems that he's on the mark (as he usually is, of course). But we need a source that goes into a little more detail to be sure of what the IPA chapter is trying to describe. kwami 05:04, 2005 Jun 9 (UTC)

I dont think that the Japanese flap is usually retroflex & that this is perhaps not the best choice of a phonetic symbol to represent this sound. Ladefoged & Maddieson (1996) dont really have much to say about this sound except that they quote a Japanese article and note that there are many langs which do not have a lateral-central contrast involving rhotics. They also note that consistent articulations of these sounds may often be (mis-)perceived as either lateral or central (as mentioned by kwami above).

The passage that Ladefoged & Maddieson quote is the following:

"Some Japanese use both a lateral approximant [l] and a flap [ɾ] as completely free variants. Some Japanese use a lateral approximant [l] in the word initial position and use a flap [ɾ] in the intervocalic position. Some use a lateral approximant [l] in each position. Others use a retroflex voiced stop [ɖ] in addition to these sounds." (Shimizu & Dantsuji 1987:16)

Akamatsu (1997) presents a slightly different picture. He calls this sound a voiced apico-alveolar tap (note that he doesnt distinguish between tap and flap, but his description is of a flap). He notes that when describing taps/flaps (in any language) most people do not focus on whether there is contact between the sides of the tongue and the alveolar ridge. In intervocalic position he says that the tap is usually central and not lateral. In utterance-initial position, the tap is realized as (1) "a variant of [ɾ]" or (2) a lateral flap (Akamatsu's symbol is [l ̌]). Akamatsu does not provide a phonetic symbol for the "variant of [ɾ]" since he disagrees with others' proposals. This "variant of [ɾ]" is essentially like the intervocalic flap except that the articulation starts with tongue resting lightly on alveolar ridge before being ballistically blown downwards. Akamatsu is rather opposed to descriptions of this variant being a "kind of weak [d]". He says "the apico-dental closure for [d] is relatively firm and long, which is not the case with 'variant of [ɾ]', and the characteristic hollow on the front of the tongue for 'variant of [ɾ]' is missing for [d]." Regarding the lateral flap, he thinks that what many call a lateral approximant is most likely a flap. Then Akamatsu points out the noncontrastiveness of a lateral or central articulation and makes a statement similar to Shimizu & Dantsuji:

"Although I said above that [ɾ] (voiced apico-alveolar tap) frequently occurs in intervocalic context while a 'variant of [ɾ]' or [l ̌] (voiced apico-alveolar lateral tap) occurs in postpausal context, it should be borne in mind that, depending on individual native speakers of Japanese, any of these can occur, albeit unsystematically, in either of the above-mentioned contexts, as my personal observation over the years testifies."

I note that Akamatsu is describing (so-called) Standard Japanese and not the Kansai Variety. On a more anecdotal note, I observe that when singing slow-tempo songs, the actual realization sounds rather similar to the English approximant [l]. When I once tried to produce a flap in a slow song, one Japanese person remarked that I sounded like a speaker with a Korean accent.

With all that typed, I must agree that a retroflex articulation is rather unlikely. Whether it is lateral, I am non-committal — I would like see numbers (actually, I really would!). Perhaps there is a better prototypical language example? cheers — ishwar  (SPEAK) 07:33, 2005 Jun 9 (UTC)

Hm... I wonder if the essential difference between Japanese and Spanish r is that one is a flap and the other a tap? (Though there's still some fundamental difference between an English and Japanese flap.) Maybe that's what the author of the Handbook article was trying to capture, by using a retroflex flap rather than an alveolar tap, and then denying it is retroflex? BTW, I can't recall ever hearing anything similar to English or Spanish l. Instead, it seemed to always be a flap (never a tap that I could tell) in all positions, though sometimes more and sometimes less lateral. I'd always assumed the degree of lateralness was an effect of the vowel, but perhaps I just extrapolated from the idiolects of the people I first learned from, or even from the particular tokens I imitated. kwami 08:05, 2005 Jun 9 (UTC)

re my experience: I have noticed what I suspect is lateral approximant realization is only in very slow speech, as when singing or perhaps when speaking in "silly" voice — I've never heard in "normal" conversational speech. But this observation is based on only a few pop songs and only one Tokyo speaker. And I am not quite fluent in Japanese, so who knows...

re effect of vowel (& perception): Ladefoged & Maddieson do have a couple sentences on that:

"Some of the reports of alternations between ɾ and l in a variety of languages may be attributable to different perceptions of what is in fact a consistent articulation, particularly when the conditioning environment is said to be vowel environment, as in Nasioi, Barasano, and Tucano. In general, back vowels seem to predispose toward the production (or perception) of lateral variants, and front vowels toward rhotic variants. In Chumburung, the phoneme l has a rhotic variant which occurs medially in words with narrowed pharynx (retracted tongue root) vowels."

I am providing my references for any interested parties:

• Akamatsu, Tsutomu. (1997). Japanese phonetics: Theory and practice (pp. 105-116, 316-320). München: LINCOM EUROPA. ISBN 3-8958-6095-6.
• Ladefoged, Peter; & Maddieson, Ian. (1996). The sounds of the world's languages (pp. 210-211, 230-232, 237-243). Oxford: Blackwell Publishers.
• Shimizu, Katsumasa; & Dantsuji, Masatake. (1987). A cross-language study on the perception of [r - l] - preliminary report. Studia Phonologica, 21, 10-19. (cited in Ladefoged & Maddieson 1996).

— ishwar  (SPEAK) 19:16, 2005 Jun 9 (UTC)

Interesting. It always sounded the most lateral to me before /o/, which is the backest Japanese vowel, and least lateral (almost a [d]) before /i/. kwami 02:05, 2 April 2006 (UTC)

Eh, looks like somebody had the wrong IPA symbols on the ramen article, so that's why I also had the wrong symbols. I'm copying the one from here to that one. --/ɛvɪs/ /tɑːk/ /kɑntɹɪbjuʃənz/ 10:09, Jun 10, 2005 (UTC)

Recently I heard a native speaker speak Japanese on TV. Lo and behold, /r/ was an eminently unspectacular alveolar one-contact trill, the exact same sound as Spanish word-internal <r>. Not flapped*, not lateral. But again, that was just one speaker, and it may well be that all occurrences I heard were word-internal. (I don't speak Japanese – I can't tell where the words begin and end.)

(*) By "flap" I here mean the American English sound in butter and bidding, and nothing else.

David Marjanović | david.marjanovic_at_gmx.at | 01:07 CET-summertime | 2006/4/2

Initial flaps - difficult to pronounce?

Is it true that flaps are difficult to pronounce at the beginning of a word? If so, does that explain why many Koreans (especially South Koreans) don't like the r sound at the beginning? — Preceding unsigned comment added by 68.194.108.16 (talk) 02:22, 17 October 2005 (UTC)

This is an areal feature, common to Japanese, Manchu, Mongol, and Turkic. It's not uncommon elsewhere in the world either. As for why it occurs, perhaps someone else knows a definite answer. I would suggest that, rather than being difficult to pronounce, its disfavor from initial position may be a historical residue: either flaps are preferentially formed elsewhere in a word, or else initial flaps easily develop into other sounds. kwami 22:18, 17 October 2005 (UTC)

labiodental flap in IPA

The right-hook v for the labiodental flap should probably be included in the table of flaps in the IPA, instead of further below where I put it for now. However, since there's no Unicode for this symbol yet, it's awkward to include it in the table. 4pq1injbok 14:59, 3 November 2005 (UTC)

Ladefoged

However, he no longer feels this is a useful distinction to make, and prefers to use the word flap in all cases.

Page needs to be revised to reflect death of Peter Ladefoged. — Preceding unsigned comment added by 71.106.170.148 (talk) 06:04, 5 October 2006 (UTC)

Indic orthography?

Could you please add orthography or standard Indic transliteration to the examples in Flap consonant#Retroflex flaps? Thanks.--Imz 03:03, 20 November 2006 (UTC)

Iwaidja lateral flaps

There is some confusion on the web about 'lateral flaps' in Iwaidja. These are in fact quite different to the lateral flaps in Japanese, and to the presumably accurate sound file provided on the IPA's own page. Listening to this sound file you can hear clearly that the flap component precedes the lateral component, whereas in Iwaidja the opposite is the case. For this reason, we have employed symbols consisting of an alveolar lateral followed by a superscript alveolar tap and a retroflex lateral followed by a superscript retroflex tap for the two flapped lateral phonemes in Iwaidja. Bruceabirch (talk) 23:53, 12 November 2011 (UTC)

To whom exactly refers the "we"? Can you cite what you are claiming with a reliable source? If not, it would be original research and then it cannot be included in Wikipedia. --JorisvS (talk) 16:15, 13 November 2011 (UTC)