ศาลาชุมนุม

[r2d2]

RID provides the following entry:

ไทย ๑ [ไท]

The phonetic entry [ไท] is not fully clear for the following reasons:

Lak Phasa Thai by Kamchai Thonglo




suggests the following:

ใอ = อะ+ย
ไอ = อะ+ย

Accordingly,ไอ = อะ+ย, or ไทย ๑ [ไท =ทัย].

It is discussed whether the final yo yak, as well as the final wo waen, should be i or j and w or o respectively in IPA.

The following picture suggests - in principle - that ไทย ๑ [ไท =ทัย] is /thaj/ in IPA:




But, "a) where are the rules how to apply IPA on sattha akson?"

Any suggestions?

[Richard Wordingham]

Don't expect an answer. For English, European pgoneticians hear <I> as [ai] and American phoneticians hear it as [aj] (and write it as [ay]).

I'm strongly tempted to say it's /j/, pronounced as [i]. My ears hear a vowel, but it patterns like resonants, including in such important matters as segment length aswell as in syllable structure. Incidentally, surely the other uncertainty is /u/ v. /w/ rather than /o/ v. /w/. The spelling uses /o/ because to an Englishman, <ao> is /au/ whereas <au> is likely to get pronounced as /??/.

[r2d2]

Thanks a lot for your assessment.

When I take this example
IPA Transcription Result for เหยี่ยว
jìːo

... oops there is something wrong with the algorithm ... เหฺยี่ยว I read jìːo
--> yiaaoL! = jìːo

No sorry, was not searching for bugs. Btw, found it in my own dic with jio.

Honestly, what to do with the final semivowels ย and ว (o, w?), and with : in diphtongs? I miss some published guidance on this topic (e.g. from RID?).

P.S.: A real mess with these diacrits needed for IPA, if Firefox is used?!

[chevil]

I can't remember the reference off the top of my head, but Gedney prefers writing final /j/ and /w/ as glides. The primary reason is simply that /j/ and /w/ can't co-occur with other consonants - e.g., sequences *CVjC and *CVwC are unacceptable (i.e. there are no words in the syllable canon - aside from possible recent English loanwords - of the form /kajt/, /bawn/, etc.). Also, they can combine relatively freely with the rest of the vowels (aside for the backness co-occurence restriction - *ej *ow are not allowed, etc.). If /aj/ and /aw/ (and /iw/, /oj/ etc.) were true diphthongs, then syllables of the form /kajt/ and /bawn/ would be allowed.

[DonSena]

The system of transcription used by Mary R. Haas, Richard B. Noss , Robert B. Jones and others makes the most efficient use of the Roman alphabet, requiring only four special characters (in addition to the four supra-segmental tone marks). ยืม Is transcribed as /jyym/, for instance, freeing up the /y/ for the vowel อึ, อื. The four special characters are those needed to denote the vowels in กอด, เลย and แดง, and the consonant in งู. This system was in wide use during the mid and latter half of the previous century and probably should be retained even during this century, as well.