Please forgive duplicate messages, but I appear to be having difficulty posting to the list.
Elizabeth
On Jun 10, 2011, at 4:51 AM, André Szabolcs Szelp wrote:
> Hello,
>
> thank you for your attempt.
>
> 2011/6/9 ejp10 <ejp10_at_psu.edu>
> I do agree with Everson's comment that IPA is essentially the Latin alphabet with lots of characters. Specifically, the common sounds (e.g. /a,e,i,o,u/ and /p,t,k, s,f,x b,d,g, m,n, l,r.../ are all from the old ASCII range. Many more are from the rest of the Latin A block (the top half of the old Latin-1 encoding). This is by design since many of the most common sounds are represented by characters from ASCII,
>
> Any phonetic transcription a linguist makes will be generated by inputting standard Latin characters with a smaller percentage of characters coming from the "IPA phonetic block" or one of the other more extended blocks (see this paragraph transcribed below)
>
> /ɛni fənɛtɪk trænskrɪPtʃən ə lɪŋgwɪst meks wɪl bi dʒɛnɨretɨd baj InpUtɪŋ stændərd lætɪn wɪθ ə smɔlər pərsɛntɛdʒ ʌv kærɨtətrz kʌmɪŋ frʌm.../ (phonemic, USA standard Mid Atlantic)
>
> Well, I counted in your very example:
> ASCII 72
> Other 41
> Assumptions for this count: your capital P were taken as [p], your capital I as [ɪ], your capital U as the correct IPA (non-ascii) char.
The substitution for /U/ is /ʊ/. Pardon my input error.
> <by> which you transcribed as [baj] was corrected to the more common practice of [baɪ], just for the record.
There is a lot of variation in how English vowels and diphthongs are transcribed, partly because. At the phonemic level /aj/ is adequate for American English.
> For your claimed pronunciation, probalby a great number of your <r>-s would be given with an other symbol than [r], further changing the relation (to ca. 66:49).
>
>
The phonemic level means that not all phonetic detail is recorded. It is common practice in the U.S. to transcribe American [ɹ] as /r/ UNLESS it is critical to record the phonetic details of said /r/
http://www.uiowa.edu/~acadtech/phonetics/#
If I were doing a detailed phonetic transcription, I would have to include stop aspiration, vowel nasalization, vowel lengthening, /t/ flapping & glotalizations and other details
> That's a whopping 57% ASCII vs. 43% non-ASCII!
BUT...many additional IPA characters are also used in Latin orthographies including /æ,ð/ (Icleandic), /ç,œ/ (French) /ŋ,ʒ/ (Sami) and /!/ (English).
Finally, there are lots of conventionalized, but non-IPA transcription systems which use characters such as /ñ/, /č,š,ž,ǰ/, /ā,ē,ă/ and /ü,ö/ used in plenty of other spelling systems. Distinguishing transcription versus orthography can get very complicated.
We need to tag the passages as a separate orthographic convention (with a language variation code per Peter Constable), but not necessarily separate the script.
>
> Clearly, IPA has a similar relation to Latin (even in your example!) as Cyrillic to greek or Latin to Greek.
>
> not to mention that no linguist wants to use exotic input methods for common sounds and typesetting non-standard characters has always been tricky
>
> What you describe very much sounds like a problem of input methods, rather than character encoding.
>
> If phonetic transcription were separated out as a "separate script", there would be (another) duplication of glyphs from the Latin block (probably almost all of the ASCII range).
>
> That's a common problem with related scripts, but it's not unheard of. That's why we have three (graphically) A-s, three K-s, three H-s three E-s, etc., two a-s, two e-s, two y-s etc.
>
> While texts could be converted to a "proper" separate script for transcription, I suspect it would very rarely happen.
>
> This really is an input method problem. Actually, I'm quite sure a keyboard layout containing all IPA symbols (pointing to the correct characters) would be readily used. Switching to it for IPA input is actually less bothersome to use, than using the language-specific Latin keyboard, then change to a character map program every second-third character, there constantly switch between character ranges (as you mentioned, Latin-1 (e.g. for æ) and _several_ phonetic extension blocks... I know what I'm speaking about, as I have done it several times.
>
> | A final issue is that linguists are notorious for inventing new transcription symbols informally.
>
> What you describe corresponds to the unlikely scenario, that one invents an ad-hoc character, it becomes widespread, and it does not only become part of official IPA, but of some Latin orthography as well... Yeah, this happens all the time.
In fact. the invention of ad-hoc characters does happen all the time and their usage does propagate. There are characters used in various academic disciplines (including the sciences) which are still NOT in Unicode proper.
You could use the methods you suggest to segregate characters into separate scripts, but I am not sure how well it would translate into the field. I do think linguists feel that phonetics is based on Latin. They would continue to equate IPA [s] as Latin "s", and it seems like taking cultural history into account can be important.
Elizabeth
>
>
> Hope this helps,
>
> Szabolcs
=-=-=-=-=-=-=-=-=-=-=-=-=
Elizabeth J. Pyatt, Ph.D.
Instructional Designer
Education Technology Services, TLT/ITS
Penn State University
ejp10_at_psu.edu, (814) 865-0805 or (814) 865-2030 (Main Office)
210 Rider Building (formerly Rider II)
227 W. Beaver Avenue
State College, PA 16801-4819
http://www.personal.psu.edu/ejp10/psu
http://tlt.psu.edu
Received on Fri Jun 10 2011 - 10:36:19 CDT
This archive was generated by hypermail 2.2.0 : Fri Jun 10 2011 - 10:36:20 CDT