Re: USAGE: di"f"thong (was: Tetraphthongs, Triphthongs, Dipht..)

>I think it's kind of weird to design a writing system phonemically when
>it's for a language with many dialects that are strongly disparate in
>pronunciation, with different sets of distinctions made in each.
"Phoneme" is probably too strict a term. "Metaphonemes", each with dialectal
variation even on the phonemic level, is what might work the best. But you
already seem to proceed to that direction.


>Rather, maybe it'd be better to lay out a set of spelling rules that are
>more like those of French; many spellings converge on one pronunciation. In
>the case of English, we want to lean toward "many spellings converge on
>similar pronunciations", possibly. Then we can build some rules clusters
>that hint at other information (someone suggested that there is a graphic
>minimalism rule for non-function words; that's a good example; so is the
>use of different spelling schemes for words with different origins)
This sounds workable. Except if by the "origin" part you mean "origin
language", I'll have to disagree there. If there's one single thing to be
blamed for the current situation of English spelling, it's borrowing words
(esp. from French) and retaining the original spelling even if the existing
rules would suggest a different pronunciation.


>and with sufficiently diverse spelling options, make homophonic words
>non-homographic. It's more important to convey the word's identity than its
>precise dialectal pronunciation.
>
>So, for a word like /mE4l=/ we can have spellings...
><metal> for...metal
><medal> for that sort of insignia
><mettle> for wherewithal
><meddle> for interference
>all with the COMMON ENGLISH RULES CLUSTER and happening to be assigned to
>different items
>maybe <metalle> with the FAUX OE RULES CLUSTER for the heraldic colours OR
>and ARGENT
>maybe <maital> with ANGLICISED FRENCH RULES CLUSTER for some other word
>entirely, perhaps one having to do with food; since many food terms are
>French borrowings, neologisms can use that rules cluster to indicate that
>field of interpretation.
>
>Seems much more interesting and fun to do it this way, too.
>
>--
>Shreyas
That's pretty much the current system, then. To be precise, that part of the
current system that works. It could be fine-tuned, of course... (one
example: regularizing the alternation of i/y and u/w in vowel digrafs)


>Which is allegedly easier: remembering to spell homophones
>differently, or remembering to spell non-homophonic homonyms the same?
>  If the former, then we should standardize on a dialect with the
>maximum number of distinctions (including rhoticity); if the latter,
>the minimum (which implies non-rhoticity).   If the two varieties of
>orthographic discrepancy are considered equally undesirable, then we
>should go for somewhere in the middle.
>
>--
>Mark J. Reed <markjreed@...>
My personal experience suggests the latter, but then again I deal primarily
with written English. I've had a lot of "wait, these AREN'T pronounced the
same?" experiences over the last three or so years. People who use English
mostly orally (proceeding from speech to writing) I'd expect to choose the
opposite. So I can't see either extreme as being optimal.

John Vertical

Reply