One thing you must remember is that the system I designed must be noted as being provisional in nature, because I specifically did not design things in which I myself could not tell what would be used in any given case were a distinction to be maintained in certain places. Trying to figure out what should be what across the board is easy for "father" versus "bother", as the two are quite distinct from each other in the current orthography, but others are not so obvious. Hence, many of the less common distinctions in North American English are not included if they don't exist locally, simply because I really am working off a very limited set of data myself.
And yes, it cannot distinguish /@/ from /V/ or /@`/ from /3`/, simply because it is designed for a set of dialects in which [@] and [V], and [@`] and [3`] are allophones, with the former of each occuring in unstressed environents, and the latter of each in stressed environments, so hence such distinctions were simply not designed in. Of course, I would have a hard time designing any orthography which distinguished both members of both pairs, both not speaking a dialect which distinguishes either pair and also not having access to data which would allow termining what form should be used in what places if such distinctions are to be preserved. The only thing is that it does seem like the current orthography does preserve some indications where unstressed [V] might be present.
On the other hand, though, it does distinguish in all cases /l/ and /l=/, and as well may indicate /n=/ and /m=/. One note though is that various cases of [l=] in many dialects are actually as a result of allophony of /l/, such as in "tile" here, which is /taIl/ -> [t_haIl=]. Locally, it is clear that this is allophonic, as adding the morpheme "ing" in the case of "tiling" results in /"taIlIN/ -> ["t_a:I.lIN]. On the other hand, "dial" here by itself is /daIl=/ -> [daIl=], but when one adds "ing" to it as "dialing", it results in /"daIl=IN/ -> [daIl=IN]. Consequently, even though in the realizations of the two verbs' infinitives there are no distinctions in realization between the two, once one brings morphology into play, it is clear that there is a phonemic differentiation present, nonetheless. Similarly, with the other two pairs you mentioned, "royal" and "roil", and "towel" and "owl", once one adds another morpheme after them which starts with a vowel, in cases like "roiling", "toweling", and "owlings", and the hypothetical word "royaling", also are differentiated here. Consequently, I would write these as:
"dial" : daiyl
"tile" : tail
"royal" : roiyl
"roil" : roil
"towel" : tauyl
"owl" : aul
>>Quote-''One important note, though, is that the grapheme <s> may represent multiple different phonemes, based on context and like; it is not ambiguous in its usage, but rather one just has to understand a number of rules that determine how /s/ and /z/ are to be represented.''
Why not just use ''s'' for the /s/ sound and ''z'' for the /z/ sound, rather than making a bunch of needless complex rules determining whether an ''s'' is pronounced /s/ or /z/?<<
Actually, I initially had that, but unfortunately people tend to be picky about how things look, even if the changes to make things more palatable make something logically more complex. Of course, if I did not care about what people thought, I would use an eth (Ð ð) for "dh", a thorn (Þ þ) for "th", and an ash (Æ æ) for "ae"; as much as I myself prefer the aesthetics of these, and these characters did historically exist in written Old English and some extent Middle English, and are used today in various languages, many might not want to try to type them, and one of the main goals of the system which I devised provisionally was to compromise between logical cleaness, aesthetics, and practical matters such as typing. Of that, the only part that was not compromisable is that, for the targeted phoneme inventory and distribution, it would be purely phonemic, and could represent any possible target phoneme in any place where it could occur in theory in the first place. One note though is that when I'm on a machine which will easily allow such, I will still write in this orthography using eths, thorns, and ashes nonetheless, even though they aren't required parts of said orthography.
>>How would this sentence come out in your system:
''I was pulled into the pool by something.''<<
It would be:
"Ai was pyld intu dha puul bai samthing."
or if I feel like using extra characters:
"Ai was pyld intu ða puul bai samþing."
>>Including the /A/-/Q/ father-bother distinction in your system would also have the benefit of not making a bunch of words such as ''not'', ''cot'', ''on'', ''bother'', ''pot'', ''mom'' etc. become undesireably longer, as spelling them ''naat'', ''kaat'', ''aan'', ''baadher'', ''paat'' and ''maam'' would do.<<
Again, I wanted to make sure things kepts being purely phonemic, as I did not design this as trying to target a particularly large dialect range in the first place, and did not mean for it to be one which tries to maximize the number of distinctions present even if they are not present at all in the target dialect range. Hence, I did not include the /A/-/Q/ distinction, even though I very well could have if I wanted, primarily because I was specifically not being etymological from a design perspective here. There is one distinction that many lack in said range which is still included, which is /w/ as in "Wales" versus /W/ as in "whales", simply because many older people here still have it, but even then, such is meant to be purely optional in nature (so hence one may write "what" as "hwat" or "wat" for such in the dialect here in such).
|
