Saturday, December 26, 2009

comparison english/chinese diphthong phonotactics. hioxian...

Standard English diphthongs
RP (British) Australian American
GA Canadian
low [əʊ̯] [əʉ̯] [oʊ̯]
loud [aʊ̯] [æɔ̯] [aʊ̯] [aʊ̯]
lout [əʊ̯]1
lied [aɪ̯] [ɑe̯] [aɪ̯]
light [əɪ̯]1
lane [eɪ̯] [æɪ̯] [eɪ̯]
loin [ɔɪ̯] [oɪ̯] [ɔɪ̯]
loon [uː] [ʉː] [ʊu̯]4
lean [iː] [ɪi̯]4 [ɪi̯]4
leer [ɪə̯] [ɪə̯] [ɪɚ̯]3
lair [ɛə̯]2 [eː]2 [ɛɚ]3
lure [ʊə̯]2 [ʊə̯] [ʊɚ̯]3

Chinese (Mandarin).

Just lemme cut 'n paste here. Macs- I hate 'em.
SIlvia is rebuilding my PC so I can load up the Font design software for me again.
I get by with a little help from my friends...

D: OK so diphthongs- dual vowels- with phonetic meaning and lexemic content are important for Decimese.

Keep in mind that our vocabulary generating, based on the consonant-vowel (CV) syllable versus just a single C or V makes brevity difficult.
Most taxonomic languages suffer from being unclear when heard. If only one sound stands between 2 different meanings, then any background noise can be a problem.
E.g. if roboba is banana, while robobe is orange, then the rest of the word that is not different is of no help.
My CV-syllable basis is a compromise to retain the benefit of 'Ro' while minimizing the shortcomings.
Witness: if robocabam is banana, and robocatim is orange, then we have a much better chance of hearing the difference.
Metronyms seem to be a theme in Decimese.
I studied the top 1000 words of English - the most common ones- this month.
A taxonomic philosophical language like Ro fares pretty well with them.
I found I was trapped in my English language while trying to suss out a few concepts behind most words.
I kept digging up synonyms, or other common words. I can see 'thinking outside the box' will prove hard.
Things like 'kinship', 'social status/organization hierarchy', social prestige and such kept repeating as concepts.

Special syllables.
Anything that starts with an H can be used in this fashion. Hi, Ha, He-...
Similarly, mid-word, embedded inside a word, -hi-, -ha- ... can also denote 'special' meanings.
It's hard to explain with Decimese, since the premise is different that other languages.
Instead of starting at, say, a list of pronouns, I attempt to build them up from one iteration earlier.
I.e. primitive math, geometry (and therefore space) concepts.
MELTS- math, ethics, logic, time, space.
Lots of languages claim they have some special design basis or emphasis. How many deliver?
Math concepts - the space/time - really DO form the foundation behind my closed-class words.

HIOXian. Insight.

I knew that English meaningfully uses voice/voiceless consonant pairs. E.g. P, B.
Chinese instead uses aspiration. P, Ph (breathy). They just use 'B' to stand in for Ph.
I just assumed this meant I'd need to constantly denote voiced/voiceless pairs and standard/heavily aspirated ones.
I was wrong. It means - and this is just in reference to Decimese - that I DON'T need to bother with either!
The HIOXian characters I was making were overly complicated to hand-write in cursive script.
Too many swirls and flourishes. It took too long.
Look at the basis for Decimese words. CV-CV-CV... (end with nasal consonant) - for standard lexicon.
A little more: The very first -word-initial- consonant *can* be voiceless, whereas the internal ones are not.
E..g bapan. (B-P pair).
Keep in mind that word boundaries will still be clear even without this. It is just one additional aid.
Enough of the world cannot use voiced/voiceless consonant pairs that to differentiate between them is a non-starter.
So now the HIOXian cursive script becomes much simpler.
I imagine that the computer font could also neglect both the English voiced/voiceless and Mandarin standard/aspirated dual aspects.
It would be less phonetic and accurate, strictly speaking. But would make for a more spartan and minimalist appearance.

I was having an additional problem with HIOXian - the opposite, in fact.
A character that is TOO minimal tended to have a series of disconnected parts. Worse, the letters could be on either the right OR left hand side of the character space. This is NOT a big deal with COURIER font (regular even spacing)- though a series of such unfortunate characters side-by-side could still cause confusion.
I found myself looking at a much earlier proposed design I had explored.
Let us look at the labial aspect - lips.
Trivia: labia is already plural. Labium is singular. Our bilateral human bodies favour the existence of a 'dual' plural option.
Decimese will be able to capture this nuance with its single/dual/plural system. Decimese will be looting various natural languages to increase the power and scope of its closed class function words.
Ergo, Decimese can express:
1) labium. one lip.
2) set of labium - labia.
3) labia-ez? Presumably multiple sets of the same kind. Or one human's grand total.
I put wayy too much thought into um language LOL. What can I say - I'm a cunning linguist! <:
Cuz the HIOXian letter is really nothing more than a stylized human head in cross-section, only slightly stylized so that it can almost claim to still be pictographic (a picture) instead of ideographic (an idea).
OK. so imagine this human head looking to the left. The lips are obviously the left-most anatomical part that is involved in speech. Logically, the teeth would be next.
When the lips are engaged with each other, for example in the bilabial nasal "M' sound, or the initial part of the plosive 'P', we need to show the two vertical left-side 'meaning segments - bars' in contact.
Here's where it gets interesting. The HIOXian letter, being a static and unchanging figure like any letter (unless we dress up a font with video animation) will have difficulty expressing a change of anatomical positions.
For example, M shows no change in bilabial lip position. Whereas the plosive P obviously has the lips part in the second part o the movement. The air flow bar segments indicate that the P is a plosive movement, with a sudden forceful exhalation.
However we see that the M and P sounds require different later expressions of lip position compared to each other.
I initially considered indicating when lips and teeth were retracted-not in contact. I could also do this with the tongue.
Anyway, instead of denoting retracted, the plosive P sound can use this second position of 'retracted' to denote the second half of the movement.
Note: I struggled with wide/narrow tongue positions. I realized this week that I can express this as an air-flow aspect, with the implication that the tongue is behaving in said fashion.
So the HIOXian figure (think old calculator display) has two bar segments on each facing.
The left/top is top lip/ retracted after movement. By extension, the right/top would be dentition - the top teeth - retracted after movement.
The top 1/2 of relatively vertical bar segments are reserved to indicate tongue position of contact on the mouth top.
(Lip) (Teeth) velar ridge - hard palate- soft palate.
I also struggled with this. Do I need to show adjacent physical contact between body parts?
I.e. if the fricative "F" is top teeth-bottom lip, then should those bar segments be in contact.
I eventually ruled this out cuz I need to free up some segments for additional meanings.
Ergo F would show the lower lip - left/low/vertical and upper teeth- mid/high/vertical bar segments.
There is no change in position during a fricative. Air resistance is denoted via the leftmost bar segments.
Back to the tongue.
I eventually realized that I really DON'T need to denote tongue contact point very specifically.
Only certain combinations are possible. For example, nobody will use the rear raised part of the tongue to make contact with the teeth. I suppose this IS possible. Certainly. an L sound can be had via multiple positions. Different phonete, yes, but same phoneme - same meaning. This is much like how aspiration is unused in English, while remaining present as a physical side effect of bioarticulation.

It is not phonemic in English. In English we say that the aspirated [p] of 'pill ' and the unaspirated [p] of 'spill' are allophone

D: whereas Mandarian Chinese DOES use the P/P(aspirated) pair for phonemic and hence lexical functionality.

Ok so back to the tongue.
So far I think I'm reserving the bottom half of the fairly vertical /|\ internal bar segments for the tongue.
It may be the I will need them to denote details about air flow. If so, I just use one - | - to denote the tongue and let the user figure out the rest.
A theme that is emerging about the HIOXian phonetic, ideographic, anatomical letter system is how optional many details can be. Want a fancy cursive? Indicate all optional details.
In a rush to short-hand yer notes in class while the prof drones on? Less detail.

An interesting side effect of HIOXian is the ability to mechanically compare to figures or a series in a word, and guess how the sounds will deform. This would be based on both sound sequence, as well as syllable stress. Rules to apply could be used without the user even understanding anything about co-articulation or any other aspect of linguistics.
Take, for example, the word "elephant"
e- le- phant. e-le-fant.
E! (primary stress) le (no stress) fant (secondary stress).
We can expect that the second syllable -le- will become truncated and deformed.
The 'eh' sound in colloquial speech become instead the 'schwa' sound from 'the' - think a really brief and weak 'uh'.
The diacritic in HIOXian - strictly optional BTW - would indicate syllable stress.

Origin of HIOXian - it was for a sci-fi short story I am writing. Basically about an after-the-holocaust 'Rosetta Stone'.

The Rosetta Stone is an Ancient Egyptian artifact which was instrumental in advancing modern understanding of Egyptian hieroglyphic writing. The stone is a Ptolemaic era stele with carved text made up of three translations of a single passage: two in Egyptian language scripts (hieroglyphic and Demotic) and one in classical Greek. It was created in 196 BC, discovered by the French in 1799 at Rosetta and contributed greatly to the deciphering of the principles of hieroglyph writing in 1822 by the British scientist Thomas Young and the French scholar Jean-François Champollion. Comparative translation of the stone assisted in understanding many previously undecipherable examples of hieroglyphic writing.

D: how do you get illiterate primitive savages to self-teach how to read???
With a picture of a human head in cross-section. And so HIOXian was born. That was a few years ago.
I'll post the story.

HIOXian diacritic and English syllable stress system.
There are various ways to denote syllable stress. Obviously a mono-syllable language has no need of it- Vietnamese?
English happens to use 3 parts:
1) pitch
2) duration
3) volume.
Normally, writing short-hand suggests a simple mark for primary and secondary stress would suffice.
Because in English this is linked to duration.
The diacritic is UNDER a consonant and OVER a vowel.
An option to speed up reading speed.
CV-CV is then very clear.
Anyway, a syllable MUST have a vowel. While a consonant is optional.
In Decimese, this is NOT true. No syllable is smaller than CV-.
Ergo, in Decimese we have the option to denote syllable stress on ... the CONSONANT.
Weird huh.
Why would we want to do that?
Tonal languages.
Lexical-tone, to be precise. Such as Mandarin Chinese.
Roughly speaking, a high or low or mid pitch changes word meaning.
ma ma ma ma - scold, hemp, horse, mother.
We then have the option later to explore this aspect on the vowel.

Anyway, so on a Decimese consonant, the diacritic (subcritic versus supercritic) would appear as:

| /|\ |
- -

Nice artwork, huh? <:

OK so the simplest thing to do would be to simply use the central post marked to show stress.
But then for large enough words in, say, English, we could not show both secondary and primary stress.
So we use - -. That way, we show timing- duration.
In a language like French, it does not change duration as part of syllable stress.
The additional vertical bars can be used to express *some* pitch and volume information, with 2 levels.
|/ \|
The central bar, well, again it is a nice 'cheat' to short-hand stress info. Primarily for cursive writing to keep it brief.

Suprasegmental meaning - for example, interrogative.
We raise the pitch on the last word in English to show a question.
The presence of 2 levels of pitch means - wait for it- we can start showing prosody date if we wish to also.

So we have a rough outline, in summary of
1) the main square HIOXian character which
a) contain initial anatomical position
b) shows change/movement.
c) shows air resistance and qualities
With a diacritic that shows
a) volume
b) duration
c) pitch.

My room is gonna be a mess. Make lotsa sheets of HIOXian blank characters, explore schemes.
Try to map 50-60 phonemes onto them. !
I am primarily interested, in descending order, in
1) Decimese
2) English
3) Chinese
being mapped well onto the characters.
4) IPA is a consideration, I'd like to be able to express more nuance.
I've finally accepted that I cannot supplant International Phonetic Alphabet in the core character plus diacritic.
I'd hafta use other variations of the HIOXian shape to do so.
For example, the left OR right half of the HIOXian letter plus sub/super diacritic.
OR: one of three tiers, if we treat the HIOXian character space as a stack of 3 diacritics.
But that is for later. I have wayy to much work as it is.
Practically speaking, my first use will be taking notes myself in HIOXian minimalist cursive writing.
That means 2) English is in fact what I should focus on for now.
Only by using my system will I get a feel for its strengths and limitations. Then I can maybe refine it.
One day, maybe it will be ready for primetime.
One day... year.... decade.... lifetime?


Dino Snider said...

My apologies for the 3 typo mistakes so far.
I'm sleep deprived for two days now, and must work again tonight too. I'll make some mock-up rudimentary HIOXian letters this holiday using the Paintbrush utility LOL.

dino snider said...

Take the time to read the actual Story of the Rosetta Stone. It is so exciting, they could make a movie of it.

dino snider said...

Thinking on the P/B/M phoneme set. All look the same to lip reading. Well, once we indicate what the air will do, denoting that the lips part midway thru P/B but not for M is in fact redundant. This insight condensed down the character considerably.
Recently, I've realized the cursive form will necessarily be a hybrid between the narrow and wide transcription versions. It must be. It needs to possess as many simple continuous lines (curves really) as possible, to make writing them simple and fast. There will as much art as science in making the right judgements.