Tuesday, November 08, 2011

Shuulinjii-in: A Practical, Phonetic Means of English Spelling

I've created my own orthography for the English language. Many people have done this before, and it wasn't adopted, not even when Shaw gave it a go, nor will this be. Nonetheless, my orthography is eminently practical and sensible.

As anyone who's learned a foreign language knows, it takes a while to learn an alien way of representing sounds, no matter how simple the rules underlying it. Which means, this will sound more complicated than it is.

Attractive features of my spelling system include: (1) it uses less letters than the current English alphabet; (2) it uses no new letters; (3) it's entirely phonetic; (4) each letter, digraph, or trigraph represents a single sound; (5) the sounds of the letters are largely derived from traditional English spelling; (6) it has a sensible arrangement of voiced and unvoiced consonants.

Let us begin with the consonants. The consonants come in the following groups (1) the stops; (2) the fricatives; (3) combinations; and (4) the liquids. We start with the stops as they are generally regarded as the "strongest" consonants.

A stop is any consonant that completely stops the flow of air in the mouth. We'll present them based on where they are formed in the mouth (the so-called "point of articulation"), starting at the back of the mouth and moving forward.

Velar stops.

These are formed with the tongue touching the back of the mouth.

Without a pitch, the sound is of the "c" in "car." This is represented by "k." (A pitched sound is one when the vocal cords vibrate. Hold a finger over your throat. Make a "k" sound. Now make a "g" sound. Hear the difference?)

With a pitch, the sound is of the "g" in "gun." This is represented by "g."

Finally, with the nasal passages opened, the sound is of the "ng" in "hang." We represent this with "ng."

So this gives us the series: k, g, ng. All are articulated at the same point in the mouth, with the first non-voiced, the second voiced, and the last nasalized.

Now at a different part of the mouth.

Alveolar stops.

These are formed with the tongue touching the alveolar ridge, behind the teeth, and occur in the same order.

Unvoiced: "t" as in "ton." Represented by "t."

Voiced: "d" as in "dam." Represented by "d."

Nasal: "n" as in "note." Represented by "n."

Finally.

Labial stops.

These are formed with the lips.

Unvoiced: "p" as in "pit." Represented by "p."

Voiced: "b" as in "beat." Represented by "b."

Nasal: "m" as in "moose." Represented by "m."

This completes the collection of stops. It forms this pleasantly ordered square.

k g ng
t d n
p b m

The first column is unvoiced, the second column is voiced, the third column is nasalized.

The first row is velar, the second is palatal, the third is labial.

The next group is the fricatives. Again, we group this by point of articulation, starting farthest back in the mouth.

Glottal fricative.

There is only one fricative at this location, the "h" as in "hot." It is formed in the throat, and is unvoiced, and is represented by "h."

In some languages this fricative also occurs voiced, Sanskrit, for example, but in English we only have the unvoiced version. Moving forward towards the lips, we arrive at the palatal.

Palato-alveolar fricatives.

Unvoiced: The "sh" sound in "shin." This is represented by "sh."

Voiced: The "sh" sound in "measure." This is represented by "zh." Note that as "z" is typically the voiced variant of "s," so "zh" is the voiced variant of "sh."

Alveolar fricatives.

Unvoiced: The "s" sound in "sin." This is represented by "s."

Voiced: The "z" sound in "zoo." This is represented by "z."

There's a slight difference in the position of the tongue here (it reaches back for "z") but the location is pretty close.

Dental fricatives.

Unvoiced: "th" as in "thin." This is represented by "th."

Voiced: "th" as in "the." This is represented by "dh." These two sounds, typically spelled the same in English, have to be distinguished. As "d" is the voiced variant of "t," it makes sense that "dh" would be the voiced variant of "th."

Finally, as we move forward, we reach the lips.

Labial fricatives.

Unvoiced: "f" as in "fun." This is represented by "fun."

Voiced: "v" as in "vine." This is represented by "vine."

This completes the fricatives.

h
sh zh
s z
th dh
f v

Combinations.

There are two sounds in English that are combinations of a fricative and a stop. They occur in a voiced unvoiced pair and are articulated at the alveolar ridge. One option would be to simply write these consonants as a combination of the fricative and the stop that make them up, but since (1) it's closer to current English usage and (2) it's easier to type one letter than two or three, I've represented them as single letters.

Unvoiced: "ch" as in "cheer." This is represented by "c." This is a combination of the stop "t" and the fricative "sh," and thus could be written "tsh." But I've chosen "c."

Voiced: "j" as in "judge." This is represented by "j." This is a combination of the stop "d" and the fricative "zh" and thus could be written as "dzh." But I've chosen "j."

This gives us the following combinations.

c j

Liquids.

The final group of consonants is the liquids, which are tougher to order, so here they are in no particular order.

"r" as in "run." This is represented by "r." "r" is a weird consonant, since after a vowel it consists of bending your tongue backward, but in front of a vowel, it's more done with the lips. This "r" represents the front of vowel version.

"l" as in "laugh." This is represented by "l."

That completes the consonants.

Vowels.

We could arrange the vowels by where they are in the mouth, but it's generally easier just to go by the typical English arrangement, which roughly goes from the top to the bottom of the mouth anyway. The vowels are as follows.

"a" represents the vowel in "fun."
"aa" represents the vowel in "watt."
"e" represents the vowel in "fed."
"ee" represents the vowel in "they."
"i" represents the vowel in "bid."
"ii" represents the vowel in "meat."
"o" represents the vowel in "caw."
"oo" represents the vowel in "foe."
"u" represents the vowel in "wood."
"uu" represents the vowel in "food."

Generally speaking the double vowels in the list are farther forward than the single vowels. In the cases of "ee" and "oo" they actually represent diphthongs, but few English speakers have any sense of diphthongs vs. monopthongs, so the distinction isn't terribly important.

Unfortunately this doesn't exhaust the vowels.

The vowel sound in "cat" was represent in Old English by the letter "æ." This isn't easy to reach on a modern keyboard though, so it's easier to just spell it as "ae."

The vowel sound in "fight" is represented by "ai." This vowel is a diphthong, and the spelling makes phonetic sense, since, using the vowels we've already established, it's a combination of "a" and "ii." I've simply omitted an "i" for simplicity's sake.

The vowel sound in "five" is represented by "aai." This vowel is a diphthong, and the spelling makes phonetic sense, since, using the vowels we've already established, it's a combination of "aa" and "ii." I've simply omitted an "i" for simplicity's sake.

The vowel sound in "boy" is represented by "oi." This vowel is a diphthong, and the spelling makes phonetic sense, since, using the vowels we've already established, it's a combination of "o" and "ii." I've simply omitted an "i" for simplicity's sake.

The vowel sound in "south" is represented by "au." This vowel is a diphthong, and the spelling makes phonetic sense, since, using the vowels we've already established, it's a combination of "aa" and "u." I've simply omitted an "a" for simplicity's sake.

To my ear, a vowel plus an "r" produces a sound so different from the vowel and consonant that makes the sound up, that we might as well just count the combination as a new vowel, in much the same way in French we count nasalized vowels as a new vowel.

So, the r-vowels, or "rhotic vowels" are the following.

The vowel sound in "car." This is represented by "ar."

The vowel sound in "wear." This is represented by "er."

The vowel sound in "fear." This is represented by "ir."

The vowel sound in "core." This is represented by "or."

The vowel sound in "fur." This is represented by "ur."

The vowel sound in "bower." This is represent by "aur."

The vowel sound in "foyer." This is represent by "oir."

The vowel sound in "fire." This is represent by "air."

So that completes the consonants and the vowels. This leaves the so-called semi-vowels, which are vowels used like consonants. That means that the vowel is said very quickly while leading to another vowel. Again we'll start from the back of the mouth forward.

The "y" sound in "yarn" is represented by "y."

The "w" sound in "warm" is represented by "w."

That's it, the complete palette of English sounds.

Leftover letters.

"x" wasn't used, but it could be used as a substitute for "ks" for the sake of speed, or alternatively represent foreign glottal sounds, like German "ich" or "nach" or Scottish "loch." "q," since it always occurs as "qu," which we can spell just as easily "kw," is completely superfluous and not to be used.

Hiatus.

It may be necessary at times to distinguish between two separate vowel sounds and diphthongs. In this case, as has historically been done, an umlaut can be used. This is somewhat difficult in modern keyboards, so feel free to substitute an n-dash.

Allophones.

Certain sounds are different in English but the difference is generally ignored by native speakers. The "t" sound in "ton," for example, is aspirated, meaning it is followed by a puff of air. The "t" sound in "stun," on the other hand, has no aspiration. Yet we spell them the same, and unless made of aware the distinction, native speakers are generally unable to distinguish the two. This is to be contrasted with, for example, Sanskrit, where the distinction between the two sounds is perceived by speakers.

As native speakers ignore the distinction, it hardly seems necessary to represent it. However, for foreigners learning to properly pronounce the language, the distinction might be beneficially marked, in which case, an apostrophe could easily be used, as in Ancient Greek.

"Pun" would thus be spelled "p'an." "Spun" would be spelled "span."

There's also the matter of the dark and bright "l" sounds, but that's far too abstruse to be represented as this level.

Stress.

Another distinction that isn't typically marked in English is the stress of a word. Again, this needn't be marked, but for purposes of teaching foreigners, an accent mark on the vowel can easily be used.

Alphabet.

For the purposes of ordering the alphabet, using digraphs as actual letters (e.g., treating "sh" as a letter by itself rather than simply a combination of "s" and "h") simply makes things more difficult. So to order the alphabet, we will simply use single monographs, and organize them by phonological position and quality.

Kk Gg Tt Dd Nn Pp Bb Mm Hh Ss Zz Ff Vv Cc Jj Ll Rr Yy Ww Aa Ee Ii Oo Uu

The order being:


Consonants
Stops (Velar[Unvoiced, Voiced], Alveolar[Unvoiced, Voiced, Nasal], Labial[Unvoiced, Voiced, Nasal]),
Combinations,
Fricatives(Glottal, Alveolar[Unvoiced, Voiced], Labial[Unvoiced, Voiced]),
Liquids,
Semivowels,
Vowels

Summary.

k, as the "c" in "cat."
g, as the "g" in "gun."
ng, as the "ng" in "ring."
t, as the "t" in "ton."
d, as the "d" in "dirt."
n, as the "n" in "not."
p, as the "p" in "pear."
b, as the "b" in "bear."
m, as the "m" in "mile."

c, as the "ch" in "chop."
j, as the "j" in "joke."

h, as the "h" in "hoe."
sh, as the "sh" in "shoe."
zh, as the "sh" in "measure."
s, as the "s" in "soap."
z, as the "z" in "zoo."
th, as the "th" in "thin."
dh, as the "th" in "that."
f, as the "f" in "fun."
v, as the "v" in "vote."
l, as the "l" in "loan."
r, as the "r" in "row."
y, as the "y" in "yell."
w, as the "w" in "woke."
a, as the vowel in "fun."
aa, as the vowel in "watt."
e, as the vowel in "pet."
ee, as the vowel in "they."
i, as the vowel in "pin."
ii, as the vowel in "sheet."
o, as the vowel in "raw."
oo, as the vowel in "show."
u, as the vowel in "book."
uu, as the vowel in "lewd."
ae, as the vowel in "cat."
ai, as the vowel in "fight."
aai, as the vowel in "five."
au, as the vowel in "cow."
oi, as the vowel in "toy."
ar, as the vowel in "car."
er, as the vowel in "care."
ir, as the vowel in "cheer."
or, as the vowel in "sore."
ur, as the vowel in "fur."
aur, as the vowel in "tower."
air, as the vowel in "fire."
oir, as the vowel in "sawyer."

A Robert Frost Poem.

Sam see dha wurld wil end in fair.
Sam see in ais.
Fram wat aaiv teestid av dazair
Aai hoold widh dhooz huu feevur fair.
Bat if it haed tuu perish twais,
Aai thingk aai noo iinaf av heet
Tuu noo dhaet for distrakshan ais
Iz olsoo greet
End wud safais.

Raaburt Frost

Note how easy it is to tell which lines rhyme.

Skaat Shuul

6 comments:

rob said...

...and all this time I thought your name was Scott shoila.

You might consider treating "spun" and "stun" as not containing "p" or "t" but [b] and [d] respectively. The differences between deaspirated voiceless stops and voiced stops are so minimal that in many, perhaps most, enunciations they are not distinguished. I once wrote a blog post arguing that the standard view in linguisitcs of this allophony is preposterous, at least in regard to the perception of the sounds. (There may be more than one way to produce a single sound.) In case you're interested:
http://languageandphilosophy.wordpress.com/2007/06/11/three-theories-of-english-plosives-the-myth-of-deaspiration/

Scott said...

Rob, I'm sorry, tell me where I know you from.

I'd never thought it about it that way but I think your post makes an interesting and suggestive point. I'll think about it.

rob said...

Also, just in passing, all phonetic-based writing systems face this dilemma:
consistent word spellings=>inconsistent phonetic spellings
consistent phonetic spellings=>inconsistent word spellings
across speech communities distinguished by physical, social or temporal distance.
So, I distinguish Mary, merry and marry, but some Chicagoans don't. One needed reach across the Pacific to find differences between "mate" and "might", "day" and "die" -- just stroll from one neighborhood to another in NYC.

Since to some extent we perceive written words as a whole shape (rather than sounding out the word from left to right), there's an argument for preserving standard spellings, which means that some speech group would have to be shafted, sooner or later. Currently, all groups are equally disadvantaged, which gives it a bit of justice to its insanity.

rob said...

From First Things, a couple of summers ago, discussing Hart's New Atheist article.

Scott said...

Yeah, I'm aware of that tension. Nonetheless, one rational spelling would be a net plus, even were it not precisely in line with one's own dialect. I bet it would be easier to learn consistent, say, Australian spelling, than current English spelling insanity--even though I'm American. Obviously, not as easy for as for Australians, but on net, easier.

Another option is simply to alter spelling with pronunciation, so British English would be spelled (even more) differently than American English. Is that a huge problem? My guess is no--I can read eye dialect in various English dialects--RP, black, southern US--without any real problem.

David Leon Gil said...

Though I'm responding rather late:

I think that the old-fashioned linguistics solution is in the notions of diaphonemes and diasystems. Long long ago, when I was interested in linguistics as a computer scientist, the way I recall these being described was along these lines:

For each dialect of some language, determine a complete set of phonemes. Then determine rules connecting dialects (e.g., car => caah). If applying these rules would result in a one-to-one mapping between phonemes in different dialects, collapse them to a single diaphoneme. Otherwise, map each phoneme to a diaphoneme. And voila, the set of diaphonemes, or the diasystem.

Wikipedia actually uses a fairly good diasystem (though limited to the prestige dialects of English). See http://en.wikipedia.org/wiki/Help:IPA_for_English

Alas that IPA is so unpleasant to look at.

(Oh, and should you read the Wikipedia article Diaphoneme: Ignore the first paragraph, which is confusing and perhaps misleading. And read the talk page after, which is much clearer than the article.)