An orthography for English is simply a way to code the 12 pure vocalic phonemes and 22 pure consonant phonemes of speech. The simplest code is an alphabetic one.

An alphabetic code allows the 12 pure vowels to be combined to form diphthongs or blended vowels. In addition to the 12 uncombined vowels, there are about 20 vowel combinations (diphthongs) in common use.

With a simple code, FILE would be spelled [fah-eel], TAPE would be spelled [teh-eep] and TOWN would be spelled [tah-oon].  These sounds are all vowel glides or blends commonly referred to as diphthongs (Greek, 2 sounds). 

Instead, FILE is spelled fill-eh (as in filet) and TAPE is spelled tap-eh (as in tappet).  TOWN is actually a viable spelling [taw-uun] the problem is that the spelling could be pronounced tone (tow+n) or (t+own).  In fact this was the way it was pronounced before 1400.  ow and ou, originally [oh], are essentially the same but all words containing these digraphs do not rhyme.  About half of the words with this spelling shifted.  The vowel in TOWN and OUT came to be pronounced /au/ as in kraut. 

Pronunciation dictionaries will spell these words FILE /fail/, FAIL/feil/, TAPE /teip/, TOWN /taun/, TONE /toun/.  The IPA digraphs (shown in bold face) are simple combinations of the single letter grapheme-phoneme correspondences.

To avoid contradictions, IPA adds a few new letters such as the schwa [] and hook were added.  Other letters such as the turned c []  were added primarily to avoid confusion.  If o was always associated with the sound in [awe] and never [oh] or [ah] there would be no need to add a special [] character.  The stragety doesn't always work since the letter patterns in FAIL and OUT already have traditional or historical interpretations:  They are associated with the sounds  /feil/ and /aut/.

The problem with logical spelling is that they collide with traditional spelling  [fah-eel] would be spelled in Spanglish and IPA as fail.  FAIL would be spelled [feh-eel] or feil.  These two shifted sounds are among the most problematic in the traditional spelling of English.  See the number of ways to spell the sound in DAY.

Speech cannot be fully reduced to graphic representation.  However, a writing system can capture enough cues to enable native speakers to reproduce it.  The goal of reform spelling has been to index speech sounds in such a way that non-native speakers who know the sounds associated with the letters can read it aloud and be understood.  This is possible with the orthographies of other languages such as Spanish.

It also works when the orthography used for Spanish is used to transcribe English speech sounds. When this is done, one ends up with a set of spellings that are just as consistent as those in Spanish [see Spanglish].  The problem is that while we can read these spellings, they often look odd when matched with historical spellings. [see fail and teip above]

Learning how to pronounce strings of letters on a page is facilitated by having some consistent relationship between the graphemes and the phonemes (or significant speech sounds).  Learning an inconsistent code is possible but it takes longer.  Decoding (reading) takes so long that some people never learn to do it well.  Encoding (spelling) is even more difficult.  Most English readers cannot spell unfamiliar words in the inconsistent traditional code without the aid of a dictionary or spell checker.  [See Valerie Yule's Spelling Test]

Even when keyed with IPA as below, many of those interested in orthography have a difficult time understanding notations other than the one they are accustom to.  Some of those on the Spelling Mailing List have suggested taping conversations so that "we can sound the phonemes in the dubious cases."  Ian and Paul think that this would be a good idea.  The audio clips below are the next best thing.  [See also the audio clips for old and middle English which are essentially the same as Spanglish]

Vowel Chart - 15k gif file - This compact chart is keyed to the multiple notation chart below. It shows all of the vowel phonemes and the graphemes used by Chekt Spelling. The abbreviated chart below shows the same arrangement of graphemes-phoneme correspondences for UNiGRaF.

Minimum number of vowel phonemes for a broad transcription of English
Unigraf A        B        C        D  a  q  I qr Ir e  R  A   er i  E  Q   ir q  o  O   or c  C  U Cr Ur u  ^  V   Vr	The 25 vowel phonemes are illustrated in Unigraf notation on the left.  Unigraf limits the use of digraphs to r-combinations. The new sound-signs are Q-oi, and V-owl. The familiar U symbol has been turned to create 2 new sound signs C & c [lazy U's] for /u:/ & /u/. The first two columns [A & B] display 12 pure vowels: six short (checked) vowels and six extended vowels. The remaining columns [C & D] display six diphthongs and seven shwa combinations. A more detailed table is available for Chekt Spelling. The 25 speech sounds have been captured as audio clips and can be played if your computer can interpret .wav files. A wave player is included with Windows. Simply click on the hot links in the Uni column below.
UnifOn 2 short  long  i-comb   r-comb a   q   I  ar air e   R   A   er  i   E   oi  ir q  Q o  O   or  v   u   U   ur Ur x c c   au  aur	U2 was developed in 2001 as part of an effort to come with the best possible phonogram locations on a standard keyboard. It was a minor refinement of Unifon.  Changes included adding a british short o [Q] and avoiding the spelling of army as ormy.  o=awe, a=aa, and UnifOn as YnifOn.  The lazy u's use was expanded and x was used only when stress needed to be indicated as in abut [cbxt].
ROMIK short  long  i-comb   r-comb ae  aa   ai ar air e   rr   ei   er  i   ii   oi   ir o   oo   ou   or  u   uu   iu   ur  u.  a 'e au   aur	Romik World English notation is a compromise notation developed by the saundspel group.  It borrows ideas from several ascii-ipa notations [ nuu romaji, spanglish, winglish, zinglish, ...].  In Romic, a lone [a] is a mid lax vowel and slightly ambiguous as in TO. It can be ae, aa, or @ but never /ey/.  Romic manages to represent 25 vowels with 6 symbols counting the syllabic R. Wans apon a taim ther waz a biutiful princes hu wisht tu faind hrr prins

View chart as a graphic  (58kb gif file)     another letter matrix alternota   Spanglish vowel chart

Alternative Vowel Notations Compared
Clik-n-Lisn tu th saunds (Click-on Uni letter to play the digitized audio clip)
aulterna'tiv vaul no-taytions compaired (OGD)  Can u reed this?
olt'rna'tiv vaul no'tei5nz ka'mperd (Ka.n yu rid thi.s?) (chekt speling)
Olt'ernativ vaul notacions comperd. (Can yu rid this?) (spanglish)
oLTrNuTiV VmL NOTAsuNZ KuMPerD (KaN U RED xiS?)   (Unigraf -upstyle)
oltRn^tiv vMl nOtAS^nz k^mperd  Kan U rEd xis  (Unigraf-downsize) code overlaps shown in red
 

ROMIK notation uses the apostrophe to mark unstressed vowels and the schwa sound. penc'il litt'l pens'il yun't. u.th'r ath'r The organization of this chart is the same as in the one below.  In other words, B-3 in both charts will refer to the same sound.

Clik on dh' letr or simbl in dh' Uni col'm tu plei dh' saund.             [Romik]
Click on the letter or simbal in the Uni colam tu pley the saund.   [Spanglish]  [sp3]
Click on the letter or simbal in the Uni colam tu play the sound.   [Rong]

Globish Madjukar Gogate writes: I do not prefer capital-small mix (Angel), diacritics, or numerals to
indicate  peculiar sounds.

With only 26 letters for 40 sounds, you have to choose the lesser of the evils. None of the choices are that great. Madjukar merges short and long vowels and uses a few digraphs in his notation.  Digraphs are OK for true blends but are problematic when used to represent sounds which have nothing to do with the sound values assigned to the single letters.

Unigraf:  There are 52 ASCII letter characters but most scripts make use of only 26 as unique sound signs. Unigraf employs the lower case letters as distinct sound signs. Each letter in TO has 2 allographemes, eg. [G g] making it a dual redundant system. [K k] in Unigraf refer not to one but two sounds /k/ and /tsh/, commonly known as [ch]. [C c] refer to vowels not consonants. These sound-signs are referred to as "lazy-U's" /u:/ and /u/. The upright U is resevered for "you".

 

Unifon	I bxt Dc uthcr yelO kOt wen I sx Dc prIs
UNi	I BoT tu utr YeLO KOT weN I So tu PRIS.
Unigraf	I bqt xu uxR yelO kOt Wen I so xu prIs.
Follick	Ai boat dhu udher ielou cout huen ai soa dhu prais.
NF	`I bo:t th uthr ielo' co't wen `I so: th prys.
TSns	Ie baut thu uther yeloe coet wen ie sau thu pries
TS	Ie baut thu uther yeloe coet wen ie sau thu pries
CS	I bot th othr yelo coat wen I saw th price.
CKS	'I bot th othr yelo' co't wen 'I so th pr'ys.
SS	Ai bot tha other yellow cowt wen ai sao tha prais.
CSS	Ai båot dh athr yelw cwt wen ai såo dh prais
Romik	Ai bot th'e u.th'er yelou cout hwen ai soo th'e prais.
OGD	I baut the uther yelo coat wen I soaw the pryss.
OGD2	I baot xe uxer yelo koat wen I sao xe prys. caost

SoundSpel is another IPA-EuroSpel writing system. SS is more consistent with IPA than NF which makes it a more divergent from traditional English orthography TO. SaundSpel insists on having a unique symbol for every pure vowel. The vowel in *her is one of the primary vowels and therefore deserves a symbol distinct from the consonant [r]. Ascott uses the capital R for the sound in *her [hR].

A better system for representing 40 or so sounds could be devised if there were 18 more character shapes available. Such solutions are called augmented alphabets. Sound English or SaundSpel was originally an augmented alphabet which used Greek letters for unigraphic sound signs. Here the augmented forms have been replaced by digraphs and limited to what is available in the ASCII and Latin I character sets.

Nu Folik adopts the TO practice of substituting o for ah, but the a = /a:/ correspondence is retained in digraphs such as ar and ai. Nu Folic also substitutes y for /ai/ because this has long been a dominant Middle English and Modern English spelling pattern. ar for [ar & a:r] is retained because or is associated with another sound [o:r & o'r]. This strays a little from the Semitic distinction between yod and ayin where the y is associated with the /i:/ sound in eel and ayin both means and sounds like "eye". ¹

ai doesn't make sense as a sign for the vowel in eye unless a is /ah/. ai is pronounced /ah-ee/. Other systems, such as New Spelling use combinations that cannot be reduced to component sound signs: e.g., ie /ih-eh/ is not related to the sound in *eye except in TO where is is also related to the sound in *companies.

IPA uses au as the sign for the vowel in *owl /ah-ool/ and *sound. Most phonemic systems use the same two letters au for /awe/ (*hawk would be hah-uhk). It can be confusing when the same letter combinations are associated with more than one sound. Nu Folik uses ao /ah-oh/ and aw /ah-uh/ -- not perfect but a little better.

New Spelling uses oe /awe-eh/ to represent the sound /owe/. Obviously the e is silent and used as a marker. EuroSpel systems are opposed to the inclusion of silent letters. IPA uses ou and Follick uses both o and ow. ²
As a concession to TO, Follick reverses the w and the u.

Spelling and Etymology - the current books do not deal with spelling.

LINKS to The Sounds of English                & Their Graphical Representation  Saxon, Chekt, SaundSpel,   Spanglish, Nu Folik, & Anjel Click on the button of your choice Local (relative)links Spelling Links SiteMap-index OGD Remote (absolute) Spelling Links SiteMap-index Delphi.com/spelreform Use back button on browser Back - Previous Page Next Page CKS-checked spl ©1998 BETAWebcDesign