CHARACTER SET FOR AKSES

Selection of Phonemic Word Elements and Characters

Phonemes are language units postulated to mediate conversion of thought/words into spoken words and of spoken words back into word/thoughts. Their nature and location in the brain are unknown, but they link natural English (defined as the oral tradition based upon phonemes) to natural written English (a writing system based upon the same phonemes). Noah Webster (Appendix to Dissertations on the English Language, 1789) devised a system "to reform English spelling," and noted that others had addressed the task before him, most notably Sir Thomas Smith, Secretary of State to Elizabeth I. Many other spelling reform schemes proposed since then were based upon vague understandings of phonemes and their functions. The pronunciation guides in American dictionaries are a logical place to look for phonemes as conceived in these pages.

In the following discussions, reference to spelling is to text in traditional (spelled) orthography. Written phonemic characters represent word elements and are shown between slashes; for example, the name of the AKSES character /a/ is spoken as the vowel sound in "cat".  The spoken word "cat" is written /kat/.

The American College Dictionary, Random House, 1965, and Webster's Seventh New Collegiate Dictionary, Merriam-Webster, 1967, display tables of pronunciation symbols with spellings of words illustrating the sounds. (Current editions provide about the same information.)  Most entries are obviously phonemic; for example, both tables list the same 18 consonant letters, each representing a single phoneme: b, d, f, g, h, j, k, l, m, n, p, r, s, t, v, w, y, and z. Both list a "singing" n identified by ng or ‘eng’ (IPA). All other AKSES consonant characters are represented by common digraphs: ch, sh, th (voiceless), ŧh (voiced), and zh (a "shushing" z). The last 2 characters are never spelled that way, but the suggested digraphs are easily distinguished. The 24 consonant word elements and characters should not be controversial.

Selection of vowel elements and characters is more complex. Lexicographers do not agree on symbols and occasionally include symbols that distinguish pronunciation effects such as stress or inter-phoneme and foreign influences. Short vowels /a/, /e/, / i/, /o/, and /u/ are usually identified by the corresponding Roman letter. Long vowels are named for the Roman letters a, e, i, o, and u, and are often printed with the macron (a line above the letter).  [If macron is not available in a font, the underline might be used although not as clear as macron.] Corresponding characters for ASCII are:  ae, ee, ie, oe, and ue (from the VCe “silent e” spelling pattern). These 10 vowel phonemes should not be controversial; they are accepted elements of spoken English and the suggested characters are widely used and intuitively understood.

Remaining vowel elements and characters are addressed in related groups. No "natural law" requires phonemes to be single sounds. ZH in measure, for example, is two sounds blended simultaneously, and ue, long-u, consists of sequentially blended /y/+/oo/. It could be represented as the separate characters, but character u-macron is included in AKSES as a phonemic “short-cut” symbol named “you.”

 [Letters X and Q represent possible phonemic vowel blends /eks/ and /kyoo/.  Q is never spelled without a following U except in few uncommon, obviously foreign, words.  It is not used alone in shortcut spellings to represent the vowel blend of the letter's name.  Authors realize, for example, that "qt" is unlikely to be interpreted as "cute."  Instead, most conventional readers will “see” an abbreviation for "quart" or the slang shortcut for "quiet" as in "on the QT."  X is a more useful phonemic symbol.  "ex" is a fairly common syllable or fragment.  Written words / hx/, /flx/, /txt/ and /xit/ are easily identified as phonemic written equivalents of "hex," "flex," "text," and "exit."]

Vowel sounds are often affected by a consonant that precedes or follows. Lexicographers recognize that vowel letters followed by r produce especially different sounds. For example, er (in her, fir, murder, nervous) is an r-modified vowel sound. Others are: ar (car, armor, start), or (for, tore, order), and the trigraphs air (fair, terrible, bear) and ear (fear, merely, queer). These vowel elements are always linked with r, so r should be part of the character. They cannot be represented by simple phoneme blends; /ar/ is not like /a/+/r/ any more than /er/ is like /e/+/r/. Likewise /or/ does not resemble /o/+/r/ or even /oe/+/r/. The pronunciation of /air/ is somewhere between /ae/+/er/ and /a/+/r/, while /ear/ lies between /ee/+/r/ and /i/+/r/. The 5 vowel characters ar, air, ear, er, and or are subject to many differences in individual pronunciation, but they tend to be pronounced consistently by an individual and, despite individual differences, are interpreted as the correct word element by listeners except when spoken in extreme dialects. They may be controversial in linguistic circles, but these phonemes and characters are useful for introducing written English to children and help readers and writers transition to AKSES.

Of the remaining 5 vowel sounds, 3 are easy to accept:  aw (awful, all, saw), ow (how, bounty, owl), and oy (toy, oil, oyster). Characters aw and ow (with w, not u) and oy (with y, not i) make each recognizable as a phonemic digraph rather than adjacent vowel elements. The remaining 2 vowel elements are often spelled oo, but that spelling is most characteristic of a "rounded" /o/, oo (moon, June, cruise). The other is like a "rounded" /u/, uu (book, could, pull), although never spelled that way. The 5 vowels aw, ow, oy, oo, and uu complete the vowel characters and should be acceptable to all but the most tradition-bound individuals.

Table 1 lists 44 phonemes (24 consonants plus 20 vowels) by "name."

Table 2 lists AKSES characters in alphabetical order and is titled "DHu Foenubet" to distinguish it from the English Roman “Alphabet.”

Return to Introduction (Links)



Published October 20, 1998. (Last worked on 10/12/07) - James H. Kanzelmeyer.