[Spelling Progress Bulletin, Summer 1982 pp13-15]
[Note: this page is set to display International Phonetic Alphabet characters. If your browser over-rides this setting, by default or by choice, the IPA characters may not show.]
[See English IPA chart and link to full IPA chart.]

Position Paper, by Robert L. Trammell, Ph.D.*

Why is it so hard to write a good R-controlled vowel rule?

* Florida Atlantic Univ., Roca Raton, FL.

It is not surprising that educators have not succeeded in writing very useful or meaningful rules for r-controlled vowels. The situation is so dialectally complex and various that even linguists and phoneticians are not in agreement on all points, even if we limit ourselves to regional dialects of standard English. Phonetically, vowels before /r/ do take on various degrees of "r-coloring" which may make the identification of the vowel phoneme difficult, or even arbitrary. For example, the /o/ of horse and /ō/ of hoarse may, in some dialects, be merged into a single sound half-way between /o/ and /ō/ for both words. On the other hand, both words may have /o/, or both may have /ō/ depending on the regional accent. In all three cases, the difference between /o/ and /ō/ before /r/ has been "neutralized", making horse and hoarse homonyms for those speakers with one vowel sound (/o/, /ō/ or the half-way-between sound) in both words.

Another type of r-coloring which creates perceptual difficulties in vowel identification is the occurrence of "centering diphthongs" in some standard English dialects. Certain vowels and diphthongs may add a glide to the schwa position (a central articulation) before syllable-final /r/, e,g. /hiər/ here, /paər/ pair, /fiər/ fire and /åuər/ our. At the phonetic level, the vowels become new diphthongs and the diphthongs become triphthongs, which makes equating them with particular non-r-controlled sounds difficult.

Since these problems do not occur in all dialects, r-control phonics rules which are correct for one speaker may not be correct for another. But the question rarely arises, because adults tend to hear other speaker's r-controlled vowels in terms of their own speech. Children are more phonetically perceptive and may be confused by a rule which contradicts what they hear.

While the situation is complex, a little knowledge of r-coloring, neutralization, centering diphthongs, and standard English dialectology would go a long way in improving the writing and teaching of r-control rules.


Position Paper, by Robert L. Trammell, Ph.D.*

Are all vowels before R really R-controlled?

In standard English dialects vowels before /r/ may follow, depending on the word and dialect, the regular phonics patterns; and thus, they are not r-controlled for many speakers. Only pronunciations of vowels before /r/ which violate the regular patterns are truly r-controlled pronunciations. In order to determine whether a vowel is r-controlled or not, compare the sound of the vowel before /r/ to the sound of the vowel in a similar word which follows a regular phonics pattern, e.g., tire/tide, boar/boat, moor/moot, our/out, hear/heat. If the vowel sound is the same in both word, the /r/ is not controlling the vowel. If the vowel starts as the same sound in both words but ends as a centering diphthong (i.e. with a little schwa) in the r-word, the pattern is still maintained, because this type of r-coloring is phonetically automatic. In my pronunciation, only hear is r-controlled, because it has /i/ not /ē/. For other speakers, more of the words may be. In the pairs far/fad, for/fop, pair/pain, the vowels will be different for most speakers, and hence, r-controlled.

Some speakers' pronunciation of the following words will follow the patterns while others will be r-controlled:

Long vowel, silent e:
Regular pattern: cōre, hēre, hīre, befōre, cūre, pōōr

R-controlled:/ker, hir, befor, kyur, pōr/

(Long i is generally immune to r-control.)

Regular pattern: mărry, mĕrry, Mŭrry, hŭrt, fŭr

R-controlled: /merē/ for marry, /marē/ for merry,

(Short u is generally immune to r-control)

Open-syllable, long-vowel:
Regular pattern: Māry, pūrify, stōry, Mōōrish

R-controlled: /merē/ or /marē/ for Mary, /pyurifī,storē, murish/

Long-vowel digraphs:
Regular pattern: āir, hēar, bēer, bōard

R-controlled: /er/ or /ar/ for air, /hir, bir, bord/

In many cases the r-controlled vowels fall into r-controlled patterns or phonograms. The pronunciation of some of these phonograms differ in a predictable way according to the dialect of the speaker.

-ar as /ar/ - far, car, party, star, tar, bar
-are as /ar/ or /er/ (depending on the dialect)-care, stare, scare, rare
-or as /ōr/ or /or/ (depending on the word and/or dialect) - for, short, or, fort
-er and -ir as /ər/ - her, after, herd, sir, first, fir

[Katherine Betts: see Bulletins.]

Vowels + r: Methodology,

by Katherine P. Betts, Ph.D.*

* Florida Southern College, Lakeland, FL.

An effective methodology for teaching pupils vowel + r: situations in word forms derives from these premises:

1. The phoneme /r/ is a consonant only in initial-syllable position (e.g. (r)un, co(rr)ect, (br)ing in spoken words. When pronounced in isolation, /r/ shifts to /ər/ a vowel.

2. The phoneme /r/ after any vowel in the same syllable is also a vowel, and therefore, with the adjacent vowel, is considered a centering diphthong (e.g., ar /är/ in park).

3. Words to be learned are identified by the pupil from a selection at his instructional reading level and are returned to that contest with their meanings (syntactic and semantic).

d. Pupil need to learn is a potential motivational factor assuring his participation in instruction.

Therefore, instruction for these and other word-perception skills begins with helping the pupil discriminate sounds as pronounceable units in spoken words (e.g., saying the whole word park, the last part - V+C - p(ark), the first part - C+V - (par)k, the vowel p(ar)k, and the whole word again, park). The procedure is repeated with the written word park, facilitating the chunking of graphic units representing phonemes. Next, by using initial and final consonant substitution (e.g., park-dark-dart), guided by the teacher to maintain a consistent pronunciation of ar in these words, the pupil can make a useful generalization for the analysis of similar words (e.g., harm, farm, cart), including the development of meanings in every instance. In summary, this methodology - phonics countdown - has several applications: enhancing auditory discrimination skills (e.g., vowel sounds embedded in syllables), relating phonemes to the graphemes representing them, chunking graphic units with transfer potential to the analysis of other unknown words. Thus, we foster independence in the pupil, guiding him to eventual automatic use of word-perception skills and freeing him to attend to the message, to comprehend - the purpose of reading, after all!


Position Paper, by Robert L. Trammell, Ph.D.*

What is the relationship between syntactic meaning and word perception?

* Associate Prof. of Languages and Linguistics, Florida Atlantic Univ., Boca Raton, FL.

Syntactic meaning makes word perception easier, because following the syntactic pattern enables the reader to assume the right mental set for predicting what part of speech is likely to be next, greatly reducing the number of possible words to be perceived at that point. The fact that we can silently read and understand:
The none tolled hymn she had scene a pare of bear feat inn hour rheum
in spite of the "wrong" spellings is attributable in part to the power of syntactic meaning as an aid to word perception. The syntax tells us the spellings are wrong and enables us to perceive the correct meaning by leading us to the appropriate homophones (nun, him, seen, etc.)

Word-callers with little or no comprehension are not sufficiently using syntactic meaning, which would help them to perceive individual words more quickly and enable them to read fast enough to "chunk" for long-term memory storage.


Position Paper, by Robert L. Trammell, Ph.D.

How do initial teaching alphabets compare and are they necessary?

Initial teaching alphabets may be compared as to the number of symbols and the degree to which the symbols facilitate transfer to the traditional spelling system. Unifon is almost a perfect one symbol-one sound system, but the symbols for most of the vowel sounds and some of the consonant sounds have little or no transfer value to traditional spellings or letters. The lack of a lower case is also a problem in transfering to regular spelling and the use of all capitals destroys the word configuration patterns established with L. C. ascenders and descenders. Finally, three pairs of Unifon letters are mirror images of each other - increasing the liklihood of reversals in perception and writing. The i/t/a has more symbols than there are sounds in English, but most of the symbols are related to traditional spelling patterns. Diacritical marking systems work better with vowels than with consonants or consonant-vowel combinations which indicate a consonant sound (e.g., ocean). The transfer value of such systems is great, however.

The wisdom and necessity of initial teaching alphabets is questionable, since many one-symbol-one-sound relationships can be established by the careful selection of words in traditional orthography; also, the spelling system to be learned is based on patterns and morphophonemic relationships as much as it is on one symbol-one sound correspondences. The three different pronunciations of "y" in cycle, cyclical, gym, my and funny are perfectly predictable by spelling rules and patterns. I/t/a spellings hide these patterns. The i/t/a spellings for fœtœ, fœtugraf, futogrufee present three recognition problems for the morpheme "photo," and two for "graph," destroying the semantic unity of the traditional spellings.


[John Henry Martin: see Journal, Bulletins.]

Position Paper, by John Henry Martin, Ph.D.*

The use of an initial teaching alphabet.

*Windjammer, Stuart, FL.

Tracing the changes introduced and made by Pitman and Dewey in constructing an initial learning alphabet, this paper (SPB, Winter, 1981) presents a new alphabet based upon the results of four years of research in beginning writing and reading with kindergarten and 1st grade children. This alphabet eliminates the new letter forms of Pitman, drops the ligatures in the digraphs and diphthongs and uses single graphemes to designate the two sounds each of "r", "z", "th" and "oo." Unlike Dewey, the macron is used to distinguish the long vowel sounds from the short vowels, which are presented without diacritics. 42 phoneme-graphemes are employed in harmony with a simplified form of the pronunciation key found in the Thorndike-Barnhart dictionaries. Thus all the graphemes are instantly recognizable and are useable for life.

These simplifications are the result of research based upon the intensive observations of children learning to write as the introductory process to learning to read. Children are encouraged to change to traditional spelling when it differs from the phonemic as rapidly as their reading discovers for them the conventional ambiguities and seeming irregularities of regular orthography. Thus they write the sound of "oo" in you, through, threw, to, two, too and glue as "oo" until they "see" the "way it looks in books." We found once more that the use of simple consistency in the material placed befote children will facilitate their understanding of the concept and the mastery of the skill. Finally, the paper rejects the need for absolute consistency between all phonemes and their graphic representations as a dogmatic overstatement of children's learning needs. This finding eliminates the need for new alphabetical symbols and unusual digraphs irritating to the sensibilities of parents and teachers whose acceptance of the process is critical to its adoption and maintainance.


1. Pitman, James. Alphabets & Reading, Pitman & Sons, Ltd., 1969.

2. Dewey, Godfrey. English Spelling: Roadblock to Reading. Teachers College Press, Columbia Univ., 1971.


Position Paper, by Robert L. Trammell, Ph.D.

Why do linguists resist a major spelling reform for English?

Since the pronunciation of English will continue to change over the centuries, many "phonetic" spellings would become irregular in time. Any reform would only be temporary. The choice of one English speaking country and one standard regional dialect within that country as the model for "phonetic" spellings would be difficult to agree on.

The present spelling system has certain advantages that would not exist in a more phonetic system. Speakers of very different dialects can communicate in writing without being able to readily understand each other's accents. Reading for meaning is facilitated in three areas by the present system. Related words keep the spelling of roots and stems the same even though the pronunciation changes (c.f. photo, photograph, photography). Longer spellings (e.g., though, through) are easier to perceive than shorter ones (tho, thru) because of the redundancy of the former. The written language is often less ambiguous than the spoken (The book was red/read).

In short the principal benefit of a major spelling reform would be the improved spelling for a relatively small number of people who speak the chosen dialect. We must better understand what is regular, predictable, and advantageous about the present spelling system before launching any major reforms.

Minor reforms, such as marking stress and indicating phonics syllables with hyphens, would take much of the ambiguity out of the pronunciation of vowels.

Back to the top.