[Spelling Progress Bulletin, Spring 1983, pp2-8]
On another page: part 2.

The Principles of Reforming our Spelling, [part 1]

by Edmund V. Starrett, Ed.D.*

*Excerpted from the Ed.D. Thesis of Edmund V. Starrett, Wayne State Univ., Detroit, Mich. 1981.
*Dearborn, Mich.

Beginning with an Augustinian monk named Orm at the start of the 13th century and extending through the following seven centuries; eminent linguists, educators, writers, statesmen & organizations have developed plans, schemes, proposals and ideas on how to change the spelling of English to make it more phonemic.

Over the centuries, these spelling reform proposals have varied greatly in purpose, nature and extent of change. They have been referred to by a myriad of different names: simplified, augmented, phonetic, phonemic, rational, revised and amended spellings, to name but a few.

Despite the nature and extent of these proposals, they fall rather naturally into six distinct categories according to the types of change.

One type

 of proposal uses new letters or characters to augment or supplement the present alphabet. These new letters may be modifications of present letters or completely new symbols. John Hart, in 1569, and Alexander Gill, in 1619, were two of the earliest writers to recommend the use of additional letters. In the 19th century, Sir Isaac Pitman popularized Phonotypy, an augmented alphabet of 42 letters. Almost a century later, Sir James Pitman, grandson of Isaac, devised a new 45 symbol alphabet called the Initial Teaching Alphabet (i.t.a.) to assist beginning readers to learn to read.

A second type

 of spelling reform proposal includes those that advocate supplementing the present alphabet by the use of diacritical marks placed over, under, or next to a particular letter to indicate pronunciation. Early enthusiasts of this type of change are 16th century educators William Bullokar and Richard Mulcaster, and American patriot Benjamin Franklin.

A third,

 and most popular way to achieve spelling simplification involves :the use of digraphs, or new letter combinations, to represent certain consonant, vowel and diphthong sounds, The first recorded spelling reformer, Orm, about the year 1200, doubled consonants following short vowels. Examples of more modern alphabets are Nue Spelling by the Simplified Spelling Society in England, Anglic by the Swedish linguist Robert Zachrisson, and World English by Godfrey Dewey.

A fourth

 category includes those proposals such as advocated by The American Philological Association at and after the turn of the century, and the Spelling Action Society a decade ago, which advocated and promoted systematic; limited and step-by-step reform beginning with a certain group of words and extending beyond that to other word groups as acceptance dictates.

A fifth,

 and least common, means of achieving spelling simplification is through the use of colors to signal certain sounds of the printed word. Color coding proposals, such as those popularized by Caleb Gattegno and Alex Bannatyne are not, strictly speaking, intended as examples of permanent reform of English orthography; rather they are, proposed as a means of teaching beginning reading to children and adults.

The sixth,

 and most radical, type of spelling reform involves the, total replacement of the present Roman alphabet with new letters or characters which may or may not have some visual relationship to the present alphabet. Such radical new alphabets are based on the idea that the present alphabet is so irregular and inconsistent that a complete overhaul of the alphabet is necessary; and they are devised so as to save space, time of writing, and may be considered as forms of shorthand. The alphabet of the George Bernard Shaw contest and the Deseret Alphabet of Brigham Young are examples of this type of reform.

Seven hundred years of spelling reform proposals seem to indicate a consistent desire on the part of some scholars to try to do something to improve the state of the orthography of the English language. It shows a need for such improvements. It also points out the failure of this group to bring it about. Perhaps some other tactical means is needed to bring about such changes.

One of the most debilitating characteristics of most spelling reform efforts; however, is the inability, to agree among themselves as to the nature and extent of the reform. Some have, for example, limited themselves to methods of overcoming the special difficulties faced by children learning to read; others have been concerned with devising 'a phonetic alphabet, subsidary to or augmented to the Roman one, in order to indicate the "correct" pronunciation of dictionaries, for use in grammars or learning textbooks. Still other reformers have devised and sought to promote radical systems of spelling and alphabet reform for the ultimate benefit of the whole English speaking world. Their solutions range from minor modifications which affect a few words to completely new alphabets which bear little or no visual relationship to the present Roman one. While most proposed that the reform be introduced as a single dose, others preferred a series of gradual installments.

Classification of Spelling Reform Proposals.

Several authorities have made efforts to classify the different types of spelling reform proposals. Godfrey Dewey, for example, stated that English spelling could be simplified in three ways:

1) Standardizing the Roman alphabet by assigning to each single letter, and to each digraph selected to represent those sounds for which the available single letters do not suffice, a single sound.

2) Supplementing the Roman alphabet by assigning to each of the 23 useful letters a single invariable value and creating some 18 appropriately designed new letters.

3) Supplanting the Roman alphabet by creating and making available on typewriters and composing machines throughout the world at least 41 wholly new characters."


Tauber, in his Ph.D. Thesis, Spelling Reform in the United States, presents five classifications for spelling reform: Type I which has to do with systems of reform that rely on diacritical marks or modifiers to obtain a greater degree of regularity; Type II which has to do with those systems of spelling reform that seek minimal changes such as the elimination of silent letters or substitution of a letter or letters; Type III is an extension of Type II and uses digraphs to form new letter combinations; Type IV which has to do with adding new symbols to the alphabet, and Type V which has to do with systems of spelling reform which replace the letters of the present Roman alphabet with a whole new set of symbols.

This study will divide spelling reform proposals into the following categories for the purpose of a more logical presentation and comparison:

1) Augmented alphabets: Those alphabets proposed to add to the present Roman alphabet additional characters, or letters, resulting in an augmented or increased English alphabet. These new letters may be modifications of the present letters or completely new symbols or characters.

2) Diacritic Marks: Those systems of spelling reform wherein special marks are placed over, under, or next to a particular letter to indicate pronunciation generally with little change in the traditional spelling of the word. Sometimes these systems have been proposed merely to assist children in the beginning stages of reading. At other times, the proposals were serious attempts at permanent spelling reform.

3) Digraphic Alphabets: Those alphabets proposed to replace the present spelling system by omitting silent letters and using digraphs to represent certain consonants, vowels or diphthongs.

4) Word Simplification: Those proposals for spelling reform that rely chiefly on the omission of certain silent letters and the respelling of words to bring about simplification of certain groups of words.

5) Color Coding Systems: Those systems of spelling that use different colors to signal certain sounds of the printed word. Color coding is generally proposed as a means of teaching beginning reading:

6) Non-Roman Alphabets: Those alphabets proposed to replace the present Roman alphabet by creating new letters or characters. The new characters may be entirely arbitrary with little, if any, relationship of one character to another. They may be inverted, mutilated, or Roman letters with appendages.

Limitations of this Study.

It is impossible to present information on all or most of the proposals for spelling reform because their number is legion and because of the nature of promoting most of these proposals.

Spelling Reform Proposals using Augmented Alphabets.

Most observers of English spelling' point out the fact that the English alphabet contains only 26 letters, 3 of which are redundant (c, q, x) for the forty or so sounds of the language. As a consequence, one letter must stand for more than one sound. As if this situation were not bad enough, one finds that approximately 1/6 of all English words contain unnecessary silent letters and that over the years, alphabet or spelling change has not kept pace with sound change. English is thus left with a system of orthography in which no one can tell with certainty how a word is pronounced if he only sees it written but never hears it pronounced; and, inversely, cannot tell with certainty, how a word is spelled if he only hears it pronounced but never sees it written.

Throughout the history of the language, many individuals, scholars, and laymen have advocated the need for new letters to supplement the present system so that one symbol can stand for one sound only and each sound be represented by one symbol only.

Although there is some disagreement regarding the exact number of sounds for English, it is generally conceded that an alphabet of approximately 40 letters would be necessary to arrive at the desired effect of one sound for one symbol. In order to achieve this, approximately 17 new symbols would be necessary.

When it comes to the question of supplementing deficiencies of the present alphabet, this can be accomplished in several ways: according to P.A.D. McCarthy:
"... by simply taking over existing letters from other alphabets (e.g. Greek, Cyrillic), with such typographical modifications as may be demanded by the need to harmonize with Roman fonts; by altering Roman letters sufficiently to constitute additional independent shapes; by designing new letters to go with the Roman (these might of course have resemblances to existing letters, and so come optionally under the preceding head); and by combinations of the foregoing."
The reputed advantages of augmented alphabets over other systems of spelling change are as follows:

1) An augmented alphabet would present a one to one relationship between sound and symbol without the addition of unnecessary diacritics or unnecessary letters, as in the case of phonemic alphabets based on digraphic spellings.

2) An augmented alphabet would retain most or all of the symbols presently used in the traditional alphabet, thus establishing a pattern of familiarity, and understanding for those who already know the present spelling.

3) An augmented alphabet would allow for more consistent spelling of sounds and thus be more helpful for those learning to read and write the language.

4) An augmented alphabet is based on the logical principle of retaining what is consistent and usable and adding to it rather than starting over with a totally unfamiliar non-Roman alphabet as other reformers have proposed.

5) Augmented alphabets, without sacrificing essential characteristics, generally use less characters in writing which afford greater saving of time and cost in the reproduction of material.

Problems in the use of Augmented Alphabets.

A one sound, one symbol phonetic notation by supplementing the present Roman alphabet with new symbols has several major drawbacks that hinder its acceptance as a viable alternative.

The first objection has to do with the uncertainty regarding the exact number of phonemes contained in English and the number of symbols needed to express these phonemes.
(

Ed. comments:

 This is true of all phonetic alphabets, not just augmented alphabets. Franklin came up with a lot fewer sounds because he could not hear the subtle differences in some pairs of sounds. Pitman has 3 duplicates in order to make easier the transition to T.O. The I.P.A. splits some sounds into too small a unit to be useful, such as the dark and light 'l', which have no difference in meaning. Edwin Leigh's system had hairline type for silent letters, with the sounds superimposed by bolder type. The greatest advantage of Leigh's system is that it preserved the conventional spellings, yet gave a pronunciation guide to all words. In this respect, it was superior to all other systems before or since, and had the easiest transition to conventional spelling of any system ever proposed. It's a pity it isn't still in use.

You say there isn't agreement on the number of speech sounds. But there is agreement among the best qualified phoneticians, such as Pitman, Dewey, MacCarthy, Ives, Rondthaler, Tune and others. You cannot expect agreement with uninformed alfabeteers.)
In the alphabets presented in this thesis, there is little agreement as to the number of symbols needed. Benjamin Franklin eliminated six of the present symbols and added six new ones so his system contains the same number of symbols as the present alphabet. William Thornton used 30 symbols, the Initial Teaching Alphabet has 45 symbols, the International Phonetic Assoc. Alphabet contains over 48 symbols, and Edwin Leigh found that 70 symbols were needed to express his spelling system because he wanted to retain all present spellings, but superimposed on them was a phonetic system in bolder type.

Sir James Pitman expresses the opinion that between 35 and 41 different characters would be adequate to express all the basic consonant and vowel sounds without employing combinations of letters or digraphs. Obviously, an alphabet with as few as 35 symbols would be inadequate, and an alphabet of over 50 characters, while producing more accurate pronunciation, would run beyond the need for sounds that distinguish meaning. Altho many spelling reform proposals fit this 35 to 50 symbol requirement,, there is certainly a wide range of disagreement on just what symbols ought to be present in the new alphabet and what symbols should be excluded. Until there is some agreement, at least on the minimum number of symbols, proponents of augmented alphabets should not be surprised that their ideas are met with suspicion or outright rejection.

A second obstacle to the acceptance of augmented alphabets is the great difficulty of devising new letter shapes that harmonize with present shapes in upper and lower case, and in all fonts of type, as well as in handwriting. According to P.A.D. MacCarthy, "the new letters should remain sufficiently simple in their basic shape so as not to add much to the number of pen movements required to form them by hand." He also added that they should have good lateral extension.

A third objection to augmented alphabets lies in the acceptance of the number of new characters needed to make the alphabet completely phonetic. If only a few characters were needed, they would not cause too much relearning for the average reader, could be quickly assimilated into the present system and require minimal changes in our general overall system. Not two or three new symbols for English are needed, but more like 17 new ones are needed, or approximately 40% of the new alphabet. One has to want the change badly to accept these kinds of changes. As P.A.D. MacCarthy expressed it:
"To produce nearly a score (or more) of new shapes such as would commend themselves to expert and public alike, sufficiently powerful to secure wide approval and eventual official adoption - this has so far defeated all those calligraphers and type designers, whether amateur or professional, who have from time to time made an attempt."

(

Ed. comment:

 Not true. It's merely his opinion)
Another difficulty with a new phonemic alphabet for English would be to find a phonemic pattern acceptable to the various dialectal groups comprising the English speaking world. The system should be acceptable to the total English speaking world, not just mid-Western Americans, Bostonians or Welchmen. Since pronunciations differ, not only between English speaking countries, but within the countries themselves, the task of picking the correct symbols to represent certain sounds becomes a difficult one. Some words such as dance, because, father, been, going, etc., could be represented by different spellings according to the various dialectal and/or ideolectical differences among groups or individuals. Monson presents additional examples of this difficulty:

"Not all dialects of English have the same number of phonemes: In the words pants and dance, some people have different vowels, some do not; and the phoneme occurring in the dance of the first group does not exist as a separate phoneme in the speech of the second group. In some regions, starry, merry and Mary have three distinct vowels; in others, two, and in still others, the words are homonyms. A similar problem would occur in transcribing dog, fog, and frog. Thus a standard phonemic orthography acceptable to all different dialects of English would be difficult to devise."
(

Ed. comment:

 Not difficult to devise. It only requires that all 41 or 42 sounds be identified and symbolized. The difference comes with books printed in America (past) and in England (paast). But we already have printing differences, viz: colour, centre, petrol, lift, which in America are: color, center, gasoline, elevator; These differences are such a minor difficulty as to not be worth considering. Merely a 'red herring' intended to obscure the more important issues.)
But the most important obstacle is the need to change all typewriters, printers' fonts, remote teletype machines, and dictionaries; library indexing systems, and any system that uses the 26 letter Roman alphabet. Where are you going to insert these 17 new symbols into the Roman alphabet? Think of the disruption this will cause!

The final obstacle to acceptance of a new phonemic augmented alphabet lies in the degree of change and difficulties in getting people to switch to the new system. Aside from the fact that new printing fonts and new typewriters would be needed, with all the difficulty, confusion and expense this would involve, is the tremendous problem of educating the general public to the need for such changes. The educated and illiterates, the old and young, would need a large degree of re-education or instruction, and practice. Asking everyone to change deeply ingrained habits developed over many years is considered a difficult, if not impossible task by spelling reformers and spelling reform associations. Dewey concluded: "Because of the, enormous practical difficulties of making new characters everywhere available, such a notation is unlikely to achieve adoption for general use for several generations to come."

Problems in the use of Diacritical Markings.

At first glance, supplementing the present alphabet by use of various kinds of diacritical marks would seem to be a good answer to the problems of an inconsistent orthography. Many of the advantages claimed for using diacritical marks, however, break down upon closer examination.

It is only a half truth, for example, to say that diacritical marks remove the necessity for new letters to obtain a more exact sound-symbol relationship. One might well ask: is a bar, a dash, or some other mark over the a, to indicate length, a new letter or isn't it? Is it already in the printer's fonts? Some would agree, with good reason, that it is more difficult to remember to put a bar over the vowel letters than it would be to learn a completely new symbol that sound. It is a gratuitous assumption to suppose that it would be easier to learn a new marking system than to learn totally new symbols.

It is true that diacritical marks can save printing space, but it is also true that most typewriters and printing shops are not equipped to add these marks in the normal process of reproduction. Going back to add the necessary diacritical marks can become a nuisance in both typing and handwriting. The relatively small saving in printed space is more than offset by the loss of time required to go back and add diacritical marks and the resultant loss of continuity in the writing and thought process.

It is often difficult to teach the child to dot his i's and cross his t's. Imagine how much more difficult it would be to teach the child to go back and add three or four or more additional marks. There is a good chance that the kind of diacritical marking system that is required to make our alphabet more phonemic would cause as many, if not more, spelling mistakes than currently exist with the present alphabet.

Provided they are not too complicated, diacritical markings can be very helpful to lead the beginning reader to a more exact pronunciation of an unfamiliar word or help him to sound out the word. To this extent, diacritical marks seem to have a useful place. But as a system of permanent spelling reform, it lacks validity and to that extent is hardly acceptable as a means of bringing about a more consistent spelling system.

Spelling Reform Proposals using Digraphs.

An ideal spelling system would have one symbol for each phoneme, but English has only 26 letters and over 40 phonemes. Unless diacritical marks are used or new letters are added to the alphabet, it is impossible to get this optimal one-to-one relationship for English. The most common way to make up for this deficiency is to use digraphs - a combination of two letters to stand for one sound.

The technique employed by most spelling reformers who use digraphs to make up for the deficiencies of the alphabet is to establish a principle of consistency in their use of digraphs. This may be done in one of three ways. The most common method is to attempt to discover the common spelling digraphs used to represent a particular sound and then select one of these digraphs to represent consistently that sound. This method considers major spelling patterns and does not exclude alternative spellings of the same sound. The second method is less selective or systematic.

It consists of establishing digraphic spelling based in part on current spelling patterns and in part on other considerations. The third method is arbitrary and based primarily on the subjective desires or fantasies of its creators.

With almost all systems of digraphic spelling, the authors apply rules for eliminating silent letters. Again, the degree of change varies with each proposal depending on the needs and purposes of the particular system.

The general advantages of digraphic spelling over other systems of spelling change may be summarized as follows:

1) No new letters or diacritical marks are necessary. This makes it possible to use present methods of printing, typing and reproduction of printed material without addition al expense or inconvenience,

2) Much of the current spelling remains unchanged. This makes for greater ease of learning the new system for those who are already familiar with traditional spelling; and, makes the transition from the new to traditional for those who are not familiar with traditional spelling less confusing.

3) Most digraphs used in these proposals are currently in use or are familiar to the literate public. This means they do not clash with present familiar spellings and can be easily remembered and retained.

4) Digraphic spelling produces considerable continuity with the past and with other languages using the Roman alphabet.

5) Digraphic spelling initially produces less psychological and technological disturbance. For this reason alone it has a better chance of being accepted by the general public:

Back to the top.
On another page: part 2.