[Spelling Reform Anthology §12.9 pp178-180]
[Spelling Progress Bulletin Fall 1978 p16-18]
[Walter Gassner: see Bulletins.]

The Essential Requirements for a Reformed Spelling,

by Walter Gassner, L.L.D.*

*Randwick, Australia.
*Presented at the First International Conference of the British Simplified Spelling Society, London, Aug. 1975.

The purpose of this paper is to present, in order of importance, various aspects of spelling reform which would provide a maximum of benefits to future users and entail a minimum of inconvenience to those accustomed to the traditional spelling system. It is designed for the English language but the principles stated also hold for other languages.

(1) Only the 26 letters of the Roman alphabet should be used

 - with no augmentations and no diacritical marks which are not in use at present.

The introduction of a new system of spelling will, of necessity, impose a burden on everyone who uses the English language, no matter whether English is his native language or a language that he has acquired. To this burden, no further burden should be added, and a weighty additional burden it would be if it should become necessary to learn new letters or to acquire the habit of putting written accent signs over letters. Indeed, if it were not for these considerations, the most reasonable thing to do would be to adopt the International Phonetic Alphabet, which is widely used for teaching the pronunciation of a large number of languages and is known to scholars throughout the world. This alphabet contains, in addition to the letters of the Roman alphabet, a number of specially designed or adopted signs to provide a sign for each sound (or, to be precise, for each phoneme, a phoneme often encompassing a family of similar sounds whose exact nature in each instance is determined by its phonetic environment ).

The deficiencies of the Roman alphabet can, however, also be overcome by continuing the practice of using groups of two letters - so called "compound symbols" or digraphs - for certain sounds, such as sh for the initial sound in ship, or th and ng for the initial sound and the final sound in thing. The use of compound symbols does not contradict the phonetic principle as long as these symbols are used in a consistent manner and as long as a separation mark (say, an apostrophe) is inserted where letters that usually make up a compound symbol are pronounced separately - e.g. as in mishap.

(2) Writing should be so designed that speech can be deduced from it

 - or rather, so that there can never arise any doubt regarding the pronunciation of a word.

The emphasis is on "word" -, for there are limits to the things which an orthography can indicate. An orthography cannot, for example, do justice to sentence stress or allow for alternatives for certain short words usually referred to as "weak forms." Every other aspect of the pronunciation should, however, be clearly indicated, and thus it becomes necessary to have available an appropriate symbol for the Neutral Vowel (i.e., the initial sound in about) and two distinct sets of symbols for the majority of vowel sounds, so as to make it clear which syllable is stressed and which syllables are unstressed. The last point is extremely important, for a word wrongly stressed is a word wrongly pronounced - and indeed, often misunderstood.

If this principle is applied, the difficulties of those who have to deduce the spoken form from the written form will disappear. Those who will benefit most from the implementation of this principle will be foreign learners who study from textbooks. Such persons, once they have mastered the significance of the symbols, would be in a position to read any text - even if they do not understand it - almost faultlessly, the only shortcoming being that their speech might appear slightly pedantic.

Foreign learners would of course, not be the only persons to benefit from this principle. Native English speakers, too, including highly educated ones, are at present, frequently in doubt as regards the pronunciation of a word - say, a scholarly or a technical word, or a proper name or just a word they happen to not have heard before. To clarify these cases, pronouncing dictionaries are available, and the entries in such reference works should supply the basis for the written forms under the reformed system. Obviously, once the new system is implemented, pronouncing dictionaries will no longer be needed.

(3) Nothing should be introduced that would impair clarity by obliterating the individuality of words.

 Hence, provision should be made for the distinction of homophones which are distinguished at present.

Opponents of phonetic spelling claim that clarity would suffer if homophones - words with the same pronunciation, but different meanings - were spelt alike; they might use the existence of homophones to discredit phonetic spelling and to justify the traditional system with all its inconsistencies and archaisms. Whilst their other arguments can be ascribed to prejudice, this particular argument is a sound one. For, if homophones were merged, there would be no way of distinguishing a "mail clerk" from a "male clerk", or the "fore legs" of an animal from all its "four legs." The passage in Shakespeare's Hamlet: "I scent the morning air" would have the same written form as "I sent the mourning heir"; and, as there is, indeed, a "mourning heir" in the play, such an interpretation is by no means strained.

The author of this paper has devised a system of spelling in which provision is made for the distinction of homophones wherever such a distinction is considered desirable. He cannot agree with the argument which alleges that since spoken language can afford to use like forms for words with different meanings, written language can do so likewise. The answer to that argument is: in spoken language there exist numerous ways of making the meaning of an ambiguous word clear, such as differences in intonation, pauses, gestures - all of them devices which written language lacks. Moreover, through habit we instinctively make a choice when hearing a word that allows for alternative interpretations, and this choice would usually be the right one, no matter whether we are aware of having made a choice or of the very existence of homophones. For written language, this habit of making a choice does not exist and such a habit would take a long time to develop. Even where there is no danger of an actual misunderstanding, there is still the possibility of a ludicrous effect - for instance, if "reign" were spelt the same way as "rain." On the other hand, the fact that in spoken language homophones do not create a ludicrous effect unless such an effect is intentionally brought about in a pun, demonstrates that what is tolerable in spoken language is not necessarily so in written language.

(4) Writing should be made so that it can be deduced from speech

 - subject to the limitations imposed by the alphabet and the need to distinguish homonyms.

This principle is complementary to the principle stated in the second place. The beneficiaries of it are those who know the pronunciation and are supposed to deduce the written form from the sound. They are chiefly native English speakers, schoolchildren in particular, but of course, also adults, and by no means only illiterates or semi-literates; foreigners only to the extent that they have "picked up" the language without at the same time learning how to write it.

In the minds of most people who give some thought to spelling reform, this principle should rank first. That it has fourth place, not first, is of course, due to the limitations stated above. If it were not for these limitations and for uncertainty regarding the standard pronunciation where the speaker's pronunciation is a regional variant - there would never be a need to memorize the spelling of a word. Under the author's proposed system, the spelling of some words would have to be memorized, but the proportion of such words would be small, and learning to write would no longer be the drudgery that it is today, because in every instance there will be a reasonable explanation for an exceptional written form.

It goes without saying that the exceptional written form would in no instance suggest a pronunciation other than the correct one; indeed, the device for clarifying the stress position - availability of two sets of symbols for vowels - is also the device for providing distinct forms for homophones.

(5) Continuity with the traditional system should not needlessly be disturbed.


The choice of symbols should, whenever compatible with the basic principles of phonetic representation, follow current English usage, not only for single letters, but also for letter groups such as sh, ee, ai, oa. This is proposed not in deference to tradition, but with the sole motive of making transition from one system to the other as easy and as smooth as possible and to allow for the probability that for a long time to come documents written under the traditional system will call for an interpretation, so that many persons will be required to have a reading knowledge of the system at present in use, which will then have become an obsolete one.

The occurence of a particular written form for one word in the new system which is identical with the written form of another word in the old system, is particularly objectionable. Whilst such a situation cannot entirely be avoided (for example, warm will have to be spelt worm), the symbols have been selected in such a way that cases of this nature have been reduced to the unavoidable minimum.

It is not an unwarranted disturbance of continuity if letters, which otherwise would be redundant, are allotted to sounds for which there is no suitable way of representation under the traditional system.

(6) The existence of variant pronunciations,

 to the extent that they are pronunciations used by educated speakers, should be taken into account.

The new system should be suitable for use throughout the entire English speaking world, despite differences in pronunciation. There a particular sound used in one speech variant is, in some other speech variant, consistently replaced by some other sound, the key-word given will automatically be variously interpreted, but the alloted symbol will be applied in exactly the same manner by the different types of speakers. Thus, if a symbol (no matter which) is alloted to the "a as in paper," this description would appear to speakers of Southern British and many other types of English as referring to a certain diphthong (phonetically ei), whilst to Scottish speakers it would appear to refer to a certain long vowel (phonetically e:), and to some Australian speakers as referring to a different diphthong, similar to but not identical with the i in time. These Australians think they are saying long-A, and would not confuse it with the diphthong i in time.

Where sounds overlap, the system should, in principle, be based on Southern British speech (Received Standard) as recorded in pronouncing dictionaries that use the International Phonetic Alphabet, such as that of Prof. Daniel Jones, but other pronunciations should be taken into account in cases where a great many speakers make a distinction which is not made by speakers of Southern British and which is supported by the written form of the traditional spelling.

This is so with certain vowels for which cases "with r" and "without r" are being kept distinct from each other, the cases "with r" being those where the letter r-silent in Southern British speech, but pronounced in Scottish speech and phonetically significant in American speech - occurs before a consonant or at the end of a word, as in farm and farther, as contrasted with the word father, which has the same sound "without r." In Rec'd Standard, farther and father are homophones.

In situations which do not come under any of the above descriptions, alternative written forms should be permissible, as with the words now written either and clerk. The word clerk should be written klark in England and klirk in the United States. Such written variants do not constitute a threat to uniformity any more than the use of different words for the same concept, such as autumn and petrol in England and fall and gasoline in the United States.

There are also differences in pronunciation depending on the style used. Pronouncing dictionaries often give more than one pronunciation of a word. In such cases, the written form should reflect the most careful variant - assuming that though only a small proportion of speakers actually use it, most speakers would have an opportunity of hearing it. However, spoken forms that exist only in the minds of those who allow themselves to be influenced by the written form under the traditional spelling should be ruled out. This would apply chiefly to unstressed vowels.

(7) The system should be economical

 - achieving a saving of time and space.

This issue is not a fundamental one. Elimination of silent letters and of needlessly doubled letters will evidently bring about the shortening of a large number of words. The lengthening of certain other words should not detract from the overall effect, since cases in which a word is shortened occur more frequently than cases in which a word is lengthened.

(8) The suggestion that spelling should reflect etymology is to be rejected.

The ordinary user of words (reader or writer) is not interested in etymology and cannot be expected to be an expert on it, and the scholar has other means at his disposal. Spelling is often incorrect in indicating the true origin of words, as these who were involved, along with Dr. Samuel Johnson, in finalizing the spelling were often badly informed of the origin of some words.

However, the choice of symbols for the sounds on the principle of continuity and preference given to a "more careful" pronunciation where several variants exist might result in preserving etymology in some instances.

(9) Some observations relating to other languages might have a certain relevancy.

Whilst, in theory, it might appear desirable to have the same orthography for all languages, limitations of the Roman alphabet and the need to preserve continuity with traditional usage rule out any suggestion of devising reformed English spelling in such a way that letters are given particular values solely on the strength of compliance with usage in other languages or international phonetic practice. Thus, it goes without saying that the letters a and u must, in English, be alloted to sounds which are quite different from the sounds which these letters represent in most other languages. It stands to reason that similar considerations also apply to compound symbols, such as au and eu.

If thought is given to spelling reform in languages other than English, the principles applied should, broadly speaking, be the same as those applicable to English, but different approaches are needed for each language to do justice to different situations. Such diacritical marks as are used at present, e.g., the acute, grave, and circumflex accents in French and the umlaut sign in German, would continue to be used (and, of course, like everything else, in a consistent way), but no new written accent signs should be introduced. Stress presents no problem in French, but there are other factors that complicate the issue, such as the large number of instances in which up to 5 or 6 words have the same sound, and the words with consonants at the end which are normally silent, but pronounced when the next word begins with a vowel. The stress problem is satisfactorily solved in Spanish and Portuguese, but not completely in Italian; it leaves a lot to be desired in German, and still more so in Russian, where there is the additional complication that in an unstressed position, certain vowels merge and become obscured.

Spelling reforms on a minor scale have been carried out in the last few decades in a number of languages such as Russian, Polish, German, Dutch, Norwegian, Portuguese, Malay and Indonesian (both of which derive from the same stock, & have now the same system of writing). These reforms were usually designed to take into consideration the needs of native speakers, but not those of foreign learners, and were needlessly concerned with etymology. (Thus, in German, th has been replaced with t in native German words, but has been retained in foreign loan words).

For the Mandarin dialect of Chinese, now the National language of China, a romanized orthography has been devised, which has received official sanction. This orthography has not superseded the traditional ideographic notation. It is merely used as a means to teach the National Language with the correct pronunciation to speakers of other dialects and to overcome illiteracy, or rather as a first step to do so. Now, there is the fact that Chinese abounds in homophones - to a far greater extent than English or French-and by using the ideographic notation, the meaning of a word is always made clear, whilst this would not be the case if the romanized form were used. It is this aspect which explains why it must have appeared unreasonable to abolish the ideographs altogether; and since the situation with regard to homophones allows an analogy with English, these observations should strengthen the argument in favor of maintaining a distinction of homophones in English.

(10) Some thought should be given to the idea of introducing reformed spelling gradually.

The concept of a gradual approach is championed by W. Harry Lindgren of Narrahbundah, Australia. He proposes that the first step (termed SR-1, short for Spelling Reform Step 1) should encompass one single sound: the short vowel sound in "bet." This sound would invariably be rendered by the letter e -thus frend, helth, eny, meny, sed, etc.

Campaigning for this modest reform is supposed to create awareness of the desirability of a more logical system of spelling, without interfering to any sizeable extent with established usage. A decision on subsequent steps is left in abeyance and is not supposed to be made a subject of discussion until SR-1 is well established.

The author of this paper is not very much in favor of the gradual approach, but is prepared to concede that such an approach might be the only way to get something done. It might be feasible, first to introduce SR-1 in all English-speaking countries, and after this step has been achieved, to proceed further. To be sure, future steps will, on each occasion, have to cover a more extensive field, otherwise implementation of the ultimate reform would take a century.

A succession of steps, numbering more than, say 4, is bound to create confusion, not only because at any point of time documents would come into existence which have to be meaningful for future generations, but also because after each step dictionaries and reference works would have to be revised and reprinted. A succession of a large number of steps would cause a strain on schoolchildren, on the teaching profession and, of course, on the general public, not to speak of the confusion it would create in countries in which English is taught as a foreign language, with the result that even among those teachers and students who are in favor of phonetic spelling, many would give up in despair and revert to the system now in use, pending total implementation of the ultimate reform.

If an opinion regarding a gradual introduction of reformed spelling is required, it would appear that four steps should make up the maximum needed. The steps following


 as described above would be the following:


 covering all the consonants. The letters chiefly affected are k and s (replacing c), i (replacing g), z (replacing s), and the groups sh, hw and zh.


 covering the remaining short vowels a, i, o, and u, and abolishing silent letters that serve no useful purpose. This would produce such written forms as prity (for pretty), gluv (for glove), wosh (for wash).

It stands to reason to assume that disagreement among spelling reformers on any of these points would be unlikely.


 would be the reform that produces the ultimate shape and would encompass the long vowels and diphthongs, the Neutral Vowel and a device to indicate the position of the stress. This step can evidently be taken only when agreement has been reached among spelling reformers, and it should be the task of interested persons and organizations in all English speaking countries, as well as certain instrumentalities of the United Nations Organization to prepare the ground for such an agreement.

Back to the top.