[Ian Martin: see Newsletters.]
Also on this page: Axel Wijk.

The 24th Annual Thailand TESOL International Conference
Khon Kaen, Thailand.
29-31 January, 2004.

Axel Wijk remembered: The problem of spelling.

Presentation by Ian Martin.

AUA Thonburi Language Center.

Axel Wijk (1902-1979) was a Swedish linguist who proposed a reform of English spelling: Regularized English (1959). His work remains of continuing relevance and usefulness especially the light of recent studies such as those on reading by Seymour (2001) and spelling by Spencer (2000). Both reveal the learning difficulties children experience because of English spelling.

Wijk gave courses on English pronunciation at the University of Stockholm. In these courses he paid a great deal of attention to the intricate question of the relationship between spelling and pronunciation. His doctoral dissertation was on Henry Machyn, the 16th century London diarist. Thanks to a very detailed statistical investigations of Machyn's curious spellings he was able to prove that the Diary was not written in the speech of London, Cockney, at all but instead in a more or less pure South-East Yorkshire dialect. It was the methods used in investigations of this kind that enabled him to devise his spelling reform proposal.

Wijk's interest in spelling reform was sparked by the difficulties that his 6 year old son and the son's classmates had in learning to read English. His son attended a public school in New York in 1950 at the time Wijk was a lecturer in Swedish at Columbia University. His son was taught only 58 different words by the whole word method, without any reference to the connection between individual letters and sounds. The rate of progress contrasted with his 7 year old daughter who, three years later in her first year in Sweden learned 1,900 different Swedish words by the phonic method. Wijk's observations of the difficulties children experience when learning to read English are borne out by Seymour's conclusions.

Wijk began to investigate spelling reform and the many proposals. He came to reject the idea that spelling should be purely phonetic i.e. one symbol for one sound. In his analysis of the present written form of English, however, he states the language is far too difficult to learn for the great majority of foreigners. For foreign students who generally learn the language from books and not so much from hearing it, pronunciation can constitute a major difficulty since the confused, irregular spelling offers such poor guidance to its pronunciation. There is hardly a letter or a combination of two or three letters in the alphabet that cannot be pronounced in two or three different ways, and a good many of them actually have from half a dozen to a dozen different pronunciations. Moreover we often find a great many different spellings for one and the same sound especially in the case of vowel sounds. Wijk states there are 46 different speech sounds (phonemes), 21 vowels and diphthongs and 25 consonants in British Standard English. He identified 60 symbols that are normally used to represent the 21 vowels and diphthong sounds in the written language. The 25 consonant sounds are normally represented by 44 symbols. Many of these occur double. All consonant symbols can occur as silent letters in some words. If we add up the vowel and consonant symbols, we find that apart from the double consonants, the 46 sounds of the spoken language are normally represented by 102 symbols in the written language (60 plus 44 minus 2, since u and y stand for both vowel and consonants sounds, as in: cup, persuade, pity, yes.)

Wijk noted that having many varying spellings for one and the same sound is not necessarily a defect. It enables the spelling of the large number of homophones which is a feature of English. However within this category of irregularity is a large number of anomalous spellings. These can be divided into exceptional irregularities, deviations from general rules that are only found in one or a few words and certain larger or smaller groups of words which all display the same type of deviation with regard to spelling and pronunciation.

The general impression of excessive irregularity in English spelling is very largely due to the fact that so many of the irregular words are among the commonest in the language. This irregularity obviously makes it difficult to learn to read and write.

Wijk proposed a reformed spelling based on the systematic preservation of all the various sound symbols of the present orthography in their regular, i.e. in their most frequent usage.

In dealing with the so-called long sounds of the five simple vowels, the greatest problem of any would be spelling reformer, Wijk considered with stressed syllables, the present spelling offers sufficient guidance as to the pronunciation when supplemented by certain not very complicated rules. For pure vowels and diphthongs he restricted each particular symbols to denote one sound only and to change the spellings of those words which show deviation from this pronunciation into one of the regular spellings for the sound in question.

The various consonantal sounds being a stable element in the English sound system are nearly always preserved in Regularized English.

Wijk hoped his reform could be implemented through being taught in the schools. He devised a method of teaching it. He also hoped it could be used experimentally in the teaching of reading.

Wijk's reform proposal was generally well received but it has shared the fate of all spelling reform proposals, lack of interest and the broad change in perception about the non-phonemic spellings of English: "English orthography, despite its often cited inconsistencies, comes remarkably close to being an optimal orthographic system for English" (Noam Chomsky and Morris Halle 1968).

In conclusion, mention should be made of the criticism leveled at Wijk by Taylor (1981). The main reason for reform, as Wijk argues, is that traditional English spelling makes it difficult , for natives and non-natives alike, to learn to read and write, so that English-speaking children take 'from one to two years longer' to learn to read and write their language. Taylor says there does not appear to be any evidence to support such a proposition. The research of Seymour and Spencer tells us otherwise. Both produce evidence of the difficulties of the English orthography. Taylor calls for a reform of the teaching of spelling. The use of spelling reforms proposals as initial teaching alphabets and dictionary pronunciation guides may well be part of such a reform!


Chomsky, N. and Halle, M., The Sound Pattern of English. 1968.

Seymour P. How do children learn to read? Is English more difficult than other languages? BA Festival of Science Glasgow, 2001. [See Media, Bulletins.]

Spencer, K. Is English a Dyslexic Language. Dyslexia 2000; Vol. 6, No. 2. [See Journal, Newsletter, Media.]

Taylor, D. English Spelling: A Help Rather Than A Hindrance. English Language Teaching 35, 3:316-21 (April 1981).

Wijk, Axel. Regularized English. 1959 and 1977. [See Bulletins.]

Wijk, Axel. Rules of pronunciation for the English Language. Oxford University Press, 1966.

Presentation given at the 21st Thailand TESOL International Conference.
Bangkok 18-20th January 2001.

The work of Daniel Jones:
phonetics, spelling reform and the classroom.

Ian Martin.

AUA Thonburi.

Daniel Jones was the leading British phonetician in the first half of the twentieth century. He developed the concept of Cardinal Vowels, "reference" vowels for phonetic description and transcription.

Much of his work still influences phonetics and language teaching today. The term 'phoneme' is at the heart of phonetics. It was Jones who did much to describe its nature and use. Its definition and the question of what is a phoneme has been a source of perhaps the biggest disputes in the field of linguistics. However, the value of this notion as claimed by Jones works 'well in practical language study'. Indeed his main preoccupations were with practical language teaching and the elaboration of transcriptions and orthographies.

This is evident in his book The Pronunciation of English (Jones 1956) and his famous English Pronouncing Dictionary (Jones 1967), both aimed at the language learner. The dictionary provided a description of Received Pronunciation, the standard British accent, which was also useful for those interested in an analysis of the English language. The notation used in the dictionary was a modified form of the International Phonetic Alphabet.

By referring to the dictionary the complexities and absurdities of English spelling come to light. The 40-odd sounds of English can be spelled in hundreds of ways, and one spelling can represent many sounds. Simple words like once, who defy all logic.

The same ending has different spellings in burglar, teacher, actor, glamour, acre, murmur, injure, martyr. The same stem varies in high/height, speak/speech, precede/proceed, defence/defensive (U.S. defense is consistent). The endings -ant/-ent, -able/-ible switch bewilderingly. How can we tell which of afraid/affray, inoculate/innocuous, omit/commit, have double consonants? Letters are inserted for no good reason, eg. C in scythe, G in foreign, P in receipt, S in island. Foreign words are altered: -ance/-ence are reversed from French correspondance/connivence, and Spanish M, RR often become MM, R in English incommunicado, guerilla. Several forms compete in loanwords like borshch, lychee, popadum,yoghurt.

Much greater mismatch between sound and spelling can be found in the spelling of names and place names, many examples of which can be found in the English Pronouncing Dictionary. Inconsistency is rife because English has no strategy for ensuring consistency. No other language tolerates such alphabetic chaos.

The reform of English spelling was a lifelong interest for Daniel Jones, who was for many years actively concerned in the affairs or the Simplified Spelling Society. The society was formed in 1908. In 1911 Jones became a committee member. In 1946 he became President of the society.

The society decided in 1937 to re-write the book Proposals for a Simplified Spelling of the English Language by Ripman and Archer. It was renamed New Spelling (Ripman & Archer 1948) and was adopted by the society as a proposal for spelling reform.

New Spelling is based on the principle of having one letter for each distinctive sound of English using the existing roman alphabet. In order to compensate for the absence of the extra letters of the complete 40 or so sounds, a system of digraphs (sequences of two letters denoting single sounds) is employed.

An example of New Spelling is as follows:
We instinktivly shrink from eny chaenj in whot iz familiar and whot kan be mor familyar dhan dhe form ov wurdz dhat we hav seen and riten mor tiemz dhan we kan posibly estimate?
Jones considered New Spelling to be a remarkably good solution to the problems of English spelling considering the limitations the Simplified Spelling Society imposed upon itself. But since his time little progress has been made in implementing spelling reform.

Back to the top.