[Journal of the Simplified Spelling Society, 1994/2 pp13-20 later designated J17]
[See Journal, Newsletter, Anthology, Bulletin articles, Personal View 10 and website by Valerie Yule.]

Spelling and Society: Orthografy and Reading.

Summary of a research thesis [1].

Valerie Yule, Monash University, Australia.


This doctoral thesis (1991) sought to provide a comprehensiv account of the grounds for spelling improvement, to open a wide field for empirical research, and to back this up by a network of theory and research. Twenty six experiments, small-scale studies and other original investigations filled in apparent gaps and tested a specific spelling modification, the deletion of surplus letters in words, that was supported by the theory and evidence presented. The following is a reorganized and updated summary of content and conclusions. [2]


Britons and Americans have been leaders in devising hundreds of new writing systems that are now used by previously illiterat peoples - but they have not put their own spelling in order. Language planning is a thriving field for national efforts world-wide - but not for the written English language. Almost every major language in the world has made major or minor improvements in its writing system over the past century - including Spanish, Chinese, Russian, Japanese, German, Dutch, Turkish, Portuguese, Korean, Hebrew, Malaysian, Indonesian, Norwegian, Czech, and even French - but not English (except in the United States, 1898-1920, ed.).

One reason for this is that English spelling has got itself into such a mess that rationalising it has seemed too difficult. Intending reformers have usually regarded a clean sweep to some form of direct fonemic sound-symbol correspondence as the only way to go. Indeed, that is Crystal's definition of spelling reform in the Cambridge Encyclopedia of Language (1987).

Empirical research in spelling improvement is therefor necessary. All assumptions must be checkd, whether tacit or explicit. Until recently, most attempts at English spelling reform have been conductd from armchairs or the equivalent of barricades, with fussilades of arguments and new systems or lists of revisions presented as à priori constructions demanding implementation. Altho children have at times been taught to read successfully by various fonemic spelling scemes, there has been no mainstream reserch that compared these scemes or explord other issues involvd in improving a writing system. The ERIC database of 1982-1993 listed 32,293 articles on READING and 2,172 on SPELLING, but it reported only 12 articles on SPELLING REFORM, and of these, only Beech (1983) and Citron (1981) were positive All other spelling reform reserch during that period has been effectivly censord out from future researchers simply by being omitd. The Modern Language Abstracts data-base 1981-92 listd nine items on spelling reform - all on reform in other languages except for one on the Renaissance and an articl of mine in a Spanish journal.

Only recently has reserch begun to investigate the question of what might be optimum to meet the real needs and abilities of users of the English writing system. The Harvard Educational Review was one of the first major lerned journals to take this question seriously without the aim of dismissing it (Yule, 1986), and an empirical comparison of responses to different types of spelling reform was publishd in Human Learning in the same year (Yule & Greentree). The topic had a hard time in 1987 to gain acceptance in a University as a respectabl field of study for a thesis. But now a mainstream international literacy journal, the Journal of Reserch in Reading (1992) has broken ground in publishing a reserch article written thruout in a reformed spelling (Christopher Upward's pioneering experiment, 'Is traditionl Inglish spelng mor dificit than jermin?'), showing that a door is now open, and progress is possible.

PART I. An overview of writing systems.

The variety of possibilities for the relationship of written and spoken language.

The first step of the argument of the thesis is a cross-cultural review of writing systems, how they relate to the languages that they represent and how they meet the needs and abilities of users. It describes how and why writing systems other than English have changed over time, the role of planned interventions and the conditions for national reforms. Other contemporary orthografies beside English have their own ways in which to be difficult for learners and users, but most are relatively user-frendly. Other countries also attempt and often succeed in deliberat orthografic improvement, with trends towards simplification and consistency. Cross-disciplinary data can be reanalysed from the perspective of its relevance to English.

The four major possibilities so far inventd for writing to communicate language are 'logografic', syllabic, alfabetic, and mixtures and messes of these possibilities. The messes result from language change over time, and the accidents and expediencies that accrue. The language may be living, but the writing system may be deadening.

Logografic writing systems (characters represent words) and alfabetic writing systems (characters represent the speech-sounds of a language) both have advantages and disadvantages. They are compared and contrastd because they bear on the 'Whole- Word' and 'Fonics' controversies in English education which are currently unresolved, but have significant implications for the direction of spelling reform.

Chinese is the outstanding exampl of writing that is logografic, or ideografic (characters represent ideas). Since it does not represent speech sounds, written Chinese was able to unite an empire of many languages. It is particularly suitd to its original mandarin spoken language, which is made up of around 400 monosyllabls, with multipl homofones and distinctions of up to five tones, making it difficult to represent by any alfabetic or syllabic principl. Lerning and recognition is difficult but not impossible becaus the characters are not entirely arbitrary: they have components of radicals and significs, made up of simpl and compound ideograms with semantic associations, pictograms and rebuses that are often now unidentifiable and some fonetic clues.

However, acquiring full mastery of from 40,000 to 70,000 complex characters was always considerdd work for a lifetime scholar. It is a writing system for an elite. All efforts at simplification down to a bare minimum of around 4,000 characters, and the intensiv literacy campaigns undertaken by Communist China, faild to achieve universal literacy. Now roman alfabetic 'pin-yin' is commonly used to introduce beginners to reading.

The alfabetic principl is that each character should represent a speech sound, as in Latin, but this is rarely found intact in any living language, since spoken languages change over time and place, accidents and expedients occur in the writing of it, and other aspects such as morfemes and gramatical markers may assume importance. The great advantage is that most languages are made up of only 15 to 40 speech-sounds to be represented by letters, so that these should be as easy as ABC to lern, and it should be easy for readers and writers to work out what any unfamiliar combination of letters should represent. All you have to do is to segment the artificial speech-sounds from the stream of speech - and this skill can be taught, even if it is usualy not.

Writing systems in which each character represents a syllabl, as in many Indian languages, are probably the easiest of all, since the spoken syllabls in words are easy to segment. However, they are only suitabl for languages with a limitd range of syllabls, like Japanese, where beginners need lern only 48 hiragana characters, increasd to 71 by diacritics. English however, would need hundreds, if not thousands of characters to represent its extensive range of syflabls - around 70 in this paragraf alone.

Writing systems in society.

The nature of a writing system is bound up with the nature of a society. Orthografies have roles in society as bearers of literacy, with functions in maintenance or change of social structure. Orthografic change is usualy related to social change.

When a writing system is very difficult, it can be the property and governing instrument of a ruling elite, or the secret of the priests who alone have the opportunities for the scholarship to master them. (Hieroglyfics = sacred carvings.) Indeed when the ingeniously simple Korean alfabet was inventd in the 15th century, the Chinese-cultured Korean mandarins suppressd it as soon as possibl on the express grounds that its simplicity could enable ordinary peple to read and write, whereas the Chinese script was their monopoly. It was the Korean court ladies, debarrd from education, who maintained hangul for their own use among themselves over the following centuries.

The invention of the alfabet in the Middle East made possibl the trading success of the Phonecians and the transmission of Greek culture, becaus anyone with a little effort could lern from 20 to 40 rudimentary characters to represent the speech sounds of their language. The alfabet has made democracy possible. Where, however, over time the spelling system no longer represents the spoken language in any consistent way, then universal literacy is no longer an easy matter.

Two apparently very successful orthografic solutions, Japanese and Korean, are mixtures. Foreigners tend to call Japanese the 'devils own language', and the Japanese put intense and determind effort into lerning the written form - but then can often read and write it extremely fast. A single sentence of Japanese could in theory contain up to six writing systems - Chinese characters or kanji for the key vocabulary, the 'syllabic' hiragana that children learn first that carries the Japanese grammar, furigana a little fonetic key to give clues for some kanji characters, arabic numerals, and for imported vocabulary either syllabic katakana or the roman alfabet romaji. It is possibl that so much visual distinctivness between types of language may aid fast access to meaning.

Korean is more systematic and simpl, but also ingenious. It shows what can be done when a writing system is purpose-designd rather than developd over a period. It is constructd of symbols which represent speech sounds, which in turn are combined into the semblance of gestalt-like characters which represent syllabls, giving it the visual advantages of logografs rather than the visual disadvantage of linear strings of letters, while it directly represents the spoken language. Hangul (the 'Great Letters') can be used to represent the Korean grammar embedding traditional Chinese characters, but today is more commonly the complete script.

New orthografies designd for existing and inventd languages are also relevant to English spelling design. A fifth of the world's languages have writing systems that were only inventd in the last 50 years or so. These show the fruits of experience and experiment, as well as how practical and social issues thwart ideals of theoretical perfection. They also scotch the popular myth that orthografies have to develop mysteriously and organicly without human intervention (e.g. Henderson, 1986: 68). If English spelling had been inventd this century, it would he very different from what has developd over the past thousand years.

How orthografies change.

Writing systems reflect the changing cultures in which they operate, and fill strategic functions in societies. There has been a tremendous variety of challenges and responses to the pressures to reform. Today's changes to writing systems reflect modern trends to simplification and greater efficiency and accessibility. Some specific national reforms are described, to illustrate themes and theories that counter Anglo-Saxon orthografic parochialism.

Reasons for failures make a wry comment on the chequered career of English reform campaigns. Some exampls are the continuing German controversies over whether it is important to capitalise nouns, the Danish dilemma over preserving their very small language area, French linguistic xenofobia, Hebrew and Arabic reverent attitudes to their sacred scripts, and the unfortunat Indian delusion that if the roman alfabet were used as a shared alternativ script for their many languages with their many regional scripts, it would have to be as difficult as English spelling, which is their only experience of the roman alfabet in action.

Successful reforms carried thru by stable governments include Finnish, Norwegian, Dutch, Afrikaans, Greenlandic, and the intercontinental orthografic reforms of Spanish and Portuguese.

The most radical reforms have usualy been among the first actions of revolutionary or radicly reforming new governments, as in Turkey, Russia, China, and postwar Indonesia, Malaysia, Israel, and Korea. It is in fact strange that spelling reform has never been on the agenda of the Left in English-speaking countries, when in other lands so much priority has been given by idealists and seekers of justice to reform of the writing system.

PART II. English spelling.

Part II examins the nature of English spelling, its basic structure and deviations from that structure, its adequacy as a vehicl for the English language, and current theories that describe it. English spelling has advantages, but also disadvantages that could be reduced. Trends to chanje reflect and are caused by social chanje and the chanjing functions of spelling in society. [3]

An alfabetic writing system suits the English languaj, which idealy would require only about 40 symbols, including around 19 for vowels. However, it has to make do with 26 roman letters, with only five for vowels, and the uncoordinated expedients that have been resorted to to get around this have added to the other impositions of history. Other considerations investigated in detail for an 'optimum' English spelling include the unreliability of etymology, the limits of Chomsky's concept of lexical structure, the morfemic nature of the languaj, the place of grammaticl markers, the importance of backwards compatibility with our heritaj of print, and the representation of homofones. (There are 11 homografic homofones in this paragraf, compared to 5 heterografic homofones.)

Theories of English spelling are mainly attempts at classified description. The theory of orthografy that is proposed regards English spelling as a basically diafonic conventionalised representation of the sound system of the languaj, modified by morfemic principts to represent the structural principls of English, but currently obscured by inconsistencies and elaborations. English spelling improvement may require only clearing up the inconsistencies and what is in effect clutter, from that basic structure.

English spelling has chanjed since Johnson's standardisation in 1755, which he tried to base on etymology to bypass the dialect problem. The chanjes have reflected chanjes in society, and in the function of education to screen out or to raise up.

In the 18th century, the complexity and elaboration of English spelling were valued, becaus these made it an instant screening test for social Gentility. The sociologist Veblen (1899) described it as a striking exampl of the 'conspicuous consumption' of the time. Victorian England valued dificult spelling as a vocational screening test for intelijence and diligence. Today an accessibl spelling is needed for an eficient society mor than a lexical barrier to protect an hierarchical one.

There have been dictionry as well as popular chanjes which follow the tecnolojici trend towards greater user-frendliness. When dictionaries list alternativ spellings, the simpler versions tend to gain popular support (e.g. lacquey to lackey). Seventy per cent of the spelling chanjes in commercial spellings are shortenings, and importd vocabulary is now often given simpler spellings. Even the spelling reformers achievd some chanjes, as in the 'American spellings' that have been adoptd from the lists of Noah Webster and the 1905 proposals of the Fonemic Spelling Council.

Today's chalenj.

The great importance of English spelling today is that it can promote or destroy English as the international languaj of the world. The poor link between the spoken and written forms risks losing it the universality that the Empire gaind for it. Even the most pragmatic rising nationalisms often reject the continued use of English because the difficulties of the writing system undercut the advantages of English languaj for education, technology and international communication. For forin lerners, English is a dubl languaj, with its relativly easy spoken form sabotaged by its very dificult written form.

An international angl on how English spelling could chanje is provided by the spellings of 'International English', as in Franglais, Japlish, Russlish, and so on, in loan words from and into other languages, and in the new pijin-English orthografies. Papua New Guinea has chanjed from English to Tok Pisin as the official languaj to unite its 900 or so tribal languages mainly because the spelling of pijin-English is easy to lern and read and write. The practice principls of pijin English spellings could be usefuly considerd in the design of English spelling.

In the 20th century, universal literacy is necessary for a democratic and tecnolojicl society; the most powerful forces against it are economic and political interests that want dumb passiv consumer fodder. The direction of our interests in spelling improvement can follow this century's trends for simplicity, streamlining, and eficiency.

PART III Needs and abilities of the users of spelling [4]

This section investigates and refutes the educationaly-popular claim that lernrs and readrs do not use spelling becaus eficient readers operate on context and whole words - so that improving spelling wil not help them. 'Orthografy' and 'Spelling' may not even appear in the index of books about reading theories and teaching. A vigorus critiqe is made of Frank Smith's wilder and most popular claims (e.g. Smith, 1982) about the danjers of fonics and the value of gessing or asking others in ordr to identify a word (e.g. Weckert, 1994). Theory and reserch is revewd to show that the needs and abilities of readers and lerners must be considerd in orthografic design.

Skilld readrs

 are shown to recognise most words with visual automaticity, but they use many strategies in reading text, including context and prediction, with fonic decoding for unfamiliar words. The skilld readr neither 'barks at print' and misses the meaning (the result of poor 'Fonics' teaching) nor gesses and skips words and misses the meaning becaus incorrect gesses cannot be checkd nor new vocabulary identified (the result of teaching without Fonics).

A good readr appears to use a whole cognitiv organization with recognition memory, visual serch, short-term memory, freed and cued recall, intentional and unpland lerning, semantic and lexicl decisions, imajery, prose comprehension, the activation and utilization of knolej, and long-term memory (Puff, 1982). Both visual and fonological features of orthografy are processd at an abstract level in the brain - so that the visual features are independent of letr case and font that distinguish words in space, and the fonolojicl features are independent of the chronolojicl constraints that distinguish words in time. Skilld readrs can play the 'psycholinguistic gessing game' of reading (Goodman, 1982) both 'top-down' using context and 'bottom-up', thru detaild processing of words, which involvs the importance of spelling regularity, use of sound-symbol relationships and spelling analogies, representation of morfemes and gramaticl markrs, etymology, context and visual features such as distinctiv visual elements of words, their frequency, length, shape, distinctivness, letr information and sequences, and 'sequential redundancy', which means no interfering letrs, not 'as much clutr as posibl'.

Poor readers

. English spelling is shown to be a criticl addition barrier for all lerners facing any other typ of handicap, and altho not the only factor, it is a final barrier preventing adequat literacy in English- speaking countries despite costly universl education. Excellent methods are now availabl to teach lerners to 'hear sounds in words' - removing one criticism of using spelling in reading, that 'fonologicl awareness' is often dificult to aquire.

'Information overload' is a major factor in reading dificulties - too much has to be done at once, and ineficient readrs may focus on one aspect or another, at the expense of comprehension. Since multipl levels of operation are required in reading, any reduction by simplification in the amount of simultaneus processing required in reading can help to save them.

Short studies to iluminate these issues included Asian students' reading in diferent writing systems, the correlation of spelling and reading abilities, and a demonstration that even highly educated peple usualy cannot spell 16 common words correctly.

Lerning to read.

 All readrs continue to be lernrs, and many adults nevr reach the staje of fluent easy reading. The discussion here includes theories of lerning to read, criticl longitudinl reserch studies which follow children's cognitiv development, the fonic intuitions of erly 'spontaneus' readrs, and experience with initial lerning spellings. Studies of lerners show that spelling is relevant to them at three stajes - when they need to undrstand the nature of reading and how to relate the ritn languaj to the spoken languaj that they know, to ensure accuracy in identifying unfamiliar words, and to facilitate practice top fluency.

The fonics-taught poor lerners who merely bark at print without regard for meaning, and the 'whole-word' taught poor lerners who wrongly gess 'whole words' cud all be assistd by a mor custom-designd spelling that gave them mor oportunity to reach rapid automaticity in acurat word recognition, with their limitd resources. Fastr checking of the acuracy of gessd words means greatr fluency, and greatr fluency increases the ability to follow the meaning of a text, as short-term memory can carry mor of it. An orthografy that makes it easy for beginrs to be establishd as confident lerners, must also make it posibl for able lerners to race ahead. But young lerners ajed five to six, and most adults with reading dificulties - indeed most adults - do not know the history of the English languaj, nor etymolojy, nor do they hav explicit knolej about morfemic relationships of words to explain the English spelling they must lern by rote.

A mor consistent, economic and predictabl representation of languaj in spelling would also make possibl improved teaching methods, and less rote-lerning, whethr of whole 'sight-words' or of fonic drils. It wud assist movement from stratejies of limitd logografic and random feature recognition, and make 'decoding' a fastr process on the way to automatic recognition. Children cud read erlier at their mental aje level, and this wud assist the development of their stil immature spoken English, as well as their reading for subject content. More lerners might be able to 'teach themselves to read' than the few who do so today - especialy now that video and CD-ROM make posibl independent home lerning. [5]

Originl experimentl studies on lerning to read include a comparison of whole-word and alfabetic lerning stratejies, a demonstration of why the pedagogy of Paulo Friere for teaching South American pesants to read cannot be used with English spelling, children's silent and oral reading in TO and in Surplus-cut spelling, and three studies of orthografic structure and analojicl strategies in word recognition. Children's own 'invented spelling' is shown to provide leads for a user-frendly English spelling; spelling is shown to be a major factor in the graded reading dificulty of a standard reading test, the Neale Reading Analysis, and five case-studies show how different are the ways by which children may lern to read so that there are risks in using group data to make jeneralisations about 'stajes' for all children. All these confirm the necessity of improving English spelling, to improve literacy in the English languaj for lerners, skilld readrs, riters, disadvantajd groups, foriners, imigrants, and now computers. The 'optimum' riting system wud consistently and broadly represent the spoken languaj for lerners, with consistent modifications (e.g. possibly morfemic components) found by reserch to facilitate fast and skild reading.


'PART IV reports a network of studies and experiments in spelling modification [6]. A central experimentl series tests the proposition that letrs surplus to the representation of meaning and fonolojy cud be acceptably deleted from words in English spelling, leaving the basic spelling structure of the words intact. The specific letr deletions in the experiments wer based on aplication of linguistic principls and observations of readr behavior. Results supported the à priori deletions, but ful verification and definition remain a further task. This spelling modification, at first named Surplus-cut Spelling (SC) was designated as Clipd Spelling in the thesis to distinguish it from Upwards' extension of the orijinal concept, since this is now wel known as Cut Spelling. Howevr, in this acount I revert to the first title, Surplus-cut, as this too has become known thru publishd articls (e.g. Yule & Greentree, 1986).

Deletion of surplus letrs in spelling does not solv all the problems of misleading representation of fonemes and mrfemes in English spelling, but it has the advantaj as a first step of retaining close bakwards compatibility with the visual apearance of traditionl spelling. Altho around 32% of words in text may be modified, only 5% of letrs ar deleted, so that visual familiarity is minimly afected, while visual clarity of word structure is greatly improved. Surplus letrs ar involved in the spelling of 40.3% of vowel grafemes. In primary scool children's reading, these proportions ar 35.5% and 23.7%, which ar still a hi proportion of 'irregularity' to be produced by letrs which may wel be regarded as intrusions into the ritn word. Other proposed forms of spelling modification hav been shown to lack the rapid transitional facilitation of deletion rather than vowel-letr chanje (Yule & Greentree, 1986). [7]

A core experimentl series tested the efects of Surplus-cut spelling (SC) on the reading of words and word-pairs and continuum text. The focus was on readrs' imediat adjustment to SC, since, if posibl, any orthografic chanje shud not be initaly disruptiv to those alredy literate, and shud be bakwrds compatibl with all that is now in print. The central experiment was a lexicl decision task for 30 adults which was bakd up with replication and three types of control experiment, including letr-deletions ('control-cut') which matchd Surplus-cut deletions in all except aparent superfluity.

Tle experiments wer complemented by smallr pilot experiments with children, poor readrs, and adult lerners of English as a forin languaj, and with studies of attitutes and of the needs of the riters. Erlier unpublishd reserch on practice effects in reading modified text and comparisons with other typs of orthografic modification wer included in the apendix.

Surplus-cut spellings wer recognised by naive, readrs significantly fastr and mor acuratly than control-cut - for exampl, words flashd at 30 msec in CRT experiments had a rejection rate of around 10% for TO spellings, 25% for CS, and 50% for Control-cut. Familiarity of spelling as wel as of word was found to be the major factor in speed and acuracy of recognition of words, including in the recognition of alternativ spellings accepted by dictionaries. Three practice trials cud be suficient for SC recognition rates to match TO, but control-cut recognition remaind slower and les acurat. It is stil to be asertaind whethr, with further familiarity, SC spelling wud actuary improve the eficiency of alredy skild readrs, but some individual poor readrs showd this improvement.

A less clearly establishd finding from Experiments 9.4 and 6.5 requires furthr replication and elucidation. Positiv atitudes and comprehension of the spelling system wer improved by familiarity, even tho 20 hours' practice spaced over 20 days was not suficient for all readers to fuly overcome the frequency efect that oftn gave TO the advantaj. But only a litl reading in Surplus-Cut Spelling apeared to help readers to pik up the principls of surplus-deletion and aply them in their own writing and undrstanding of spelling. This sujests that if Surplus-cut spellings wer permitd as acceptabl alternativ spellings, readrs cud pik up the principls of Surplus-cut simply by reading, and cud then aply them to both reading and riting - which wud be a mor efectiv and inexpensiv way of transfering to modified spelling than by lerning sets of lists or undertaking special costly training. The ability too jeneralise languaj principls by analojy, which is so important in children's lerning of the spoken languaj, cud then also be exploited to greater advantaj in the teaching of reading. That is, our demonstrated abilities to switch code and chanje set wud facilitate transition to a consistent spelling system.

It is crucial that beginning readrs ar not demotivated by initial decoding confusions. Experiments and studies gave some suport to two claims - that those alredy literat wud not be disturbd in accuracy or comprehension in their first exposure to reading continuous text in Surplus-cut Spelling, and many wud not even be initialy disturbd in reading rates, and that lerners might read mor new words independently and acuratly if presented in Surplus-cut Spelling.

Pilot maskd-priming experiments (Appendix 9.3) did not confirm the proposition that for skild readrs the mentl representation of words might be a morfofonemic core as represented in Surplus-cut Spelling rather than the visual features of th orthografy, but further investigation is indicated, since this typ of experiments procedure tests only one typ of recognition and/or lexicl entry.

Two furthr tendencies cud be significant. For both reading text and reading singe words, readrs tend to make use not only of th visual configuration and/or th visual patrns and redundancy in th orthografy, but also profitd from the morfofonemic representation of th spoken languaj in print. This also held for children who wer lerning to read successfuly, altho not for those who wer failing. This has important implications for th teaching of reading, confirming th evidence and arguments presentd for helping children to awareness of sounds in th spoken word and how they are represented in print. Apendix 6.2 sets out some 'childish' tecniqcs for children to help them to this criticl awareness.

Furthr small rating experiments used ratings to assess atitudes, and letr-canceling and similar tests to study comprehension of Cut-Spelling principls.


The studies and experiments disprove some traditional and Chomskyan claims for advantages in conventional English orthografy, while suporting others, and hav implications for theories of reading in education and cognitiv psychology. The experiments bring out some of the unsolvd issues in th nature of reading processes, and th problems that must be faced in reserch on spelling design. It is concluded that a Surplus-cut spelling wud clarify the morfemic and fonemic structure of th English languaj and be minimly disruptive as a practicabl step towards an optiml English orthografy. All categories of readr wud benefit by spelling modifications that increasd predictrability and reveald th underlying structure of th orthografy mor consistently.

Furthr cross-cultural studies of th eficiency of riting systems for other languages can clarify useful features for orthografy that may be incorporated into our own spelling, and others may be workd out from fullr knolej of th psychoneurolojicl processes involvd in comunication by languaj. Some posibl directions for orthografy ar discussd. Reserch in spelling design also offers a point of entry towards solving some of th puzls that stil tantalise scolars in th cognitiv psycolojy of reading.

Th goal is Chomsky's concept of 'optimum orthografy' (1970), - defined as th writn representation most favorabl for readers in a particular languaj, and th best fit to meet the diferent and sometimes incompatibl needs and abilities of users and lerners, readrs and riters, nativ-speakrs and th forin-born, th bright and th dul, th norml and th handicapd, humans and machines, while maintaining access to our heritaj of print (bakwards compatibility). Following Venezky's (1970) principls for th design of a practicl riting system, it shud harmonise with linguists' fonolojicl classifications and with nativ speakrs sometimes conflicting perceptions, it shud be compatibl with th linguistic, politicl and sociocultural aspects of its cultural setting, and it shud be psycholojicaly and pedagogicly apropriat for its users. This investigation shows that these ideals can be realisabl, and that acceptabl trade-offs wud be posibl, refuting the common à priori argument that orthografic chanje useful for one purpose or group might disadvantaj another (e.g. Frith, 1980).

Implementation by action-reserch is feasible with popular acceptance following familiarity as the criterion for what is most 'user-frendly'. Methods of introduction include 'trial by television' in subtitlitig, initial lerning spellings, and alternativ 'demotic' spelling for those groups currently cut off from conventional literacy.


The apendices include details and data from th experiments, and furthr short studies:-
The bibliografy contains aproximatly 860 refrences, and there is an extensiv glossary.


[1]. Copies of the thesis are held by the Simplified Spelling Society, Manchester University Mont Follick Library, and the Library and Education Faculty at Monash University, Clayton, Vic., Australia, 3168.

[2]. The manuscript of a book for the general reader based on the first two parts of this thesis is still looking for a publisher. This updated and rewritten version includes the oddities and fascinating miscellanea that an academic thesis must leave out, and corrects its defects.

[3]. Chapters on 'The politics of spelling' and 'The politics of international English spelling', using material from this section, apear in The Politics of Literacy in Australia and the Asia Pacific Region, ed. David Myers & Nicholas Walker, Melbourne; Australian Scholarly Publishing. In press, 1994.

[4]. Articls using material from this section:-
Learning to read without effort. Reading (U.K.), 26.2 1992, 12-16.
Children's abilities and 'Surplus-cut' spelling reform. JSSS, J13 1992/2, 3-8.
Improving English spelling for readrs: th necessity for reserch. JSSS, 7.1 1993, 10-18.

[5]. The half-hour 'take-home' computer grafics cartoon video Teach Yourself to Read or Find out where you got Stuck (Melbourne: Literacy Innovations, 1993) is my attempt to exploit the potential of th new media for Do-it-Yourself literacy. One of its inovativ features is its explanation to lerners of how and why English spelling is silly, and how to cope with this problem. Spelling reform is promoted as the eventual solution.

[6]. See 'Experiments in public response to Surplus-cut spellings in text' JSSS J16 1994/1, 7-16, for an article based on this material.

[7]. In this articl the extent of letr deletion has varied, so that readrs may monitor their own responses. Substitutions of < f > for < ph > and <j > for soft <g> ar also made, since these ar now familiar thru spelling in advertising and in new vocabulary.

[8]. An articl using this material apears in Visible Language, 28.1 1994. 26-47. 'Problems that face research in the design of spelling.'


Campbell, Stuart. 1982. On Esperanto. Spelling Progress Bulletin. 22. 3: 4- 5,

Chomsky, C. 1970. Reading, writing & phonology. Harvard Educational Review, 40:287-309.

Chomsky, N. & M. Halle. 1968. The sound pattern of English. NY: Harper & Row.

Crystal, D. 1987. The Cambridge Encyclopedia of Language. Cambridge: CUP.

Frith, U. 1982. Cognitive processes in spelling and their relevance to spelling reform. Spelling Progress Bulletin, 22.1 6-9.

Gogate, M. N. 1988. Roman Lipi Parishad Proceedings. Bombay. See India.

Goodman, K. S. 1982. Language and Literacy: the selected writings of Kenneth S. Goodman. Vol II Reading, Language and the Classroom Teacher. F. V. Gollasch (Ed.) London: Routiedge & Kegan Paul.

Morris, Joyce M. 1990. The Morris-Montessori Word List. London Montessori Centre.

Puff, R. C. (Ed.) 1982. Handbook of research methods in human memory and cognition. NY: Academic Press.

Smith, F. 1971, 1978, 1982. Understanding reading. NY: Holt Rinehart & Winston.

Upward, C., P. Fletcher, J. Hutchins & C. Jolly, 1992. Cut Spelling: A handbook to the simplification of English spelling by the omission of redundant letters. Aston UK: SSS.

Upward, C. 1992. Is traditionl english spelng mor dificit than jermn? Journal of Reserch in Reading, 15.2. 82-94. Summary J13 1992/2 p22.

Veblen, Thorstein. 1899. Theory of the leisure class. Republishd 1925, London: Allen & Unwin. Last page.

Venezky, R. L. 1977. Principles for the design of practical writing systems. In: Joshua A. Fishman, (Ed.) Advances in the creation and revision of writing systems. The Hague: Mouton.

Weckert, C. 1994. Teaching reluctant readers in the mainstream classroom. Gostord NSW: Bookshelf Publishing, and Multimedia International U.K.'

Yule, V. 1986a. The design of spelling to match needs and abilities. Harvard Educational Review, 56: 278-297.

Yule, V. & S. Greentree, 1986b. Readers' adaptation to spelling change. Human Learning 5: 229-241.

Back to the top.