[Journal of the Simplified Spelling Society, J21, 1997/1 pp10-12]
[Valerie Yule: see Book, Journals, Newsletters, Media, Anthology and Bulletins, Personal View 10 and web.]

Teaching Reading and Spelling Reform.

Valerie Yule.

Valerie Yule is a reserch psycolojist and associat in the English department, Melbourne University, working on literacy and the development of imajination. She is the author of Orthografy and Reading: Spelling and Society, an investigation of writing systems and experimentl reserch on improving English spelling to betr match user needs and abilities.

This articl is ritn in a transitionl staje of Internationl English Spelling (Interspel). [1] Chanjes ar introduced during the course of the articl.

A lively issue during three decades of post-war reconstruction was the reform of writing systems to promote literacy. Other countries went ahead to amend or radically change their orthographies, including Korea, Indonesia, Malaysia, Japan, Vietnam, China, Spain, Netherlands, Israel and regions with pijin and creole Englishes. But the British spirit failed (Follick 1965, Reed 1975).

Mont Follick MP, who led the British Parliamentary spelling push, put up untested proposals that lacked face validity for the public. [See Hansard account pdf file and Spelling and Parliament.] People thought the new 'rational spelling' lookd dificult, ie, worse not betr, and the considerabl support that had been aroused collapsd. A P Herbert MP printed as a joke Follick's opening paragraf on 'rational spelling', in an article signed ei. pii. eitch, in Punch (1949). This is a sample:
Ool dhis shouz dhat in dhe bilding up ov dhe North Atlantik Iunion, dhe rwling serkels ov dhe Iunaited Steits and Greit Britan endever tw drag intw dhis afair dhe greitest posibel number ov Steits and dhus teik dhem in hand.
'APH' had no difficulty mocking the chanjes. Any acceptabl improvements on current English spelling wer understandably overlookd. In the event, government investigation of spelling reform was passd up as a trade-in for instituting in schools the Initial Teaching Alphabet of Sir James Pitman MP. The ocasion passd. The winds of chanje blew in other directions, and spelling reform in Britain was blown into a dusty cornr.

Forty years on the media stil stir and parents stil worry about reading standards, but spelling reform has not yet re-surfaced in popular concern for litracy. One reason for this was that educators' next expedient was to try to solv the English spelling problem by bypassing it. Spelling, they claimed, was not needed for reading or lerning to read anyway - no spelling reform therfor required. As spelling was not needed, teachrs did not hav to teach it - hoorah - and children did not hav to undrstand the spelling system or fonics - wasn't that lovely? Unfortunatly cutting the Gordian knot cut an essential life-line to reading.

Ful-blown Whole Languaj (alias 'Real Books') teaching is now in retreat in the UK and America, tho stil top practice in Australia. Many stil maintain that children do not need explicit teaching in lerning to read - they absorb it from their environment as 'naturaly' and easily as they lern to speak. The claims ar falacious.

a) Litracy is not 'natural' and universl like human languaj. Only a handful of societies hav inventd riting systems. Most languajes in the world had no riting until this century and hundreds stil hav none. Like most advanced skils of civilisation, reading and riting must be taut, and lernt with aplication and practice.

b) Children do have an inbilt motivation for languaj, but they do not lern to talk as easily as they lern to brethe. Observ the eforts of small children from their first babbling. When they arive at school, they ar stil not expert.

'Whole Languaj without Spelling' is beginning to colaps like a baloon. Cognitiv reserch shows that a necessary step to becoming an expert independent readr is ability to decode the ritn languaj into the spoken languaj, and it is betr to discovr how to do this erlier rathr than later (see, for example, Ehri & Wilce 1987, Stuart & Coltheart 1988, Bryant & Goswami 1987, and th revew in Yule 1991.) In the grim past, too much 'fonics' produced some lernrs who only 'barkd at print' and coud not read for meaning, but also, thruout history until recently, all successful Westrn readrs lernt to read for meaning starting off with the alfabeticl principl. With no-fonics Whole Languaj, lernrs ar stil found who do not read for meaning. Their problem is that they gess rongly from context or pictures, and they ar unable to chek their gesses.

It can be dificult to persuade comitd Whole Languaj litracy educators that eficient litracy needs spelling, and so spelling needs spelling reform. Like their pupils, they read by predicting what they expect to read. So they do not read accuratly whatevr may difr from their predictions. A recent exampl ilustrates th defect.

The Australian state of Victoria has set up a world-first Ministry for Multimedia with over $A40 milion to spend. I sujestd investing $20,000 on a half-hour take-home cartoon video with computer-animated grafics that gave an overvew of the English riting system and demonstrated how to lern to read, how to hear sounds in words, and how to use context and other stratejies. (See Yule 1995) Th oficial reply from th Ministry of Education (21/11/96) set out its Whole Languaj filosofy, and then itself demonstrated its basic weakness. The letr began:
...reading is a complex process... Current practice in early years classrooms focuses on reading to, with and by the student. Teachers provide students with many opportunities to engage in the reading process through shared reading activities, specific teaching of the visual features of print, independent and guided reading for students who are matched to text and language experience for those who are not. In guided reading attention is paid to the ways that students use and attempt to integrate the cueing systems.
The project to provide take-home repeat-viewing independent lerning for adults and teenagers was then rejected on grounds none of which applied - such as that a school class of small children wd not be able to remain on task for 30 minuts or to undrstand it one-off.

Whole Languaj and post-modern literary theory.

It is partly a corolary of this recomended method of reading by gessing that education has become so enthusiastic about litrary theories that state explicitly that readrs do not read what authrs rite - they construct their own meanings from th text. To some extent this has always been tru. Readrs employ a gradient of personl reconstructions, that shoud be at zero when reading an instructionl manual or othr factual information, that increases when reading fiction acording to th narativ's levl of ambiguity and complexity, and that may become almost total re-creation when trying to undrstand som modrn vers. But th Whole Languaj aproach approves gessing for all genres - altho an adult litracy course is likely to warn that it is betr not to gess when reading labels on medicin botls. Ask someone else insted.

Lerning to read without spelling.

One reason why teachrs may encuraj lernrs to make gessing a primary tactic is reaction to a previus extreme. Children cd be traumatised by teachrs insisting that each word had to be decoded corectly befor attempting th next. Poor readrs might nevr read a ful sentence or paragraf or story without interruptions to the sense. How much easier, then, to giv cues. "Well, what do you think Jane and Peter wer going to fly in?" The ansr is easy:"a plane" if it is a short word, "a helicoptr" if it is long.

Children practice this sort of gessing with many worksheets of 'cloze' exercises - that is, gessing th missing words in a story or even a sentence. [2] Gessing in 'reading' erly picture books can be very successful - predicting one line of text per paje from th picture that takes up most of th space. A bright child might make up a betr story than the actual text, but this creativity wil not be encurajd. Howevr, as soon as th books cease to contain mainly pictures, children's gessing habits may stil rely too much on what they expect to find, based on their own limitd knolej of th werld. And as for 'checking bak to ensure that yr prediction is corect' - what slo reading habits this must produce, if readrs realy do chek bakwrds all th time. Unless prose is realy complex, it is surely mor eficient to read strait forwrd, knowing that th reading has been acurat in th first place.

'Whole Languaj' reading confirms readrs' own ideas about th world, rathr than teaching mor about it. Readrs without spelling nolej read unfamiliar words as ones they alredy kno that look similr. And so howlrs apear even in the daily press, as 'cohesive' is confused with 'coherent' and 'constellation' with 'catastrophe'. This is serius enough for children, but even mor so for adults. [3]

When peple can only read vocabulary that they recognise from past experience, and lak decoding skils or understanding of the spelling system, they wil be jitry about any deviation from the familiar. Nor do they become good readrs, even aftr internalising numerus spelling patrns to use by analojy. An ongoing study of how peple read airport thrilrs and blokbustr paperbaks is confirming my theory that averaj adults do not read wel.

The return of fonics, and spelling reform.

'Fonics' is making a come-back. That is, teachrs can let beginrs kno that th ritn languaj is based on th spoken languaj and letr sequences represent sounds. Accurat reading of th words can again be made part of th ajenda in 'reading for meaning' without th previus terrorism.

But this aproach to teaching reading needs an emfasis that is not on series of drills, but on understanding at an erly aje what th basic English spelling system is and how to operate it. Fonics teaching faces th dificulty of demonstrating th rationality and user-frendliness of TO. It must be openly recognised in schools that th need to 'fuj the blends' in decoding is due mainly to th distortions that now clutr up th underlying system. Spelling reform may cease to seem politicly incorrect

'Only connect,' rote E M Forster, and it is not just passion that needs connecting to prose. We need educators who can connect th masses of reserch findings - who can play music with mor than one string at a time, so to speak. Thus they wil be able to read about spelng reform without necessarily assuming it can only be 'spelng just as u speak'. Thay wil be able to read acuratly th definitiv reserch that shows, for exampl, that readrs who can decode unfamiliar words ar not slowd down by having to decode them perpetualy: once th words ar familiar in print, automatic recognition takes over. U don't need to read a map for familiar places.

Teachrs hav feard spelng for too long, altho with reason. It is time for them to be able to teach it as a coherent system, so that they and their pupils can realise that English spelng cd be reformd by cleansing. The underlying system can be stripd of its inconsistencies and incongruities, even for th 19 or so vowel sounds. [4] There shd be no need for speling lists to lern eithr by fonics or visual rote, nor for trying to invoke pictografic-style 'reading without a riting system' in ordr to remembr th linear letr-strings of an alfabetic orthografy. English spelng is not a spider's web of majic origin, but only a human artefact.

We now can hope for mor rational teaching of literacy. To make it completely rational, 'User-Frendly Spelling for the Year 2000' must be set firmly on th educationl ajenda.


 [1] 'Internasionl English Spelling'- InterSpel. English spelling coud be updated for internationl use by public experiment with three main steps and three principls.

i. Delete letrs in werds that mislead or do not serv to represent meaning or pronunsiasion.
ii. Consistent consonant spellings.
iii. Redusing spelling patrns for the 19+ vowel spellings from an estimated 318 to around 40. Much of this rationalising is obvius but some questions ar stil unresolvd, eg, finel s/z, and silent e for long vowels.

Principls: Morfemic and gramaticl convension to be consistent.

Public use of Internasionl English Spelling can be adjustd to 'what th market can bear' during transision. Dictionaries alredy giv two or mor spellings for sevral thousand words, and this policy can be extended.

Some 12 very comn iregularly spelt werds may be betr retaind at first, eg, one, all, one, was, what, who.

See: 'A Transitional Spelling Reform for Adults and Learners. Spelling Progress Bulletin, xx .3. 7-10, 1980.

[2] Context as a basic reading techniqe to gess meaning. Th sampls of 'cloze' text below ilustrate that relying on context to gess meaning is mor risky when th text is informativ (a) than when it is a child's story (b). As in 'cloze' exercises in school, one word in eit has been deleted.
(a) To convert documents from different file expansion..., you must install the appropriate converters. To ... and export graphics contained in documents with ... you must install graphics filters. If you ... a complete installation when you installed the ... , converters and graphics filters were installed with ... . If not, you can run the designated ... program again. For instructions, double-click the ... button on the standard toolbar, and type ... .

(b) The ... was coming up hot and golden over ... hill-side now, and the heath was ... and the bracken and gorse green and ... . Down below in the valley the children ... see smoke rising from the chimney-tops and ... red roofs like scarlet caps among the ... . They were walking along a little path ... which was steep and slippery and set with ... pebbles, and which was carved into the ... side of the hill. Sebastian slipped more ... once and Tom had to hold his ... very tightly or he would have fallen.
[3] Good spelrs ar usualy good at undrstanding what thay read. Whole Languaj adherents comonly make assertions such as that if readrs blend th sounds in stop, s-t-o-p, in ordr to read th word, they wil not kno what stop means. They can only kno what a word means if they read it 'as a whole'. But obervationl evidence shows that good spelrs who can use their spelling knolej to read unfamiliar werds, do imediatly recognise them if thay ar in their spoken vocabulary. This is so taken for granted that nobody (I think) publishes reserch to demonstrate the relation of spelling and reading abilities eny mor - they may, like Frith (1982) simply quote Malmquist (1958) and correlations of 0.50 to 0.80.

Anyone can replicate th findings of th present author's unpublishd studies of this reading-spelling connection by comparing th spelling ability on th 16 Word Spelling Test of good readrs who comprehend wel, and poor readrs who comprehend badly. Sixteen words ar set out with surplus letrs deleted, and th task is to rite them out correctly in TO.
This experiment has three secondry advantajes. It demonstrates to th subjects:
[4] The basic 19 Australian-English vowel sounds (fonemes). In TO each vowel sound has some distinctiv representations but meny that overlap and require rationalisation.
oo (book)
(Welsh w)

oo (boot)


Bryant P & Goswami U (1987). Phonological awareness and lerning to read. In J Beech & A Colley (Eds) Cognitive approaches to reading. Chichester. John Wiley.

Ehri L C & Wilce L S (1987) Does learning to spell help beginners learn to read words? Reading Research Quarterly. 22, 47-65.

Follick M (1965) The case for spelling reform. London: Sir Isaac Pitman & Sons, chaptrs 21-24.

Frith U (1982) Unexpected spelling problems: good readers who are atrocious spellers. In U Frith (Ed) Cognitive Processes in Spelling, London: Academic Press.

Herbert A P (1949) The cosmic mess. Punch, CCXVI, no. 5648, March 9, 262-263.

Malmquist E (1958) Factors relating to reading disabilities in the first grade of the primary school. Stockholm: Malmquist & Wiksell.

Reed W (1975) Spelling and Parliament. Spelling Progress Bulletin, Winter.

Stuart M & Coltheart M. (1988) Does reading develop in a sequence of stages? Cognition 30 (2) 139-181.

Yule V (1991) Orthography and reading: spelling and society, unpublished thesis, Monash University. Paperback copy held by SSS.

Yule V (1995) Teach yourself to read at home by video: problems and promises. JSSS, J18 1995/1 pp11-18. [See web links.]

Back to the top.