[Journal of the Simplified Spelling Society, 1992/2 pp3-8 later designated J13]
[See Journal, Newsletter, Anthology, Bulletin articles, Personal View 10 and website by Valerie Yule.]

Children's abilities and 'Cut' spelling reform.

The advantages and disadvantages of a 'Cut' spelling reform for children aged 5-7 learning to read and write, in view of what is known of their needs and abilities.

Valerie Yule.

Faculty of Education, Monash University, Clayton, Victoria, Australia

The prime purpose of spelling reform is to benefit the users of English spelling.

This is not the same as the benefit of spelling reform in producing a rational spelling system to replace the present conglomeration of contradictory principles and exceptions. Hundreds of rational systems for English spelling can and have been devised. Once you start asking "Who will benefit from these improvements?' the ball game changes its nature and even its playing field.

A reform of English spelling must benefit many different categories of users who have differing needs, and within each category there are extremes of expertise and incompetence. (Yule 1986). There are differences between the requirements of
Many opponents of spelling reform have claimed that it would be impossible to produce a reform to benefit one group of users that did not disadvantage others, and hence spelling reform is not feasible. However, my research and reviews of research (Yule 1991) present evidence "to the point of overkill" (examiner's comment) that
Rather than immediately trying to implement a spelling system that on the face of it is theoretically ideal, we must first ensure that it will be the best practicable spelling for the English language to meet the needs and abilities of these groups. The metric system is an object lesson - its measures are perfect in theory, but need to be supplemented to be really practical for housewives, carpenters, farmers, and baby-weighers. An example of the sort of practical compromise that may be necessary is the decision to retain the present Roman alphabet, despite its imperfections, because it has world-wide currency in other languages too. The examples of writing systems devised in this century for illiterate peoples also show how modifications of the ideal are always necessary for practical purposes.

What, then, are the needs and abilities of human beings that are not met by present English spelling, and are of prime concern to spelling reform?

The first distinction must be made between the needs for writing and the needs for reading. Curiously, spelling reforms have traditionally been designed more for writers than for readers. "Spell as you speak" has been an ideal to help writers. The difficulties this presents for readers have not been fully recognized.

The next question is to consider the nature of the English language - since a writing system that suits one language may not be as suitable for another. We can look at a range of possibilities by surveying mankind from China to Peru as the great lexicographer Dr Johnson would have recommended. Reform proposals must be appraised for ease of immediate adjustment by those already literate, case of learning to read/write, international facilitation of communication, effects on the universal literacy problem - the priority of whose needs should be served most, and whether a simplified solution might be accessible to all, that led into a spelling that gave the greatest efficiency possible to the experts.

Children learning to read and write.

The biggest argument for spelling reform is not that English spelling is imperfect, but that it is a barrier to literacy for millions. If a spelling reform does not make it easier to learn to read, its other aims have little force. In English-speaking countries, most people are taught to read between the ages of 5-7. A significant portion fail to learn, and may even leave school in adolescence as non-readers.

What sort of spelling reform could help prevent this by matching the needs and abilities of normal children aged 5-7? This article considers the needs of child learners, and focuses on the advantages and disadvantages of "Cut Spelling" proposals for them.

It is essential to realize that there is a strong and even "established!' line of thinking and practice that spelling is to be avoided in teaching young children to read, and so spelling reform is completely irrelevant. This influential doctrine has swung in and out of fashion for the past 150 years (see Chall, 1983). Thousands of teachers laugh out of court anyone wanting children to learn the alphabet and the relation of letters to spoken English. They believe words are to be recognized as wholes, and context helps to identify novel words (see influential works of Frank Smith, e.g. Smith 1982.)

There are good reasons for these waves of fashion for "Look and Say", "Sentence Reading", "Paired Reading" and other non-phonic approaches to the teaching of reading.
The critical problem is that altho all peoples naturally learn the complex cultural artifact of spoken language, the complex artifact of written language is not a natural development. It is an invention - or discovery - more closely resembling mathematics. Most languages of the world have never been written down, and most written languages today have been produced within the last 200 years, with outside help. Most writing systems until then have followed the most obvious principles - one symbol = one word, or one symbol = one syllable, with various permutations. The ingenious idea of the alphabet, one symbol = one speech sound, which allows any word at all to be encoded and decoded, has apparently been invented only once, in the Middle East, whence the various existing alphabets have derived.

There is good evidence that most children have to be taught to recognize sounds in words (see e.g. Liberman & Shankweiler, 1985), and not only that, but that illiterate adults cannot intuitively segment words into their constituent speech sounds (see e.g. Morais et al, 1979). It is quite a conceptual achievement - as the singularity of its invention suggests.

Nevertheless, there are many great advantages for literacy with the alphabetic principle, once it is understood. It is so economical and flexible, and more suitable for the English language than a character system like Chinese or a "syllabic" system like Japanese kana, that it may well be the writing system of choice for English spelling reform.

Even with English spelling as capricious as it is now, the "Look and Say" types of reading instruction are fashions that wash out as well as in, since they still do not solve the learning problem. Learning to read new words in English without any phonics is extremely difficult, if not impossible, for all except those with exceptional visual memories and language skills.

Few children can remember more than 40 "sight words" without clues to distinguish them.

Chinese, which is acknowledged to be much harder to learn than English in any case, would be quite impossible without the clues of stroke order, radicals, phonic elements and other classifiers and mnemonics, and its gestalt qualities which contrast in visual memorability with linear strings of letters. "Look and Say" learners in English who succeed are those who can apply their alphabetic understanding from being taught to write, to learning how to read. Cognitive psychologists today have demonstrated convincingly the vital importance of "phonology" in learning to read, altho educationalists have not generally been aware of current work (see e.g. Bradley & Bryant, 1983; Ehri & Wilce, 1987; Fowler et al 1979; Liberman et al, 1985; Stuart & Coltheart, 1988).

It is essential therefore that English spelling reform is linked with instructional reform, that teaches children HOW to hear the sounds in words. Even the Initial Teaching Alphabet was not as successful as it might have been because the initial hurdle of recognizing sound in symbol is only partly overcome simply by removing confusing exceptions to general principles.

Researchers have demonstrated the importance of teaching children HOW to recognize sounds, not just presenting them with what they must learn (see John Downing's work on linguistic awareness, e.g. Downing & Valtin (1984) - altho sometimes psychologists' prescriptions have been pretty dull and drill-like.

However, others like Bradley (1980) and Bryant & Bradley (1985) have shown how very simple and clear the early introduction can be for young children, using singing, plastic letters, and principles of association, chunking and cognitive understanding. See also my video Preparing to Read through Play (1981), and Anderson et al's 1985 review of practices in successful classrooms that show how simple and basic the phonic comprehension needs to be in order to learn to read. Ninety-five percent of drills and worksheets in much "phonic" teaching is unnecessary. Even the simple strategy of "Sing as slowly as you can" has been the clue to understanding how to bear sounds in words and hence to learn to read for a surprising number of adults and teenagers I have seen who had failed previously to learn.

Establishing how children can easily be taught to recognize sounds in words is an essential step for spelling reform that regularizes the relationship of spoken and written language.

The next steps are to understand what sorts of sounds young learners do easily recognize in words, and what sorts of spelling they can read most easily. A "regular" relationship may still be difficult for them if it does not take account of their needs and abilities.

Children's writing.

Difficulties children may have in relating sound to symbol may not be obvious to adults. A letter written by Jessica, a bright five-year-old to the Queen illustrates children's principles of phonology that have been found by all researchers into pre-school writing, e.g. Carol Chomsky (1971) and Charles Read (1975).

DEAR QUEEN ELIZABETH I AM JESSICA LEWIS I HAV IN MI STAP BOOK LOTS OF STAPS OV YOU AND ON YOUR ENVELOPE THEIR IS A STAP FROM MI STAP BOOK BUT IT IS SKAP DO YOU NO SOPHOIN BIKOS I DO I LOVE MATNETAK ILOD DO YOU LIK THE WELD I DO MI HAWS IS IN BULEKPUPUN.

(Translated this reads - I have in my stamp book lots of stamps of you and on your envelope there is a stamp from my stamp book but it is scrap. Do you know [?word unknown] because I do. I love Magnetic Island. Do you like the world? I do. My house is in Balikpapan. (Frequent letter reversals and switches of letter case are not shown.)

This letter illustrates - apart from royalist sympathies:
Children have problems sorting out consonant strings and tend to omit or misrecognize consonants in them e.g. ILOD MATNETAK STAP SKAP. This difficulty can appear in their spoken language as well, showing that they are not always aware of the real "form of the word!". They are still learning vocabulary, and may waver in their versions of it. "Cut Spelling" therefore should not cut out any schwa ("obscure") vowel, however minimal, if it may change the nature of the word for children who are still learning vocabulary, and/or add to the complexity of consonant strings.
From the evidence of how young children naturally write, when, a child-oriented spelling reform would cut out superfluous letters in order to prevent confusions, shorten words, simplify vowels, allow a range of phonemes to be represented by a single vowel, represent consonants and vowels consistently, have no more consonant strings than are absolutely necessary, retain representation of schwa vowels that separate consonants, and allow a limited number of very common sight words and suffix conventions (< -ION> for example is a quickly learnt convention) if it would assist compatibility with present spelling or international usages.

Further features of young children's writing, but not illustrated in Jessica's letter, are the use of letter names as spellings (e.g. RITN) and morphemic intactness i.e. they do not modify spellings when adding suffixes unless the sound also changes - e.g. PARTYS HAPYNES - i.e. Chomskian theory (C. Chomsky, 1970) applies to morphemes rather than to a lexical base that disregards phonology. This is part of children's remarkable linguistic capacity to generalize and reason which many spelling reform proposals ignore. Sound-symbol correspondence may require consistent modifications to allow for morphemic representation that shows the relationship of word families, particularly for built-up polysyllabic words.

This is evidence from how children write. However, it is well known that children can have difficulty in reading even their own writing. Spelling that is easy to read may not be identical to spelling that is easy to write. What features of words facilitate beginning readers?

Children's reading.

This question is bound up with the extent of vocabulary knowledge of beginning readers. It must be remembered that at each age level literates usually have far greater vocabulary than non-literates, largely because reading is how so much of their vocabulary is acquired. A spelling reform must make it easier to increase their reading vocabulary and also to be able to use it in their spoken language.

As everyone would expect, the longer the word, the harder children find it to read, unless it has particular salience and visual distinctiveness for them - e.g. most children learn quickly to recognize on sight memorable words like LITTLE, ELEPHANT, CROCODILE, CHOCOLATE, ROCKET that stand out among short words. In general, however, economy in word length is an advantageous principle for young readers. Superfluous letters are a nuisance. In these respects, "Cut Spelling" is excellent. For example, to take a page at random from SSS Cut Spelling Handbook (1992) p153.

1. Twenty-three Cut Spellings on p153 would probably make the words easier for beginners to read, since their main effect is to cut out superfluous letters that can confuse them, and the resulting spellings follow regular principles that are found in other words too:

AJECTIV
ALREDY
ALTERNATIV
ALYND
AR
DUBLET
FOREN
GIV
IGNORD
ILUSTRATED
IMEDIAT
NO
PRACTIS
PROMIS
SERIUS
SWICH
SYTHE
TH
THRE
U
WEL
WER
WICH

2. Eleven Cut Spellings on p153 would also probably make it easier for beginners to read the words since the shortening results in "letter pronunciation" for the final syllables, which is often used by children in trying to read as well as write:

COMN
RITN
EITHR
LETR
FURTHR
HOWEVR
PRACTICL
RADICL
READR
UNDR
WETHR

3. Cut Spelling inevitably leaves some spelling difficulties because it does not by its nature attempt to address all the initial problems of TO for readers - some of which are fairly common conventions, once they are learnt e.g. /u/ spelt as <o> and <c> pronounced as /s/ cf. eleven words on p153:

SOM
TWO
ABOV
OTHR
VOICED
APEARANCE
ASOCIATED
INCONSISTNCIS
INTRODUCED
CONCERND
OCURENCES

4. Some Cut Spelling conventions can be taught to older learners, since they are consistent and not confusing - but not for 5-7 year-olds.

RULE
CHANJES
ACORDNG
SPELNG
ENDNG
INCONSISTNCY

5. Some Cut Spelling-, apply morphemic principles, and so in theory should not pose problems,
e.g. SIMPLIFYD RELYBLY.

Some Cut Spellings would require explanation on initial exposure, since they are cuts that do not result in direct sound/spelling relationships
e.g. WUD AL.

Some cuts add to possible homophonic confusions - e.g. DISCUSD as DISCUS-D - altho the existing high proportion of homographic homophones in English spelling suggests this is not the big problem often supposed. This article itself has many such spellings that will not have been noticed
e.g. MAY SORT SOUND HIGH WILL.

Affixes which beginners should not be exposed to until they are familiar with the base words, since they include 3-letter consonant strings difficult for them to sort out, include
RESOLVD CONSIDRD POINTD COMPOUNDD.

Spellings with alternate possible pronunciations - NUMERUS COZY FOLOS POSESD - may not be a major problem if the correct version is established in one-trial learning, but this is subject to proof. It is seen as a problem in TO.

The major problem that Cut Spelling in its present form could present to learners of reading is in how to segment many of its polysyllabic words and relate them to spoken language. Experienced readers automatically use the strategy of segmenting polysyllables in novel words in order to decode them easily, read them aloud, or work out meanings from their roots and affixes. However, young children find it hard to segment words for decoding - another reason for shorter words, but also an argument for retaining CVC constructions (syllables in which consonants are separated by vowels) that make it visibly and phonologically easier to read the words. Cut Spelling therefore should not make "short cuts" that in fact can add to learners' difficulty, unless these are a version for more advanced readers (cf. Hebrew pointed scripts for learners and unpainted scripts for adults).

Cut Spellings may add a burden of consonant strings that learners could find difficult to sort out
e.g. on p153, DIFICLTY, INCONSITNCIS, INTRCHANGEBL, EQUIVLNT, SIMLRLY.

Spellings that could confuse learners up to late secondary level, who still are learning new vocabulary, as to where the schwa vowel separates consonants include SIBLNT which could be reconstructed as SIBLANT, PATRNS as PATRONS, ASIMLATION as ASIM-LATION, DIFRENTIATION as DIF-RENTIATION, also AVAILBL NORMLY ORGNIZE REGULRIZE SIMLRLY.

We might find that new "spelling pronunciations" would develop from Cut Spelling, that cut the spoken words as drastically as printed forms like these that drop obscure vowels that are already very weak. Pronunciation changes could also be promoted by spellings such as THER WHER THERFOR RELYBLY.

How might Cut Spelling help or hinder young child readers at the very beginning of reading?

There is a current trend to use "Real Books" rather than artificially constructed "readers" in order to make first reading more interesting. Beyond the HOP ON POP level of Dr Seuss, children are given stories such as Eastman's The Best Nest (1968) of around 706 words, in which the 116 word tokens with Cut Spellings would be as in Table 1.

Note that 14 CS word types have sequences of 3 letter consonant strings, and three words have 4 letter consonant strings - i.e. in CS over 25 % of the word types in this story for child beginners have consonant strings of three or more letters, whereas in TO the only such word, MATTRESS, is triple-string only by reason of its double letters, and is easily segmented into two CVC syllables.

Comment.

Cut Spelling in general follows "natural" children's writing and early reading preferences in deleting superfluous letters and reducing many digraphs to single letters. However, there are some cautions to make.

In view of the known difficulty that beginning readers have with consonant strings, empirical research is essential into what additional degree of difficulty might be introduced by such a high proportion of CS multiple consonant strings.

It is also necessary to be sure that there is a great compensating value to other users in removing all letters that represent minimal obscure vowels.

Future articles will consider how CS would affect other users of English print, and also what research suggests would be the advantages and disadvantages of phonemic reforms such as New Spelling 90 and American Spelling, to the different categories of users. The final discussion considers overall what features are required of a spelling reform that might prove the "best fit" to human needs and abilities, and the sorts of experiments and other empirical research that could settle the issues.

TABLE 1.

Simplification Simple
Convention
Harder
Convention
3-4 letter
consonant
strings:
Ambiguous pronunciation:
ALREDY
AR
BAK
BAREL
BEL
BILD 3
EG
I'L 3
LO
OCLOK
PIK
REALY
RONG
TH 23
U 3
WIL 2
TWELV
ANOTHR
EATN
FETHRS
FOREVR
HARDR 2
NEVR
PARKR
SWETR
UNDR
LOOKD 7
RAIND
POPD
SPELNG
CRYD
CHANJE
ASKD
BUMPD
GASPD
WANTD
WORKD
GETNG
SINGNG 2
SITNG
SPELNG
STUFNG 3
STOKNG
AL 12
CUD 4
EVRYWHER
GESS
LOV 5
MATRESS
MOTHR
NEEDD
PUL
SOM
SOMETHING
THER 10
WHER

REFERENCES.

Anderson, Richard, et al. (1985). Becoming a nation of readers: the report of the Commission on Reading. Washington DC: National Institute of Education.

Bradley, Lynette & Peter Bryant (1983). Categorizing sounds and learning to read - a causal connection. Nature. 301, 419-421.

Bradley, Lynette (1980). Assessing reading difficulties. A diagnostic and remedial approach. London: Macmillan Education.

Bryant, Peter & Lynette Bradley (1985). Children's reading problems: psychology and education. Oxford: Basil Blackwell.

Chall, Jean (1983). Learning to read. the great debate. NY: McGraw Hill.

Chomsky, Carol (1970). Reading, writing and phonology. Harvard Educational Review, 40, 87-309.

Chomsky, Carol (1971). Write first, read later. Childhood Education. 47. 296-299.

Downing, John & R. Valtin (Ed's) (1984). Language awareness and learning to read. IL: Springer.

Ehri, L. C. & L. S. Wilee (1985). Movement into reading. is the first stage of printed word learning visual or phonetic? Reading Research Quarterly. 20, 163-179.

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

Fowler, C. A., D. Shankweiler & 1. Y. Liberman (1979). Apprehending spelling patterns for vowels: a developmental study. Language & Speech, 22. 243-252.

Liberman, 1. & D. Shankweiler (1985). Phonology and the probems of learning to read and write. Remedial and Special Education, 6. 8-17.

Morais, J., L. Cary, J. Alegria & P. Bertelson (1979). Does awareness of speech as a sequence of phones arise spontaneously? Cognition, 7. 323-331.

Read, Charles (1975). Lessons to be learned from the pre-school orthographer. In: E. H. Lenneberg & E. Lenneberg (Eds.); Foundations of language development, vol. 2. NY: Academic Press.

Shankweiler, D. & I. Liberman (Eds.)(1989). Phonology and reading disability: solving the reading puzzle. Ann Arbor: University of Michigan Press.

Smith, Frank (1982). Understanding reading. 3rd edition. NY: Holt, Reinhart & Winston.

Stuart, Morag & Max Coltheart (1988). Does reading develop in a sequence of stages? Cognition, 302. 139-181.

Upward, Christopher (1992). Cut Spelling. a handbook to the simplification of written English by omission of redundant letters. London: Simplified Speling Society.

Yule, Valerie (1981). Preparing to read through play. Video of a nursery school. Aberdeen University Television Service. Now available at Aberdeen University Library.

Yule, Valerie (1986). The design of spelling to match needs and abilities. Harvard Educational Review, 56. 278-297.

Yule, Valerie (1991). Reading and Orthography: Spelling and Society. Doctoral dissertation, Monash University.



Note: C. Upward will reply to the above observations about CS in a future publication.

Back to the top.