[Journal of the Simplified Spelling Society, J31, 2002/2, pp1,2]
[Steve Bett: see Journals, Newsletters, Personal View, Web link.]

Editorial.

What is wrong with a logographic writing system?

'English orthography turns out to be rather close to an optimal system for spelling English'
Chomsky, 1968, 1970
.

Several scholars have argued that the English writing system is close to ideal and that there is nothing wrong with a logographic writing system - or as they refer to it - a historical-morpho-lexical writing system. [Stubbs, 1980; Taylor, 2000; Davidson, 1999; Chomsky, 1968; Chomsky, 1970]

I have said that there is nothing wrong with a logographic writing system such as Chinese or the 60% logographic system such as the English writing system, other than being difficult to master, expensive to support, and probably too difficult for about 30% of the population.

Most Anglophone children take three years to reach a literacy standard that children in languages with relatively consistent spellings can reach in one. P. Seymour, 2001.

I also think that requiring non-native speakers to learn a new Britannic shifted vowel alphabet, which only works 40% of the time, will limit the level of mastery and the spread of English as a world language. Many non-native speakers of English who have picked up enough words to carry on a conversation will not take the time to learn traditional English spelling [tradspel]. See Gogate's article on Globish or Global English, p. 11.

Apologists for retaining tradspel, including many advocates of explicit phonics, suggest that with better teaching, all of the alleged difficulties involved in learning a partially logographic system would vanish.

California WL (whole language) teachers routinely accept a 25%, failure rate (Boden, 2002). Phonics teachers often claim that they can get the failures below 10%. Some say below 1%. If they are right, then there might not be much point in regularizing English since 10% is the typical failure rate reported in Italian and Spanish elementary schools. There are cases where no child was left behind in an individual classroom. There is no evidence that this success can be generalized to the nation at large.

Most spelling reformers would be happy with a writing system as consistent as Spanish or Italian and do not think that the impact on teaching success would be much different than found in these countries. My expectation is that 2 years of schooling could be saved and that the percentage of those who fail to achieve full literacy could be reduced from around 30% to around 10%.

Dr. Richard Venezky (author of The American Way of Spelling) and other scholars scoff at such expectations. "The claims that we lose one to two years of education because of spelling irregularities or that international business is hampered by the same cause are quite hallow and are rarely bolstered by any empirical evidence." (Savant, p. 88) [see Campbell's references in JSSS30 for the evidence].

My only doubts are with respect to the level of phonemicy that is required to make the writing system significantly easier for those who struggle to achieve more than a 4th grade reading level. Is it enough to move from 40% to 75%? Or do we have to match the phonemic regularity of the Spanish and Italian writing systems to reap the benefits?

Types of Writing Systems:

While actual writing systems tend to be mixes [see Shaffer's article, p. 31], there are 3 basic types:

1. Ideographic or Logographic.


The marks represent an idea or whole word. A symbol stands for a particular lexical unit [e.g. a morpheme]. It does not represent the phonetic shape of the morpheme.

The best known system of this kind is Chinese where a character may correspond to widely different pronunciations but will always have the same meaning.

English uses some logograms such as the number system. "4" means the same in English, Spanish, and French, but it will not be pronounced the same. [mobile phone] Texting may use "4sale" but this use of 4 as a phonogram would be pronounced "kwaa-tro saal-ley" in Spanish. Quatro sale

The abbreviation "i.e." [id est] will be understood across different languages but will not be pronounced the same. To the extent that the symbol has a shared meaning without a shared pronunciation, it is logographic.

2. Syllabic.


In a syllabic writing system, there is a definite relationship between symbol and sound that is lacking in logographic systems. Japanese has a partly syllabic writing system. Hotsuma, an ancient Japanese writing system, is classified as alphabetic but it has the look of a syllabary. A single block can contain a vowel and several consonants. "about" would use two letter spaces instead of five: e.g. [a][bout] Korean has a similar structure and may share a common Indian origin.

3. Alphabetic.


In a pure alphabetic system, each letter would correspond to one and only one sound. As explained in Ralph Emerson's article [p. 4], writing systems may start out as alphabetic or phonemic but they never stay that way. The pronunciation of words will change over time. The only way to preserve the alphabet is to periodically respell the deviant words. This is the way that Spanish and Italian have retained a high level of phonemicy and an alphabetic regularity of over 85%.

While words have been respelled in English, this has not been done systematically. The result is a writing system that is arguably only 40% alphabetic today. While the new conventions added after 1066 by Norman scribes did not help, most of the damage was due to the failure to respell words affected by the great vowel shift in the 15th Century.
Letters, the most useful invention that ever blessed mankind, lose a part of their value by no longer being representative of the sounds originally annexed to them. The effect is to destroy the benefits of the alphabet.
- Noah Webster
D.S. Taylor (a scholar from the University of Leeds, UK) argues that spelling reformers are wrong in trying to restore the alphabet and regularize spelling. Many languages (French, Hebrew, Greek, etc.) have some words that are spelled historically rather than as currently pronounced. [Barr, 1976, 81]. The list includes languages said to have highly phonemic writing system such as German and Arabic. (Cf. Shaeffer, p. 31)

Taylor says that there is no reason for the writing system to reflect speech. There is no reason why one independent symbol system should be judged by the standards of the other.

Taylor says, "Since an alphabetic writing system can only reflect a particular dialect, the first question that would arise is whose speech?" Taylor then digresses into listing the problems of phonetic spelling which he claims would have to be revised every two or three years.

Nobody recommends phonetic spelling for a writing system. What is needed is a broad phonemic or diaphonic writing system based on broadcast English.

Such a system would have to be revised at the same rate as the pronunciation guide in the dictionary: about every ten or twenty years. This does not affect that many words since we are talking about broad approximations. The shift would have to be a major one to warrant a change in the pronunciation guide spelling.

Taylor observes, words are not always pronounced the same way in different contexts within the same dialect. The word <and> for instance, has 11 pronunciations.

This too is a kind of a pseudo problem. What we spell in a reform spelling, is either an over-pronounced segmented word [citation spelling] or an abbreviation. "And" has three spellings: and, ənd, or n. Most reform writing systems keep the traditional spelling of this word. Spanglish uses /ənd/ which is spelled <and>. Shavian uses <n>.

Taylor wants to preserve the morphemic regularity of written English. Valerie Yule agrees and incorporates many morphemic regularities in her reformed writing systems. Joe Little, J29 2001, showed that ALC SoundSpel has more morphemic regularity than tradspel.

The Saxon Alphabet reform preserves many morphemes because the conditions of respelling are so high. If the tradspel word can be understood as written, it is not respelled. *photograph and *photography are not changed in a Saxon reform. However, the Saxon Spanglish pronunciation guide spelling is foatagraef and fotaagrafy corresponding to IPA: /'foutəgræf/ /fə'ta:grəfi /

Conclusion:


So what is wrong with the traditional 60% logographic spelling? Nothing as long as you are willing to waste time and money on teaching it and can accept a high percentage of failures. A majority of the 70% who learn to read and write cannot spell with confidence because they have yet to over-learn the spelling of the whole words.

Taylor argues that a historical-morpholexical writing system is just as easy to learn as an alphabetic one. Phonics advocates claim that everyone would learn to read and write if their methods were widely adopted. There is no hard evidence to support either of these claims.

Spelling reformers may exaggerate the benefits of removing half of the irregularity in English. The writing system may need to be as regular as Spanish before the advantages of consistent spelling begin to show up. However, the results of cross cultural studies of orthographic regularity [e.g., Seymour, P. New Scientist, 2001, p.18f and Margaret Harris & Giyoo Hatano (1999) Learning to Read and Write: A Cross Linguistic Approach, Cambridge University Press] seem to support the reformers position more than the position of the apologists for historical spelling. - Steve Bett.

References:

Chomsky, Carol. 1970. 'Reading, Writing, and Phonology.' Reprinted from the Harvard Educational Review 40:2. In Smith, Frank, ed., 1973. Psycholinguistics and Reading. New York: Holt, Rinehart & Winston, Inc.

Chomsky, Noam & Halle, Morris. 1968. The Sound Pattern of English. New York: Harper & Row.

Davidson, Keith. 1999. Breaking the Spell. English Today, Vol. 15, No. 3, July, Cambridge University Press.

Stubbs, Michael. 1980. Language and Literacy: The sociolinguistics of reading & writing, Ch. 3, 'Some principles of English spelling.' London: Routledge.

Venezky, Richard. Quoted in Savant, M.V. 2001. The Art of Spelling. p. 88f. Similar statements are made in Venezky's The American Way of Spelling. 2000.


Back to the top.