Comparing reading education in English and other alphabetic languages

Many will claim that failure to learn to read in English is due to a failure of the students to apply themselves properly to the task, or to various societal problems, or to inadequate teaching. The lack of the alphabetic nature of English is the real culprit, however, in this sense: (1) although most students can learn to read English, it requires significantly longer to learn to read than in alphabetic languages Frank C. Laubach, Teaching the World to Read (New York: Friendship Press, 1947), p. 103 and 108; Sanford S. Silverman, Spelling For the 21st Century (Cleveland, Ohio: self-published, 2003), pp. vi-vii. This is the Preface by Steve Bett, Ph.D., Editor, Journal of the Simplified Spelling Society; Rudolf Flesch, Why Johnny Can’t Read—and What You Can Do About It (New York: Perennial Library, 1983), and (2) an unknown but substantial number of students are so resistant to the lack of logic and inconsistency of English spelling that they cannot learn to read without the extensive help of a one-to-one tutor for a year or more. Different students have different abilities. Some people—-particularly young girls—-are very good at memorizing. Young boys and most adults prefer to learn new things by comparison to what they already know-—i.e. they like to learn by logic. Unfortunately, the lack of logic is a complete “turn off” to some of the most intelligent students who are looking for logic in what they learn. Sir James Pitman, Alphabets and Reading (New York: Pitman Publishing Company, 1969), pp. 38, 54, and 161;

Ever since alphabets were first invented, alphabetic languages have used a letter or letter combination to represent the sounds in the words. The easiest alphabetic languages to learn are those that use one specific grapheme (a single letter or a specific letter combination) for each specific phoneme (the smallest sound in a language or dialect used to distinguish syllables or words). English uses at least 1,768 graphemes to represent the 40 English phonemes. Although these 40 English phonemes could be spelled with 26 single letters and 14 digraphs (two letter combinations), they are spelled with all 26 single letters in the alphabet and at least 153 two-letter graphemes, 98 three-letter graphemes, 14 four-letter grapheme, and 3 five-letter graphemes, for a total of at least 294 different graphemes. This is less than the 1,768 mentioned above because every English phoneme is spelled with more than one grapheme. The number of spellings of the phonemes varies from at least four (for the TH phoneme in words such at this) to at least 60 spellings of the U phoneme in words such as nutty-—which is exactly what English spelling really is.

Some phonics spelling advocates claim that English is more than 80 percent phonetic. This is only possible, however, if you allow more than one grapheme for a phoneme. If you allow only one grapheme for every phoneme as logic and ease-of-learning demands, English is only a little more than 20 percent phonetic. The problem is that there is absolutely no way of knowing which word is spelled phonemically and which is not. There are no invariable spelling rules in English—-every rule has exceptions and some of the exceptions have exceptions. Edward Rondthaler of the American Language Academy in a personal letter to Bob Cleckler, author of Let’s End Our Literacy Crisis, stated, “A 1986 round table of British linguists called by eminent scholars to discuss the underlying pattern of English spelling concluded, not surprisingly, that only one rule in our spelling is not watered down with exceptions: No word in English ends with the letter V.” Since Webster’s Ninth New Collegiate Dictionary includes the words rev and spiv there are therefore no invariable spelling rules.

In addition, Dr. Diane McGuinness’ book Why Our Children Can’t Read explains the complex logic that is required to learn to read English. Unlike many alphabetic languages, there are tens of thousands of different syllables in English, with sixteen different syllable patterns in English: (C=consonant, V=vowel) CV, CCV, CCCV, CVC, CCVC, CCCVC, CVCC, CVCCC, CCVCC, CCVCCC, CCCVCCC, CCCVCC, VCCC, VCC, VC, and V. There are two or more syllables in most English words. Each syllable can have one of the sixteen syllable patterns. If each vowel and each consonant in each of these patterns consistently represented the same phoneme (one-to-one mapping), there would be nothing in the logic of these syllables that would be beyond the abilities of most four- or five-year-olds. But they do not. English spelling also has one-to-many and many-to-one mapping. This requires a type of logic that most children do not develop until they are eleven or twelve years old.

The types of logic required for one-to-many and many-to-one mapping are: (1) the logic of “classes” (categories where objects or events that are similar are grouped) and “relations” (where objects share some features but not all features, e.g. all poodles are dogs, but not all dogs are poodles) and (2) “propositional logic,” which involves combining both the classes and relations types of logic. This requires the ability to think of the same item in more than one way at the same time. These combinations require the use of relational terms such as “and,” “or,” “not,” “if—then,” and “if and only if” in formal statements of propositional logic, e.g. if an H follows the T, then say /TH/ as in thin or then; but if any other letter or no letter follows the T, then say /T/ as in top or ant.

The eyes of the fluent reader skip easily over a multitude of traps for the beginner. Most fluent readers who learned to read as a child have long since forgotten the difficulty they had in learning. Due to the difficulty of English spelling there are basically three ways of learning to read (a more precise explanation of the time required for learning is in the section “Time required to learn to read English vs. other alphabetic languages” below):

    Young children can learn (a) the very limited number of common English words that are phonemically regular (one-to-one mapping) either by phonics teaching, by whole word, or by whole language teaching, (b) learn a few hundred sight words (most of which are almost totally unphonemic) by the whole word or whole language method, and (c) if being taught by the phonics method, memorize—-without understanding the logic involved—-the hundreds of many-to-one and one-to-many phoneme-to-grapheme mappings. Then—-with constant practice in reading that extends past the age when they can understand the logic required—-using their knowledge of phonics, they learn one-at-a-time all 20,000 or more of the words in their reading vocabulary required to be a fluent reader. This process requires at least two-and-one-half years to give young children the foundational knowledge and confidence to continue reading long enough to become fluent readers and—-for most students-—extends past their eleventh birthday.
    Begin learning to read after age eleven or twelve—-when they can understand the logic involved—-and spend at least one to one-and-one-half years learning strictly by phonics. Then with additional reading experience, learn one-at-a-time all 20,000 or more of the words in their reading vocabulary required to be fluent readers. This method, of course, is totally impractical. Children should begin learning to read at the age of four to six years of age when they are best able to learn to read. Furthermore, students of almost all other school subjects need the ability to read to be able to do the classwork, homework, and testing required to learn each subject. Delaying this instruction would place students at a serious competitive disadvantage with students of almost every other nation.
    If taught only by the whole word or whole language method and if they do not learn phonics—-on their own or with help outside the classroom—-they can learn 2,000 or so words (or perhaps as many as 5,000 or a little more if they have a superb memory) and join the ranks of the functionally illiterate (see U.S. statistics on functional illiteracy below). Those who can only read 2,000 to about 5,000 simple words they learned in the first four grades in school have difficulty competing in our increasingly complex and competitive world as well as they should.

As stated in the Whole Word method section above, the human mind cannot remember more than about 2,000 symbols,.Students of whole-word-only or whole-language-only teaching cannot become fluent readers unless they also learn phonics-—either on their own or with help outside the classroom—-as a tool to help them “decode” new words. When phonics knowledge or contextual clues do not reveal the word they must consult a dictionary or ask someone.

The problem is that learning words individually until one knows enough words to be able to “get by” in life as well as they should—-as well as is required in our increasingly complex society-—takes much longer than is required in alphabetic languages. Although some phonics advocates have recently designed much-improved methods of teaching phonics, learning to read in these programs still requires a year or more longer than a perfect one-grapheme-for-one-phoneme spelling system.

The problem with whole word (or whole language) only type of teaching is that most of today’s adults cannot hold an above-poverty-level wage job if they only know 2,000 common words (or perhaps as many as 5,000 if they have a superb memory) they learned by sight in the first four grades in school. Although there are several ways of determining functional illiteracy, due to the fact that very few U.S. adults can afford to accept a job that pays less than they are capable of earning, the average yearly earnings is the best method—or certainly one of the best methods—of determining functional illiteracy. The most comprehensive study of U.S. functional illiteracy ever commissioned by the U.S. government proves that 46 to 51 percent of individual adults earn significantly less than poverty level wages. This percentage of families is not in poverty only because most families have more than one employed adult and most low-income families receive governmental and charitable assistance.Irwin S. Kirsch, et al., Adult Literacy in America (Washington, D.C.: U.S. Department of Education, Office of Educational Research and Improvement, 2002), pp. xvi, 63, 65, and 66, available for free inspection and download from http://nces.ed.gov/pubs93/93275.pdf. This is a 199 page report on the most comprehensive study of U.S. adult literacy ever commissioned by the U.S. government. It consisted of lengthy interviews of 26,700 U.S. adults. The interviewees were statistically balanced for age, gender, ethnicity, and location (urban, suburban, or rural) in a dozen states across the U.S. to be representative of the U.S. population as a whole. It used statistically rigorous methods to ensure accuracy and was reviewed by an outside testing agency before it was released. No other persons had access to the study before it was released. The same group who prepared this study did a less statistically rigorous study with a slightly smaller database of interviewees and issued a report in 2003 that showed little or no statistically significant improvement from the earlier report. It is available at http://nces.ed.gov/NAAL/PDF/2006470.PDF.