The Dishonest Opposition to Improving Literacy

American reading education has been a catastrophe for decades. The latest report from the National Assessment of Educational Progress found that just 32 percent of public school fourth graders are “Proficient” readers, and only 61 percent achieved even the “Basic” level, and scores have been nearly this low for the last thirty years.¹    

What has caused this ongoing disaster? One crucial factor is the fact that many educators do not use the systematic phonics method. This method entails systematically teaching children the connections between letters and sounds so that they can learn to connect text to the spoken language that they already know.

This approach gives children who already speak fluent English the key to understanding it in written form. Researchers have found abundant evidence of the effectiveness of systematic phonics. A landmark meta-analysis of this research, the 2000 report of the National Reading Panel, found that according to available studies, “Students taught systematic phonics outperformed students who were taught a variety of nonsystematic or non-phonics programs.”²

Influential arguments against phonics

Many educators, though, seem to think that systematic phonics doesn’t work. Where did they get that idea?

They got it from many of the authorities who are supposed to train teachers and inform them about how best to do their jobs. But although these authorities are esteemed professors, the quality of the argumentation they offer completely undermines any claim that they are giving genuine expert guidance. The arguments they give fail to consider facts that anyone who seriously studies the nature of writing in general and the English system in particular would learn.

One of the most influential of the expert claims is that English spelling is too complex and too irregular to be learned phonically. Many opponents of systematic phonics have made this argument, going back at least eighty years. For example, Frank Smith, one of the intellectual leaders of the anti-phonics Whole Language movement, articulated his critique like this: systematic phonics is impractical because “[i]t is difficult to identify an unfamiliar word on the basis of its component letters because of the complexity and unreliability of rules for ‘sounding out.’”³

Looking only at the very surface, without investigating further, this argument has at least some plausibility. The English system of sound-letter correspondences really is more challenging to learn than that of some other languages, like Spanish.

Spanish has a highly regular spelling system; most of the genuine exceptions are words adopted from other languages. And its system of sound-letter correspondences is pretty simple. Someone learning to read Spanish needs to learn just one letter for most of the sounds and one sound for most of the letters, with only a few additional wrinkles. For example, the sound “ah,” like the “a” in the English word “swan” or the Spanish word “azul” can only be written in Spanish using the letter “a.” And the letter “a” always corresponds to the “ah” sound. Even one of the few wrinkles in Spanish, the “ee” sound, has only two possible spellings: “i” (as in “si”) and “y” (the one-letter word for “and”).

In contrast, as anyone who has learned to read and write it knows, English has common words whose sounds seem not to correspond to their spellings according to a predictable set of rules, e.g., “does,” “been” or “knot.” Phonics opponents seize upon this obvious fact and another widely known fact, that the system of sound-letter correspondences in English is more complex than in Spanish. English really does have many sounds that are represented by numerous letter combinations (e.g., the “ee” sound in “reed,” “meat,” “field,” “eve,” “key” and “happy”) and many letter combinations that represent several different sounds (e.g., “gh” in “through,” “laugh” and “ghost”). The phonics advocate Rudolf Flesch advised teaching some 180 phonetic rules for reading English.

Addressing claims about irregularity and complexity

But the argument loses all semblance of plausibility upon closer examination. To begin with, there’s the observation that adults who have learned to read by the phonics method are not overwhelmed by complexity and irregularity when they try to sound out an unfamiliar word. They’re usually able to come up with, at most, two to three fairly similar candidate pronunciations, which they can easily identify with a spoken word they know. Children learning to read by phonics can often be heard doing the same thing.

Most people who know English realize that its spelling system is far more regular than it initially seems, once they realize that there are recurring patterns behind many apparent irregularities. For example, the “k” at the beginning of “knot” is easy to deal with once one learns that “kn” forms a single phoneme when it begins a syllable, as in “know” or “knife.” Almost no English readers are stumped upon encountering an unusual “kn” word (like “knell”) for the first time. It turns out that once you account for cases like this, the language is mostly regular. Flesch found that 97.4 percent of seventeen thousand common English words are decodable using the rules in his reading program.⁴ Similarly, reading experts Louisa Moats and Carol Tolman argue that only about 4 percent of English words are truly irregular.⁵ This puts the challenge of memorizing the true exceptions into proper context.

We can easily see this by taking a cursory look at the first sentence of a recent front-page New York Times article: “Former President Donald J. Trump glided to victory in Nevada’s Republican caucuses on Thursday, an outcome all but guaranteed because he was the only major candidate on the ballot.”

Of the twenty-eight words in that sentence, twenty-six can be read using typically learned phonics patterns — including the proper name “Trump” and the unusual word “caucuses.” Only two — “was” and “guarantee” — deviate from phonics rules enough to cause any difficulty. “Was” is irregularly spelled, but common enough that a child would learn it early on. “Guarantee” is less common, but its pronunciation is not that different from the “garrantee” or “gwarrantee” that phonics rules would produce. A reader who pronounced it either way could still easily recognize the word.

The irregularities of English are infrequent and easy to overcome by using phonics to generate something very close to a familiar spoken word. They don’t represent a compelling reason to abandon the rules.

What about the claim that English, with its 100-plus rules, is too complex for a child to learn? The reality, which the critics of phonics don’t dare to examine, is that children in other countries learn writing systems that are far more complex than English. For example, we know that Japanese children learn to read a writing system that uses two different sets of characters that operate on different principles. There are ninety-six kana characters, most of which each represent a full syllable, but also thousands of kanji characters (two thousand to three thousand in common usage) adapted from Chinese writing to represent one or more words. For example, 椅 is a kanji character that represents the word for “chair,” and 雨 is one that represents the word for rain. Each kanji often has multiple pronunciations and multiple meanings, and Japanese writing mixes kanji and kana together.

In comparison, learning twenty-six English characters and 180 or so rules for decoding them is easy. The 180 or so rules of the English writing system serve to greatly simplify the thousands of words that one encounters, enabling one to learn how to read most words with, at most, a few years of lessons and practice. Anyone who bothered to look at the different writing systems of the world would learn facts like these. And no one denies that Japanese children learn how to read. It’s hard to see how an expert in the field of reading could honestly conclude that English phonics is too complex to learn, but Japanese somehow is not.

Non-phonics methods make reading far more difficult

It’s also hard to see why, if they genuinely care about the challenge of complexity, phonics opponents advocate “solutions” that greatly exacerbate the complexity of the reading experience. Instead of proposing some alternative way of simplifying reading, they advocate methods that make reading far more complex.

For example, the Whole Language approach involves providing children with texts to read while not teaching them the system of English sound-letter correspondences. This leaves children with no alternative but to try to memorize each word individually and guess those they have yet to memorize. In essence, this means treating each word as an exception and treating every word in English — hundreds of thousands of words⁶ — as though it is irregular, instead of 3–4 percent of the words being irregular.

Whole Language in effect means treating each English word like a Japanese kanji to be memorized or guessed. And while learning kanji takes a lot of work, Japanese children manage to learn it because there are a limited number of characters and kanji, at least, represents a whole word with a single character that can be taken in at a single glance. But English words consist of multiple characters that would be difficult to memorize wholesale, especially the longer words.⁷ And one would need to deal with not two thousand symbols, as in kanji, but forty-two thousand or more of these multi-letter English word symbols (the typical vocabulary of an American adult).⁸ In practice, since it would at best take many years to memorize that many symbols (especially since Whole Language rejects deliberate memorization and practice), that means doing a lot of guessing. And all this to avoid the supposed difficulty of learning the 180 or so rules of English phonics!⁹

To grasp the needlessly, crushingly difficult challenge of learning to read English by the Whole Language method, consider a hypothetical scenario. Imagine an authority figure tells you that you must start “reading” a system that pairs each spoken word with a unique bouquet of flowers. Like words, bouquets have thousands of unique, complex arrangements of their elements. They vary in the number of flowers, in the color and shape of each flower, and the flowers’ relative placement. Think how difficult it would be to try to memorize the visual image of a bouquet for each word. It would be hard enough to deal with twenty or thirty of these, let alone two thousand or forty-two thousand. You couldn’t memorize many and would be left trying to guess many of those you encountered, an equally impossible task. Yet this hopeless experience is akin to what so-called experts say teachers should put children through by telling them to read entire books written in an alphabet that they have not been taught to understand.

The dishonesty of attempts to discredit phonics

Since systematic phonics is by far the best way to simplify English writing and deal with its irregularities, why does anyone think it has a complexity problem? Teachers who’ve done less to study the complexity and irregularity of English might trust experts they think would have the relevant knowledge.

But are phonics detractors true and honest experts? Consider the facts we’ve examined here. True experts in reading should know the mechanics of the English writing system, how it compares to the writing systems of other languages, and how different educational approaches deal with these facts. So they should know that English is less complex than other learnable languages. They should know that phonics greatly simplifies navigating these complexities. And they should know that Whole Language offers an obviously more complex approach. Anyone who concludes that because the rules of systematic phonics are too complex and irregular, students would be better off not learning any system of rules to simplify reading English, is either not genuinely an expert or not seriously engaging with the evidence.   

The originators of the complexity and irregularity arguments demand that systematic phonics be held to a standard that they have no intention of applying to their own methods. They do not raise it as part of a genuine attempt to solve the challenge of teaching reading. They raise it as part of an attempt to discredit a method that they have decided to oppose regardless of the evidence.

Most of the teachers who are still skeptical of systematic phonics today have been deceived by leading figures in their profession who present dishonest arguments like this one. It’s an encouraging development that more teachers have recently learned truthful information about reading education thanks to advocates of systematic phonics. The next step should be to replace the old guard whose deceptive arguments muddled the issue, and bring in new, honest, intellectual leadership for the teaching profession.

Image credit: Evtushkova Olga/Shutterstock.com.

1. “2022 Reading State Snapshot Report: Nation Grade 4 Public School,” National Center for Education Statistics. U.S. Department of Education, 2022.
2. Eunice Kennedy Shriver National Institute of Child Health and Human Development (U.S.), Report of the National Reading Panel: Teaching Children to Read: Reports of the Subgroups (00-4754, Washington, DC: U.S. Government Printing Office, 2000),  2–93.
3. Frank Smith, Reading Without Nonsense (New York: Teachers College Press, 4th ed., 2006), 6.
4. Rudolf Flesch, Why Johnny Still Can’t Read (New York: Harper & Row, 1981), 95.
5. Louisa Moats and Carol Tolman, “English Gets a Bad Rap!” Reading Rockets, accessed March 26, 2024.
6. “How many words are there in English?” Merriam-Webster, accessed May 15, 2024.
7. Also, there are certain meaningful elements that recur across multiple kanji symbols. A skilled teacher of Japanese reading and writing will teach students these components to reduce the amount of raw memorization necessary.
8. The number comes from a 2016 study that did not treat inflected variants as different words — e.g., “walk,” “walks,” and “walked” counted as one word, not three.
9. Many students do not need to be taught all 180 rules explicitly, since they will figure some of them out on their own once they get started in a high-quality program.