Fighting Words.

Why the reading wars aren't over.

Suddenly, everyone's hooked on phonics. After the reading scores of California students dropped precipitously, a bipartisan majority of the state legislature voted to mandate that teachers be trained in phonics. In Texas, the other key state for influencing textbook publishers, Gov. George W. Bush pushed through a similar law. Massachusetts, too, now mandates phonics instruction in every public school. Pro-phonics bills are pending in many other states as well, and phonics products are selling briskly. Two major recent reports-one by a division of the National Academy of Sciences, the other by the Learning First Alliance, which includes the top two teachers' unions and the national PTA- recommend phonics instruction.

The triumph of phonics would be great news, if only it were true. The news accounts may give the impression of triumph, but parents need to read between the lines.

Ever since Rudolph Flesch's 1955 blockbuster Why Johnny Can't Read, proponents of phonics have been arguing that the schools should resume teaching kids to read by connecting the written word to its sound. They believe children should receive systematic instruction in the alphabetical code-in how letters represent the 44 sounds of the English language and how combinations of letters blend those sounds and string them together. Then children will be able to read, and to decode unfamiliar new words, by sounding them out. Before too long, they will be doing this so quickly that they will not even realize that's how they're reading.

In recent years, however, the dominant approach to reading in the schools and, even more, in the teachers' colleges has stressed not phonics but "whole language." The debates over phonics and whole language came to be called "the reading wars" because of their intensity, but it's important to understand that these debates have not concerned how to teach reading. The question has been whether to teach reading at all.

Whole-language theorists such as Kenneth Goodman believe that just as children learn to speak naturally, by exposure to the speech of others, so can they learn to read and write by exposure to literature and "print-rich environments." Goodman, a professor emeritus at the University of Arizona, says that parents and teachers should read to children and encourage them "to experiment with reading and writing," "to build on their natural tendency to make sense of the world around them, including the print around them." Context cues will help children make sense of words, a process Goodman has notoriously called a "psycholinguistic guessing game."

I asked Goodman whether spelling errors in children's early experiments should be corrected. His response: "What's an error? . . . Language is a social invention and it's also personal. My voice isn't yours, my dialect isn't yours. . . . You learn to spell through misspellings. A skilled teacher can look at a child's writing and see that though some of the spellings aren't conventional they show the children's growing skills at the spelling system."

Phonics advocates such as Robert Sweet, founder of the National Right to Read Foundation, describe whole language derisively: "Put somebody in a house and put a hammer in his hand and he'll become a carpenter." Whole language is, however, not a new fad but an ancient heresy. Phonics instruction has not been the norm in American schools since at least the 1930s. Hostility to phonics goes back all the way to Horace Mann, the father of American public education, who inaugurated a tradition among educational theorists in disparaging phonics and its supposedly deadening drills. As Andrew Coulson points out in Market Education, Mann either misunderstood or misrepresented phonics-another tradition that continues to this day.

Influenced by Thomas Gallaudet, the pioneering teacher of the deaf, Mann thought that reading instruction should begin with whole words- with units of meaning, not units of sound. His wife published a reader that taught words by illustrating them, an approach that came to be called "look and say" or "sight word" and to be employed in the vapid Dick-and-Jane books many of today's adults grew up with. Whole language takes this approach, and progressive education's disdain for authority and structure, to a logical conclusion.

Says Goodman, the whole-language theorist, "Learning is something people are uniquely capable of, and when the time comes they learn to respond to the need for written language." In the history of the species, though, that time didn't come for tens of thousands of years. When MIT linguist Steven Pinker wrote a book called The Language Instinct a few years ago, his point was that human beings appear to have a universal grammar, rooted in biological affinity, for spoken language. Only a minority of societies has devised written representations of language, through glacially slow trial and error.

Children are unlikely to recapitulate that process in a few years. Building up a written vocabulary is hard enough. For most of this century, American schoolchildren have been introduced to fewer and fewer words per year because the prevailing method of whole-word memorization cannot accommodate more. Whole language makes the problem worse because now children are supposed to start out reading "quality children's literature" rather than readers designed to acquaint them with everyday words.

No one method of teaching reading will work for all children. But research as well as logic and historical experience suggest that instruction in phonics works better than whole language-both in fostering literacy and in getting kids to understand and enjoy their reading, which comes more easily when kids are not stumbling over every other word. Another research finding is that the best readers rely least on the contextual guesswork recommended by whole language. Much of this evidence was summarized in Jeanne Chall's 1967 book Learning to Read: The Great Debate, updated in 1983; though her findings have been rebutted, they have not been refuted. Federally funded longitudinal studies of schoolchildren and recent neurological research have confirmed and filled out these findings. As Keith Stanovich, a researcher at the University of Toronto, has written, "That direct instruction in alphabetic coding facilitates early reading is one of the most well-established conclusions in all of behavioral science."

A large increase in the proportion of high-school graduates who are illiterate or barely literate has coincided with the eclipse of phonics in this century; more than 40 million Americans are illiterate today. Poor teaching methods may also explain the recent explosion in special ed: Forty-five percent of kids labeled "learning disabled" have reading problems but no deficiencies in intelligence, physiology, or perceptive acuity.

Yet the prevailing reaction among journalists and academics to the mounting evidence of the failure of whole language and the success of phonics instruction has been to call it a draw: Teachers should take a balanced approach combining the best of both philosophies; there's no need to choose. Convergence is the message of most of those stories announcing an end to the reading wars.

When researchers call for balance, they usually mean phonics plus such components of whole language as reading good books to children, paying attention to how they are reading, having books around them, etc.-in other words, as two scholars put it in the Annals of Dyslexia of 1990, "suggestions about instruction that sensible people like your grandmothers and ours would have regarded as truisms." Thus the Learning First Alliance's report calls for a "proper balance" but makes clear that this does not mean "a little phonics on the side" or "stopping from time to time during reading or writing instruction to point out, for example, a short a or an application of the silent e rule."

But a little phonics on the side is exactly what many others who speak of balance mean. A typical formulation can be found in a position paper of the International Reading Association, an influential whole-language redoubt, which says that "phonics instruction should be embedded in the context of a total reading/language program." Goodman himself has complained that "administrators have become nervous about even using the term 'whole language,' substituting euphemisms or dubbing their programs 'balanced.'" It's rare to find a school that does not claim to teach phonics, but whole-language programs such as Reading Recovery remain more prevalent.

An even bigger problem than obstruction and obfuscation by schools is that most teachers have never been trained in phonics themselves. Lynne Cheney, a tough critic of the education establishment, notes, "The war is over at the teachers' colleges and whole language won." Most ed- school textbooks promote whole language and denigrate phonics; a small few are merely neutral between them. Professional associations such as the National Council of Teachers of English help exert peer pressure for whole language.

From their trenches in the ed schools, whole-language theorists have attempted to counter the political tide now running in favor of phonics. A common tack is to reject quantitative research altogether for research that is "not replicable" because it "focuses on a reader in a certain context." With the notion of objective evidence and thus the possibility of accountability out of the picture, the theorists are then free to concentrate on the motives of their opponents.

Opposition to whole language is cast as a plot by the "far right"-the phrase is omnipresent in In Defense of Good Teaching, a collection edited by Goodman and released last year-to exploit the fundamentalist fanaticism and "overwhelming fear of change" of Christian conservatives. Proof of their zealotry is that they say it's important to be able to read in order to have access to the Bible.

The far right's purpose in "frightening and politicizing rural and working-class parents," according to Goodman, is to reduce confidence in the public schools and, ultimately, to privatize them. The paranoia cannot be overstated. He writes, "It is the visible success of whole language, not its weaknesses, that has made it the target of a powerful coalition of forces." Reporters who attack whole language are encouraged in advance and rewarded afterward by the CEOs of their parent publishers. "Efforts are highly coordinated; often one sees identical wording in bills in states far apart geographically." Call the police!

Why have so many moderate, non-fundamentalist parents been joined this far-right crusade? Three whole-language theorists, writing in the March 1 Phi Delta Kappan, suggest that "some parents are unconsciously terrified of their children's dawning independence, as symbolized by their learning to read and write. . . . As long as a child spends most of her time enunciating t's and d's and decoding only synthetic, denatured texts, she will never encounter troubling or dangerous ideas, or begin to think and read for herself."

Such rhetoric is evidence of desperation in the whole-language ranks. As Sandra Stotsky, a researcher at the Harvard Graduate School of Education, succinctly puts it, "They're going off the cliff." The most recent example: In a remarkable article in the latest issue of Reading Research Quarterly, a publication of the International Reading Association, Mark Dressman strives mightily to insinuate that respected scholarly critics of whole language are promoting a racist and capitalist agenda.

The reading wars are far from over. Indeed, in Losing Our Language, Stotsky directs our attention to a second front that progressives have opened. To the whole-language demand for rich, authentic literature has now been added a multiculturalist definition of authenticity. Among the many dolorous consequences is that students learn even fewer standard English words so that they may learn Native American words such as parfleche and travois. Academic goals have been sacrificed to political ones. As Stotsky explains, "The fusion of the anti-intellectualism of the multiculturalists and the antiteaching philosophy of whole language advocates is an educationally deadly combination in the elementary grades."

Still, Robert Sweet, the phonics booster, takes an optimistic view: "I would say although the mop-up crew hasn't finished yet, I believe we have truly turned a corner." With fewer than 15 percent of American children learning phonics, by his rough estimate, that mop-up crew has a lot of work ahead. Members of the far-right conspiracy to dismantle the public schools, meanwhile, can take heart that the Democratic party of California and the linguistics department at MIT have joined them.