Teacher Alert!

Phonics Fads Sweep Nation’s Schools

By Harvey Daniels, Steven Zemelman, and Marilyn Bizar

The so-called “Great Debate,” a 40-year war of words pitting phonics-centered instruction against more holistic approaches to reading, has reached a furious new intensity this year. The opposing sides constantly launch research studies, invective, and accusations at each other, making primary teachers feel even more nervous about whether they are doing the right thing for the beginning readers in their care. The most recent version of this battle has erupted around the repudiation of whole language in California and the ascendancy of “decodable text,” the latest silver bullet in the war on illiteracy.

Based upon a dubious and un-replicated study of seriously disabled readers in Houston (Foorman et al, 1997), instructional mandates are being issued by state legislatures and education departments across the country. Even though the Houston findings are highly controversial, and at most concern only the lowest-achieving 20% of young readers, their “decodable text” methods are now being required for all beginning readers in California, Texas, North Carolina, and Illinois, with many more states undoubtedly to follow. A bill now before the U.S. House, dubbed the “Reading Excellence Act,” would make decodable text the law of the land.

We’ve written elsewhere about the California debate and the questionable kinds of “scientific proof” being used to support various partisan and extreme reading programs (Daniels, 1996; Zemelman, Daniels, and Hyde, 1998). Here we want to offer some perspective on the latest phonics flap for educators and parents who are just trying to do the right thing for children. We think that everyone needs to take a deep breath and let the hyper-phonics bandwagon pass on by. After all, our profession has 60 years of solid research on effective reading instruction. Honoring this deep knowledge base, teachers must resist precipitous and ill-founded reversals of mainstream reading practices.

The current commotion is so severe that we should probably begin by reminding ourselves that phonics is only one ingredient in a balanced K-12 reading program. To begin with, phonics has nothing to do with reading instruction much above second grade level (for most kids, first grade), except for students with specific learning disabilities or other special needs. Phonics is a tool for unlocking the meaning of print, and once the sound-symbol code is learned, it is learned. We would no more teach phonics as part of, say, a fifth-grade “reading” program, than we would teach the alphabet to high schoolers in the name of “writing.” We want students to read and write whole texts just as soon and as much as they can.

Even during the first couple years of school, phonics is still only one part of an effective reading program. Lessons in word analysis must share time with other key activities that help young readers grow, including hearing good literature read aloud, experimenting with writing, and talking about the ideas found in books. After all, no child ever fell in love with books, or took that critical step into life-long reading habits, by repeatedly enunciating “t” and “d” phonemes. All the skills in the world can never make the connection of the heart that active literacy requires.

Most reading experts agree that phonics, as one of several necessary daily activities, should occupy no more than a fraction of total reading time. Further, even given this limited role in a narrow span of grades, we must remember that the majority of children, 60% to 80% by most estimates, will learn all the sound-symbol strategies they’ll ever need from ample natural reading experiences and from practicing spelling in their own writing. Indeed, many children internalize the alphabetic principle at home, before ever coming to school (Weaver, 1994). All this puts the role of phonics in perspective: in a responsible and balanced K-12 reading program, phonics amounts to less than five percent of the instructional efforts that teachers make.

Phonics may be a proportionally small element of an overall reading program, but it is certainly a key part. Phonics is a gateway skill; if you don’t crack the alphabetic code, you can’t read. Happily, most kids do break the code, and do so early — many through rich, informal exposure to reading at home. However, a fraction of children do not acquire sound-symbol awareness in this embedded, automatic way, but they can learn and use these strategies when they are taught directly in school. To be certain that we give all children what they need, without over-teaching phonics to some and under-serving others, wise teachers carefully balance time, activities, and student groupings.

A consensus already exists among mainstream reading educators about what ingredients need to be balanced, and in roughly what proportions (Flippo, 1998; International Reading Association, 1996; Braunger and Lewis, 1997). Based upon that consensus, here is one way that a first grade teacher might divide her language arts time:

Possible Time Allocations for First Grade Reading

  • 20% — The teacher reading good literature aloud; students discussing ideas in text.
  • 20%— Phonics and word study (word sorts, word walls, word families, spelling patterns, using developmental spelling).
  • 20% — Shared reading (teacher-guided lessons, including language experience stories, big books, other literature).
  • 20% — Independent reading at child’s fluency level (wordless books, picture books, or chapter books as appropriate).
  • 20% — Writing (journal-keeping, stories, responding to literature; using age-appropriate developmental spelling and drawing).

In this model, explicit, separate instruction in sounds and words constitutes about one-fifth of the reading and language arts time — a fractional but important component. Nor are the recommended phonics activities dull, mechanistic drills, but lively strategies like word sorts (when kids arrange words on cards into different kinds of sets), word walls (where the students create complex inventories of words, which they can group and regroup by what they mean, how they sound, or other relationships), and personal dictionaries (in which students record words as they learn how to say, spell, and read them), along with a variety of whole-group choral singing and chanting games (Cunningham, 1995; Moustafa, 1998). Further, sound-symbol relationships also get explicit and embedded attention any time that children are spelling words in their own writing, as well as when the teacher draws their attention to sounds, letters, or syllables as part of a guided reading lesson.

Once young children are reading fluently, which is to say consistently applying their phonetic knowledge in making meaning from print, phonics instruction should be over. Children will continue to practice their newly acquired phonetic understandings by reading age-appropriate texts and by writing using developmental spelling. The portion of the day previously devoted to word analysis in the form of phonics can now be used for vocabulary study. Further phonics instruction may well be needed for a small fraction of children who still need help and should certainly be offered to them individually or in small groups but not as whole-class phonics lessons that would waste most classmates’ time.

None of this seems so very radical or controversial. Indeed, we sometimes think that “The Great Debate” should be renamed “The Fake Debate.” No responsible educators — including the most dogmatic whole language zealots — have ever advocated withholding phonics instruction from children. The genuine questions are the same as they have always been: how should we teach phonics, for how long each day, to which students, and in what groupings? And how does phonics fit into the larger picture of a complete, well-balanced literacy education?


The latest phonics fad to sweep the nation, and the subject of hyperbolic claims as this article goes to press, is the idea of “decodable text” — the notion that young children should be exposed only to text specially synthesized to include limited sets of sounds which children have previously been systematically taught (NICHD, 1997). In the Houston study, for example, first graders were placed in the “Open Court” commercial basal program, which requires children to master 70 separate phonics units be- fore ever reading any real books — a task that takes up nearly the first half of the school year. In fact, decodable text labels traditional children’s literature as contra- band: neither Mother Goose nor Goodnight Moon qualifies as decodable, not- withstanding their generations of previously unharmed young readers. Indeed, in some Sacramento schools where decodable text has recently been imposed, teachers have been required by school officials to literally lock away all children’s books previously used in their classrooms: fairy tales, poetry, novels, picture books, big books — the works.

The doctrine does permit real children’s literature to be read aloud, but only if teachers and parents hide the book from their children. The theory says that looking at real print before mastering the 70 official phonics skills would confuse the kids. Taken to its logical conclusion, decodable dogma means that responsible parents of preschoolers must shield their youngsters’ eyes from virtually all environmental print, including such subversive, undecodable texts as “McDonalds,” “Sesame Street,” or “Disney World.”

Well, the theory of decodable text obviously leaves its advocates with some serious explaining to do. Ever since written language was first invented around 2500 B.C., children have been learning to read just fine by sitting in their parents’ laps, looking at real print. And so far, archaeologists have unearthed no clay tablets of “decodable text” from the ruined schoolhouses of ancient Sumeria. Now we are told that some psychologists in modern-day Texas have suddenly discovered that no one in human history learned to read “the right way.” But is it even slightly plausible that if a “right way” to teach reading were suddenly discovered in 1998, it would involve dumbing down stories in order to present children with the textual equivalent of baby talk? Just think what would happen if the decodable text theorists took over oral language acquisition! They’d probably make parents stop using real language in front of their children and require them to babble instead. Decodable talk, you know.


So where do crazy ideas like this come from? The current contagion of decodable text comes from ambitious researchers and demagogic politicians vastly over-generalizing the preliminary results from some limited research on a narrow subset of disabled readers in Houston. The objectivity of the researchers and the validity of their methodology have been widely challenged (Allington, 1998). Their procedures for assigning students to experimental groups, calculating scores, and reporting results have incited wide debate and considerable condemnation. The investigators, long on record as pro-phonics partisans, even claim to have scientifically tested the validity of whole language by personally selecting and training certain teachers to mimic its approaches under their monitoring.

Much more problematically, the studies focused on a very small and atypical population — significantly disabled and disadvantaged readers — rather than “normal” beginning readers. Many of the experimental strategies were used in supplemental, pull-out tutoring programs, rather than in regular classroom instruction. In the end, some of the study’s phonics-intensive interventions seemed to work better than the researchers’ special version of whole language, as measured by the word reading tests favored by the researchers. Still, even with all

these factors arranged to favor phonics approaches, the “whole language” group matched the scores of phonics groups on comprehension subtests — arguably the only genuine measure of real reading. The credibility of the Houston research has been further compromised by the authors’ refusal to submit their testing data for peer review, which has been requested by several nationally prominent reading researchers.

None of the flaws in the Houston study prove that phonics isn’t important. Indeed, it is entirely plausible that delayed or disadvantaged readers may benefit greatly from extra attention to sound-symbol relationships. However, even if all its methodological flaws were resolved, the Houston research would still offer no reliable guidance about instructional practices for the great majority of children. It is scientifically and ethically indefensible that the Foorman study is now being used to justify unbalanced reading instruction in regular classrooms, with mega-doses of phonics for everyone, regardless of need, and the withholding of real reading practice from all children, regardless of their love of books or their ability to read.

One colleague of ours who teaches in the Chicago area confided that she is now required to teach one hour of systematic, intensive phonics every day to her 28 third grade students, all of whom already read at or above grade level. Texas governor George W. Bush, an enthusiastic convert to decodable text, has just awarded the Foorman team a half-million-dollar contract to design a statewide assessment to identify students who need — guess what? — more phonics.

The “decodable text” fad reminds us that the seemingly straightforward segment of the public school curriculum called reading is perennially afflicted with crazy ideas, crackpot fads, instant cures, political vendettas, professional posturing, and profiteering juggernauts. Nor, unfortunately, does this tangled state of affairs seem likely to change in the immediate future. This leaves all of us raising or educating children to find our own way through the adamant claims and counterclaims. In the end, wise teachers and responsible school districts will ground their reading instruction in a deep understanding of children and in the recommendations of reputable, enduring professional organizations which are guided by the preponderance of responsible research and have shown that they have children’s best interests at heart.

The authors are the full-time faculty at the Center for City Schools at National-Louis University in Evanston, IL.


Allington, Richard L. and Haley Woodside-Jiron. 1997. “Adequacy of a Program of Research and of a ‘Research Synthesis’ in Shaping Educational Policy.” National Research Center on Literature Teaching and Learning. Report Series 1.15. http://cela.albany.edu/allington/ index.html.

Braunger, Jane and Jan Lewis. 1997. Building a Knowledge Base in Reading. Portland, OR: Northwest Regional Laboratory.

Center for the Future of Teaching and Learning. 1997. “Thirty years of NICHD research: What we now know about how children learn to read.” Effective School Practices, 15, 33-46.

Cunningham, Patricia. 1991. Phonics They Use. New York: HarperCollins.

Daniels, Harvey. Spring 1996. “Whole Language: What’s the Fuss?” Rethinking Schools.

Daniels, Harvey and Marilyn Bizar. 1998. Methods that Matter: Six Structures for Best Practice Classrooms. York, ME: Stenhouse.

Foorman, B. R., Francis, D. J., Beeler, T., Winikates, D., & Fletcher, J. M. 1997. “Early interventions for children with reading problems: Study designs and preliminary findings.” Learning Disabilities: A Multi-disciplinary Journal, 8, 63-71.

Flippo, Rona. December, 1997. “Sensationalism, Politicism and Literacy: What’s Going On?” Educational Leadership.

International Reading Association and National Council of Teachers of English. 1996. Standards for the English Language Arts. Newark, DE: International Reading Association.

Moustafa, Margaret. 1998. Beyond Traditional Phonics. Portsmouth, NH: Heinemann Educational Books.

Weaver, Connie, 1994. Creating Support for Effective Literacy Programs.

Portsmouth, NH: Heinemann Educational Books. For current updates, visit www.heinemann.com.

Zemelman, Steven, Harvey Daniels, and Arthur Hyde. 1998. Best Practice: New Standards for Teaching in America’s Schools. Second Edition. Portsmouth, NH: Heinemann Educational Books.