Reading Is Anything But Natural

Published July 1, 1999

Must children be taught phonics as part of learning how to read? Or do they learn to read in a natural process, simply by being exposed to the “whole language” of written literature?

With an estimated 44 million adult Americans who are dysfunctional readers, and with four in 10 U.S. fourth-graders who lack basic reading skills, it’s clear that the prevalent method for teaching reading–the “whole language” approach–is not working. To understand why, we turn to research psychologist G. Reid Lyon PhD, the nation’s leading expert on how children learn to read.

Lyon is chief of the Child Development and Behavior Branch within the National Institute of Child Health and Human Development, which in turn is part of the National Institutes of Health. In that position, Lyon is responsible for research programs in reading development, reading disorders, learning disabilities, and other related language disorders. In addition, he is responsible for communicating NIH scientific discoveries concerning the health and education of children to the White House, the U.S. Congress, and other government agencies.

Lyon first became involved in reading research in 1975, after being trained as an experimental psychologist and spending two years teaching third-grade students and children with severe learning disabilities. His experiences led him to ask three basic questions about reading:

  • What does it take for children to learn to read?
  • What is happening when children don’t learn to read?
  • What can we do to help children learn to read?

Prior to joining the NIH in 1991, Lyon served as an associate professor of neurology at the University of Vermont from 1983 until 1991. He also served on the faculties of Northwestern University in Evanston, Illinois, and the University of Alabama-Birmingham. Lyon recently spoke about his research with School Reform News Managing Editor George Clowes.


Clowes: What is going on mechanically when I read?

Lyon: Even though it’s relatively unconscious, there are a lot of mechanical things going on. First of all, you are going to scan and fixate on the letters of the text and, even though you’re an expert reader and do it very quickly, the eye movements that we study tell us that the eye does move across every word. People don’t skim, but the eye actually fixates on most words and, as quickly as that happens, you also translate those visual symbols into sound.

Clowes: So the eye momentarily settles on each word?

Lyon: Yes. Surprisingly, and in contrast to what conventional wisdom has suggested in the past, expert readers do not use the surrounding context to figure out a word they’ve never seen before. The strategy of choice for expert readers is to actually fixate on that word and decode it to sound using phonics. That’s an important finding because for the past 20 to 30 years, a number of teaching methods have been developed on the assumption that you don’t need to learn how to decode letters to sound or speech to sound–that is, you don’t need to learn phonics.

These methods say you use the holistic or surrounding context to predict unknown words, but we have found that’s clearly not the case. No more than 20 percent of words in the text can be predicted from the surrounding context. If we insert a word you’ve never seen before–a nonsense word–you will decode it.

Clowes: With Dr. Suess, isn’t it much easier to decode the words as you go along than to memorize each individual nonsense word?

Lyon: It’s literally impossible for a reader to memorize the visual configuration of every word. That’s why our alphabetic language–with 26 letters that correspond to about 40 to 44 sounds–provides the reader with enormous capability to crack open words they haven’t seen before. You don’t have to memorize every word.

Clowes: What is it that enables me to derive meaning from what I read?

Lyon: One of the most critical skills in reading is to bring the words off the page very rapidly. That’s another mechanical skill. The longer it takes someone to recognize word patterns and decode them, the less energy available to relate what they’ve just read to what they already know.

Comprehension entails comparison. As you read, you bring the vocabulary off the page in its textual format and compare whatever is being said to what you already know. If you don’t have the background knowledge, comprehension is very difficult.

For example, if you’re reading a physics text and you’re not a physicist, even though you can decode the words and bring them off the page, it’s not going to mean very much to you, because you don’t have the vocabulary understanding and experience with the concepts.

Clowes: Could you explain the steps involved in learning how to read?

Lyon: We now have a very clear understanding that reading is not a natural process. If it were, we wouldn’t have so many youngsters having difficulty learning to read, and we also would have more cultures on this planet with a written language as well as an oral language.

A number of building blocks have to be mastered as one comes to the written word. A major prerequisite for learning to read an alphabetic language like English is to understand that the words we use in our speech are actually composed of individual sounds. While that may seem like common sense, it’s very difficult for some children and adults to get that. As I speak, the words coming out of my mouth are obviously composed of different sound elements, but your ear never hears them.

If I were to say to you the word “cat” and ask you how many sounds were in there, you’d probably say three: “c” – “a” – “t.” But, because you and I have to communicate rapidly and efficiently, the minute I say “cat” to you, the “a” and the “t” sounds fold right up into the initial consonant, so it comes by the ear not as three sounds in sequence, but as one sound. The brain has to tear those sounds apart but the ear never hears them. So, we have a written language that’s based upon sounds the ear never hears, and we build our written language by coding these written symbols to their individual speech sounds.

Clowes: So the sounds are like atoms, and the words are molecules.

Lyon: Exactly. And they come by the ear as a burst of acoustic energy, not as individual sounds. The individual sounds are called phonemes, which beginning readers learn before proceeding to text. But those squiggly lines on the page really don’t mean anything until the brain hooks them to sounds and then links those sounds together to form words. As youngsters are learning to read, they’ll use this phoneme awareness to develop what’s called the alphabetic principle, which is an understanding that the print in front of you is, in fact, linked to various sounds.

As a child’s coming along, he’s never seen the word “bag” before, and what he has to learn is that the “b” written symbol goes with a sound like “b” and the “a” goes with a sound like “a” and so forth, and that’s how a child learns to decode the sound.

Clowes: The linking of sounds to letters is the phonics part?

Lyon: Right. But that is not the same as phoneme awareness. When we’re assessing children to determine how robust they are with phonemic awareness, we always do that without any print present. For example, I may ask a five- or six-year-old to give me some words that rhyme with “cat.”

That simple task is very illuminating in terms of phonemic awareness capability because, to rhyme, you have to strip off that first sound and replace it with another sound. That’s a good window on phonemic awareness because the ear doesn’t hear the different sounds. By stripping off that first sound “c” and replacing it with “s” to make “sat,” you can see that the child’s brain hears the three sounds and can manipulate them, even though the ear doesn’t hear them.

So phoneme awareness is a prerequisite, and it allows one to develop phonics skills. One then applies those skills in ever-increasingly rapid and accurate ways to text until they are all applied unconsciously. As you develop your reading skills, oral language interaction and reading builds more and more vocabulary, which is crucial for the comprehension process. Most of the vocabulary that we develop is from reading after about the fourth grade, learning words that aren’t much used in oral language.

Clowes: Is whole language where you recognize the word but try to do it without the phonics or the phonemic awareness?

Lyon: Apparently. In its purest form, whole language is a philosophy that argues against deconstructing anything into its parts. It argues for a holistic look at all academic content, not just reading. A whole language philosophy would suggest that teaching children phonemes and how those sounds link to letters is fragmenting the language. Their proposal would be that reading is, in fact, a natural process, and that children learn to read by reading. That’s simply not the case.

Reading is anything but natural. It’s a fairly new skill in evolution, relative to oral language, and it is contrived. Why 26 letters? Why 44 sounds? Other alphabets have more or less sounds or more or less letters. Much as we may not like it, reading is initially a technical skill, and it requires a learning of a number of fundamentals to do it well.

Clowes: Why isn’t reading being taught well in so many of our schools?

Lyon: A great deal of it is related to the tendency in education to develop teaching practices based on philosophy rather than on science. When you look at reading practice over the last 15 to 20 years, that practice has been based more on constructivist or whole language philosophy, which eschews the more direct instruction on the technical side. As a result, the majority of the teachers who have been trained in the last generation have not been provided with the information to ask those fundamental questions like “What does it take to learn to read?”

What you have are teachers who want to do well by children but who come out of their training typically either philosophically driven or method-driven. So when they look at a child, they tend to apply a one-size-fits-all approach, or they apply a philosophy that they learned without having the benefit of being able to analyze a difficulty or a problem the child might be showing them.

Clowes: So if a child is struggling to read, the teacher needs to know how to help?

Lyon: Exactly. If they see a very slow, labored reader who’s sounding out the words in a laborious way and being inaccurate in the process, that tells the teacher that the child probably doesn’t really understand the sound system and how it relates to print. Another child might also be a slow reader but sounds out the words accurately. In that case, the child probably just needs to practice reading more to gain speed. But early reading problems need to be corrected early because, as poor readers grow up, their vocabularies decline; as good readers grow up, their vocabularies improve.

Clowes: What can parents do to make sure their children learn to read properly?

Lyon: Parents should understand how much they need to interact with children before they get to school, from birth onwards. A lot of good teaching goes on in homes where children are read to, where there’s a lot of language being used, where there’s a lot of language play, doing nursery rhymes, Dr. Suess, playing pig Latin, and so forth. All of that really helps children understand that our language is very dynamic and flexible, which helps them when it comes to reading.

As the youngster gets to the end of kindergarten and through first grade, parents need to make sure that their children can handle rhyming and that they can play with the language. That’s foundational to applying phonics to the written script.

As the children are getting into late first grade and second grade, Mom and Dad need to be looking at the fluency and ease with which youngsters read. Because, again, if we just go back to the basics, the best indicator of how well someone will enjoy what they’re reading, and how well they’ll understand it for meaning, is the ease and the accuracy by which they pull that print off the page.

What parents can do if their child is struggling with the print is to read the story to them and determine whether in fact they can comprehend it that way. If the information is getting in with facility and ease, what can the child do with it? Do they have the vocabulary? Do they get the main idea? If you read the story to them and they comprehend it, that indicates that it’s a data-in problem.

If parents see their child struggling, they have to seek out an informed professional who knows what it takes to learn to read. The majority of the children we look at with reading problems are bottlenecked at this letter-sound stage, not at the comprehension stage. They are slow and labored in getting the print off the page.