A whole new language

The evidence is clear,… that direct systematic instruction in phonics during the early years of schooling is an essential foundation for teaching children to read. (Executive Summary, p1.)

Turning to the question of whole language, the committee went on to say:

The Inquiry found strong evidence that a whole-language approach to the teaching of reading on its own is not in the best interests of children, particularly those experiencing reading difficulties. Moreover, where there is unsystematic or no phonics instruction, children’s literacy progress is significantly impeded, inhibiting their initial and subsequent growth in reading accuracy, fluency, writing, spelling and comprehension. (p. 2)

Rather than further describe the large body of empirical evidence referenced in the report to support these conclusions, we will look instead at learning to read from an information processing perspective.

Most researchers working from this perspective have concluded, like the Rowe inquiry, that the whole-language approach is more difficult for the child.

Why? Let’s begin at the beginning.

Most children come to school at about the age of five or six already fluent in spoken English. They have already learned the associations between the sounds and the meanings of anything up to about 5,000 root words. That is about 5,000 “pieces” of information, where a piece of information is just the association between one sign (the sound of the spoken word) and another sign (the meaning of the root word).

This is an oversimplification of course. In reality there are many complexities. A single sound can have multiple meanings, and a single meaning can have multiple sounds, and as linguists and philosophers will tell you, meanings themselves are anything but simple. Nevertheless, this number of word-meaning associations gives a rough benchmark for the complexity of learning a new language.

Now consider a six-year-old learning to read English using a whole-language approach.

The emphasis is on reading whole words in a meaningful context, so the child will not be drilled in the ways particular letter groups are typically pronounced. In some whole-language programs, there will be an “embedded” phonics component. This is an indirect approach to phonics which will not usually occupy too much time, especially at the front end of the program. However, the front end is precisely where we should expect phonics to be most effective, as we will see.

As you would expect, students vary in the learning strategies they adopt. Some quickly pick up the rules relating letters and sounds by induction from examples. They start to do what whole-language says they should: construct their own model of the relationships between letter groups, sounds, and meanings.

But not everyone does this easily. Struggling students may construct only a fragmentary phonics model or maybe even none at all. In these cases it could mean that in effect the student has to learn a whole new language - about 5,000 new pieces of information, as he is not making proper use of the information that written words in English have an internal structure which can be correlated with their phonetic structure.