Whole Language vs. Code Emphasis: Underlying Assumptions and Their Implications for Reading Instruction*

Promoters of Whole Language hew to the belief that learning to read and write can be as natural and effortless as learning to perceive and produce speech. From this it follows that there is no special key to reading and writing, no explicit principle to be taught that, once learned, makes the written language transparent to a child who can speak. Lacking such a principle, Whole Language falls back on a method that encourages children to get from print just enough information to provide a basis for guessing at the gist. A very different method, called Code Emphasis, presupposes that learning the spoken language is, indeed, perfectly natural and seemingly effortless, but only because speech is managed, as reading and writing are not, by a biological specialization, that automatically spells or parses all the words the child commands. Hence, a child normally learns to use words without ever becoming explicitly aware that each one is formed by the consonants and vowels that an alphabet represents. Yet it is exactly this awareness that must be taught if the child is to grasp the alphabetic principle and so understand how the artifacts of an alphabet transcribe the natural units oflanguage. There is evidence that preliterate children do not, in fact, have much of this awareness; that the amount they do have predicts their reading achievement; that the awareness can be taught; and that the relative difficulty oflearning it that some children have may be a reflection of weakness in the phonological component of their natural capacity for language. Various studies have estimated the number of children who fail at reading to be 20-25 percent of the school population (Stedman & Kaestle, 1987). While it is generally agreed that this presents a serious problem, opinion is deeply divided about its underlying causes and inevitably, therefore, about the proper route to its solution. In this pa-per, we will explore two current views. One of these is commonly referred to by its partisans as Whole Language; the other, which we embrace, we call Code Emphasis, borrowing the name given it by Jeanne Chall (1967). At the level of their most fundamental assump