By the time most children are five, they generally can speak thousands of words in their mother tongue. At this point (and probably earlier), it is possible to teach them the alphabet, the literary symbols that make up the verbal sounds. In a matter of months, children between the ages of five and six can master the written alphabet and begin to sound out words.

For centuries, billions of children in dozens of nations around the world have mastered the all-important art of reading by this method. Even in English, a so-called irregular language, only approximately 13 percent of words are pronounced differently than they are spelled. This means that children trained in phonics can sound out about 87 percent of words in the language. Then, as proficient readers, they learn by experience to match irregular spellings—for example, “rough”—with the spoken word that they well know.

But throughout the 20th century and continuing now well into the 21st, the majority of American educators have rejected phonics, opting instead for some variant of the “whole word” method.

This method teaches children to look at the whole word, to recognize its shape, to examine the context in which it is deployed, and, if necessary, to guess. The first iteration of this method, known as “look-say,” held that children learn to recognize a written word by seeing it repeatedly. The theory was: Students need only master a core group of commonly used words and then employ context cues to decipher the rest. The children cannot be overloaded with thousands of new word shapes at a time. Therefore, the look-say method, starting in the first grade, entails introducing children to only several hundred new words per school year, which are then relentlessly repeated. Here, for instance, is a sentence from the kind of books commonly used: “‘We will look,’ said Susan. ‘Yes, yes,’ said all the children. ‘We will look and find it.’ So all the boys and girls looked. They looked and looked for it. But they did not find it.”