However, even advocates of look-say highlight its overriding, intractable problem. As one advocate put it, children

should receive praise for a good guess even though it is not completely accurate. For example, if a child reads “I like to eat carrots” as “I like to eat cake,” praise should be given for supplying a word that makes sense and follows at least some of the phonic cues.

Of course, countless sentences have multiple words that might satisfy the context cues. A reader unable to sound them out can only guess. Indeed, proponents of this method acknowledge the undeniable role of guessing in the process. As Martin Gross details in The Conspiracy of Ignorance: The Failure of American Public Schools, proponents of the look-say method claim that “reading is ‘a psycholinguistic guessing game.’ Students are encouraged to ‘create’ and are not marked wrong for guessing wrong.” Philosopher Leonard Peikoff writes archly of this method:

How would you like to see, at the head of our army, a general with this kind of schooling? He receives a telegram [today an instant message, if not a tweet] from the president during a crisis ordering him to “reject nuclear option,” proceeds to make a good guess, and reads it as “release nuclear option.” Linguistically, the two are as close as “carrots” and “cake.”

Advocates of the whole word method hold that phonics overloads a child’s mind with too many letters and sounds that must be memorized. So, in place of this, they endorse a system that requires readers to memorize the shape of every word in the language. Memorize not twenty-six letters and forty-four sounds that enable readers to decode untold thousands of words in the English language—but memorize the shape of every word in the entire language.