PubMed2001
Learning to read is much more difficult than learning to speak. Most children teach themselves to speak with little or no difficulty. Yet a few years later when they come to learn to read they have to be taught how to do it; they do not pick up reading by themselves. This is because we speak in words and syllables, but we write in phonemes. Syllables do not naturally break down into the sounds of letters and letter units (i.e., phonemes) because these do not correspond to physiologically distinct articulatory gestures (Liberman, Shankweiler, & Studdert-Kennedy, 1967). Alphabetic writing was only invented when people realized that syllables could be artificially divided into smaller acoustically distinguishable phonemes that could be represented by a small number of letters. But these distinctions are arbitrary cultural artifacts, and their mastery was originally confined to a select social class. And until about 100 years ago it did not matter much if the majority of people could not read; the acquisition of reading probably had no serious disadvantages. Reading requires the integration of at least two kinds of analysis (Castles & Coltheart, 1993; Ellis, 1984; Manis, Seidenberg, Doi, McBride-Chang, & Petersen, 1996; Morton, 1969; Seidenburg, 1993). First, the visual form of words, the shape of letters, their order in words, and common spelling patterns, which is termed their orthography, has to be processed visually. Their orthography yields the meaning of familiar words very rapidly without needing to sound them out. But for unfamiliar words, and all words are fairly unfamiliar to the beginning reader, the letters have to be translated into the speech sounds (i.e., phonemes) that they stand for, and then those sounds have to be melded together in inner speech to yield the word and its meaning. Reading exclusively by the phonological route is more time consuming than if words can be accessed directly without requiring phonological mediation.