ABSTRACT
This chapter attempts to chart the linguistic and behavioural data that might be expected to constrain models of the translation of print into sound in the unimpaired reading of English words. It sets out a memorandum of the pertinent characteristics of English orthography. The chapter reviews the literature on the speeded naming of isolated English words. It introduces and evaluates the main theoretical options that have been considered in constructing models of orthographic translation for English. The limiting case of mechanical translation would be exhibited by a device that translates each letter into a speech sound without use of any contextual information. Polysyllabic words introduce further problems, notably the assignment of lexical stress. Decisions about what syllable to stress carry implications for vowel quality. Morphophonemic adjustment in vowel values and shifts in lexical stress are parallel but largely independent processes that depend on suffixation.
