Abstract

The study is an empirical, text-based analysis of Latinate elements of the English lexicon, taken as representing a more formal, literate level of language use than words of its native Germanic stock, and hence as diagnostic of linguistic register across the variables of age and text type (narrative versus expository genres, spoken versus written discourse). An innovative procedure was devised for automatic coding by tagging all the open class vocabulary items for Latinate/Germanic origin in 192 texts produced by 16 native English speakers in four different age groups (9-year-old grade-school pupils, 12-year-old junior-high students, 16-year-old high-school students, and graduate-student adults. Results indicate that the Latinate-Germanic divide is a valid diagnostic of register level in English. The study reveals a statistically significant increase in the proportion of Latinate terms across all the variables considered. This ratio is across the board higher in expository than in narrative texts and written compared with spoken texts, and it rises as a function of age, with a clear cut-off point between junior high and high school. These findings, which confirm an independently established hierarchy of text types ranging from oral narratives at one extreme to written expository discussion at the other, are discussed in relation to other measures of lexical usage (e.g., word length, lexical density, and diversity), in crosslinguistic comparison with Hebrew as a typologically different language, and in terms of their implications for the study of register in later language development in general.