The Lambda-structure of Texts. Ioan-Iovitz Popescu, Radek Cech, and Gabriel Altmann.

Abstract

The book under review, jointly written by three authors with a pronounced expertise in the field of quantitative linguistics and text analysis, is primarily devoted to the analysis of lexical frequency structures in texts. In a way, the book may be said to represent part of an ongoing research project, integrating earlier findings (Popescu et al., 2009), and being elaborated in ongoing research (Popescu et al., 2010). The authors consider it to be a major task of the book to develop, present, and introduce new methodological instruments to quantitative analyses of texts. First and foremost, they concentrate on the frequency occurrence of words, i.e. word forms and/or lexemes, or lemmata. On the one hand, this concerns aspects of lexical rank frequency distributions, on the other hand, issues of lexical richness, derived from such data for purposes of individual stylistics, authorship attribution, quantitative text typology, etc. As is referred to in detail in the introductory part (p. 1ff), a whole variety of indices and indicators have been developed in the past, to calculate lexical richness on the basis of type-token ratios, percentage of hapax legomena, or various specific ratios based on lexical inventory used in a given text. However, all of these indicators suffer from a major theoretical flaw, namely their dependence on sample size, i.e. on the length of the text(s) under study, which renders them rather useless, or eventually even misleading, for any kind of comparative approach.