Lexical Analysis: Norms and Exploitations by Patrick Hanks (review)

Let me state at the outset that this is an important book that should be read by everyone interested in the study of word meaning. Patrick Hanks is well known to people interested in lexicography and English, as he has worked on many large-scale dictionary projects for major British publishers (Collins Dictionaries and Oxford University Press). He is particularly known for his contributions to corpus lexicography and onomastics and has also established himself as an important figure in computational linguistics: Church and Hanks (1990) is required reading in the field. For several years now, Hanks has worked in academia and has been an invited researcher and lecturer both in Europe and the United States. (Full disclosure: In 2010, at my invitation Hanks spent a month at Pompeu Fabra University in Barcelona and since then has been in contact with me and several members of the research group I belong to. I trust that my familiarity with his work serves to improve, and not cloud, this review.)

Lexical Analysis consists of thirteen chapters, each of which starts off with a paragraph that specifically sets forth the objectives of the chapter and ends with a summary that often takes the form of a list with bullets (the concluding chapter, understandably, is an exception to this organization). This practice is an excellent way to keep the reader on track and should be adopted by others. The organization of the material in the book very much reflects the bottom-up approach to lexical analysis advocated. After an introductory chapter justifying the need for a new approach to the study of word meaning, the book builds up what Hanks terms the Theory of Norms and Exploitations from one chapter to the next by presenting analyses of corpus data, as opposed to the more general practice of presenting the theory first and then supporting it with a few examples. The first ten chapters of the book are devoted to exploring the nature of lexis and to presenting corpus-driven analyses of words to give shape to the Theory of Norms and Exploitations. Chapters 11 and 12 briefly compare Hanks’s proposals with those of others, and Chapter 13 concludes with a brief summary of the theory and with suggestions pointing at its possible application in language pedagogy, lexicography, and computational linguistics. The theory is exemplified not only with “regular” English prose as found in informative texts but also with phrases taken from literary texts. The breadth of examples considered is a strength of the book and of Hanks’s work in general.

Lexical Analysis sets forth a theory of lexical meaning that draws much from Hanks’s experience as a lexicographer. In his own words, “TNE is a ‘double-helix’ theory of language: the set of rules that govern normal, conventional use of words is intertwined with a second-order set of rules that govern the ways in which those norms can be exploited and that contribute very largely to the phenomenon of language change” (411). It is grounded in the view that words have “meaning potentials” as opposed to “meanings”; words only acquire their full meaning when used in context. This view necessarily entails exhaustive consideration of “repeated and reinforced authentic usage” (4) and thus is dependent on close examination of data from large corpora.

Hanks envisions norms and exploitations not as two closed categories that oppose each other but rather as endpoints on a cline (4, 147). Norms are the manifestation of conventional use, whereas exploitations are the manifestation of exceptional use, of pushing a word’s meaning to the limit. A norm is associated with a conventional meaning of the word and is identified by a pattern containing linguistically relevant semantic categories (as shown by analysis of corpus data) that form lexical sets (such as [[Human]], [[Object]], [[Event]] or [[Attitude]]). Some words have many conventional patterns (they are polysemous), while others have only one (often the case for nouns used in scientific terminology...