LINGUIST List 25.4161

Tue Oct 21 2014

Review: Lexicography; Linguistic Theories; Text/Corpus Ling: Hanks (2013)

Editor for this issue: Sara Couture <>

Date: 07-Apr-2014
From: Gill Philip <>
Subject: Lexical Analysis
E-mail this message to a friend

Discuss this message

Book announced at

AUTHOR: Patrick Hanks
TITLE: Lexical Analysis
SUBTITLE: Norms and Exploitations
YEAR: 2013

REVIEWER: Gill Philip, Università degli studi di Macerata

Review's Editor: Helen Aristar-Dry


Patrick Hanks' ‘Lexical Analysis: Norms and Exploitations’ is the fruit of a career's research into and analysis of the English lexicon. The scope of this work is impressive, and on the whole the work lives up to expectations. Running at 430 pages plus notes and references, this is a book that demands close and careful reading. Six of the thirteen chapters are re-worked versions of previously-published papers, while the rest expands on the Theory of Norms and Exploitations (henceforth, TNE) and contextualises it within the discipline of linguistics as a whole. The writing is engaging and erudite, featuring many insightful comments and observations and a plethora of worked examples, all based on corpus evidence.

In Chapter 1, 'Words and Meanings: The Need for a New Approach', Hanks introduces the rationale for the book and for the theory of meaning which is to be expounded over the following chapters. A comment made early on sets the tone for what is to come: ''In linguistic analysis, many things seem obvious, and only some of them are true'' (p. 4). A case study of the verb 'hazard' serves as an initial illustration of the sorts of information that TNE is able to shed light on, but which most current approaches to linguistic analysis fail to address adequately: the unexpectedly complex relationship that holds between lexical sets and semantic types.

In Chapter 2 the author deals with the possible answers to an apparently straightforward question: 'What is a Word?' A number of competing definitions are discussed: 'type', 'token, 'lemma' (or 'lexeme'), 'multiword expression' (or 'praseme'), and 'lexical entry' (or 'headword') in a dictionary, plus the variable element of proper nouns (which may or may not be included within each of the other categories). Hanks explains the main ways in which new words are formed from the existing resources of the language, suggesting that rather than finite, the lexicon is instead a ''small infinite set'' (p. 30). The section dealing with multiword expressions (MWEs), the most productive area of the lexicon, raises thought-provoking issues about the nature of meaning in compounds, in particular regarding the mismatch between compositional meaning and meaning in collocations.

Chapter 3, based on Hanks (2000), asks 'Do Word Meanings Exist?' The issue raised here is not, as the title might suggest, to ask whether or not we can use words meaningfully, but rather to draw a distinction between meaning in actual communicative situations ('meaning events') and meaning in the memory ('meaning potential'). Understanding this distinction is crucial to the appreciation of what TNE has to offer linguistics. The argument is that meanings are tied to context, and that by examining words used in context (i.e. in corpus data) it is possible to identify how those meanings are carried by words. Words shorn of context cannot have 'a meaning' as such but have (usually more than one) 'meaning potential', potential that can only be realized when the words are used in combination, in context, to achieve some communicative purpose. Once again, a case study lends support to the argument; more detailed contextualization within the history of linguistics can be found in later chapters (particularly chapter 11).

Chapter 4, 'Prototypes and Norms', begins the section of the book dedicated to TNE, and, in the author's words, ''may be regarded as the seed paper for the whole enterprise'' (p. xiv). Drawing on Hanks (1994) this chapter expands on the argument that meanings are either potential or realized events, tying the discussion into conversational implicatures (Grice 1957, 1975) and relevance theory (Sperber &amp; Wilson 1986). Systemic grammar, the framework proposed for carrying out syntactic analysis with corpus data, is also outlined in this chapter, and we also discover exactly what the author means by a ''norm'', and how to identify one in linguistic data.

Chapter 5, based on Hanks (1996), is dedicated to the topic of 'Contextual Dependency and Lexical Sets'. Through a series of worked examples the author illustrates and discusses how recurrent uses of words in particular collocations and phrasal structures contribute to the construction of cognitive profiles; in other words, he explains how our memories of repeated meanings-as-events are transformed into beliefs about meanings.

Extending the discussion of norms further, Hanks addresses diachronic change in Chapter 6, 'Norms Change Over Time', where he focuses on the words 'enthusiasm' and 'condescension'. These words seem to be used with the 'wrong' polarity in a 1762 monumental inscription: 'enthusiasm' is something to be 'exposed', while 'condescension' is painted as a virtue. The author compares contemporary data from the British National Corpus with historical data from the Oxford Historical Corpus, to uncover late 18th C norms (both lexical usages and, by extrapolation, beliefs about meaning) for the two words. The diachronic meaning change presented here illustrates the point that different word senses co-exist, usually in different spheres of experience, and that as society's norms change, so too do the norms that govern word meaning and use.

The stage between norms and exploitations is alternation, and 'Three Types of Alternation'- lexical, semantic-type, and syntactic - are discussed in Chapter 7. Lexical alternations involve relations of synonymy and co-hyponymy, including metonymy; the alternative word choice perhaps adding emphasis or specificity, but not changing the message conveyed. Semantic-type alternation operates at a more abstract level, involving changes in the semantic types which normally populate a verb's clause roles. Again, for it to be alternation rather than exploitation, the meaning conveyed should shift the focus but not the overall sense, and here metonymic relations appear to be key, i.e. the semantic type 'human' can alternate with 'human institution' (e.g. bank, government) but not 'vehicle'. The long section dedicated to syntactic alternation lists the main grammatical transformations which can occur to change the focus of a clause without causing a change in meaning.

Chapter 8 is dedicated to 'Exploitations'. While acknowledging that the boundary between alternation and exploitation is fuzzy, the essential difference is that an exploitation changes the meaning conveyed. As well as outlining different types of exploitation, this chapter provides a useful overview of rhetorical tropes commonly involved in lexical exploitations as well as an outline of Pustejovsky's (1995) notion of 'semantic coercion', whereby context forces new meanings onto words so that they are made to 'fit'.

Chapter 9 deals with 'Intertextuality', not only in the more commonly-received sense of deliberate citation, but also in the everyday recycling of words and phrases that every language user engages in. Hanks takes the reader through all forms of literature high and low to illustrate how writers exploit the existing resources of the language, modifying patterns and creating memorable forms which in turn are reused by other speakers and writers.

Chapter 10, 'Word and Pattern Meaning: A Complex Linguistic Gestalt', looks at the conceptualization of words and meanings in terms of gestalts. The gestalt for a word is a mental agglomeration of the patterns and meanings which it participates in, and which contributes to the individual's beliefs about its meaning and awareness of how to use it. Hanks discusses words of minimal, medium and maximal complexity, outlining for each how a gestalt can be built up from typical lexical patternings, and then goes on to discuss how secondary norms can emerge out of exploitations of (primary) norms.

Chapter 11, 'Meaning, Philosophy of Language, and Anthropology', marks a turn in the book's focus. The author draws together the theoretical points which have emerged during the previous ten chapters analyzing language data. This comprehensive overview of how meaning has been dealt with by various academic traditions over the years allows the reader to appreciate where TNE sits with regard to the study of words and meaning. In particular, the influence of Wittgenstein (1953) is evident in using word use to determine concepts, that of Grice (1957, 1975) and Austin (1962) on 'doing things with words'; whereas Rosch's (1973, 1975) work on prototype theory lies at the basis of the view of language as being constructed from norms (lexical prototypes) and exploitations of these.

Chapter 12 addresses the context for TNE from another angle, this time tracing 'The Role of the Lexicon in Linguistic Theory'. This long chapter (60 pages) surveys the treatment of lexis in the main strands of linguistic research from European structuralism to systemic linguistics, including work which is relatively unknown in the English-speaking world, notably the German Bedeutungsfeld (semantic field theory) and Russian lexicology. Central to the arguments raised in this chapter is the use of invented linguistic examples vs. the use of attested instances of language in use. These contrasting practices reflect two opposing views of what linguistics is: the study of what is possible in language, as opposed to the study of what is normal (or probable). For each of the theories discussed, the author makes explicit those aspects which have contributed to the development of TNE, stressing where TNE can continue, improve or refine what has gone before.

As the title to Chapter 13, The Broader Picture, suggests, this final chapter serves not only to recapitulate on the main points discussed in the book and to reiterate the central aspects of TNE, but to suggest its potential applications in various linguistics-related fields, primarily lexicography, natural language processing, machine translation, and language learning and teaching. The book ends with a short description of the author's work in progress, the Pattern Dictionary of English Verbs.


The volume is well-conceived structurally. Chapters 1-3 provide the background to TNE, outlining the issues at stake, presenting the theory and contextualising it within the study of lexis in general. Chapters 4-9 deal with the main aspects of TNE, from defining norms to explaining how exploitation works; while the final portion of the book addresses the details of how the relationship between words and meanings has been (and is now) conceived in different fields of linguistics and in related disciplines. Chapter 6 ('Norms Change Over Time'), might have been better positioned after Chapter 10 ('Word and Pattern Meaning: A Complex Linguistic Gestalt'), since these two chapters seem to form a pair, discussing on the one hand how the semantic features associated with words come together to form gestalts (chapter 10), and how these are subject to change over time (chapter 6). This quibble aside, the sequencing is appropriate.

In a book of this length and complexity, ease of navigation is important, and sometimes the publishers' 'clean page policy' (for want of a better term) interferes: the contents pages detail only the main numbered sections of each chapter; there is no separate listing of figures and tables; and endnotes are used in preference to footnotes. While this makes for a neater page layout, it does entail some effort and will on the part of the reader trying to (re)locate information. Worthy of note, however, is the decision to 'bookend' each chapter with an abstract at the start and a bullet-point summary at the end. These facilitate the reading (and re-reading) of the text, keeping the reader's focus fixed on the main arguments that are built up as the volume progresses, and help the reader refresh his or her memory before embarking on subsequent chapters. The bibliography is clearly-set out (following the lexicographic convention of listing dictionaries, corpora and 'other literature' separately). The only potential addition to the volume, given its impressive scope, would be a glossary of technical terms used - these are all highlighted and defined within the text, but gathering them together in glossary at the end would make it easier to keep track of them all.

With TNE, Hanks addresses the search for regularity ''lurking beneath the surface of everyday usage'' (p 410). This regularity seems to be accessible via gestalts in the mind of the language user, available for use whenever the need to communicate arises, but locating prototypical examples of word use (norms) in corpus data can be nigh-impossible except by generalizing and abstracting out from the data. TNE offers the theoretical framework for these prototypical forms to be identified, expressed in conceptual terms, and illustrated with genuine examples. In one sense this is nothing new: norms are already available in the minds of sensitive language analysts, indeed their existence in the mind is precisely what makes it possible for linguists and language teachers to invent example sentences ad hoc to suit their exemplifying purpose. In another sense, however, TNE is quietly innovative. It carves a consistent, empirically-sound distinction between lexical use and concepts, which has implications for all areas of linguistics which deal with the representation of meaning in the mind and its representation in language.

The analysis of genuine examples of exploitations can tell us a great deal about how language users make creative use of the resources available to them, but one of the thorny issues raised (but not resolved) is this: when does an alternation become an exploitation, or, conversely, when is an exploitation not an exploitation but 'simply' an unusual alternation? At this point, the individual and social aspects of language clash. Linguistics always treads a precarious path between the language use of the individual (usually the linguist him- or herself) and that of the collectivity of language users. While corpus evidence can be used to posit norms for the community of language users, there is always an underlying tension between what is normal for all and what is normal for the few. These norms can conflict, even to the extent that what for one speaker or even speech community is familiar and repeatedly encountered, may be unfamiliar to another. For the former, it is a norm; for the latter, an alternation or exploitation, depending on how far it deviates from other familiar patternings.

Even setting matters of the individual vs. the collectivity to one side, deciding whether a variant is an alternant or an exploitation remains something of a grey area. Hanks repeatedly mentions that the border between these two classes is fuzzy, and that more analysis is required, and on a very large scale, to resolve it. Having said that, he is very clear on how, in theory, to distinguish alternants from exploitations. The difficulties that are likely to be encountered in analysis are therefore not due to inadequate classification, but from problems relating to its application. The degree of subtlety and sensitivity required is considerable, much hanging on the analyst's judgment and on his or her ability to distinguish between metonymy and metaphor, which is itself a specialized area of study.

One of the recurring themes in this book is the need to differentiate between word meaning and concept. The two are very often treated as one and the same thing, but Hanks argues (along a similar train of thought to that of the 'common-sense' philosophers, but also following Saussure) that words are not concepts, and concepts are not words: words are used by people to encapsulate concepts. In some areas of linguistics the distinction is necessary. Lexicography is one, foreign language learning and translation are others, because in these fields the connection between words and meanings in the mind of the language user is, frankly, irrelevant if the lexical patternings that are chosen fail to activate the intended meaning in text. TNE therefore has a lot to offer all branches of applied linguistics; it is also relevant to other areas of linguistics, provides an analytical framework that makes it possible to isolate meanings in text from meanings in the mind.

This book is recommended reading for those who are serious about studying lexis and/ or the interplay between structure and meaning, especially students and scholars working in lexicography, text linguistics, cognitive linguistics and natural language processing. It provides a straightforward methodological framework for analysis, explains the necessary technical terms and provides a plethora of illustrative examples. It not only presents the Theory of Norms and Exploitations, but encapsulates in its discussion all the major trends in twentieth-century linguistics and philosophy of language. In this sense, it is also reference work, to be consulted time and time again.


Austin, John.L. 1962. How to do things with words. Oxford: Clarendon Press.

Grice, Henry Paul. 1957. Meaning. Philosophical Review 64, 377-388.

Grice, Henry Paul. 1975. Logic and Conversation. In P. Cole &amp; J.L. Morgan (eds), Syntax and Semantics, vol. 3, Speech Acts. New York: Academic Press.

Hanks, Patrick. 1994. Linguistic norms and pragmatic exploitations, or why lexicographers need prototype theory and vice versa. In F. Kiefer et al. (eds), Papers in Computational Lexicography: Complex '94. Research Institute for Linguistics, Hungarian Academy of Sciences.

Hanks, Patrick. 1996. Contextual dependency and lexical sets. International Journal of Corpus Linguistics 1 (1), 75-98.

Hanks, Patrick. 1998 'Enthusiasm' and 'Condescension': changing norms of meaning and use. In T. Fontenelle et al. (eds) Euralex '98 proceedings, 151-166. Liège: University of Liège.

Hanks, Patrick. 2000. Do word meanings exist? Computers and the Humanities 34, 205-215.
Pustejovsky, James. 1995. The Generative Lexicon. Cambridge (Mass.): MIT Press.

Rosch, Eleanor. 1973. Natural categories. Cognitive Psychology 4 (3), 328-350.

Rosch, Eleanor. 1975. Cognitive representations of semantic categories. Journal of Experimental Psychology, 104 (3), 192-233.

Sperber, Dan &amp; Deirdre Wilson. 1986. Relevance. Oxford: Blackwell.

Wittgenstein, Ludwig. 1953. Philosophical Investigations. Oxford: Blackwell.


Gill Philip is a lecturer in English Language and Translation and TEFL at the University of Macerata, Italy. Her research interests cover all aspects of figurative language and phraseology, with particular emphasis on the interplay between conceptual knowledge and its lexical realization.

Page Updated: 21-Oct-2014