Introduction It is now widely recognised that metonymy plays a crucial role in language, and may even be more fundamental to human speech and cognition than metaphor. One of the benefits of the use of metonymy is economy of language. Metonymy is such a ubiquitous feature of our everyday linguistic behaviour that people do not experience any problems making inferences about metonymic expressions, the production and comprehension of metonymies being largely automatic. That is because people not only speak, but also think in metonymies, and patterns of metonymy in language reflect patterns of metonymic thought (Gibbs 1999: 74). Poets and writers create spectacular examples of metonymy, but in everyday communication people follow conventional metonymic pathways, such as CONTAINER-CONTENTS, CAUSE-EFFECT or CONTROLLER-CONTROLLED, which can be easily understood on the basis of our knowledge of the world. The present book does not include a discussion of the approaches to metonymy before the advent of cognitive linguistics (henceforth CL) in the early 1980s. As is widely known, the earliest extant definition of metonymy comes from the Rhetorica ad Herennium, written before 82 AD (Anonymous 1894: 337, as quoted in Koch 1999: 140-141). Still earlier, Aristotle had recognised the existence of metonymy as a rhetorical trope, subsuming it under metaphor in his Poetics and Rhetoric. However, the pre-cognitivist literature on metonymy is scarce in comparison with the number of publications on metonymy which have appeared since the beginning of the 1980s. Moreover, before cognitivism, metonymy tended to be overshadowed by metaphor, which was considered a more central rhetorical and linguistic device. The cognitive framework is adopted in this work as it is the predominant approach to metonymy in the current semantic theory. It is also the most fruitful for the present research project. The advantages of the cognitive perspective in the dictionary-making process are recognised by increasing numbers of (meta) lexicographers (e.g. Adamska-Sałaciak 2006, 2008; Atkins Rundell 2008; Moon 2004, 2005; van der Meer in numerous publications). As Philip (2006: 896) notices, one of the more recent innovations in advanced learners dictionaries has been the move to mark metaphorical and figurative language more explicitly. This trend is particularly noticeable in the approach to metaphor in the MEDAL dictionary. As the only monolingual learners dictionary (henceforth MLD), MEDAL contains metaphor boxes which explain the conceptual processes underlying metaphorical expressions. It should be stressed that the latest edition, MEDAL2, has added twenty metaphor boxes to the existing forty-plus in response to the demand from its users (MEDAL2: viii). The movement towards explicit marking of figurative language in MLDs is supported by studies which suggest
16 Introduction that new lexical items are learned and recalled more effectively when their metaphorical underpinnings are explicated (e.g. Boers 2000). There is a growing body of evidence for the contributions that CL can make to second or foreign language teaching, especially with regard to polysemy and idiomatic expressions (for a survey of these contributions, see Boers Lindstromberg 2006). The main benefit of CL-inspired pedagogy is the quest for linguistic motivation, e.g. in order to explain meaning-meaning relations, such as metaphor and metonymy. The pivotal role of motivation in the learning process is emphasised by Lakoff (1987: 346): It is easier to learn something that is motivated than something that is arbitrary. It is also easier to remember and use motivated knowledge than arbitrary knowledge. Not only is the need for the interaction between lexicography and CL stressed by (meta)lexicographers, but it is also recognised by some cognitive linguists, though still to a lesser extent. The main advocate of the application of the cognitive theories in lexicography among cognitive linguists is Geeraerts (1990 [2006], 2007). He claims that CL provides a theoretical framework that is highly congenial to the actual practice of dictionaries, more so than any structuralist theory. Geeraerts (2007) also notes that some aspects of the cognitive conception of lexical semantic structure, such as prototypicality, the intractability of polysemy, and the structured nature of polysemy are vindicated by lexicographic practice. He postulates, moreover, that there should be a relationship of mutual inspiration between CL and practical lexicography, as both disciplines can learn something from each other. The contribution of lexicography to CL is visible mainly in the more widespread use of corpus materials, and in the research into historical lexicology taking inspiration from the great historical dictionaries such as The Oxford English Dictionary (henceforth OED). The present work aspires to add support to the assumption made by Atkins (1992/1993 [2008]: 48) that theorists and practitioners need to work together if dictionaries are to be improved. It attempts to be a contribution to the enquiries into the interplay between CL and lexicography, as conducted by cognitivists and (meta)lexicographers alike. This study aims to show how the tenets of the cognitive theory of metonymy can benefit the representation of metonymic lexemes in pedagogical lexicography, so that the semantic connections between basic and derived meanings become more transparent and motivated. By following this aim, the book sets out to take a step towards filling a niche in the metalexicographic research into figurative language. The treatment of metonymy in dictionaries has so far received little attention from (meta)lexicographers, who preferred to study the lexicographic representation of metaphor (e.g. van der Meer in his numerous publications; DeCesaris Alsina 2002; Moon 2004, 2005; Philip 2006). Given that metonymy is an integral part of everyday communication, its presence in pedagogical lexicography should not be disputed. The present study
Introduction 17 investigates the representation of metonymy in the Big Five MLDs (CALD2, COBUILD4, LDOCE4, MEDAL2 and OALDCE7), pointing out certain problem areas, and hopes to suggest possible improvements in line with the cognitive approach. Admittedly, the intricacies of the cognitive theory of metonymy prevent it from being fully implemented in lexicographic practice. Therefore, the most fundamental premises of the theory have been selected as tools of evaluation. It is assumed that the cognitive perspective should be favoured in MLDs as long as it serves the user and presents metonymic senses in a transparent way. Chapter 1 traces the development of the cognitive theory of metonymy since Lakoff Johnson (1980 [2003]). It provides a discussion and a critical analysis of the latest accounts of metonymy which have emerged in response to the growing dissatisfaction with the standard view of metonymy as an intra-domain mapping. A stance is taken on such crucial issues in the cognitive theory of metonymy as metonymy vs synecdoche, the concept of a domain, the process of mapping, the typology of metonymic relationships, and the metaphor-metonymy interplay. The chapter closes with a proposal of a model of metonymy which is meant to be eclectic with respect to the discussed theories. While Chapter 1 is entirely theoretical, the following three chapters report the findings of empirical studies into the representation of metonymy in the five MLDs. Each of the three chapters focuses on a different element of the dictionary entry: sense arrangement, definition, and the correlation between codification and exemplification. These features are evaluated against the background of both the cognitive theory of metonymy and the widely accepted principles of lexicographic practice. Chapter 2 delves into the disambiguation and arrangement of metonymic senses in the dictionary microstructure. It discloses a number of conventional metonymic senses which are not covered in some of the dictionaries under scrutiny. It also reveals discrepancies in the arrangement of metonymic meanings both between the dictionaries and within an individual dictionary. The issue of ordering of multiple metonymies is also brought up. Recommendations are made about the arrangement of metonymic senses in the microstructure of a MLD which best reflects the development of metonymic meanings from the basic senses of a word. Chapter 3 provides an analysis of the dictionary definitions of metonymic lexemes. It is assumed that the definition should present the metonymic sense as a semantic elaboration of the source sense, thus enabling the user to make associations between the basic and derived meanings. It is checked to what extent the implications of the connection between the source and target meanings of a metonym are built into the definitions themselves. The ways of referring to the source sense in the definiens of the metonymic target are identified and compared in terms of the strength of forging the semantic relationship with the definition of the source.
18 Introduction While Chapters 2 and 3 discuss the lexicographic representation of metonymy from the semantic perspective, Chapter 4 takes a grammatical viewpoint. It deals with metonymisation which results in a change of the noun s countability. It is maintained that being a frequent source of learners errors, the count-mass distinction should be encoded in the most explicit way. The chapter reports the findings of an enquiry into codes, examples and definitions, and evaluates their effectiveness as grammatical indicators. The conclusions sum up the problem areas detected in the empirical studies reported in the book. Suggestions are made about the possible improvements in MLDs, so that the representation of metonymy corresponds more closely to our conceptualisations of metonymic meaning extensions. Finally, the five dictionaries are graded according to the various aspects of their treatment of metonymy.